id
stringlengths 3
9
| source
stringclasses 1
value | version
stringclasses 1
value | text
stringlengths 1.54k
298k
| added
stringdate 1993-11-25 05:05:38
2024-09-20 15:30:25
| created
stringdate 1-01-01 00:00:00
2024-07-31 00:00:00
| metadata
dict |
|---|---|---|---|---|---|---|
27978680
|
pes2o/s2orc
|
v3-fos-license
|
Toll-like receptor 4 (TLR4) deficient mice are protected from adipose tissue inflammation in aging
Adipose tissue (AT) inflammation is a central mechanism for metabolic dysfunction in both diet-induced obesity and age-associated obesity. Studies in diet-induced obesity have characterized the role of Fetuin A (Fet A) in Free Fatty Acids (FFA)-mediated TLR4 activation and adipose tissue inflammation. However, the role of Fet A & TLR4 in aging-related adipose tissue inflammation is unknown. In the current study, analysis of epidymymal fat pads of C57/Bl6 male mice, we found that, in contrast to data from diet-induced obesity models, adipose tissue from aged mice have normal Fet A and TLR4 expression. Interestingly, aged TLR4-deficient mice have diminished adipose tissue inflammation compared to normal controls. We further demonstrated that reduced AT inflammation in old TLR4-deficient mice is linked to impaired ER stress, augmented autophagy activity, and diminished senescence phenomenon. Importantly, old TLR4-deficient mice have improved glucose tolerance compared to age-matched wild type mice, suggesting that the observed reduced AT inflammation in aged TLR4-deficient mice has important physiological consequences. Taken together, our present study establishes novel aspect of aging-associated AT inflammation that is distinct from diet-induced AT inflammation. Our results also provide strong evidence that TLR4 plays a significant role in promoting aging adipose tissue inflammation.
INTRODUCTION
Adipose tissue inflammation has become widely accepted as a major contributor to metabolic dysfunction and disorders [1,2]. Previous studies on diet induced obesity (DIO) mice have shown that adipose tissue is primed for inflammatory changes prior to other metabolic organs [3]. There is a plethora of research investigating factors in obese adipose tissue inflammation to identify valuable therapeutic targets for metabolic dysfunction. However, much less is understood about aging adipose tissue inflammation and dysfunction. A better understanding of the cellular and molecular mechanisms of adipose tissue inflammation in aging will be crucial in the development of therapeutics for metabolic diseases beyond cases of diet-induced adipose tissue inflammation and insulin resistance.
Both age-related adiposity and diet-induced obesity are characterized by immune cell infiltration and a sustained inflammatory cycle. Among these various immune cells, adipose tissue macrophage (ATM) accumulation, proliferation, and polarization are major contributors to adipose tissue inflammation and dysfunction [4,5]. Interestingly, recent studies suggest AGING that changes in preadipocyte function during aging also lead to dysfunctional adipose tissue, eventually progressing to chronic inflammation [6]. These changes include reduced preadipocyte replication, decreased adipogenesis, increased lipid toxicity, increased proinflammatory cytokines, chemokines, extracellular matrix (ECM)-modifying proteases, and stress response elements. Our group have recently shown that elevated endoplasmic reticulum (ER) stress response in aging also contributes to greater inflammatory responses [7], in part due to compromised autophagy activity in the aging adipose tissue [8]. Recent studies have also indicated that with aging there is increased accumulation of senescent cells in many organs including fat depots, which contributes to aging pathologies [9]. Moreover, removal of senescent cells by senolytic drugs rescue adipose tissue from the aging phenotype [10,11]. However, the detailed molecular mechanisms that lead to increased inflammation in aging adipose tissue are poorly defined.
During the last decade, major advances were made in identifying the molecular mechanisms by which lipidderived products promote inflammation in different cell types [12]. These lipid derived products, which include ceramides and Di-Acyle glycerol (DAG), induce insulin resistance via de-phosphorylation of protein kinase B (PKB) and phosphorylation of a serine residue of insulin receptor substrate 1 (IRS-1) [13,14]. Another lipid-derived product, non-esterified fatty acids (NEFA), elevates tissue inflammation through interaction with the pattern recognition receptor Toll-like receptor 4 (TLR4) via its endogenous ligand Fetuin-A (Fet A), a liver derived glycoprotein [15]. Fet A is considered a biomarker of chronic inflammation due to its ability to stimulate the production of inflammatory mediators from both adipocytes and macrophages [16]. Studies examining the role of Fet A in obesity have shown that free fatty acid (FFA)-induced insulin resistance is dependent on the presence of both Fet A and TLR4, where Fet A serves as an adaptor for FFA to stimulate TLR4 signaling that results in the release of pro-inflammatory cytokines through the TLR4-NF kB pathway [15]. These findings were further supported by studies that showed TLR4 -/mice fed with high fat diet (HFD) have alternative macrophage polarization, reduced AT inflammation, and decreased hepatic steatosis [17][18][19]. Interestingly, Fet-A null mice were also protected against obesity and insulin resistance with aging [20].
The involvement of Fet A-mediated activation of TLR4 pathway in adipose tissue inflammation in diet-induced obesity is well explored. However, the role of this pathway in age-associated adipose tissue inflammation is unknown. We undertook this study to test the hypothesis that age-related adipose tissue inflammation is dependent on the Fet A-mediated TLR4 signaling pathway. We first evaluated the expression of Tlr4 and Fet A gene products in adipose tissue, liver, and plasma samples derived from young and old mice. We then exploited the TLR4-deficient mice to investigate the role of TLR4 in age-associated adipose tissue inflammation, ER stress response, autophagy activity, cellular senescence, and metabolic status (glucose tolerance).
Adipose tissue expression of TLR4 and Fet A are elevated in DIO but not in aging
Consistent with previous reports [15,21], our present analysis indicates that DIO increases the expression of both TLR4 and Fet A, and contributes to adipose tissue inflammation as indicated by higher mRNA expression of Mcp1 (Fig.1). We then analyzed and compared the expression of Tlr4 in epididymal adipose tissues from young and old mice and observed no significant differences at either mRNA or protein levels ( Fig. 2A and 2B). However, there is enhanced activation of NF-κB (elevated serine-311 phosphorylation) in adipose tissue lysates of old mice compared to young (Fig. 2C). Since Fet A has been reported to be an adaptor of FFA and a ligand of TLR4 in DIO, we analyzed the serum Fet A levels in both young and old mice. Surprisingly, serum levels of Fet A in old mice were significantly lower than those in young mice (Fig. 2D). Since liver is the main source of Fet A, we examined Fet A gene expression in liver of young and old mice. Our analyses indicated diminished Fet A mRNA levels compared to the young cohort (Fig. 2E). We then performed mRNA analysis on adipose tissue samples from young and old mice and the results indicated that Fet A mRNA level was also diminished in old mice (Fig. 2F). Finally, we did not observe significant difference in Fet A protein expression between young and old adipose tissue lysates (Fig.2G). We also analyzed mRNA expression of fibronectin and Tenacin C1, which are reported to serve as endogenous ligands for TLR4 to promote inflammation in HFD-induced obesity [22] and in arthritis model [23]. We observed diminished expression of both fibronectin and Tenacin C1 in the aging adipose tissue (Fig. S1). Our results therefore indicate that, contrary to DIO, adipose tissue inflammation in aging is not dependent on the Fet Amediated TLR4 activation pathway.
Reduced adipose tissue inflammation in TLR4 deficient mice
To understand the role of TLR4 in aging adipose tissue inflammation, we conducted studies using a TLR4 AGING deficient (Tlr4 Lps-del ) mouse model. We first evaluated the expression of Tlr4 in wild type (C57BL/6) and Tlr4 Lps-del/ /TLR4-KO mice and validated that TLR4 expression was absent in the KO mice (Fig. 3A). We observed diminished p-NFκB expression in the aging adipose tissue of TLR4-KO mice compared to the WT age-matched controls ( Fig. 3B and 3C). Our analyses indicated that the expressions of three major proinflammatory cytokines (Il-6, Mcp1 and Tnf-α) in the adipose tissue were significantly reduced in old TLR4-KO mice compared to the old wild type mice (Fig.3D). Protein analysis on adipose tissue lysates also indicated decreased levels of IL-6 and MCP1 in old TLR4-KO compared to age-matched WT mice (Fig.3E). Of note, we were unable to detect TNF−α in the adipose tissue lysates by ELISA method. This data suggests that TLR4 contributed to aging adipose tissue inflammation in a Fet A-independent manner.
ER stress response and autophagy activity in the adipose tissue of TLR4 deficient old mice
Our recent studies have demonstrated that impaired autophagy leads to elevated ER stress in aging adipose tissue [8]. We sought to determine if TLR4 deficient mice were protected from the elevated ER stress response and reduced autophagy activity in aging. Consistent with our previous reports [7], expression of ER stress response genes Chop and Bip are reduced in the adipose tissue of TLR4 deficient mice compared to the age-matched old wild type mice (Fig. 4A). Additionally, expression levels of autophagy genes (Beclin-1, Atg7, Lc3a), except p62, are also increased in the mutant mice (Fig. 4B). To examine the impact of TLR4 deficiency on autophagy activity, we cultured SVFs (stromovascular fraction) from the adipose tissue and autophagy efficiency is measured by accumulation of LC3II/LC-I following bafilomycin (autophagy blocker) treatment. As expected, autophagy activity was enhanced in the SVFs of TLR4-KO mice compared to the WT mice ( Fig. 4C and 4D). However, no significant interaction between the variable (WT vs KO) or treatment (V vs Baf) was observed.
Reduced expression of senescence-associated markers in the adipose tissue of TLR4 deficient old mice
With reduced IL-6 and MCP-1 production in the TLR-4 KO mice (Fig. 3E), we next determined if TLR4 KO is associated with change in the expression of senescenceassociated genes. We observed reduced mRNA expres- AGING sion of p16 gene and no difference in the expression of p21 gene (Fig. 5A) in adipose tissue from TLR4-KO mice compared to age-matched WT old mice. However, protein expression of both p16 and p21 were significantly diminished in the TLR4 deficient adipose tissue lysates (Fig. 5B). These results indicate that aging TLR4 deficient mice are protected from adipose tissue senescence.
TLR4 KO mice have improved glucose tolerance
To determine glucose tolerance, we performed an intraperitoneal glucose tolerance test (IPGTT) on both WT and TLR4 deficient old mice. We found that TLR4 deficient old mice have improved glucose tolerance compared to the age-matched WT mice ( Fig. 6A and 6B). The area under the curve (AUC) of TLR4-KO mice is significantly smaller than that of the WT mice, consistent with the idea that TLR4 deficiency promotes glucose tolerance in part through the observed reduced inflammation in visceral adipose tissue.
DISCUSSION
Chronic low-grade inflammation in adipose tissue contributes to the development of metabolic diseases, Data represented in bar diagrams are mean + SD value of relative mRNA expression from three independent experiments where total RNA was extracted from gonadal fat pads of young (n=5) and old (n=5) mice and used as a template for one-step qRT-PCR reaction. Protein expressions were determined by western blotting of adipose tissue lysates from young (n=5) and old (n=5) mice. Data presented here are representative image of three independent experiments. The significance levels *p<0.05, **p<0.001 were analyzed by unpaired Student's t-test using means and SDs.
AGING including insulin resistance in aging [6,24]. To understand the mechanism of adipose tissue inflammation, many research groups reported activation of TLR4 receptors by lipids in the context of obesity [1,15,25,26]. Data from obese animals on adipose tissue inflammation have demonstrated that Fet A serves as an adapter for FFA in the activation of the TLR4 pathway [15]. Intriguingly, both Fet A and TLR4 expressions are elevated in the adipose tissue of DIO mouse models, and adipose tissue Fet A serves as a chemoattractant for macrophage migration and polarization [15,21]. Contrary to the DIO models, we demonstrate here that Fet A level is diminished not only in adipose tissue but also in the serum and liver of old mice (Fig. 2). On the other hand, adipose tissue TLR4 expression remains unchanged between young and old mice. These data indicate that, unlike DIO, the mechanism of agingassociated adipose tissue inflammation is independent of the expression of Fet A or TLR4. To understand TLR4-mediated adipose tissue inflammation in the context of aging, we analyzed the adipose tissue of young and old TLR4 deficient mice, and observed that TLR4 deficiency protects adipose tissue inflammation in old mice (Fig. 3B), as indicated by reduced protein expressions of IL-6 and MCP-1 in the adipose tissue lysates (Fig. 3C).
AGING
We have demonstrated previously that elevated endoplasmic reticulum (ER) stress response in aging adipose tissue also contributes to greater inflammatory responses [7] as a result of age-related compromised autophagy activity [8]. The reciprocal interaction between autophagy dysfunction and ER stress has also been investigated in human adipose tissue of type 2 diabetes mellitus (T2DM) [27]. We asked whether TLR4 deficiency protects adipose tissue inflammation in aging through the ER stress response pathway and/or altered autophagy function. We observed diminished ER stress response and elevated expression of autophagyassociated genes in the gonadal adipose tissue of TLR4 deficient mice (Fig. 4). Our data are consistent with the observation that TLR4 deficient mice are also protected from ER stress response in diet-induced obesity [28]. AGING Recent studies have indicated that aging promotes the accumulation of senescent (SEN) cell burden in the VAT, which can also lead to inflammation. This is further supported by the findings of reduced SEN burden in long-lived mouse models [29], and improved health span by clearing SEN cells by drugs (senolytics) in mouse models [9,11,30]. Our analyses indicate that expressions of senescence-associated genes (p16 and p21) were diminished in the VAT of TLR4 deficient mice, which may then lead to a reduction of the chronic low-grade inflammation observed in aging (Fig. 5).
Finally, we demonstrate that TLR4 deficient old mice have improved glucose tolerance compared to the agematched WT mice (Fig. 5). This was consistent with the observation that TLR4 deficiency protects the animals from glucose intolerance induced by HFD [28].
Taken together, our Fet A and TLR4 expression data in aging adipose tissue inflammation are in contrast to those seen in DIO models. We have shown that TLR4 deficiency reduces age-associated adipose tissue inflammation by reducing ER stress, augmenting autophagy function, and by reducing senescence cell burden. Our study indicates an important aspect of aging-associated adipose tissue inflammation that is distinct from diet-induced obesity. This study provides strong evidence that TLR4 plays a significant role in promoting aging adipose tissue inflammation. However, we were unable to define the endogenous ligand for TLR4 activation. We have tested other known endogenous TLR4 ligands, including Tenacin C1 and fibronectin, and they both have diminished expression in the aging adipose tissue (Fig. S1).
The underlying mechanism for TLR4 activation in aging adipose tissue is incompletely understood. Recent progress in the field might explain TLR4 signaling and its regulation as follows:
a) Fet A independent activation of TLR4 by lipids
It has been demonstrated previously that lipids such as ceramide and sphingomyelin are more abundant in the AGING aging adipose tissue [31]. In addition, these lipids operate through TLR4 as specific inhibitors (myriocin or Me-SM) of ceramide synthesis pathway both in young or old adipose tissue block the productions of both IL-6 and TNF-α upon LPS treatment.
b) Endogenous activation of TLR4 via alarmin HMGB1 (high mobility group box1)
With the accumulation of SEN cells in the aging adipose tissue, there is increased abundance of high mobility group box 1 (HMGB1). It has been demonstrated that extracellular HMGB1 acts as a ligand for TLR4 to produce IL-6, and this effect can be blocked with HMGB1 antibody or TLR4 inhibition in a fibroblast culture system [32]. Interestingly, the circulatory HMGB1 is also elevated in aged mice [32].
c) Role of Cbl-b in TLR4 attenuation in macrophages
It has been reported that Casitas B-cell lymphoma gene (Cbl-b) plays a role in aging associated insulin resistance. The HFD caused increased inflammation of adipose tissue in the Cbl -/mice compared to Cbl +/+ mice. This effect is due to Cbl-b suppression of the saturated FA-induced TLR4 signaling by ubiquitination and degradation of TLR4 in macrophage cell line [33].
In summary, we have demonstrated a Fet Aindependent role of TLR4 in promoting adipose tissue inflammation in aging adipose tissue. It is likely that TLR4 activation is regulated via one or more of the mechanisms described earlier in promoting aging AT inflammation, but the concept will need further investtigation. Interestingly, the current study distinguishes aging adipose from DIO-associated inflammation in the context of TLR4 ligand Fet A, although both are mediated through TLR4 activation.
Reagents
Bafilomycin A1 (Baf), Collagenase D, D-glucose were obtained from Sigma-Aldrich (St. Louis, MO). All the chemicals were dissolved in the appropriate media solution or in dimethyl sulfoxide (DMSO) as per manufacturer's instructions and used at the indicated concentrations.
Isolation of adipose tissue
Epididymal adipose tissues were collected from the mice after euthanasia using standard procedure as described previously [7,8].
RNA extraction and real-time RT-PCR
Total RNA was isolated from 100mg of adipose tissue using Qiagen lipid isolation kit and Qiazol. mRNA fold expression changes were estimated by qRT-PCR procedures as previously described [7].
Western blotting
300mg of AT was placed in 500μL of RIPA buffer supplemented with protease and phosphatase inhibitors and subsequently sonicated to produce cell lysates for western blotting analysis. Standard western blotting techniques were performed for protein analyses as described in previous studies [7,8]. In brief, 50μg total proteins were separated on SDS-PAGE. The gel was transferred to a PVDF membrane, blocked with Superblock (Thermo Scientific) solution for 30 minutes and incubated with appropriate antibody overnight followed by HRP-conjugated secondary antibody in 5% non-fat dry milk in Tris-buffered saline for blotting and exposure of the membrane to a photographic film.
Enzyme-Linked Immunosorbent Assay (ELISA)
Quantitation of Fet A in the serum, IL-6, and MCP-1 in the adipose tissue lysates were performed using respective ELISA kits (R & D Systems).
Intraperitoneal Glucose Tolerance Test (IPGTT)
Mice were fasted overnight for approximately 16 hrs by transferring to clean cages. Mice were weighed to calculate and record the volume of 20% glucose solution required (2g of glucose/kg body mass) for IP injections: volume of IP glucose injection (μl) =10 X body weight (g). Mice were restrained with an approved restrainer device with the tail exposed to score the tip of the tail with a sterilized scalpel blade. The first drop of blood was discarded and a small drop of blood is placed on the test strip of an animal blood glucose meter (Abbott Alpha TRAK 2 Blood Glucose Monitoring System) to record the fasting glucose level (t=0). Appropriate amounts of glucose were injected into the peritoneum as previously determined. The blood glucose levels were measured at 15, 30, 60 and 120 minutes following glucose injections and recorded. At the end of the experimental session, mice were placed in a clean cage with food and water.
|
2017-11-03T01:47:28.888Z
|
2017-09-01T00:00:00.000
|
{
"year": 2017,
"sha1": "856c36c2d0b70c14b059b2816b82c8c6ed27173e",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.18632/aging.101288",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "856c36c2d0b70c14b059b2816b82c8c6ed27173e",
"s2fieldsofstudy": [
"Biology",
"Environmental Science",
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
64785474
|
pes2o/s2orc
|
v3-fos-license
|
Superresolution microscopy for bioimaging at the nanoscale: from concepts to applications in the nucleus
: Superresolution microscopy breaks the diffraction limit of light, making it possible to visualize a broad range of subcellular components with nearly molecular scale detail. The potential of this powerful tool is continuously growing since the implementation of optical configurations and data analyses compatible with the technically challenging, yet frequent in biology, thick and crowded samples. We review the principles underlying stimulated emission depletion, structured illumination, and single-molecule localization microscopy approaches, and their technical developments, with an emphasis on three-dimensional and live-cell imaging. Special attention is brought to the new requirements for probe efficiency, namely their size and their photophysical properties. Finally, recent applications exploring the interphase nucleus are described to illustrate the performance of superresolution techniques
Introduction
Microscopy has long been a valuable tool for visualizing the complexity of biological structures as well as for directly probing the dynamics of biological processes within cells, tissues, and organisms.The components of living matter span several orders of magnitude in size, ranging typically from several nanometers for individual proteins to tens of micrometers for a mammalian cell.Various microscopy techniques have been developed for the study of biological questions at these multiple scales.Electron microscopy (EM) is the method offering the highest resolution (∼nm) and has allowed the detailed study of numerous cellular nanostructures.However, EM does not provide information on the identity of molecules building subcellular structures and is unsuitable for applications in living systems, excluding the possibility to follow dynamics.Atomic force microscopy is a scanning probe method providing high resolution, comparable to that of EM, and can be used for live imaging.Specific structural information can also be obtained by functionalization of the scanning tip, but atomic force microscopy can only explore the surface of cells.
A technique that has been widely used for the specific study of dynamics and localization of intra-and extracellular components in living specimens is fluorescence microscopy (FM).The simplest method for fluorescence imaging is the wide-field configuration (ie, epifluorescence).Diffraction of light within the optical system sets a theoretical boundary for the maximal resolution of a fluorescence microscope.The theoretical image of a point source through an optical system is best described by an Airy pattern.In practice, however, aberrations and other factors modify this The diffraction of light limits the resolution of the system such that emitters closer than the width of the PSF cannot be resolved, leading to a loss of structural detail.Abbreviations: PSF, point spread function; NA, numerical aperture.
theoretical profile.The point spread function (PSF) is the real three-dimensional (3D) image of a point source obtained through the microscope and comprises both the effect of diffraction and aberrations of the system (Figure 1A).The width of the PSF in the lateral direction is ∆ ∼ λ ×0.6/NA,where λ is the wavelength of the excitation light and NA is the numerical aperture of the objective.The resolution of an optical system is defined by the distance at which two point sources in the sample can be resolved in the image plane. 1,2hen the two point sources are found closer than ∆, their diffraction patterns overlap and the two sources cannot be resolved (Figure 1B).Thus, ∆ represents the resolution of the optical system, and structures smaller than this intrinsic distance cannot be resolved optically.This limit in resolution prevented scientists from visualizing the structures and processes happening below that scale length.Confocal microscopy improves the contrast of the image by reducing the out-of-focus light using a pinhole located at the confocal image plane of the light path.The sample is illuminated with a focused spot of laser light, and images are constructed pixel-by-pixel by raster scanning.The sizes of the spot PSF and the pinhole determine the diffraction-limited resolution of the constructed image, typically 200-300 nm in the lateral and ∼500-700 nm in the axial directions.
Given the aforementioned advantages of FM, a major direction for instrumental development has been to beat the diffraction limit of light and increase the resolution up to that of EM.In the past decade, three classes of techniques that allow subdiffraction fluorescence imaging have been implemented, developed, and commercialized (reviewed in Schermelleh et al, 3 Cox, 4 Habuchi, 5 and Fornasiero and Opazo 6 ).Stimulated emission depletion (STED) microscopy is based on a confocal system and reaches subdiffraction resolution by decreasing the size of the detected PSF.This is achieved by selectively turning off molecules found away from the center of the excitation beam. 7,8Structured illumination microscopy (SIM) is a wide-field technique that beats the diffraction limit by illuminating the sample with patterned light, thus allowing the microscope to transmit higher spatial frequencies, ie, finer sample structures, than allowed by the Abbe limit. 91][12][13] Over the past few years, studies applying superresolution microscopy (SRM) have revealed that these three approaches have their specific advantages and drawbacks, suggesting their potential complementarity in unraveling nanoscale biological processes.A valuable comparative overview of SRM methods characteristics and performance, including light sources, spatial and temporal resolution, and limitations, has been provided in a study by Schermelleh et al. 3 Here, we review the principles and fundamental advances in SRM methods and discuss their live cell imaging and probe requirements.We then highlight diffraction-unlimited quantitative microscopy studies that have allowed to gain unprecedented detailed insight into the structure and inherent dynamics of fine cellular components in the nuclear compartment.
Stimulated emission depletion
The first technique that has achieved subdiffraction resolution fluorescence imaging is STED microscopy.STED was theoretically introduced in the 1990s 14 and experimentally demonstrated in 2000. 15This method relies on the photophysical phenomenon formalized by Albert Einstein, termed stimulated emission.When a fluorophore in its excited state is illuminated, it can return to its ground state through emission of a photon with the same energy as the stimulating photon.
In STED microscopy, subdiffraction resolution is obtained by shrinking the effective PSF of the diffractionlimited excitation spot in a confocal setup (Figure 2A).Stimulated emission is used to deplete the population of excited fluorophores that are located at the periphery of the excitation spot.Selective depletion is achieved by applying a doughnut-shaped beam with zero intensity at its center, aligned with the center of the excitation beam.The excitation laser has a wavelength near the absorption maximum of the fluorophore used for sample labeling, and the depletion laser has a longer wavelength than the fluorophore maximum emission wavelength.Thus, the excited fluorophores found within the minimum of the doughnut will emit at the natural emission wavelength, and those outside will emit at the depletion laser wavelength.The resolution of the system is increased when the size of the doughnut hole is reduced by increasing the depletion laser power.In biological samples, resolutions up to 20 nm have been reported. 16The spatial resolution of a STED microscope is strongly dependent on the quality of the depletion beam profile, which will define the shape and size of the STED excitation spot.Improving the spatial resolution requires a finely tuned depletion beam and a perfect alignment with the excitation line.
According to the excitation and depletion schemes used, there are several designs of STED microscopes, namely with pulsed, continuous wave (CW) and two-photon laser sources.Pulsed mode STED (p-STED) achieves the highest resolution and requires synchronization of the excitation and depletion laser pulses. 17To detect nondepleted fluorophores, either the timing 14,15 or the lifetime information 18,19 is used in p-STED.Using CW lasers for both excitation and depletion simplifies the setup since no precise time delays between laser pulses are needed. 20However, the resolutions achieved are lower compared to p-STED.Two-photon excitation has been combined with STED, 21 both in the pulsed and the CW modes in order to image thick samples, such as tissue slices, with diffraction-unlimited resolution. 22,23The different modes of STED microscopy have been widely used for both fixed and live cells, and applications have been reviewed. 7,8n STED, the use of a doughnut-shaped depletion laser beam improves lateral resolution, but the axial resolution remains that of a confocal setup, since zero depletion intensity is distributed along the optical axis.Subdiffraction axial resolution was achieved by tuning the shape of the depletion beam. 15,24Another approach has been to combine STED with either total internal reflection fluorescence (TIRF) 25,26 or with a 4Pi microscope configuration.In TIRF, the incident angle of the excitation light is highly inclined to obtain an evanescent wave with exponential decay, which restricts excitation to a thin region of 100-200 nm above the surface of the coverslip.TIRF effectively removes out-of-focus blur; however, its use is limited to imaging cellular components near the cell surface.The 4Pi setup uses two opposing objective lenses, both focused at the same point.This method improves axial resolution down to ∼80 nm and can be applied to samples a few micrometers thick, though its implementation is challenging. 27,28ulticolor imaging has also been achieved in STED microscopy.The first type of multicolor STED requires an excitation/depletion laser couple per fluorophore, [29][30][31] which is technically demanding.[34]
Structured illumination microscopy
When a fluorescent sample is observed with an optical microscope, the structure is blurred in the resulting image due to the diffraction of light (Figure 1B).In other words, features of a sample smaller than ∼200 nm in the lateral and ∼700 nm in the axial directions could not be transmitted by the optical setup.This is the case in conventional wide-field microscopy, in which the specimen is illuminated with a nearly homogeneous beam of light.SIM is a wide-field configuration capable of doubling the diffraction-limited resolution.In two-dimensional (2D) SIM, 35 this is achieved by exciting the sample with a line pattern of sinusoidally alternating intensity maxima and minima with a frequency at the diffraction limit (Figure 2B).For a given orientation and phase of the sinusoidal stripes, the resulting raw image is an interference pattern between the illumination and the sample and encodes subresolution structural information that is filtered by a conventional microscope.A high-resolution image is thus reconstructed by mathematical processing of raw images acquired with several directions of the patterned excitation. 9ypically, 2D imaging requires nine raw images (three phases along three orientations at 120°).
By modulating the illumination pattern so that it varies sinusoidally in all three directions in space, the third dimension was introduced to SIM. 36 The 3D SIM allows physical optical sectioning with axial resolution of ∼300 nm.The increased complexity of the excitation pattern requires to image at five different phases so that the resulting data can be mathematically decomposed into the constituting high-resolution parts.To be able to computationally reconstruct a high-resolution 3D-SIM dataset, each Z-section requires 15 exposures (Figure 2B).The sections have to be taken not more than 125 nm apart to allow full sampling in the axial direction.
A major disadvantage of SIM with respect to STED and SMLM is the relatively low attainable resolution.It has been shown that, in principle, SIM can reach higher resolutions if the fluorescence response is no longer linear, by saturating fluorophores in the excited state. 37The concept was applied in saturated SIM (SSIM) with lateral resolution of ∼50 nm using fluorescent beads. 38The high laser intensities required in this approach make its application in biological imaging challenging.An alternative to obtain nonlinearity is the use of reversible on-off transitions of a specific class of fluorescent probes.SSIM with the photoswitchable protein Dronpa allowed ∼60 nm resolution imaging of nuclear pores in extracted nuclei using the TIRF mode. 39he relatively large number of acquisitions per plane (∼15) in SIM can lead to photobleaching and sample drift during the acquisition.These effects can severely degrade performance and produce reconstruction artifacts.To reduce these shortcomings, it is important to correctly match the refractive indices, increase labeling contrast, and reduce sample movement during acquisition (either mechanical or biological).Particular attention must be paid when interpreting structures that are close to the SIM resolution limit, as reconstruction at these length scales is intrinsically prone to artifacts.SIM has been a popular choice to reveal various cellular structures at higher contrast. 40It offers the possibility of fast 3D imaging with most conventional fluorophores as long as they are sufficiently photostable and is highly convenient for multicolor applications.
Single-molecule localization microscopies
SMLM or probe-based superresolution imaging is a family of techniques that utilizes the particular photophysical properties of a subset of fluorescent dyes to accurately determine their individual positions and thus obtain diffraction unlimited resolution. 10These include photoactivated localization microscopy (PALM), 41 fluorescence PALM (FPALM), 42 stochastic optical reconstruction microscopy (STORM), 43 and direct STORM (dSTORM). 44Unlike STED and SIM, which tune the illumination pattern to improve imaging resolution, SMLM employs a classical wide-field configuration.The principle of SMLM methods relies on the possibility to localize a single point source of light by fitting its PSF with a Gaussian or Lorentzian function.The precision of localization is dependent on the number of photons emitted by the molecule, the background, and the width of the PSF. 45An underlying condition is a low probability of emitter overlap, ie, only a small subset of fluorophores is to be emitting in the same time over the field of view (Figure 2C).This is achieved either using photoactivatable proteins that are reversibly/ irreversibly turned on (PALM/FPALM) or through reversible stochastic photoswitching of organic dyes in the presence of a reducer in oxygen-depleted medium (STORM/dSTORM).The amount of simultaneously emitting molecules can be controlled by modulating the intensities of an excitation laser (typically in the visible spectrum), which serves to image and turn off (or photobleach) the fluorophores.7][48] STORM relies on pairs of activator and reporter dyes coupled to the same probe molecule.The activator dye absorbs at the activation laser wavelength and facilitates the activation of the reporter dye through energy transfer between adjacent molecules.The activated reporter dye absorbs light from the excitation laser, and its emission position is localized.In contrast, dSTORM makes use only of the absorption properties of the reporter dye.It is worth mentioning that both methods use similar activation/excitation schemes and imaging buffer composition.
Localization precision in SMLM is in the range of ∼10-30 nm and can be improved using brighter probes.However, the smaller the uncertainty in the emitter position, the higher the fluorophore-labeling density required to effectively increase the structural resolution. 49According to the Nyquist sampling theorem, the average distance between adjacent fluorophores must be twice smaller than the desired resolution.For the evaluation of SMLM image resolution, a Fourier ring correlation method was introduced, with the advantage that no detailed knowledge of the sample is needed for the calculation. 50or the reconstruction of a high-resolution image, the positions of all the detected single-molecule fluorescent events are overlaid, with intensities reflecting both density and localization uncertainty (Figure 2C, right).To collect a sufficient amount of localization data, most often tens of thousands of frames are needed for biological samples.The long acquisition times, typically lasting tens of minutes, lead to nonnegligible sample drift.In the axial direction, drift is corrected during acquisition with an autofocus feedback system.Lateral drift is corrected during postprocessing thanks to fiducial markers added to the sample or using spatiotemporal cross-correlation of localizations. 512][43] However, to image structures located further than ∼200 nm above the coverslip surface, several optical and computational techniques have been developed to obtain axial localization information (3D SMLM).Three categories of 3D-SMLM methods can be distinguished: interferometric approaches (including 4Pi, also used in 3D-STED and 3D-SIM configurations), 52 multiple plane imaging, 53 and PSF engineering.The last category breaks the symmetry of the PSF, thus the axial position of fluorophores can be determined using calibration curves.A widely used approach is the introduction of astigmatism in the microscope emission path either with a cylindrical lens 54 or with adaptive optics, which in addition allow optical aberrations correction. 55Axial resolutions reported with this method have reached ∼50 nm within a range of ∼750 nm.Alternatively, higher probing depth has been obtained by double helix shaping of the PSF (∼1.5 µm) with similar axial resolution. 56Isotropic resolution of ∼10-15 nm with a 3 µm axial range was achieved with the self-bending PSF method. 57 detailed overview of 3D SMLM approaches as well as a critical assessment of their performances and applicability have been recently provided by Hajj et al. 58 A further improvement of SMLM has been the optical sectioning capacity.Thick samples, such as whole cells (up to ∼10 µm above the coverslip surface) and 3D cell cultures (50-150 µm deep), have been imaged combining 3D PALM with two-photon activation 59 and light-sheet microscopy, 60,61 respectively.
3][64] Thus, the relative distribution of various molecular assemblies and cellular structures in both fixed 65,66 and live [67][68][69] cells have been revealed with remarkable detail.
Live cell imaging
A notable strength of FM is the possibility to directly probe biological processes in living samples.This allows not only the visualization of biomolecules in their nearly natural environment, but also the study of dynamics and structures of biomolecular factors, their interactions, and their transport.The high contrast, specificity, and sensitivity and the relatively low invasiveness and versatility of the labeling have contributed to the establishment of FM as a method of choice for live cell imaging.However, the time scale of a large number of cellular events is such that it remains technically challenging to obtain sufficient temporal resolution, while preserving the sensitivity of detection and the survival of the specimen. 70The challenge is even greater when in addition high-spatial resolution is needed to study smaller than the diffraction limit cell components with inherently low molecular density.In this context, the performance of fluorescence microscope configurations for a given live cell experiment is to be evaluated by taking into account the imposed trade-offs in imaging parameters, namely acquisition speed, spatial resolution, imaging depth, and the extent of light-induced photodamage, affecting both the fluorescent probe and sample viability.For instance, improving the temporal resolution demands a faster imaging rate, hence shorter exposure times for excitation.The result is a lower fluorescence signal which affects the attainable spatial resolution regardless of the superresolution technique employed.Consequently, laser power is to be increased for better signal detection, leading to phototoxicity, which generates a risk of artifactual observations.
In practice, SRM methods, while having their specific weaknesses and strengths, have been successfully applied for the study of nanoscale-sized dynamic biological phenomena, with imaging speed of tens of frames per second (fps).SIM offers the highest acquisition rates and reduced photodamage compared to STED and SMLM (10 3 -10 6 times lower light exposure), although spatial resolution is limited.Both fast imaging (28 fps) and high resolution (62 nm) have been achieved in STED in a molecularly crowded environment. 71owever, phototoxicity due to the elevated laser powers required to reach high spatial resolution remains a major limitation for live cell imaging with STED.The photon charge applied on the sample was significantly reduced with a STED variant, which uses fluorophore photoswitching in line with the concept of reversible saturable optically linear fluorescence transitions (RESOLFT). 72The imaging speed was further increased as RESOLFT was combined with multiple doughnut beams to scan the sample simultaneously. 73MLM is intrinsically slow since accurate localization of individual fluorophores requires that only a sparse subset of emitters is fluorescent in each frame within a diffractionlimited spot.Thus, a large number of frames are needed for image reconstruction, which limits the temporal resolution.However, SMLM is able to access single-molecule information, making it an attractive technique to obtain quantitative information on protein numbers and dynamics.The development of high-density localization algorithms [74][75][76] led to a considerable decrease in acquisition time.The performance of SRM methods in the context of live cell imaging, and the most recent developments have been discussed elsewhere. 4,77obes for superresolution imaging Specific identification of molecules within biological samples with low invasiveness and high imaging contrast are the hallmarks of FM.However, depending on the fluorescent probe and the individual requirements of the imaging technique, particular attention must be paid during sample preparation and the acquisition procedure to avoid potential artifacts.
Molecular tags
Specificity in fluorescent labeling is obtained either with genetically encoded tags fused to the molecular target or with affinity probes.The former strategy allows the labeling of proteins, the tag size is relatively low (∼25 kDa), and it is live cell compatible.Fusion protein labels can be either intrinsically fluorescent, ie, the well-known GFP and its variants, or coupled to a fluorescent dye by covalent enzyme-ligand binding, such as the commercially available SNAP-tag (∼20 kDa). 78When introducing tagged proteins in a biological specimen, cell physiology may be altered by overexpression, aggregation, mistargeting, misfolding, and perturbation of protein function, which constitute the main limitation of this labeling approach in conventional microscopies, and to an even greater extent at subdiffraction resolution.A powerful solution is the use of knock-in strategies, providing endogenous expression levels, especially when protein quantification is intended, as in most PALM applications. 79iological structures can alternatively be tagged with affinity probes, among which antibodies are the most widely spread.Antibodies are an accessible, versatile tool, which allows direct specific labeling of endogenous epitopes.They are particularly useful to target, among others, posttranslational protein modifications (phosphorylation, ubiquitination, SUMOylation, etc) and even to recognize methylation sites on DNA.Whereas diffraction-limited microscopy is insensitive to the large dimensions of antibodies (∼150 kDa/∼15 nm) allowing secondary antibody labeling, in superresolution imaging (SMLM in particular), probe size becomes a parameter potentially limiting the achievable structural resolution.Consequently, primary antibody monovalent fragments (F ab , ∼50 kDa) or the naturally occurring single-chain camelid antibodies (also named VHHs or nanobodies, ∼15 kDa) are a promising development, 80 though their availability is still limited.
In SRM, a nonnegligible aspect of intracellular component visualization with affinity probes is the requirement for sample fixation and permeabilization.These processing steps inevitably introduce alterations in the specimen, and structural preservation is critical for properly interpreting observations of molecular scale detail.For instance, insufficient fixation or destructive permeabilization may result in target mislocalization or degradation.In contrast, strong fixation (as practiced in EM) preserves the structures, but may also restrain epitope accessibility, thus limiting the labeling density and therefore the achievable structural resolution in subdiffraction imaging experiments.An optimized protocol for SMLM sample preparation has been recently introduced. 81
Fluorescent molecules
Imaging contrast (ie, how well the structure of interest can be discriminated from its environment) is a crucial component of FM, which relies on the performance of fluorescent molecules.Some general parameters for assessing fluorophores are brightness (calculated as the product of the extinction coefficient and the quantum yield), photostability, and water solubility.Recently, an additional property that describes the ability of fluorescent molecules to transit between bright and dark states, termed photoswitching, has become fundamental in SRM applications. 46The principle of SMLM relies on the detection of single molecules with nanometer precision.Most often this is achieved by separating emission from each single emitter in time by making use of their stochastic photoswitching behavior.In addition, the use of photoswitchable probes has contributed to considerably improve the performance of other superresolution methods such as RESOLFT and SSIM.Fluorophore photoswitching is usually quantified by the number of switching cycles, the number of detected photons per switching event, the duty cycle (fraction of time a fluorophore spends in an on state), and the on and off switching rates. 63he number of switching cycles reflects the number of times an emitter enters the bright state and can be detected.For SSIM, RESOLFT, and live cell SMLM, multiple detections are preferred to construct high-resolution images.In contrast, quantification of absolute protein numbers with PALM would ideally benefit from a single switch before photobleaching.In practice though, all known fluorophores display multiple switching cycles that must be accounted for in quantification procedures [82][83][84][85] (reviewed in Durisic et al 86 and Shivanandan et al 87 ).The number of detected photons per switching event (a metric of the photoswitch brightness) and the duty cycle (the fraction of time an emitter spends in the fluorescent state) together determine the spatial resolution achievable in SMLM methods.The former is proportional to the localization precision, while the latter limits the number of fluorophores that may be localized within the volume of the PSF.Finally, the on/off switching rates are one factor limiting the speed of image acquisition and thus the temporal resolution of superresolution methods employing photoswitchable probes.
According to their origin, fluorophores are of two types: naturally existing in living organisms and subsequently genetically engineered (FPs), and chemically synthesized (organic dyes).In the context of superresolution imaging, specific advantages of each category impact on the labeling strategy.Typically, the duty cycle of photoswitchable FPs tends to be lower than organic fluorophores and allows imaging of densely labeled structures.In addition, FPs label proteins with a controlled stoichiometry of 1:1, crucial in quantification experiments, whereas organic fluorophores are generally coupled to affinity probes, for which labeling efficiency is difficult to evaluate.In contrast, organic dyes display superior brightness and photostability, allowing higher localization precision.They are available in a broader variety of absorption/emission spectra spanning the visible and importantly the near infrared wavelengths, which makes them convenient for multicolor experiments.While FPs do not require a particular composition of the imaging medium in SMLM experiments, photoswitching of organic fluorophores has been initially obtained by depleting oxygen in the imaging buffer and by addition of a reducer (thiol), toxic for cells.][90][91][92][93] Several studies provide systematic evaluation of FPs and organic fluorophores for superresolution applications. 62,64,94hile most fluorophores have been optimized for a single superresolution technique, probes that display good performance in several of them have been recently developed, such as the photoswitchable proteins Dreiklang 95 and mMaple, 96 which will foster the development of multimodal SRM approaches.
The nuclear compartment studied with SRM
Since its first implementation, SRM has made molecular scale insight into major cellular processes, notably membrane receptor distribution and oligomerization, a critical step in cell signaling, possible. 83,97,98With the evolution of optical setups providing the possibility to image thick samples and the improvement of analysis procedures performance in lower signal to noise conditions, structures and phenomena deeper in the specimens have become accessible to quantitative analysis.In this section, we review recent SRM studies that have contributed to enrich our understanding of the organization and functioning of the nuclear compartment.Specifically, we will focus on research performed in interphase chromatin folding and transcription machinery dynamics, two crucial components of gene regulation.
RNA polymerase 2 (RNAP2) distribution and dynamics
The most regulated step in gene expression is transcription.It involves complex interactions between DNA and transregulatory elements, the latter including histone modifying enzymes, transcription factors, and RNAP complexes.The DNA-binding properties and dynamics of nuclear factors are central to the understanding of transcription and have been intensively explored with biochemical assays, or more recently with genome-wide chromatin immunoprecipitation techniques and single-particle tracking. 99RNAP2 is a well-studied transcription effector; however, its nuclear distribution and dynamics at the molecular level had not been directly probed.In particular, quantitative imaging has been lacking essentially due to the relative abundance of RNAP2 in the nucleus and microscope limitations.Recently, two elegant SMLM studies have provided molecular scale spatiotemporal insight into RNAP2 clustering in mammalian cells. 100,101ranscription was proposed to take place in RNAP2enriched foci known as transcription factories, where transcription of multiple loci can be coordinated and potentiated. 102Cisse et al 100 tested this hypothesis by investigating the dynamics of RNAP2 assembly in live U2OS cells by 2D single-particle tracking PALM, a variant of PALM allowing for the study of the assembly and disassembly dynamics of clusters with a size smaller than the resolution limit.Potential labeling artifacts were discarded by engineering a stable cell line expressing a Dendra2-fused catalytic subunit (RPB1), replacing the endogenous RPB1.Paircorrelation analysis 83 identified clusters of ∼220 nm, while time-correlated detection counting within individual highdensity clusters revealed average lifetime of ∼5.1 seconds, reflecting the transient nature of RNAP2 clustering.An analogous labeling strategy was used by Zhao et al, 101 in which RPB1 was fused to a SNAP-tag and labeled with rhodamine dyes.Localization accuracy and efficiency were improved as STORM imaging was performed in a reflected light-sheet configuration, achieving optical sections of ∼1 µm.Absolute numbers of RNAP2 molecules were determined through a novel spatiotemporal clustering analysis, which together with colocalization estimated that the majority (.70%) of detected foci are composed of single RNAP2 molecules.Quantitative SRM has thus brought arguments against a preassembled, stable organization of transcription sites in the nucleus.
Chromatin organization and dynamics
It is well established that gene regulation and cell fate determination depend on the spatial organization of DNA.Until recently, endogenous genome folding could only be addressed through genetic or biochemical methods 103,104 since nuclear substructures are typically smaller than the resolution limit of conventional optical microscopes.From this perspective, SRM is well suited to provide physical maps of gene regulation processes at molecular resolution and reveal subnuclear structures in situ.
The genetic material in eukaryotes is packed within the nucleus in the form of a nucleoprotein complex termed chromatin.The structural unit of chromatin is the nucleosome, composed of an octamer of the highly conserved histone proteins (H2A, H2B, H3, and H4) and 1.65 turns of the DNA molecule.Hence, fluorescent tagging of chromatin can be performed by labeling the core histone proteins, or directly the DNA. 105The former implies the use of immunofluorescence or protein fusions as discussed before.For instance, dihydrofolate reductase and SNAP-tag fusions have been used for live cell STORM imaging of the histone H2B in mammalian cells, potentially allowing the study of chromatin dynamics in situ. 88,89The second strategy takes advantage of a large variety of intercalating dyes available for sequence-independent DNA labeling.Some of them display SMLM-compatible blinking characteristics and have been successfully used for STORM imaging, namely YOYO-1 in DNA extracts 106,107 and more recently PicoGreen in live cells 90 (Figure 3A).Furthermore, incorporation of modified nucleotides using the DNA replication machinery combined with Click chemistry fluorescent labeling was employed for the visualization of nascent DNA fragments in live HeLa cells with STORM. 108Another SMLM approach using the DNAbinding kinetics of intercalating dyes rather than blinking is binding-activated localization microscopy. 109Alternatively, DNA can be stained in a sequence-specific manner through the fluorescence in situ hybridization (FISH) assay.However, ultrastructural preservation is a major concern in FISH experiments, particularly at enhanced resolution.Adapted protocols have been designed for SIM 110 and will likely be applicable to the higher resolution techniques STED and SMLM.Furthermore, a systematic evaluation of performance of the different histone or DNA-labeling strategies in SRM will allow the validation of the newly observed structural details of chromatin organization (Figure 3).
The global chromatin folding drastically changes throughout the cell cycle, from the ∼500 nm thick and highly compacted chromosomes with characteristic shape in mitosis to the decondensed ∼10 nm chromatin fiber in interphase.These orders of magnitude structural variations represent a specific challenge in superresolution experiments.In mitosis, the high density of DNA and histones is an obstacle to efficient labeling, and sample thickness deteriorates the signal to noise ratio due to out of focus light.Mitotic chromosome organization has been addressed both by 3D SIM and SMLM, and notable achievements have been reviewed elsewhere. 111nterphase chromatin, on the other hand, adopts a loose conformation heterogeneously spreading throughout the entire nuclear volume, resulting in low contrast in SIM and STED images or low localization event numbers in SMLM.Several groups have investigated chromatin heterogeneity and reorganization by labeling core histone proteins in mammalian cells under normal cell growth conditions, comparing differentiation states, and upon physiological stimuli.In an early study, Gunkel et al 112 applied an SMLM variant, namely spectral precision distance microscopy (SPDM) in two colors to investigate nuclear distributions of mRFP1-fused H2A and the GFP-fused chromatin remodeler Snf2H in transiently transfected U2OS fixed cells.Counting the number of neighbors in a 300 nm radius showed nonrandom distributions for both factors, with nuclear regions depleted in H2A, sites of local enrichment of Snf2H, and a partial colocalization of the two proteins.Subsequently, a radial distribution function was calculated to quantitatively explore H2B-GFP localizations with 2D SPDM, 113 uncovering chromatin nanostructures on a scale ,100 nm.The authors introduced compressibility measures to compare large-scale structural fluctuations with polymer models, which indicated nonrandom chromatin distributions even on the micrometer range.Remarkably, significant differences of H2B distribution depended on the expression method, highlighting the crucial importance of proper fusion proteins targeting.Deeper investigation of H2B nonhomogeneity at the nanometric length scale in the nuclei of fixed U2OS cells was performed using 3D PALM. 114The Ripley K(r) statistics of H2B-Dendra2 indicated clusterization without specific size in the range of 10 nm to 1 µm, compatible with the fractal globule model proposed by chromosome conformation capture (Hi-C) 115 and FISH studies, 116 and supporting the idea that chromatin organization is influenced by interloci contacts.Live cell imaging of H2B-PAGFP further revealed that this organization is highly transient. 114More recently, secondary antibody immunostaining combined with 2D STORM was used to follow the endogenous H2B heterogeneity throughout differentiation in human and mouse cells. 117The super resolved images indicated that H2B is distributed in discrete nanodomains throughout the nucleus, and clustering analysis of raw detections confirmed the lack of a characteristic size of nucleosome-enriched domains.The number of histone molecules per nanodomain was extracted using a calibration curve of H2B localizations densities, which were measured in vitro for nucleosome arrays of known length.Nucleosome density and number correlated with the pluripotency grade, indicating that differentiation leads to an increase in domain compaction.Interestingly, computer simulations showed that the observed H2B heterogeneity can be explained by the incomplete nucleosome occupancy of the DNA fiber.
In addition to SMLM, other superresolution methods have been applied to investigate chromatin structure.The transient organization of chromatin was probed with STED in immunostained rat cardiomyocytes, where pixel intensity levels accounted for the local densities of molecules. 118nduction of hypertrophy, known to cause massive gene expression changes, resulted in multilevel redistribution of endogenous H3. Furthermore, SIM imaging of the β-globin locus with FISH in mouse erythroid cells allowed for following of chromatin folding dynamics in opposing transcriptional states. 119Size and shape analysis revealed that inactive chromatin explores a wide range of conformations, while gene activation resulted in the FISH spot condensation.
Overall, whole-genome labeling methods combined with SMLM have provided a glimpse at the complexity in chromatin organization.However, two main drawbacks currently make interpretation of images difficult and functional studies complicated.The first is the lack of genomic specificity, and the second is the common appearance in the observed structures of collections of protein clusters displaying no clear continuity.Recently, a new approach based on Oligopaint technologies provided one possible solution to these issues.In this method, thousands of short, fluorescently labeled oligonucleotides are used to produce a FISH probe that covers large genomic regions. 120The application of this method to visualize topological domains has produced impressive superresolution reconstructions of the Bithorax Complex domain in Drosophila (Figure 3B). 121
Conclusion and prospects
Superresolution fluorescence imaging allows visualization of cellular components in the range of 10-200 nm, so far unexplored by diffraction-limited fluorescence microscopies.The optical configurations and analysis methods have undergone significant development over the last few years.However, several important obstacles need to be circumvented for superresolution microscopies to become widespread.
Superresolution microscopies are typically more difficult to implement than conventional microscopies, and their results more difficult to assess.Several important controls have to be performed in the quality of acquisition and analysis in order to ensure an accurate reconstruction.These are usually performed by custom-made software packages.Unfortunately, few tools currently exist that allow for quality controls, and these are often not available to the community.Ideally, future software developments should be made in a common, open-source platform easy to port, validate, and improve.In this respect, much is to be learned from software development paradigms used by other communities (ie, CCP4 package for crystallography).
Conventional microscopy can be performed in multicolor due to the large panel of organic and genetically encoded fluorescent probes available.This is currently not the case for SRM, which is in practice limited to at most two colors or less for live applications on real biological systems.In part, this limitation is due to a general lack of adapted fluorophores.Hopefully, future developments will improve our choice of available dyes.The careful study of dye photophysics will likely improve our ability to rationally engineer better dyes and devise new acquisition and analysis modes, as well as help characterize novel fluorophores found by screening methods.
Finally, an important limitation of current SRM relies on their poor performance in thick specimens (eg, embryos, tissues).This limitation is due to the increase in aberrations with the distance to the objective as well as to the diffusion of light through highly inhomogeneous media.][124]
Figure 1
Figure 1 Resolution in conventional fluorescence microscopy.Notes: (A) Light emitted by a point source (fluorescent protein or organic fluorophore) is detected by the optical microscope as a PSF of width which depends on the wavelength of emission and the light collection capacity of the objective.(B)The diffraction of light limits the resolution of the system such that emitters closer than the width of the PSF cannot be resolved, leading to a loss of structural detail.Abbreviations: PSF, point spread function; NA, numerical aperture.
Figure 2
Figure 2 Superresolution microscopy techniques.Notes: From left to right: principles underlying detection for each method, acquisition schemes, resulting images.(A) in STeD, a depletion doughnut-shaped beam is combined with the focused excitation light, thus decreasing the size of the PSF to a volume smaller than the diffraction limit (left).Acquisition (middle) is performed by scanning the two perfectly aligned light sources over the sample with the emitted light collected pixel by pixel by a detector (PMT or APD).(B) in SiM, the excitation of a structure with nonuniform light pattern results in an upshift of the sample spatial frequencies, resulting in Moiré fringes (left).A 3D SiM acquisition (middle) is performed by laterally displacing the illumination pattern (five phases) in three orientations (angles) of the sinusoidal stripes, and spatially modulated images are recorded by a CCD camera.(C) in SMLM, the position of individual emitters is obtained by fitting of their intensity profile detected by a CCD camera (left).The acquisition (middle) relies on the low density of emitting fluorophores (,1/250 nm).The single localizations are then combined to reconstruct the superresolved image (right).Abbreviations: APD, avalanche photodiode; CCD, charge-coupled device; PMT, photon multiplier tube; PSF, point spread function; SIM, structured illumination microscopy; SMLM, single-molecule localization microscopy; STED, stimulated emission depletion; 3D, three-dimensional.
Figure 3
Figure 3 Chromatin labeling strategies for single-molecule localization microscopy.Notes: (A) A 2D live cell dSTORM of DNA in U2OS cells based on direct DNA labeling with PicoGreen.Note the sparser distribution obtained here compared to the rest of the images, which may be due to incomplete labeling or detection.Copyright © 2012 wiLeY-vCH verlag GmbH & Co. KGaA, weinheim.Figure adapted with permission from John wiley and Sons.Adapted from: Benke A, Manley S. Live-Cell dSTORM of cellular DNA based on direct DNA labeling.ChemBioChem.2012;13(2):298-301. 90(B) Subdiffraction (Ba) and superresolution (Bb) image of the bithorax complex domain using Oligopaint; the squares indicate regions where linear densities can be observed; number of events (6,9369) is shown at top right of figure.Copyright © 2015, Rights Managed by Nature Publishing Group.This work is licensed under a Creative Commons Attribution 4.0 International License; please note the Disclaimer of Warranties and Limitation of Liability: http://creativecommons.org/licenses/by/4.0/legalcode.Figure adapted from: Beliveau BJ, Boettiger AN, Avendaño MS, et al.Single-molecule super-resolution imaging of chromosomes and in situ haplotype visualization using Oligopaint FiSH probes.Nat Commun.2015;6:7147.Available at: http://www.nature.com/ncomms/2015/150512/ncomms8147/full/ncomms8147.html. 121Abbreviations: dSTORM, direct stochastic optical reconstruction microscopy; hFb, human fibroblast.
|
2018-12-06T03:40:38.794Z
|
2015-09-28T00:00:00.000
|
{
"year": 2015,
"sha1": "62fee447b5ea00c1ef5de13a00a1a83c724c53e2",
"oa_license": "CCBYNC",
"oa_url": "https://doi.org/10.2147/rrb.s60501",
"oa_status": "GOLD",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "1a8063c1510626d1513e216ac3170efeb996c34d",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
48361086
|
pes2o/s2orc
|
v3-fos-license
|
Clinical Outcomes of Aspirin Interaction with Other Non-Steroidal AntiInflammatory Drugs: A Systematic Review
- Purpose: Concomitant use of some non-Aspirin nonsteroidal anti-inflammatory drugs (NANSAIDs) reduces the extent of platelet aggregation of Aspirin (acetylsalicylic acid). This is while many observational studies and clinical trials suggest that Aspirin reduces cardiovascular (CV) risk attributed to the use of NANSAIDs. Thus, the therapeutic outcome of the interaction needs to be assessed. Methods: We searched various databases up to October 2017 for molecular interaction studies between the drugs and long-term clinical outcomes based on randomized clinical trials and epidemiological observations that reported the effect estimates of CV risks (OR, RR or HR; 95% CI) of the interacting drugs alone or in combinations. Comparisons were made between outcomes after Aspirin alone, NANSAIDs alone and Aspirin with naproxen, ibuprofen, celecoxib, meloxicam, diclofenac or rofecoxib. Results: In total, 32 eligible studies (20 molecular interactions studies and 12 observational trials) were found. Conflicting in vitro/in vivo/ex vivo platelet aggregation data were found for ibuprofen, naproxen and celecoxib. Nevertheless, for naproxen, the interaction at the aggregation level did not amount to a loss of cardioprotective effects of Aspirin. Similarly, for ibuprofen, the results overwhelmingly suggest no negative clinical CV outcomes following the combination therapy. Meloxicam and rofecoxib neither interacted with Aspirin at the level of platelet aggregation nor altered clinical outcomes. The clinical outcomes data for celecoxib and diclofenac are in conflict. Conclusion: Aspirin appears to maintain its cardioprotective effect in the presence of naproxen, ibuprofen, meloxicam and rofecoxib. The limited available data suggest that the effect of interaction at the platelet aggregation level may dissipate shortly, or the reduced platelet aggregation yielded by the interaction may be sufficient for cardioprotection; i.e., no need for near complete aggregation. In addition, cardioprotective effect of Aspirin, despite reduced platelet aggregation caused by NANSAIDs, may be through its involvement in other mechanisms such as the renin-angiotensin system and/or metabolism of arachidonic acid to biologically active compounds mediated by cytochrome P450. platelet inhibit, antiplatelet effect, and or tolmetin diclofenac or ketorolac or nabumetone indomethacin sulindac or piroxicam or meloxicam or mefenamic acid meclofenamic acid rofecoxib or celecoxib veldecoxib or paracoxib or etoricoxib or lumaricoxib or cyclooxygenase* or cyclo-oxygenase* or COX* and infarction or cerebrovascular or cardioprotect* or cardio-protect* or platelet* or platelet aggregation or platelet aggregation inhibit* or antiplatelet effect* or blood platelets and Interaction or Drug interaction or Interact* Structured Provide a structured summary including, as applicable: background; objectives; data sources; study eligibility criteria, participants, and interventions; study appraisal and synthesis methods; results; limitations; conclusions and implications of key findings; systematic review registration number.
INTRODUCTION
Acetylsalicylic acid (Aspirin) is in clinical use since mid 19 th century. In addition to being an effective analgesic, antipyretic and anti-inflammatory agent, it is used, among other indications, for its anti-platelet property to reduce all-cause mortality, cardiac death, and nonfatal myocardial infarction (MI) (1). Moreover, low-dose Aspirin, alone or in combination, is recommended for the secondary prevention of acute ischemic stroke and transient ischemic attack (2)(3)(4). In general, the anti-platelet effect of Aspirin accounted for the irreversible inhibition of platelet cyclooxygenase-1 (COX-1) enzyme. COX-1 is an enzyme that catalyzes AA to produce several prostaglandins (PG), among them thromboxane A2 (TxA2), a promoter of platelet aggregation (5,6). The inhibition of the COX-1 dependent TxA2 by Aspirin, measured by plasma thromboxane B2 (TxB2) is recommended to be near completion to significantly inhibit platelet function in vivo (7-9).
The non-Aspirin nonsteroidal anti-inflammatory drugs (NANSAIDs) are among the most commonly used medications for a variety of indications ranging from headaches to arthritis. NANSAIDs bind and inhibit the COX enzymes which lead to inhibition of prostanoids biosynthesis including PGs, prostacyclins and thromboxanes (10). Thus, the concomitant use of some NANSAIDs appear to interact with the Aspirin's anti-platelet function, thereby, although unproven, _________________________________________ Study selection and data extraction Both authors examined the titles and abstracts of studies to identify studies that potentially meet the inclusion criteria. The inclusion criteria were as follows: (i) Randomized controlled trials (RCTs) or observational studies (cohort or case-control studies) that include treatment with Aspirin alone or NANSAID alone as well as concomitant use of NANSAIDs with Aspirin. The association between the treatments and risk of CV (MI), cerebrovascular events (stroke) or all-cause mortality were assessed for studies that included odds ratios (ORs), relative risks (RRs) and/or hazard ratios (HRs) with 95% confidence interval (CI). (ii) molecular interactions trials (in vitro, in vivo or ex vivo) in human addressing the interaction at the platelet level between NANSAIDs and Aspirin.
The full texts of these potentially eligible studies were retrieved and independently assessed for eligibility. Disagreements were settled through discussion and consensus. Extracted information included study information (i.e. authors, location, publication date, type of study, number of participants, and study duration), patient characteristics (i.e. age, sex, previous CV events including stroke, Aspirin use, and NANSAIDs use), intervention and comparator (i.e. drugs and doses) and outcomes (i.e. events/total for all study population or subgroups).
The identified studies were excluded if: (i) they were reviews, questionnaire, thesis, letters, simulated studies, meeting summary, conference abstracts, editorial or commentary articles; (ii) had no eligible outcomes or did not report direct comparisons of individual NANSAIDs; (iii) used extra-oral route of administration (e.g., topical use for analgesia) or used other drugs with NANSAIDs or Aspirin.
Quality assessment
The methodological quality of the included observational studies (cohort and case-control studies) was appraised using scales adopted from the Newcastle-Ottawa quality scale (NOS) (15). Based on the study design (cohort or case-control study), each study was evaluated using the appropriate scoring system. Eight items in the included cohort and case-control studies were identified and assessed. Cohort and case-control studies with 6-9, 3-5, and 0-2 points were classified as high, fair or poor quality, respectively.
Eligible studies
Our search strategy yielded 3,563 potentially relevant articles from which 3,498 were found ineligible because they were not epidemiological studies or molecular interactions experiments. Sixtyfive articles underwent full-length article review. Twenty-five of these were excluded because they did not report the outcome of interest (MI or stroke), 5 were excluded because they did not report direct comparison of individual NANSAIDs with or without use of Aspirin, and 3 were excluded because of combination other than Aspirin with NANSAIDs or use of formulation other than oral. Twelve studies (5 cohort studies and 7 case-control studies) with 80,845 events met our eligibility criteria and were included in the analysis (12,13,(16)(17)(18)(19)(20)(21)(22)(23)(24)(25). The eligible studies scored good quality based on the calculated NOS scores (cohorts, 8-9/9 and case-controls 6-8/9) ( Table 9).
Twenty molecular interactions studies addressing the interactions between NANSAIDs and Aspirin were included. The detailed flow chart of search methodology and selection process is shown in Figure 1. Table 1 compares the outcomes of both platelet effects and clinical outcomes. Data on the selected NANSAIDs are provided in Tables 2-7. The detailed characteristics of molecular interactions experiments studies are described in Table 8. The clinical data on the interactions between Aspirin and different type of NANSAIDs are summarized in Table 9.
Platelet aggregation
The 20 eligible molecular interaction studies with the information on the interactions indicated that, in general, the anti-platelet effect of Aspirin is reduced in the presence of ibuprofen, naproxen or celecoxib (Tables 1 and 8). However, meloxicam, rofecoxib and diclofenac do not interfere with the anti-platelet effect of Aspirin.
Meloxicam
No (26) No (43) No (12, 16) NA, not available A trend towards an increase in the rate of recurrent MI has been reported in one cohort study when subjects exposed to Aspirin and ibuprofen (HR, 1.50; CI 1.33-1.70) compared with Aspirin alone users (HR, 0.98; CI 0.94-1.03) (17). A retrospective cohort study has also concluded that patients with history of CV diseases had increased risk of mortality when exposed to Aspirin plus ibuprofen compared with users of Aspirin alone (24).
DISCUSSION
This is, to the best of our knowledge, the first systematic review that compares published Aspirin-NANSAIDs interaction at the platelet level with its long-term clinical outcomes. We have used broad inclusion criteria in many databases to capture molecular interactions experiments, RCTs and observational studies for a range of NANSAIDs and Aspirin users. However, no RCTs data were found.
We found that a NANSAID-Aspirin interaction at the platelet level does not necessarily amount to a loss of beneficial effects of Aspirin. Indeed, for naproxen, studies have consistently reported no negative clinical outcomes after addition of the drug to the Aspirin regimens (Table 3). Similarly, studies overwhelmingly suggest that Aspirin maintains it beneficial effects after addition of ibuprofen to the regimen. (Table 2).
As expected, the cardioprotective effect of Aspirin is not diminished by meloxicam and rofecoxib, two NANSAIDs that do not interact with Aspirin at the platelet level (Table 1). Interestingly, diclofenac for which its lack of effect on the antiplatelet action of Aspirin has been repeatedly reported appears to diminish the clinical benefit of the latter as reported by 2 of eligible 4 studies (Table 1).
Despite the limited number of eligible studies, meloxicam (12, 16) ( Table 4) and rofecoxib (12, 13, 17) ( Table 5) do not appear to diminish the cardioprotective effect of Aspirin. This is not unexpected since these drugs do not interact with the anti-platelet properties of Aspirin ( Table 1).
The data for celecoxib are not as conclusive as those available for naproxen and even ibuprofen since we found only 3 eligible studies. Two studies that suggest no loss of the beneficial effect of Aspirin (12, 13) contradict the other one (17). The reason for the conflicting results is unclear but it may be of relevance to mention that the latter study (17) stands out as the one that has also observed diminishing clinical benefit of Aspirin for ibuprofen, diclofenac as well. Nevertheless, in light of the conflicting data and the limited eligible studies, one cannot draw an unequivocal conclusion as to the clinical outcome of celecoxib-Aspirin interaction. Similarly, one cannot draw a definite conclusion regarding diclofenac as we found only 4 eligible studies, two in each side of the controversy. This is interesting since diclofenac does not interact with Aspirin at the platelet level (Table 1), thus, the loss of cardioprotective effect caused by the drug-drug interaction is unexpected. The observation that not all NANSAIDs interact with Aspirin at the clinical level despite the fact that with the exception of meloxicam, rofecoxib and diclofenac, they interact with Aspirin at the platelet level (Table 1) highlights the heterogeneity of NANSAIDs (10) that is often ignored. For, example, Arfè et al. (44) who studied the risk of heart failures causes by NANSAIDs in 4 European countries noticed that only approximately one-half of the drugs used were significantly cardiotoxic. Nevertheless, they calculated the current use of any NANSAIDs, toxic or not, and concluded that the use of any NANSAID was associated with 19% increased heart failure risk.
The heterogeneity of NANSAIDs is confirmed in a crossover study (30) in which patients received 81 mg of immediate-release Aspirin followed 2 h later by ibuprofen, rofecoxib, or diclofenac for 6 days. This was followed by a washout period of 14 days, after which the same 2 medications were administered in reverse order for another 6 days. The inhibition of COX-1 was assessed by measuring serum TxB2 level, platelet aggregation induced in platelet-rich plasma and COX-2 activity by the measuring the formation of lipopolysaccharidestimulated PGE2 in whole blood. They noticed no significant interaction between Aspirin and rofecoxib or diclofenac. However, ibuprofen significantly interacted with Aspirin given before or after the NANSAID. The Aspirin-ibuprofen interaction has been confirmed by others (26-28, 31-36).
Although we have not made a comparison between molecular interactions studies and clinical trials for all NSAIDs, it is timely to reemphasize that their interaction with Aspirin is heterogeneous in nature. For example, naproxen, celecoxib, piroxicam, indomethacin, mefenamic acid, tiaprofenic acid, nimesulide, oxaprozin, flufenamic acid and dipyrone do interact, while loxoprofen, diclofenac, rofecoxib, etoricoxib, lumiracoxib, etodolac, ketorolac, meloxicam, acetaminophen, flurbiprofen, sulindac, and sodium salicylate do not (Table 8).
It has been suggested that the Aspirin-NANSAIDs interaction is due to a competition to bind to the Arginine-120 residue of the COX-1 channel which may prevent the acetylation of the serine-529 residue by Aspirin (37, 45). Nevertheless, the interference of NANSAIDs with the anti-platelet effect of Aspirin seems to have no long-term consequences as the CV protection of Aspirin remains unaffected by concomitant use of, at least, naproxen and ibuprofen. We put forward three plausible explanations for the disconnect between the results of the short-term platelet experiments and those of observational studies. (i) The interaction at the platelet level may be short-lived so that the effect dissipates shortly after its occurrence. (ii) There is no need for near complete inhibition of TxB2 inhibition to benefit from the cardioprotective properties of Aspirin so that despite a reduction in the extent of anti-platelet effect, the beneficial effect persists, or (iii) the CV effect of Aspirin may not be exclusively due to the drug's anti-platelet properties.
For all, except one eligible study, the CV risk was assessed after > 30 days exposure to the combination while typically, the effect of NANSAIDs on the anti-platelet activity of Aspirin is studied after short exposure times. Thus, the data on the therapeutic outcome of the short-term exposure to Aspirin-NANSAIDs are limited. However, the results published by Kimmel et al. (22) based on a case-control study that assessed the risk only one week before the date of onset of MI are useful in this context. They have reported that addition of NANSAIDs to Aspirin regiment does not increase the CV risk within one week post combination therapy. To this, one may add the fact that, to the best of our knowledge, there is no published report suggestive of a quick negative clinical CV outcome in individual patients who took NANSAIDs therapy while on Aspirin. Furthermore, data from a small size clinical trial, suggest that the effect of naproxen and diclofenac on the Aspirin-induced inhibition of platelet aggregation is short-lived (38). In a randomized placebo-controlled trial, Galliard-Grigioni et al. treated healthy subjects with 100 mg aspirin daily in combination with either three doses of either 1 g acetaminophen, 50 mg diclofenac, 250 mg naproxen or placebo, and assessed the platelet function. Initially, naproxen enhanced, and diclofenac reduced the anti-aggregatory action of Aspirin while acetaminophen had no effect. After 4 days of treatment, however, the platelet aggregation was equally inhibited by all Aspirin-NANSAID combinations.
In practice, a near complete inhibition of TxB2, thereby platelet aggregation is aimed to obtain cardioprotective effects of Aspirin (27). This is while the anti-platelet action of Aspirin is shown to be dose-dependent (46), i.e., low doses of the drug may not completely inhibit TxB2. Nevertheless, Aspirin has been shown to be cardioprotective after low doses (Table 9). This may suggest that to benefit from the CV properties of Aspirin, a complete inhibition of TxB2 is not needed. Thus, a reduced platelet aggregation activity of Aspirin resulted from combination therapies with NANSAIDs, unless proven through appropriately designed clinical trials, may have no significant clinical consequences.
In addition to its anti-platelet effect, Aspirin may reduce CV risks through other mechanisms. Both inflammation and some NANSAIDs appear to increase CV risks (10). Through animal studies, it has been shown that inflammatory conditions impair the balance of vasodilator/vasoconstrictor components of renin-angiotensin system (RAS) within the heart (47). The RAS is a major regulator of human physiology and has a key role in the CV homeostasis. Interestingly, NANSAIDs appear to be void of significant effects on RAS, instead, they are able to restore the imbalances that are resulted by inflammation (47). Alternatively, an altered protective/toxic balance of the cardioactive CYP450-mediated metabolites of arachidonic acid has been reported to be involved in the cardiotoxic effects of NANSAIDs (48). Whether Aspirin influences the RAS or the CYP450-mediated metabolites of arachidonic acid, remains unknown. Nevertheless, the possibility of CV protection by Aspirin through mechanisms other than its platelet effect is plausible.
The current analysis has limitations some of which are inherent to the nature of included studies. First, we have found that the published clinical evidence was sparse and has substantial limitations. To highlight this point, we were unable to assess the heterogeneity since some studies reported RR/OR while other did HR. Second, the primary outcomes of some studies that we included in our review were not CV (MI or stroke) risks as they reported the latter as secondary outcomes. Last, we were unable to perform meta-analysis as the same reference (Aspirin alone, NANSAID alone or nonusers) or outcome (OR, RR or HR) had not been used across the eligible studies.
CONCLUSION
Low-dose Aspirin is widely used to prevent MI and other CV diseases. However, there is evidence that concurrent use of some, but not all NANSAIDs, may inhibit the anti-platelet effect of Aspirin. Naproxen, meloxicam and rofecoxib do not appear to influence the cardioprotective effect of Aspirin. Similarly, a large body of evidence supports that ibuprofen coadministration with Aspirin does not antagonize the anti-platelet effect of Aspirin. Altogether, it appears that the NANSAID-Aspirin interaction at the level of platelets does not necessarily amount to a loss of beneficial effects of Aspirin. The limited available data suggest that the effect of the drug-drug interactions on the platelet aggregation may dissipate shortly. In addition, it is plausible that the reduced platelet aggregation resulted by the interaction may be sufficient for cardioprotection; i.e., no need for near complete aggregation. In addition, the cardioprotective effect of Aspirin despite reduced platelet aggregation caused by NANSAIDs may be through its involvement in other mechanisms such as the RAS and/or metabolism of arachidonic acid to biologically active compounds mediated by CYP450.
Conflict of interests:
The authors have no professional affiliation, financial interest or conflict with the subject matter or information discussed here in this manuscript to declare.
Source of Funding: King Saud University scholarship (Z. Alqahtani) and University of Alberta Self-Directed Grant (F. Jamali).
Authors' contribution: Database search, articles screening, articles review, data analysis, and manuscript preparation: Z. Alqahtani and F. Jamali. Study design and data review: F. Jamali. All authors read and approved the final manuscript.
ACKNOWLEDGMENT
We thank Janice Kung, a librarian at John W. Scott Health Sciences Library University of Alberta, for her valuable comments. The volunteers received Aspirin (100 mg, once daily) for 6 days. Then they received either single or multiple doses of the combination of Aspirin 2 h before naproxen (500 mg, twice daily) for another 6 days. After a washout period of 14 days, the treatments were administered in reverse order.
Serum TxB2, urinary 11 dehydro-TxB2 excretion rates, platelet aggregation by aggregometry, LPSstimulated PGE2 production in whole blood Naproxen interferes with the inhibitory effect of low-dose Aspirin on platelet aggregation. Aspirin and celecoxib (alone or together) or control (saline) were added to the PRP.
Platelet aggregation (induced by AA) by aggregometry
Celecoxib interferes to a limited extent with the anti-platelet effect of low-dose Aspirin.
60s Platelets were pre-incubated with ibuprofen, naproxen, or celecoxib for 10 min. then Aspirin was added to each group.
COX-1 acetylation, TxB2 formation
A single therapeutic dose of ibuprofen or naproxen followed by Aspirin casue a potent drug-drug interaction, but not between celecoxib and Aspirin. The volunteers were randomly assigned to either Aspirin (30 mg, once daily) for 7 days, slow release diclofenac (50 mg, three times daily) or ibuprofen (800 mg, three times daily) for 1 day. Aspirin (80 mg, once daily) was given after a washout period of 14-42 days with each treatment group for 7 days.
Serum TxB2 levels Only ibuprofen interferes with the anti-platelet activity of Aspirin. The volunteers received during 4 different study periods (≥10 days washout period) either acetaminophen (1 g, three times daily), diclofenac (50 mg, three times daily), naproxen (250 mg, three times daily) or placebo plus Aspirin (100 mg, once daily) for 4 days.
PFA-100 CT Regular daily co-administration of acetaminophen, diclofenac or naproxen do not interfere with the anti-platelet activity of Aspirin. PFA-100 CT Ibuprofen, indomethacin, naproxen, and tiaprofenic acid interfere with the anti-platelet activity of Aspirin but not sulindac or celecoxib.
Healthy volunteers (aged 21-32 years, n = 10) The volunteers were randomly assigned to receive either ibuprofen (400 mg), Aspirin (325 mg) or ibuprofen (400 mg) plus one dose of Aspirin (325 mg, 2 h later). A minimum of 6 days washout period was allowed between treatments.
Platelet aggregation by aggregometry
Administration of ibuprofen before Aspirin interferes with the inhibitory effect of Aspirin on platelet aggregation. The patients were undergoing long term treatment with Aspirin (100 mg, daily), and received celecoxib (200 mg, twice daily), ibuprofen (600 mg, three times daily) or placebo for 7 days.
Serum TxB2, urinary 11 dehydro-TxB2 excretion rates, platelet aggregation by aggregometry, LPSstimulated PGE2 production in whole blood Ibuprofen interferes with antiplatelet effect of Aspirin but not celecoxib. The volunteers were randomly assigned to receive either lumiracoxib (400 mg, once daily) or placebo for 11 days. Both treatment groups received Aspirin (75mg, once daily) from day 5 to 11 (6 days
NA
Risk of bias across studies 15 Specify any assessment of risk of bias that may affect the cumulative evidence (e.g., publication bias, selective reporting within studies).
4
Additional analyses 16 Describe methods of additional analyses (e.g., sensitivity or subgroup analyses, meta-regression), if done, indicating which were pre-specified.
RESULTS
Study selection 17 Give numbers of studies screened, assessed for eligibility, and included in the review, with reasons for exclusions at each stage, ideally with a flow diagram.
4-5
Study characteristics 18 For each study, present characteristics for which data were extracted (e.g., study size, PICOS, follow-up period) and provide the citations.
18-24
Risk of bias within studies 19 Present data on risk of bias of each study and, if available, any outcome level assessment (see item 12).
22-24
Results of individual studies 20 For all outcomes considered (benefits or harms), present, for each study: (a) simple summary data for each intervention group (b) effect estimates and confidence intervals, ideally with a forest plot.
5-9
Synthesis of results 21 Present results of each meta-analysis done, including confidence intervals and measures of consistency. NA Risk of bias across studies 22 Present results of any assessment of risk of bias across studies (see Item 15).
DISCUSSION
Summary of evidence 24 Summarize the main findings including the strength of evidence for each main outcome; consider their relevance to key groups (e.g., healthcare providers, users, and policy makers).
5-9
Limitations 25 Discuss limitations at study and outcome level (e.g., risk of bias), and at review-level (e.g., incomplete retrieval of identified research, reporting bias).
12
Conclusions 26 Provide a general interpretation of the results in the context of other evidence, and implications for future research. 12-13 FUNDING Funding 27 Describe sources of funding for the systematic review and other support (e.g., supply of data); role of funders for the systematic review.
|
2018-06-30T00:51:45.498Z
|
2018-05-02T00:00:00.000
|
{
"year": 2018,
"sha1": "94b2472faafc33914089eadf271bb91eddee7fea",
"oa_license": "CCBYSA",
"oa_url": "https://journals.library.ualberta.ca/jpps/index.php/JPPS/article/download/29854/21411",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "28e94ad1b9aff331b38b75f33dec706ebde601d4",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
17419467
|
pes2o/s2orc
|
v3-fos-license
|
DOI: 10.4054/DemRes.2011.25.21 Research Article
Arguments about the spread of gender egalitarian values through a population highlight several sources of change. First, structural arguments point to increases in the proportion of women with high education, jobs with good pay, commitment to careers outside the family, and direct interests in gender equality. Second, value-shift arguments contend that gender norms change with economic affluence among women and men in diverse positions—at all levels of education, for example. Third, diffusion arguments suggest that structural changes lead to adoption of new ideas and values supportive of gender equality by innovative, high-education groups, but that the new ideas later diffuse to other groups. This study tests these arguments by using International Social Survey Program surveys in 1988, 1994, and 2002 for 19 nations to examine gender egalitarianism across 85 cohorts born from roughly 1900 to 1984. Multilevel models support diffusion arguments by demonstrating that the effects of education first strengthen with early adoption of gender egalitarianism and then weaken as other groups come to accept the same views. However, the evidence of a sequence of divergence and convergence in educational differences across cohorts appears most clearly for women in Western nations.
Introduction
Under the broad term of the second demographic transition, demographers have described the wide-ranging changes that have occurred in living arrangements, gender roles, and childbearing (e.g., Lesthaeghe 2010).The changes encompass not only new behaviors involving sexual freedom, declining fertility, childbearing outside of marriage, and a greater variety of family forms (cohabitation, divorce, blended families, living alone) but also underlying value changes of individual autonomy, social equality, and tolerance of diversity (Lesthaeghe and Surkyn 1988).Central to the second demographic transition and to the broader liberalization of values are new roles for women and more favorable values, attitudes, and beliefs toward gender equality (or gender egalitarianism for short).The progress made over past decades toward the goal of widespread support for gender equality (Fischer and Hout 2006;Jackson 1998;Thornton and Young-DeMarco 2001) has affected most demographic processes, including fertility (Goldscheider, Oláh, and Puur 2010;Rindfuss, Brewster, and Kavee 1996), childlessness (Henz 2008), combining work and family responsibilities (Oláh and Bernhardt 2008), job segregation (Charles and Bradley 2002), and family relationships (Amato and Booth 1995;Kaufman 2000).
Demographers have also noted that a good part of the change in gender egalitarianism involves processes of cohort differentiation and replacement.The cohort approach to societal change follows a long tradition (Ryder 1965) in emphasizing the importance of the economic and ideational context at the time of a cohort's youth.Relatively stable values, attitudes, and beliefs that develop during youth and young adulthood endure over the later life course, and change comes from replacement of older cohorts raised decades ago with younger cohorts raised more recently.Numerous studies have demonstrated that differences in gender egalitarianism stem in large part from cohort membership (Brooks and Bolzendahl 2004;Firebaugh 1992;Inglehart and Norris 2003;Schnittker, Freese, and Powell 2003;Scott, Alwin, and Braun 1996).
In a recent article, Pampel (2011) evaluated several cohort-based explanations of increasing gender egalitarianism.Using data from the General Social Surveys of the United States and focusing on differences across cohorts born from 1900 to 1985, support was found for a diffusion theory that predicts changes in the socioeconomic distribution as well as the level of gender egalitarianism.More limited support was found for competing theories that emphasize the importance of either structural change in levels of female education and labor force participation or broad cultural changes in values affecting all education and labor force groups.The findings build on demographic approaches to social change that emphasize cohort influences, but also, in focusing on diffusion, offer a more precise description of how cohort change occurs.
The examination of cohort changes in one nation can offer only modest support for the diffusion arguments, however.European nations show diversity in attitudes toward gender equality (Alwin, Braun, and Scott 1992), with Norway, Sweden, and the Netherlands having stronger support for working mothers than nations of central, southern, and eastern Europe (Treas and Widmer 2000).The goal of this study is to extend tests of theories of cohort change in gender egalitarianism to include European nations and thereby offer a more complete evaluation of the theories as well as their international scope.The next sections describe the competing explanations of changing gender egalitarianism as presented by Pampel (2011) and then present new tests using comparative data for 19 European or European-heritage nations from the International Social Survey Program.
Structural influences
Changes in gender egalitarianism may occur through increases in the proportion of women with high education, good-paying jobs, and commitment to careers outside the family.As jobs shift from industrial economies to post-industrial service and knowledge economies and the demand of employers for female workers rises (Huber 1990), the labor force becomes more gender integrated.Tertiary educational opportunities for women increase, as does access to professional and managerial jobs once filled by men.Trends toward later marriage, fewer children, and more divorce reinforce education and job changes (Brooks and Bolzendahl 2004).At the societal level, then, the composition of the population changes in ways that foster gender egalitarianism.
An interest-based mechanism underlies arguments about the economic stake that nontraditional women have in gender equality (Bolzendahl and Myers 2004;Huber and Spitze 1981;Plutzer 1988).Because women with high education and special job skills gain the most from equal treatment and suffer the most from gender discrimination, they tend to have more egalitarian attitudes.To a lesser extent, men may also benefit economically from more egalitarian treatment and higher pay for working spouses, children, and relatives (Smith 1985;Zuo and Tang 2000).Indeed, studies have found that men have more egalitarian attitudes when they are part of a dual-earning couple (Cha and Thébaud 2009).In contrast, women with more traditional commitments to family and children will have fewer incentives to adopt new attitudes (Glass 1992;Kane and Sanchez 1994).They may even strengthen their adherence to traditional attitudes, which widens gender-based political cleavages (Plutzer 1988).Despite structural change, then, the continued dependence of some women on men maintains inegalitarian views (Baxter and Kane 1995).Also, men's interests may partly lie in gender inequality that reinforces their advantages in job opportunities and income.
Much of the change in gender egalitarianism related to structural position occurs through cohort replacement (Brewster and Padavic 2000;Brooks and Bolzendahl 2004;Firebaugh 1992).New educational, work, and job opportunities for women-and new values, attitudes, and beliefs-emerge among younger generations.The positions and roles of older cohorts change less, as do their more traditional attitudes.The replacement of older cohorts by younger cohorts who are more affected by recent structural changes in education and work leads to greater prevalence of gender egalitarianism.If those with socioeconomic characteristics, such as high education, that predispose them toward egalitarian views become a steadily larger part of the population, particularly among younger generations, then the level of gender egalitarianism will rise as well, even as inegalitarian values persist among older cohorts.
Value shifts
Arguments focusing on broad shifts in values suggest that gender norms develop, at least in part, independently of social structural position.Women and men in varied positions of a society-those with and without advanced education, for example-come to adopt more egalitarian attitudes with material prosperity.Broad changes in values do not erase attitudinal differences, but they raise egalitarianism similarly across diverse social positions and groups.Thus, studies have found a pervasive trend toward endorsement of gender equality in the United States since the 1960s (Rindfuss, Brewster, and Kavee 1996;Thorton and Young-DeMarco 2001).Support for gender equality grew among men as well as women (Bolzendahl and Myers 2004), and among active Protestants with conservative religious views as well as less religious and liberal groups (Petersen and Donnenwerth 1998).
The pervasive value shift raises questions about why groups in different positions and with different interests similarly become more egalitarian.Inglehart and Norris (2003) argued that economic prosperity and material security foster a broad cultural shift toward postmaterialist, quality-of-life values that emphasize equality, selfexpression, and individualism (also see Inglehart and Baker 2000).Jackson (1998) argued that the shift of economic and political power from households to business and government institutions in modern societies tends to erode family authority, communal obligations, and traditional beliefs and to weaken the incentives and means needed to maintain men's power over women.Given the pervasiveness of the change, diverse social groups tend to respond with stronger support for gender equality.
These arguments treat interests in broad terms (Inglehart 1989).Despite the existence of inequality, societal economic prosperity and material security tend to affect norms and values widely.Interests thus remain a key source of egalitarian values but relate to the larger social and economic context of value change rather than to the particular positions of individuals.However, these egalitarian values tend to change generationally.Adoption of new values commonly occurs during adolescence and young adulthood, and cohorts tend to retain these values throughout later adulthood.Inglehart (1989) argued that cohorts raised during the post-World War II decades of material prosperity and economic security widely adopted postmaterialist values.In support, Inglehart and Norris (2003:46) found that generation more strongly predicts egalitarian attitudes than does sex, class, or education.
The value-shift arguments thus differ fundamentally from structural arguments in accounting for rising gender egalitarianism.They posit that, at least among post-World War II cohorts, egalitarian attitudes increase across diverse sociodemographic groups.While women in nontraditional positions maintain more egalitarian views than others, changes in gender egalitarianism occur widely in postindustrial societies.
Patterns of diffusion
Diffusion arguments highlight a sequence of influences in which the early stages of change mostly involve the attitudes of educated and working women, those in nontraditional positions and with the strongest interests in gender equality.At later stages of change, egalitarianism diffuses vertically from high status, nontraditional innovators to lower status, less innovative, and more traditional groups (Poole and Zeigler 1981).Thus, effects of the determinants first strengthen as innovative groups with strong interests in gender equality adopt egalitarian views and set themselves apart from other groups.The effects of the determinants then weaken as gender egalitarian views diffuse to larger parts of the population (Fischer and Hout 2006).
In support, several studies describe a process of catching up in which formerly wide attitude differences narrow with the spread of new values to less innovative groups.Mason and Lu (1988) found growth in gender egalitarianism among most sociodemographic groups in the United States from 1977 to 1985, except for collegeeducated women, who already had high gender egalitarianism.Bolzendahl and Myers (2004) found that as most people come to accept gender egalitarian goals, individual determinants of attitudes toward women's participation in the public sphere have declining influence.Fischer and Hout (2006) found widening and narrowing by age and city residence in measures of approval for working women from 1936 to 2000.These changes in relationships are consistent with general diffusion arguments.Montgomery and Casterline (1993) defined diffusion as the influence of adoption of innovative ideas and behaviors by some individuals on the likelihood of adoption by others.Diffusion often first occurs horizontally among higher socioeconomic groups because these groups tend to be the most innovative and have communication networks across structurally equivalent positions (Strang and Meyer 1993).Advanced education, for example, fosters gender egalitarianism at the early stage of the epidemic by promoting tolerance (Weil 1985), openness to new ideas (Rogers 2003), and cognitive skills needed to better evaluate new ideas (Ohlander, Batalova, and Treas 2005).Vertical diffusion often follows, as lower ranking groups adopt the practices and ideas of more prestigious groups (Fischer 1978;Strang and Soule 1998).Interests or the relative advantage of adoption play a role in adoption of innovations, at least initially (Rogers 2003:229).Later, acceptance of new ideas may become self-sustaining after adopters reach a critical mass (Rogers 2003:343).At that point, adoption by less innovative groups requires less risk and boldness.
The arguments predict that the early adoption of gender egalitarianism by more advantaged groups initially involves innovation that strengthens socioeconomic differences, but the vertical diffusion of the values to other groups later reduces socioeconomic differences.However, women who benefit the most from gender equality will more quickly adopt new attitudes than men (Ciabattari 2001).The diffusion of gender egalitarianism may occur for men but not as quickly or to the same extent as for women.
Hypotheses
The hypotheses focus neither on overall changes in support for gender egalitarianism nor on group differences in attitudes-topics that have been well studied.Rather, they focus on the combination of the two topics: on whether groups change their attitudes at different rates and exhibit divergence in views followed by convergence.Further, the hypotheses focus on cohort differences in gender egalitarianism.As noted earlier, cohort replacement is a crucial component of change in gender egalitarianism; the precise pattern of cohort change is less well understood, and each hypothesis posits a different mechanism of cohort differentiation.
In differentiating between the theoretical arguments, the hypotheses focus largely on education.Although other sociodemographic characteristics, such as employment, income, and family status, strongly influence gender egalitarianism, education has special value in making comparisons across cohorts.Because education is determined early in life, it relates directly to the past experiences of cohorts.Education measured at age 75 for older cohorts likely differs little from education if it were measured 50 years earlier at age 25.In contrast, current employment, income, and family status of a cohort will differ enormously from their levels decades ago.Given that older cohorts lack data on other indicators for earlier ages and periods, the stability of education over the later life course makes it most suited for specifying and testing the hypotheses.Of course, education also has value because it relates closely to other components of socioeconomic background and attainment.With this in mind, the competing hypotheses are as follows: H1: Changes in the educational composition of cohorts account for cohort increases in gender egalitarianism.Consistent with structural arguments, the hypothesis implies that differences in gender egalitarianism across groups with varied levels of education (and indirectly with varied work and income opportunities that follow from education) are maintained across cohorts, but that low-education groups decline in size relative to groups with high education.
H2: Changes across cohorts in gender egalitarianism occur similarly for educational groups.Consistent with value-shift arguments, the hypothesis implies that differences in views across innovative groups with high education and more traditional groups with low education persist but that gender egalitarianism rises across cohorts for all educational groups.
H3: Changes across cohorts first affect innovative groups with high education, thus strengthening the effects of education on gender egalitarianism and creating divergence in views, but the changes later affect less-educated groups, thus weakening the differentiating effects of education and creating convergence in views.Consistent with diffusion arguments, the hypothesis also implies that the changes occur more quickly and strongly for women than men.
Previous studies have analyzed consecutive cross-sectional surveys in the United States (Bolzendahl and Myers 2004;Brewster and Padavic 2000;Carter, Corra, and Carter 2009;Mason and Lu 1988;Rindfuss, Brewster, and Kavee 1996;Spitze and Huber 1980), and several examine the influence of cohort on attitudes (Brooks and Bolzendahl 2004;Schnittker, Freese, and Powell 2003;Wilkie 1993).Several others have examined attitudinal differences across nations (Scott, Alwin, and Braun 1996;Treas and Widmer 2000).Besides the work of Pampel (2011), however, no study has fully tested the predictions about nonlinear strengthening and weakening across cohorts of the effects of education.
Data
The International Social Survey Program (ISSP 2011), a collaboration of nations conducting annual surveys for probability samples of their populations, has completed three sets of surveys on gender roles.Eight nations participated in the 1988 surveys; 24 in the 1994 surveys; and 35 in the 2002 surveys.To examine cohort effects, however, only nations with data for at least two of the years are used. 2Counting separate surveys for East and West Germany and for Great Britain and Northern Ireland, that leaves 22 nations or national regions.However, all but three nations are European in location or heritage.The exceptions-Israel, Japan, and the Philippines-differ enough from the others in geographic location, history, and culture to exclude from the analysis.Of the remaining 19 nations, seven have surveys for all three years.Because of the longer time span and greater ability to isolate cohort effects, surveys for these seven nations-West Germany, Great Britain, the United States, Austria, Hungary, Ireland, and the Netherlands-are central to the analysis.Of the other 12 nations with at least two time points, six are former communist nations of eastern Europe (East Germany, Czech Republic, Slovenia, Poland, Bulgaria, and Russia), and six are Western nations (Australia, Northern Ireland, Italy, Norway, Sweden, and New Zealand).
Measures
All three sets of ISSP surveys ask about agreement with statements concerning gender roles and equality.Despite diverse nations in the sample and the diverse domains of gender equality covered by the questions, the 12 items appear to reflect an overall dimension of support for gender equality and form a reliable scale (Cronbach's alpha of .80).The items in abbreviated form are as follows: (1) A working mother can establish just as warm and secure a relationship with her children as a mother who does not work; (2) A preschool child is likely to suffer if his or her mother works; (3) Family life suffers when the woman has a full-time job; (4) A job is all right, but what most women really want is a home and children; (5) Being a housewife is just as fulfilling as working for pay; (6) Having a job is the best way for a woman to be an independent person; (7) Both the husband and wife should contribute to the household income; (8) A husband's job is to earn money; a wife's job is to look after the home and family; (9) A woman should work after marrying and before there are children; (10) A woman should work when there is a child under school age; (11) A woman should work after the youngest child starts school; and (12) A woman should work after the children leave home.With all items coded so that high scores indicate support for gender equality and standardized to have a mean of zero and a standard deviation of one, a standardized scale is created from summing the items.
The lack of nuance in these items means that they miss important components of views about gender equality (Brewster and Padavic 2000;Mason and Lu 1988), and the items relate to work/family issues for women rather than to politics, earnings, discrimination, task sharing at home, or men's roles.Goldscheider, Oláh, and Puur (2010) argue that attitudes about public equality relating to work and politics change at different rates than components relating to private or family equality.Also, Kane and Kyyrö (2001) make a distinction between attitudes that, on one hand, promote ideological refinement but maintain individualism and largely reproduce inequality and those that, on the other hand, espouse group-based remedies to fundamentally restructure relations of inequality between men and women's roles.The items studied here relate to private more than public gender equality and more to individual than group-based remedies for gender inequality.Still, items such as these are commonly used in national and cross-national surveys and meaningfully summarize variation in attitudes toward certain forms of gender equality across as well as within nations.The reliability of the scale used here plus the consistent relationships observed in previous studies between gender equality items and socioeconomic characteristics and period trends suggest the value of the items.
Cohort measures single years of birth and ranges from 1900 (or earlier) to 1984 (or later).Year has three values : 1988, 1994, and 2002.To avoid exact dependency with cohort and year, age is grouped into seven categories: 18-25, 26-35, 36-45, 46-55, 56-65, 66-75, and 76+.Given the focus on cohort changes and characteristics determined early in life and largely stable thereafter, the key determinants are gender, mother's work, and especially, education.For gender, females are coded 1.A mother worked measure equals 1 for respondents whose mother ever worked for pay for at least one year after the birth of a child and before the child turns 14.
The education measure is based on years of schooling completed, which is recoded to have a minimum of 8 and a maximum of 20.Years completed is more comparable across nations than measures based on the highest degree obtained.However, several nations ask only about highest degree, and in nations that do ask about years of schooling, some respondents report only their highest degree.Given that the two measures have a high correlation of .62, the degree measure is used to impute values when data for the years completed measure are missing.However, some nations in the early years truncate the measure.(For example, in 1988 Great Britain reported a minimum of 10 and maximum of 14.) To adjust for somewhat different scales, education also can be standardized within nations and years.That is, each individual's education is centered on the mean of the nation and year and divided by the standard deviation of the nation and year so that all nations and years have the same mean of 0 and standard deviation of 1. Standardized education can account for differences in gender egalitarianism within nations, but not between nations.
The remaining variables reflect current rather than past characteristics.Although less relevant to early cohort experiences, these variables affect gender egalitarianism.A set of five dummy variables measures employment status: (1) full-time worker or student; (2) part-time worker; (3) unemployed; (4) homemaker; and (5) retired.Dummy variables measure marital status by spouse employment status: unmarried and noncohabiting persons serve as the omitted group; married or cohabiting persons whose spouse is not employed define one dummy variable; and married or cohabiting persons whose spouse is employed define a second dummy variable.A measure of family income differs in wording across nations, with some nations asking about net income and others asking about gross income.With the national currency differing as well, it is necessary to create a standardized family income score that has a mean of 0 and standard deviation of 1 for each nation and year.This measure can affect differences within but not between nations.Two other measures, occupation and urban residence, are measured too differently across nations and years to use in the models.
Models
A cross-classified random-effects model for the analysis of age, period, and cohort effects (Yang and Land 2006) is well suited to testing hypotheses about cohort changes in the effects of education.The model treats cohort and cohort squared plus control variables as determinants of the outcome measures of gender egalitarianism (GE).The individual data are nested within cells created by the cross-classification of period and age group.The use of cohort quadratic terms with five-year age groups and the two or three survey years eliminates the dependency of cohort on age and period.Following Yang and Land (2006), the Level 1, or within-cell, model takes the following form: where i refers to individuals within j age groups and k years, X to m control variables (including dummy variables for nation), and e is a normally distributed error with a mean of 0 and variance of .The product terms allow for nonlinear changes in the effect of education across cohorts.With the intercept assumed to vary randomly, the Level 2, or between-cell, model takes the following form: where γ 0 is the model intercept or adjusted mean outcome; u 0j is the residual random effect of age group j on β 0jk averaged over all periods, which is assumed to be normally distributed with mean 0 and variance τ u ; and v 0k is the residual random effect of period k on β 0jk averaged over all ages, which is assumed to be normally distributed with mean 0 and variance τ v .The model thus allows for estimation of cohort effects on GE with random-effects controls for age group and year. 3The slope coefficients β 1 through β m are treated as fixed.With all variables centered, the crossed random-effects estimates come from xtmixed in Stata 11.0.4 The three hypotheses translate into predictions about the multilevel coefficients.The structural position hypothesis (H1) predicts that because changing population composition accounts for increases in GE, cohort will have little influence net of the individual determinants (i.e., β 1 = β 2 = β 4 = β 5 = 0).The value-change hypothesis (H2) predicts that because of the widespread increase in GE among all groups, cohort will nonlinearly increase the intercept or level of GE (i.e., β 1 > 0 and β 2 < 0, but β 4 = β 5 = 0).The diffusion hypothesis (H3) predicts that the slopes of education, not just the GE level, will change across cohorts as favorable views first emerge among innovative groups with direct interests in equality and then diffuse to other groups.This change in relationships across cohorts implies that the key determinants of GE interact with cohort and cohort squared (i.e., β 4 > 0 and β 5 < 0).The interactions, however, should be stronger for females than for males.1988, 19 in 1994, and 18 in 2002; pooling these data gives 45 nation-years.Most of the variance in the GE scale exists within rather than between nations and years, however.Some additional runs allow for calculation of intraclass correlations.For the group of seven nations with three time points, the between-nation and year variance in the GE scale equals 7.1% of the total; for all nations combined, including those with two time points, the between-nation and year variance in the GE scale equals 11.7% of the total.
Determinants
Table 2 shows effects of the determinants of the GE scale with random-effects controls for year and the age categories, and with fixed-effects or dummy variable controls for nation.Consider first the top panel of results for women.The initial models include only the seven nations with three time points.In the first model, GE increases with cohort, but the increase tends to level off for the cohorts born most recently.These coefficients indicate increasing GE when controlling for compositional changes, and thus favor the value-change argument over the structural-position argument.In addition, the random effects for year and the age categories show higher GE in later years and for middle-age persons.The other variables have expected relationships that match the findings of previous literature: Women with working mothers, no marital or cohabiting partner (the omitted category), high education, high family income, and full-time work (the omitted employment category) have higher GE values.More important for testing the diffusion hypothesis, the second model includes the interaction of education with cohort and cohort squared (again for the same seven nations).Differences in the effects of education, as represented by the interaction terms, fit the curvilinear pattern predicted by the diffusion hypothesis.The significant positive cohort interaction term and the significant negative cohort squared interaction term indicate that the effects of education initially increase across cohorts but eventually peak and decline.Calculations from the interaction coefficients indicate that the peak education effect occurs for the 1944 cohort.
The next two models in Table 2 use the sample of 19 nations with at least two time points.The additive determinants in the third model for the larger sample of nations are similar to those for the smaller sample.In the last model, the education by cohort interactions are again significant and in the predicted direction, although they are somewhat smaller than in the model for the smaller sample of nations.The education by cohort coefficients of 0.026 (t = 3.78) and -.003 (t = -3.67)are about half the size of the coefficients of 0.044 (t = 4.97) and -0.005 (t = -5.09)for the smaller group of nations.Both groups of nations show effects of education that rise and reverse, but the effects are weaker for the larger sample.
The bottom panels of Table 2 present the same models for males.The coefficients for males differ from those for females in several ways.For men, having a working spouse increases gender egalitarianism, while having a non-working spouse lowers egalitarianism.Also, own employment status has only weak influences for men.Of key importance, the education by cohort interactions have the predicted signs for men, but the interactions are smaller than for women in the more homogenous set of seven nations and fail to reach statistical significance.Calculations from the coefficients indicate that, whereas the reversal occurs for female cohorts born in 1944, it occurs for male cohorts born in 1985.In these nations, new gender views appear to have spread among men later than among women.In contrast, the interaction coefficients for the larger group of nations differ little between men and women.
To present the interaction effects visually, Figure 1 graphs the female and male slopes of education for each cohort as implied by the interaction models and does so separately for the two sets of nations.The four curves depict rising slopes for education that reverse, but the curve for females in the smaller sample has the highest peak and largest drop, and indicates both the greatest divisions and most convergence.For the youngest cohorts, the effects of education for women in the smaller, three-year sample of nations drops closest to zero.
Figure 1: Predicted education slope for males and females by cohort and nation group
The models for the larger sample of nations, although supporting the diffusion hypothesis, mix Western nations and eastern European nations with diverse economic, social, and political backgrounds.Table 3 controls somewhat for this diversity by estimating models separately for Western nations and eastern European nations.The regions differ in a variety of dimensions-national income, communist history, gender discrimination, cultural history-that cannot be separated with the small number of nations.Still, differences in the models across these two regions help to identify where diffusion of GE has occurred most clearly.As shown in Table 3, the two regions differ in the pattern of the education by cohort interactions.The models for the Western nations are similar to the models over all in Table 2.In contrast, the interaction terms for eastern European nations in Table 3 fail to reach statistical significance.Figure 2 graphs the education by cohort slopes predicted by the models for males and females in both regions.The reversal in education effects shows clearly for the Western nations.Likely reflecting randomness, the curves for eastern European nations fail to show the rise and decline in the effects of education that the Western nations do.These results suggest little evidence of GE diffusion across levels of education in eastern Europe.
Figure 2:
Predicted education slope for males and females by cohort and region
Sensitivity Checks
Checks of the sensitivity of the results to influential nations strengthen the findings for women but suggest less reliability in the results for men.The models in Table 4 focus on the education by cohort interactions for the smaller groups of nations with data on three time points.The largely Western nations in this group are those where diffusion has occurred most clearly.For all seven of these nations, the female education by cohort coefficient of 0.044 (t = 4.97) and education by cohort squared coefficient of -0.005 (t = -5.09)match those presented earlier.Based on jackknife procedures in which models are estimated with one of the nations deleted, the table lists seven more sets of interaction coefficients, first for females and then for males.For females, deleting West Germany and Hungary reduces the coefficients, while deleting Great Britain and the United States increases the coefficients.Yet, the interaction coefficients remain significant and similar to those for the full set of nations.The checks demonstrate the robustness of the interactions.For men, the coefficients in Table 4 show that individual nations affect the results in non-trivial ways.Among all seven nations, the coefficients of 0.017 (t = 1.90) for education by cohort and -0.001 (t = -1.52)for education by cohort squared show effects that are smaller than for women and provide limited support of the diffusion hypothesis.However, deleting the United States or Austria raises both interaction coefficients to statistical significance.This suggests that the diffusion pattern of change in the effects of education may hold across most of the nations but is weakened by the special experiences in one or two nations.The limited diffusion of GE among males in the United States and Austria does not appear typical of the other nations.Thus, these checks demonstrate less reliability of results for males.To a large extent, the pattern of diffusion is weaker and less consistent for males than for females.
Checks on alternative measures of education (not presented in the tables) similarly reveal greater robustness for the female than for the male results.Other measures include 1) educational degree rather than years of education completed, 2) educational degree standardized within nations, and 3) years of education completed standardized within nations.For women, the educational degree and standardized measures continue to show significant cohort interactions, at least among Western nations.For men, the other measures tend to produce weaker results.For example, within-nation standardization of the measure of education years reduces the interactions to below significance for the sample of nations with three time points.This again suggests less robustness in the interactions for men.
Conclusion
Based on the analysis of a diverse set of European nations plus the United States, Australia, and New Zealand, the adoption of gender egalitarian views occurred steadily across cohorts over much of the twentieth century.Less obviously, however, the results also show that adoption reflects nonlinear changes in the influence of education and the strength of socioeconomic cleavages.The effects of education become stronger across older cohorts as attitudes toward gender equality shift from largely unfavorable to favorable among women with greater education (and likely stronger commitment to work and career).Among the most recent cohorts, however, the effects of education become weaker as favorable attitudes spread widely through the population and socioeconomic differences decline in importance.These results replicate the results reported by Pampel (2011) and Fischer and Hout (2006) for the United States, but with a wider set of nations.
Theories of both structural position and value change, though incomplete on their own, play a role in explaining the pattern of results.Structural changes that increase education and related work and career opportunities for women provide the impetus to adopt egalitarianism by strengthening direct interests in equality among certain segments of the population.Value changes among other groups of women with less education (but still with interests in gender equality) occur later through processes of diffusion.According to the diffusion theory, both structural and cultural mechanisms contributed to the rising support for gender equality over the last century but did so in sequence.
The evidence is strong and robust for women in Western nations but is less supportive for men and for eastern European nations.Men not only hold less egalitarian views than women but also show less responsiveness to social change.In Western nations, the effects of education on gender egalitarianism increase and decrease for men as they do for women but appear highly influenced by a few nations and by the particular measure of education.Gender egalitarianism among recent cohorts thus depends more on level of education among men than among women.This difference in patterns of change suggests that diffusion has moved faster and farther among women than among men and qualifies theoretical claims about diffusion.The diffusion process best fits women more than men likely because of their stronger interests in and benefits from equality.Since the advantages of an innovation increase the speed at which it is adopted (Rogers 2003), the gains to women relative to men lead to quicker diffusion of gender egalitarianism among women relative to men.
Neither men nor women show a clear diffusion pattern in eastern European nations.Numerous structural, political, cultural, and historical characteristics may contribute for the differences in patterns and trends in gender egalitarianism.More complete attention to the special circumstances and forces affecting gender equality in the former-Communist nations of Eastern Europe is needed to explore the sources of the differences.The theories evaluated here focus on structural change in education and work opportunities for women and on values related to economic security in affluent nations.However, consideration of other factors not highlighted by these theories are needed to better understand the experiences of the Eastern European nations.This study describes the scope of the diffusion findings but is limited in its ability to explain the cross-national differences.
The empirical approach used here to specify and test hypotheses about the sources of change in gender egalitarianism has several advantages.Examining differences across cohorts rather than years, modeling the varying effects of education on gender egalitarianism rather than just the level of gender egalitarianism, and testing nonlinear predictions of an integrative diffusion theory all help extend the literature in new directions.Making comparisons across cohorts builds on other studies (Brewster and Padovic 2000;Brooks and Bolzendahl 2004;Ciabattari 2001;Schnittker, Freese, and Powell 2003;Wilkie 1993) but also more fully exploits the potential for change to occur across groups that are born and socialized in different historical periods.Modeling interactions between cohort and education captures both the diffusion of gender egalitarianism and the importance of cohort sources of social change.
The data and methods used to evaluate the approach, however, present some limitations.First, the data allow for only the indirect study of diffusion.The patterns of change in the education effects are consistent with initial adoption of gender egalitarianism by innovative groups and the later diffusion to other groups.Indeed, the predictions of nonlinear interactions of cohort and education are highly falsifiable.Yet, other types of data and forms of analysis are needed to more directly validate claims about diffusion.For example, better measures of the historical characteristics of cohorts over the past century can extend the approach.Gathering cohort-based measures of the economic, political, social, and cultural environment-particularly as they relate to education, work, and earnings opportunities of women-from historical sources presents a daunting task.Even so, future research can use such data to provide additional tests of predictions that the context of gender equality relates nonlinearly to the effects of the determinants of gender egalitarianism.
Second, the inability to measure SES and family characteristics of older cohorts at the time of young adulthood rather than at older ages compromises the tests.Current education, the key determinant in the analysis, reflects past education better than current employment, job characteristics, income, and family status reflect past characteristics.Measures of other socioeconomic characteristics would improve the models, but the lack of long-term longitudinal data for multiple cohorts during the formative years of youth and young adulthood warrants the reliance on education.
Third, the gender egalitarian items used here are only a subset of possible measures, and one cannot assume that they give the same results as other sets of measures would.The items on work/family issues faced by women give little attention to changes in men's roles and the need for restructuring gender roles.Studies need to do more to organize the diverse measures used in the literature into multiple dimensions of gender egalitarianism and compare results across the different dimensions.
Some of these limitations can be overcome with the release of the 2012 ISSP data, which offers a fourth module on family, work, and gender roles.The new 2012 surveys will extend the period of study for nations that participated in previous surveys and likely add some new nations to the sample.Going beyond the basic results reported here, the new surveys will allow for better measures of diffusion and a more detailed comparison of the nations of Western and Eastern Europe.
Table of Contents
Table 1 lists the means by nation and year for the standardized GE scale.The means are higher in 2002 than in earlier years, and all nations show increasing GE over time.For example, the mean for West Germany rises from -.43 in 1988 to -.15 in 1994 and to .26 in 2002, while the mean for the United States rises by less, from 0.04 in 1988 to 0.18 in 1994 and to 0.26 in 2002.For comparisons across nations in 2002, the most recent year, East Germany, Norway, and Sweden have the highest mean values, while Australia, Hungary, Poland, Bulgaria, Russia, and New Zealand have the lowest means values.Although there are exceptions, the former-Communist nations of eastern Europe generally score lower than the Western nations.As shown in Table 1, there are 8 nations in
Table 2 : Unstandardized coefficients and t values for multilevel models of gender egalitarianism with year and age group random effects
a Dummy variable controls for nation not listed.b Mother did not work omitted.c Standardized within country.d Full-time work omitted.e Unmarried omitted.* p < .05** p < .01*** p < .001
Table 3 : Unstandardized coefficients and t values for multilevel models of gender egalitarianism with year and age group random effects, by region
a Dummy variable controls for nation not listed.b Mother did not work omitted.c Standardized within country.d Full-time work omitted.e
|
2014-10-01T00:00:00.000Z
|
2011-09-30T00:00:00.000
|
{
"year": 2011,
"sha1": "30563329971bc72a3d53bc4f4704682458b8f47f",
"oa_license": "CCBYNC",
"oa_url": "https://www.demographic-research.org/volumes/vol25/21/25-21.pdf",
"oa_status": "GOLD",
"pdf_src": "CiteSeerX",
"pdf_hash": "30563329971bc72a3d53bc4f4704682458b8f47f",
"s2fieldsofstudy": [
"Sociology"
],
"extfieldsofstudy": [
"Sociology"
]
}
|
110880662
|
pes2o/s2orc
|
v3-fos-license
|
Piezoelectric d15 shear response-based torsion actuation mechanism: an experimental benchmark and its 3D finite element simulation
An experimental benchmark is proposed for piezoelectric, direct-torsion actuation using mono-morph piezoceramic d15 shear patches. This is reached by designing and assembling an adaptive plate having two identical composite faces sandwiching a core made of connected six oppositely polarized (OP) piezoceramic d15 shear patches along the length. An electronic speckle pattern interferometry system was used to measure the static tip deflection of the adaptive sandwich composite plate that was mounted in a cantilever configuration and actuated in torsion by progressively applied voltages on the piezoceramic shear core electroded major surfaces. Then, the effective rate of twist was post-processed and proposed as an evaluation criterion for smart composites under piezoelectric torsion actuation. For verification of the experimental results, the proposed experimental benchmark was simulated using three-dimensional piezoelectric finite elements (FE) within ABAQUS® commercial software. The comparison of the obtained experimental and simulation results showed reasonable agreement, but the slight nonlinear experimental response was not confirmed by the linear FE analysis. The experimentally proved torsion actuation mechanism, produced by OP piezoceramic d15 shear patches, can be applied actively to prevent torsion in many applications, such as in wind turbines, helicopter blades, robot arms, flexible space structures, etc.
Introduction
Preventing or controlling twist (or torsion) deformation is a necessity for many applications, such as in wind turbines, helicopter blades, robot arms, flexible space structures, etc.; this can be reached indirectly via bending-twisting [1,2] or extension-twisting [3] stiffness couplings or directly through the electromechanical coupling between throughthe-thickness applied electric field and twist strains via the coupling constant d 36 [4]. However, the latter is nil for mono-morph piezoelectric materials rendering the direct twist deformation sensing or actuation non-feasible; nevertheless, this coupling can be obtained indirectly using either skewed piezoelectric polymers [5][6][7], thanks to their orthotropic piezoelectric coupling properties, or piezoceramic fiber composite actuators or sensors that were made orthotropic thanks to a suitable fiber/epoxy blend [1].
In order to overcome these difficulties, researches on the torsion actuation using the direct d 15 shear response of mono-morph piezoceramic materials have attracted noticeable interest since the mid-1990s [8]. This torsion actuation mechanism (TAM) was achieved experimentally using tubular actuators assembled from eight segments that were cut from circumferentially-polarized piezoceramic tubes [9][10][11], and theoretically using symmetrically surface-bonded width-polarized patches [12,13]. A simulation benchmark (theoretical) was also recently proposed consisting of a cantilever long and thin beam consisting of two piezoceramic layers oppositely-polarized (OP) along their length and superposed along their width [14]. This concept is proved experimentally here for a cantilever plate consisting of two identical glass fiber/epoxy composite faces sandwiching a core torsion actuator constructed by the connection of six OP piezoceramic d 15 shear patches along their length and superposed along their width.
In this study, the TAM concept was first considered theoretically; then, its characteristics were studied experimentally by an electronic speckle pattern interferometry (ESPI) system with out-of-plane displacement measurement sensitivity [15] for a designed and manufactured cantilever sandwich plate test setup, made of a piezoceramic d 15 shear-based torsion core actuator and two identical glass fiber/epoxy composite faces. The piezoelectric adaptive sandwich plate experiences static torsion due to the progressively applied electric potentials on the electroded major surfaces of the piezoceramic d 15 shear core. The latter's torsion actuation capability was evaluated through the plate measured static maximum tip transverse (along thickness) deflection and post-treated maximum rate of twist and maximum effective rate of twist. The latter parameter is defined as the rate of twist per unit voltage. This new torsion actuation benchmark was also simulated within ABAQUS ® commercial code using three-dimensional (3D) elastic and piezoelectric finite elements (FE) for the composite faces and piezoceramic core, respectively. The above TAM evaluation parameters were extracted and compared to those obtained from the experimental tests. This numerical analysis is expected to help for the evaluation of the cross-section warping (axial displacement) and its importance for developing representative theoretical models for this new TAM.
The concept of piezoelectric d 15 shear-induced torsion actuation mechanism
Consider a cantilever single-layer plate assembled from six piezoceramic patches OP along their length, as shown in Figure 1. The plate is subjected to only an electric voltage (V ) applied on its electroded major surfaces so that the through-the-thickness applied electric field is perpendicular to the polarizations; in this case, it is well known [8] that only transverse shear actuation strains can be induced.
Since the polarization is conventionally along the material 3-axis, the induced actuation strains by the bottom (b) row of piezoceramic patches of Figure 1, that are polarized along the plate positive x-axis, are Hence, as stated above, only a transverse shear actuation, γ b zx = d 15 E z , is produced by the bottom row of piezoceramic patches. Due to the OP direction, the top (t) row of piezoceramic patches should produce γ t zx = −d 15 E z ; this expression is obtained by adding a negative sign to the piezoelectric constants of Equation (1). So, if the applied electric field is constant and positive, the bottom layer of piezoceramics should deform the plate positively along the z-direction, whereas the top layer should deform it negatively to this direction; the combination of these deformations should produce a global torsion of the cantilever plate because the patches are identical (have the same shear modulus) and the transverse shear stress σ xz should be continuous at the bottom/top layers interface. The latter is the location of the neutral and torsion axes due to the construction symmetry; hence, it should not move transversely.
Shear-induced torsion actuation experimental benchmark
To validate the above torsion actuation mechanism, the cantilever piezoceramic plate of Figure 1 was integrated in an adaptive plate construction as a core layer sandwiched between two identical composite faces, as shown in Figure 2. This benchmark was designed and assembled using PIC255 piezoceramic shear patches (from PI Germany), of dimensions 25 × 25 × 0.5 mm 3 , for the core, and Polyspeed G-EV 760R glass fiber/epoxy layers (from Hexcel Austria), of dimensions 75 × 50 × 0.49 mm 3 , for the composite faces; a methacrylate-based two-component adhesive (of 0.1 mm thickness) was used for bonding the faces to the core.
The bonding thickness has been controlled during the gluing process by using other glass fiber and piezoceramic patches adjacent to the glass fiber and piezoceramic patches being glued and they are raised to the thicknesses of the adhesive layers by other special glue bands of the same thicknesses of the adhesive layers. The piezoceramic shear patches are bonded to each other by using very thin adhesive layers. These special glue bands used during the process are 0.1 mm thick; this bond thickness has been selected because its value is controllable. The glass fiber epoxy has been particularly selected in this composite structure because of its non-conductive properties. The role of the composite faces was to make the assembled patches work together as a homogenous core, and their primary effect was to stiffen the whole actuator. The thicknesses of the glass fiber/epoxy composite faces and adhesive layers play a crucial role in the torsion actuation performance of the composite structure since they affect the torsion stiffness and torsion moment created by the piezoceramic patches; indeed, these are functions of the thicknesses of all layers. Besides, if the structure is longer, more transverse deflection can be obtained for the same whole thickness; whereas if the composite layers were stiffer, less rate of twist would be obtained. The proposed design is then a compromise between all material and geometric parameters effects on the torsion actuation.
A series of experiments was then carried out on the experimental benchmark by applying increasing voltages (from 45.3 to 198 V) to the piezoceramic torsion core in order to measure the maximum rate of twist of the adaptive structure. For this purpose, the maximum static tip deflection of the cross-section at the free side of the composite plate was measured by an ESPI system (Dantec-Ettemeyer ESPI Q300). Figure 3 shows a photo of the experimental benchmark (right) and ESPI system (left).
The other pieces of equipment in the experimental setup were a high-voltage (HV) amplifier (Elba Tech type T-502), laboratory power supply (EA type EA-3016) and ISTRA data processing software for the control and evaluation of the ESPI system ( Figure 4). The latter provides more sensitive and full-range measurement possibilities than strain gages and has the advantage of being able to measure deformations and deflections of engineering structures and materials without contact. It does not change, the response of the objects being studied, because of its non-contact nature. Therefore, results obtained by an ESPI Figure 3. The ESPI camera and the benchmark experimental setup. system represent reality and can be used to validate analytical or/and numerical solutions. The visualization occurs in the form of fringes on the image. From the speckle patterns, the out-of-plane and the in-plane deformations can be determined. Digital image equipment processes the information included in the speckle patterns and displays the consequent interferogram on a computer monitor. The ESPI system can be used not only for one-step measurement but also for a series of measurements. The main difference of the out-of-plane displacement measurement by the ESPI system with respect to an in-plane displacement measurement is that in out-of-plane displacement measurement, a single illuminating beam is used. The coordinate system of Figure 2 is in agreement with that of the ESPI system (Figures 3-4). Figure 5 shows the transverse deflection fringes (Figure 5a) and its through-height (y-direction) variation at x = 1.14 mm (Figure 5b) of the cantilever adaptive sandwich composite plate when torsion was actuated by a 198 V voltage. It can be seen that the tip transverse deflection is anti-symmetric with regard to the plate length (x-direction), as in Figure 5a, and that it is linear through the cross-section (in the y-z plane) height as can be seen from Figure 5b.
Due to this asymmetry, the measured and FE (with asymmetric response) transverse deflection was calculated as half of the difference between the maximum and minimum values (for example, as in Figure 5b for the measurement at x = 1.14 mm); in addition, since the length and width ranges of the torsion actuator are not totally covered by the ESPI measurement system (see the x and y ranges in Figure 5a), the transverse deflection was measured twice near the free (u 1 z at x 1 = 1.14 mm and y 1 = 48.88 mm) and clamped (u 2 z at x 2 = 72.30 mm and y 2 = 46.98 mm) ends; then the maximum transverse deflection was approximated using The maximum rate of twist of the shear-induced torsion actuated composite plate was then obtained by dividing the maximum measured transverse deflection in the z-direction by the length (75 mm) and half width (25 mm) of the composite plate (the adhesive thickness was here neglected). The effective rate of twist, which is proposed as the evaluation criterion for piezoelectric torsion actuated smart composites, is calculated by dividing the maximum 1.14 1.14 1.14 1.14 1.14 1.14 1.14 1.14 1.14 1.14 X = 1.
Finite element simulation of the experimental benchmark
The experimental benchmark described above was simulated using ABAQUS ® commercial FE code. Static torsion actuation analyses of the cantilever adaptive plate were conducted using, respectively, piezoelectric (C3D20RE) and elastic (C3D20R) quadratic brick elements for the piezoceramic core and composite faces. The FE mesh has 30 elements along the plate length, 20 elements along its width and two elements per layer through their thickness for the piezoceramic and glass fiber/epoxy layers and one element per layer through-the-thickness for the adhesive layers, leading to a total of 4800 elements and 30,166 nodes ( Figure 6). Electric potentials were applied to the piezoceramic core major surfaces that were forced to be equipotential (EP) electrodes. The EP constraints of the core electrodes were expressed by linear relationships so that the electric potentials for the major surfaces of the piezoceramic core are coupled to the electric potentials of the master nodes assigned to each surface using * EQUATION option. The OP of the top row of piezoceramic shear patches is implemented by changing the sign of the piezoelectric coefficients as negative. For this purpose, two different material properties and sections were assigned to the two rows of piezoceramic shear patches. The materials properties are given in the Appendix.
The plate global and tip transverse deflections under static torsion actuation of 198 V are shown in Figures 7a and 7b, to be compared to Figures 5a and 5b, respectively, for the measurements under 198 V.
From Figures 5 and 7, it is clear that the simulated and experimental transverse displacements have qualitatively similar plate axial and cross-section height distributions. Quantitatively, by taking the effect of 0.1 mm thick adhesives into consideration, the FE simulations provided these maximum tip displacements for which only the transverse one has been measured (= 5.267 µm): Values indicated in Equation (3) show that the transverse displacement component dominates the other ones; in particular, the axial displacement, which is governed by the section warping, is negligible compared to those in the cross-section; this can also be seen from the plate global and cross-section twist deformations shown in Figures 8a and 8b, respectively.
From results in Equation (3) and the deformations shown in Figure 8b, the displacement field can be written in a first approximation as where θ (x) and α are the twist angle and constant rate of twist (or twist angle per unit length). Hence, the Saint Venant torsion kinematics can be used, in a first approximation, to build an analytical model for the theoretical analysis of this new TAM.
From the relations of Equation (4), the effective rate of twist is defined as the maximum of the rate of twist per unit voltage; it is obtained, for the simulations and experiments, by calculating the maximum tip transverse deflections and dividing them by the applied voltage (V = 198 V) and, the length (L = 75 mm) and half of the width (b = 50 mm) of the composite plate (the adhesive thickness is neglected here); this leads to A comparison of the results obtained from experiments and FE simulations is presented in Table 1 for 198 V and in Figures 9 and 10. The results show that the adhesive layers between the active core and composite faces have an important effect on the torsion deformation produced by the piezoceramic shear actuators. It is also worth noticing that the experimental results show a slight nonlinearity, which is one of the characteristics of piezoceramic d 15 shear actuators. The comparison between the experiment and the FE simulation shows a reasonable agreement when the adhesive layers are considered. It is thought that the observed deviations in this case can be attributed to the non-realistic (nil displacements) representation of the real (softer) clamp by the FE model. Uncertainties related to the materials data used for the simulations also partly explain these deviations; finally, the bonding assembly (assumed perfect in the FE model) can influence the static deflection simulation. Nevertheless, these are usual difficulties that are met when simulating experimental benchmarks; hence, these results validate the proposed TAM concept.
Conclusions and perspectives
An experimental benchmark and its three-dimensional (3D) finite element (FE) simulation were proposed for the new piezoelectric d 15 shear response-based direct torsion actuation mechanism (TAM). For this purpose, the maximum static tip transverse deflection of a composite plate progressively actuated in torsion by the connection of six oppositely polarized (OP) piezoceramic d 15 shear patches was measured by an electronic speckle pattern interferometer (ESPI) system. The effective rate of twist was post-processed from a simple Saint-Venant torsion displacement field model deduced from the experimental and numerical observations of the proposed benchmark results. This parameter is proposed as the main static evaluation criterion for piezoelectric torsion actuated smart composites; it is defined as the maximum rate of twist under an applied unit voltage. The nonlinearity of the piezoelectric shear d 15 actuators was also observed during the experiments. The experimental benchmark was simulated using 3D FE within ABAQUS ® commercial code by taking into consideration the adhesive layers between the composite faces and the active core. The experimental effective rate of twist value was compared to its 3D FE simulation and the obtained results showed reasonable agreement.
This experimentally and numerically validated TAM concept, produced by OP piezoceramic d 15 shear patches, can be applied in preventing or controlling twist (or torsion) deformation for many applications, such as in wind turbines, helicopter blades, robot arms, flexible space structures, etc.
As an immediate continuation of the present work, experimental modal analyses of this benchmark are being conducted for the evaluation of its effective electromechanical coupling coefficient. The latter is post-processed from the measured short-circuit and opencircuit frequencies in the aim to assess the energy conversion efficiency of this new TAM.
|
2019-04-13T13:05:20.054Z
|
2010-09-06T00:00:00.000
|
{
"year": 2010,
"sha1": "c499af8c66afc5401db14d323932b07322bd7d00",
"oa_license": "CCBYNC",
"oa_url": "https://doi.org/10.1080/19475411.2010.510265",
"oa_status": "GOLD",
"pdf_src": "TaylorAndFrancis",
"pdf_hash": "a9d92dcf896f15e939325b21fd2c26c17499526d",
"s2fieldsofstudy": [
"Engineering"
],
"extfieldsofstudy": [
"Engineering"
]
}
|
236354428
|
pes2o/s2orc
|
v3-fos-license
|
Seasonal change is a major driver of soil resistomes at a watershed scale
Soils harbor the most diverse naturally evolved antibiotic resistomes on Earth that threaten human health, ecosystem processes, and food security. Yet the importance of spatial and temporal variability in shaping the distribution of soil resistomes is not well explored. Here, a total of 319 topsoil samples were collected at a watershed scale during four seasons (spring to winter) and high-throughput quantitative PCR (HT-qPCR) was used to characterize the profiles of soil antibiotic resistance genes (ARGs). A significant and negative correlation was observed between soil ARG profiles and seasonal dissimilarity, which along with seasonally dependent distance-decay relationships highlight the importance of seasonal variability in shaping soil antibiotic resistomes. Significant, though weak, distance-decay relationships were identified in spring, summer and winter, for ARG similarities with geographic distances. There were also strong interactions between specific soil ARGs and Actinobacteria, Firmicutes and Proteobacteria. Moreover, we found that the relative abundance of soil Actinobacteria, Firmicutes and Proteobacteria correlated significantly with annual mean temperature and annual mean precipitation at a watershed scale. A random forest model showed that seasonal change rather than spatial variation was the most important predictor of the composition of soil ARGs. Together, these results constitute an advance in our understanding of the relative importance of spatial and temporal variability in shaping soil ARG profiles, which will provide novel insights allowing us to forecast their distribution under a changing environment.
INTRODUCTION
Antibiotic resistance is a natural phenomenon used by bacteria to withstand the negative effects of antibiotics released by other organisms. 1,2 However, the dissemination of antibiotic resistance genes (ARGs) throughout the environment has been accelerated by extensive anthropogenic activities. [3][4][5][6][7] Consequently, increased occurrence and an accelerating spread of ARGs in bacterial pathogens has been one of the major threats to human, animal and plant health and food safety. 8 Soils are an important reservoir for ARGs and constitute an important habitat for the exchange of ARGs among bacteria, including some major clinical pathogens. 9 The continuous and intensive application of organic fertilizers and use of wastewater irrigation has led to a significant increase in the background levels of antimicrobial resistance in soils, therefore enhancing the likelihood of wider dissemination of antibiotic resistant bacteria (ARB) via surface runoff, wind, soil animal and plant. 3,[10][11][12][13] For example, in peri-urban areas, where food production occurs, ARBs in soil could directly enter the food chain via consumption of contaminated crops grown in manure-fertilized soil and subsequently impact human health. [14][15][16] Recently, several studies have investigated the distribution, fate and ecological drivers of ARGs from local to global scales. 3,10,[17][18][19] For instance, a continental scale investigation that profiled ARGs in sediments from 18 estuaries, representing 4000 km of coastal China, found that human activity was responsible for the abundance and dissemination of ARGs. 17 At global scale, Bahram et al. 20 found that the relative abundance of ARGs in topsoil was strongly related to fungal biomass and the bacterial/fungal biomass ratio. By comparison, the interaction between temporal changes and spatial variability of ARGs has received less attention. Most previous studies have overlooked potential temporal and spatial effects often using samples collected from a single point in time. 18,21 Studies that have explored temporal variation in soil ARGs provided critical knowledge on temporal and ecological drivers of ARG patterns. [22][23][24] However, many of these studies were based at microcosm scale under controlled conditions, which may not accurately predict antibiotic resistance under human disturbance and environmental change.
Here, the objective of this study was to investigate spatiotemporal dynamics of soil ARGs, and to evaluate the relative importance of space and time on shaping soil ARG profiles. To achieve this, 319 topsoil samples were collected in peri-urban areas across the Zhangxi watershed during four seasons (spring, summer, autumn and winter) with high-throughput quantitative PCR (HT-qPCR) used to quantify the abundance of 296 ARGs in order to create soil ARG profiles. While agronomic management (e.g., fertilization, crops) could be an important driver of soil ARGs, seasonal variability will be a comprehensive index of the (a)biotic changes, therefore we hypothesize that seasonal variability will be the major driver that determines ARG abundance and composition and will be more important than geographical location at a watershed scale.
MATERIALS AND METHODS Sampling site and sample collection
The study area comprised of 83 sampling sites covering a single watershed (Zhangxi, 28.85-30.55 N, 120.92-122.27 E), covering an area of 85 km 2 and an elevation range of 3-763 m, in the peri-urban area of Ningbo, Zhejiang Province, southeast China. The watershed is a mixed land-use watershed, including forests (natural secondary forest, managed forest), farmland (vegetable fields, nursery) and orchards ( Supplementary Fig. S1). Zhangxi watershed has a subtropical monsoon climate, warm, humid and windy during April-September, and the rainy season typically commences in June. Topsoil (0-20 cm) samples (n = 319) were collected in April, July, November 2017 and January 2018, representing the spring, summer, autumn and winter seasons. After removing plant debris and stones, at each sampling site seven soil cores were randomly taken, pooled, and thoroughly mixed. After collection, soil samples were immediately frozen with dry ice, transported to the laboratory and stored at −20°C until DNA extraction.
Soil DNA extraction, amplification, Illumina sequencing
Soil DNA was extracted from 0.5 g of soil using a FastDNA ® Spin Kit for soil (MP Biomedical, Santa Ana, California, USA). DNA concentration and quality were checked using a Qubit 3.0 (Thermo Fisher Scientific, USA) and gel electrophoresis with a 1.0% agarose gel. Thereafter, DNA extracts were stored at −20°C until further analyses.
The primer pair F515 (5′-GTGCCAGCMGCCGCGG-3′) and R907 (5′-CCGTCAATTCMTTTRAGTTT-3′) was used to amplify the 16S V4-V5 region. 25 Each sample was amplified in a 50 µl PCR reaction mixture with 25 µl TaKaRa ExTaq master mix, 0.5 µl bovine serum albumin (BSA), 1 µl of each forward and reverse primer, 1 µl DNA template and 21.5 µl nuclease-free PCR-grade water. PCR reaction conditions consisted of a 5 min initial enzyme activation at 95°C, followed by 30 cycles of denaturation at 95°C for 30 s, annealing at 58°C for 30 s and extension at 72°C for 30 s, with a final extension at 72°C for 5 min. Negative controls (template DNA replaced with water) were included for detection of any potential contamination during PCR. All PCR products were purified using a Wizard SV Gel and PCR Clean-Up System (TIANGEN Biotech, Beijing, China). Purified PCR products were quantified and pooled together for Illumina Hiseq2500 sequencing (Novogene, Beijing, China). Raw pair-end reads were filtered to discard raw reads containing three or more ambiguous nucleotides, reads with low (<20) average quality scores and those with short reads (<100 nt). The QIIME1.9 pipeline was used to process and analyze pre-processed raw sequences. 26 Operational taxonomic units (OTUs) were identified using the UCLUST algorithm with a phylotype defined at 97% sequence similarity. 27 Chimeric sequences, chloroplast and mitochondrial OTUs (around 1%) and singleton OTUs were discarded from the final OTU data set. Taxonomic classification was conducted using the Ribosome Database Project Classifier with a confidence threshold of 0.80, 28 against the SILVA database. 29 Raw sequences were deposited in the National Center for Biotechnology Information Sequence Read Archive under the accession number (PRJNA718353).
High-throughput quantitative PCR (HT-qPCR)
Soil ARG profiles were characterized using DNA extracts and highthroughput qPCR with the Wafergen SmartChip Real-time PCR system (Warfergen Inc. USA). 6,30 This system has previously been used for ARG detection in a range of samples including manure, sediment and plants. To ensure reproduceable and reliable data, compared to the original protocol, 6 HT-qPCR conditions were optimized for soils. A total of 296 primer sets (Supplementary Table S1) were used, including 283 primer sets targeting almost all major classes of ARGs, 8 transposase genes and 4 integron-integrase genes. Before running HT-qPCR, DNA extracts were diluted to the same concentration, and BSA added to the PCR mixture to reduce potential inhibition. The following program was used for HT-qPCR amplification: initial enzyme activation at 95°C for 10 min, thereafter 40 cycles of the following used for amplification: denaturation at 95°C for 30 s and annealing at 60°C for 30 s. Melt curves were automatically generated by Wafergen software. In addition, qPCR data were analyzed using SmartChip qPCR Software. Wells with multiple melting peaks or with amplification efficiencies beyond the range 1.8-2.2 were discarded. The relative copy number of ARGs and mobile genetic elements (MGEs) were calculated according to Ouyang et al. 31
Statistical analysis
All statistical analyses were conducted in the R3.6.1 environment (http:// www.r-project.org/). Differences were considered significant at P < 0.05. In this study, overall patterns of ARGs and bacterial communities were determined by calculating dissimilarity matrixes using Bray-Curtis distances and compared between seasons and locations using non-metric multidimensional scaling analysis (nMDS) and PERMANOVA 30 with the "labdsv" 32 and "vegan" packages. 33 Mantel tests were used to explore associations between bacterial communities and ARGs profiles and were conducted using the "vegan" package. In the present study, the values 1, 2, 3 and 4 were used to represent the spring, summer, autumn and winter, respectively, for further data analysis. The seasonal decay relationships were calculated using an ordinary least squares (OLS) model between seasonal dissimilarity (Euclidean distance) and ARG and bacterial community similarity (based on 1 ̶ [dissimilarity of the Bray-Curtis distance metric]). 34 We used a neutral community model (NCM) 35 and normalized stochasticity ratio (NST) 36 to determine the potential importance of stochastic and deterministic processes on ARG compositions and bacterial community assembly. The main predictors for the ARG profiles at the watershed scale were identified by a classification random forest (RF) analysis. The latitude and longitude data of each site (Supplementary Table S2) were transferred to rectangular data to represent spatial distance by function principal coordinates of neighboring matrices (pcnm) in R with the "vegan" package. In the RF model, pcnm1 served as spatial predictors for the ARG profile index, while the values 1, 2, 3 and 4 were served as temporal predictors for the ARG profile index. Bacteria at the phylum level were used to represent the biological predictors for ARG profiles. Network analysis was generated in the R3.6.1 environment using the "psych" package, 37 to investigate the co-occurrence patterns between ARG subtypes and microbial taxa. Network graphs were visualized in Cytoscape 3.7.1 with a circular layout algorithm 38 based on strong (Spearman's R 2 > 0.6) and significant (P < 0.001) correlations. Finally, we hypothesized that non-random co-occurrence patterns between ARGs and microbial taxa could indicate possible host information of ARGs if the ARGs and the coexisting microbial taxa possessed a strong and significantly positive correlation. OLS regression was conducted to test the relationships between the microbial taxa and annual mean temperature and annual mean precipitation (Supplementary Table S2). All boxplots, bar charts, scatter diagrams and OLS regression were generated using R with the "ggplot2" package, 39 meanwhile, pairwise Wilcox tests of soil ARGs and bacterial Shannon index in spring, summer, autumn and winter were conducted with the "ggpubr" 40 and "ggsignif" 41 package.
RESULTS
Diversity and abundance of ARGs in soil at a watershed scale A total of 193 ARGs and 9 MGEs including 8 transposase genes and 1 intI-1(clinic) integron-integrase gene were detected from the watershed soil samples. Numbers of detected ARGs ranged from 21 to 131 in spring, 41 to 99 in summer, 17 to 79 in autumn, 20 to 102 in winter (Fig. 1A). These ARGs, represent almost all major classes of antibiotics commonly administered to humans and animals.
The number of ARGs detected in spring was significantly higher than all other seasons (Fig. 1A). The relative abundance of ARGs ranged from 6.50E-05 to 1.067 copies per 16S rRNA gene in spring, 0.012-0.136 copies per 16S rRNA gene in summer, 0.009-0.242 copies per 16S rRNA gene in autumn, and 0.005-0.928 copies per 16S rRNA gene in winter (Supplementary Fig. S2). By comparison, autumn had significantly lower abundances of ARGs than the other three seasons (P < 0.05). In addition, different types of ARGs showed distinct temporal patterns. For example, spring samples had a significantly greater abundance of Aminoglycoside resistance genes than the other seasons (P < 0.05), whereas autumn had the most Vancomycin resistance genes (P < 0.05). Genes resistant to fluoroquinolone, quinolone, florfenicol, chloramphenicol, and amphenicol (FCA) showed no significant temporal variation (P < 0.05) (Fig. 1B).
Bacterial community composition in soil at a watershed scale
Across all samples, we obtained a total of 23,922,016 high quality bacterial sequences, which were grouped into 21,269 OTUs. Bacterial alpha-diversity (Shannon index) showed that winter soil samples had a more diverse bacterial community than was found in spring and autumn (P < 0.05) but not in summer (Fig. 1C). The soil microbial community was represented by 52 phyla, and was dominated by Actinobacteria, Acidobacteria, Bacteroidetes, Chloroflexi, Cyanobacteria, Firmicutes, Nitrospirae, Planctomycetes and Proteobacteria (Fig. 1D). Proteobacteria and Acidobacteria were the most dominant phyla, accounting for 35.1-82.5% of the total bacterial 16S rRNA gene sequences in spring, 34.3-86.1% in summer, 36.7-82.0% in autumn, and 37.0-78.1% in winter. By comparison, the relative abundance of Proteobacteria and Acidobacteria was significantly lower in winter than in spring, summer and autumn, whereas Actinobacteria, Firmicutes and Nitrospirae did not vary between seasons (P < 0.05) (Fig. 1D).
Seasonal and spatial variation in soil microbiome at a watershed scale
Non-metric multidimensional scaling (nMDS) ordinations indicated that seasonal change shifted the overall profile of soil ARGs (Stress = 0.1884, P < 0.01) (Supplementary Fig. S3A). Similarly, soil bacterial community composition differed significantly between seasons, in particular, the bacterial community in spring separated from other seasons (Stress = 0.1469, P < 0.01) (Fig. S3B). A significant correlation was identified between the relative abundance of ARGs and seasonal dissimilarity (R 2 = 0.0269, P < 0.001) (Supplementary Fig. S4A). Moreover, a negative relationship was observed between bacterial community similarity and seasonal dissimilarity (OLS analysis, R 2 = 0.0701, P < 0.001) (Supplementary Fig. S4B).
We estimated distance-decay relationships for each season for both soil ARGs and bacteria (Figs. 2, 3). Except for ARGs in the autumn (P = 0.18), most distance-decay patterns were significant for both ARGs and bacteria. Fitness values were remarkably low (R 2 < 0.1), indicating a weak distance decay of both ARG profiles and microbial community similarities with geographic distance.
These results were consistent with the nMDS visualization, which showed that ARG profiles and bacterial community composition did not clearly cluster geographically (Supplementary Fig. S5).
A NCM was used to estimate the potential dispersal limitation of ARG in soil ( Supplementary Fig. S6A). There was a higher immigration rate (m) of ARGs in spring, indicating that the dispersal limitation of AGRs is lower in spring than the other seasons ( Supplementary Fig. S6A). We further used the NST to quantify the role of deterministic and stochastic processes in soil ARG profiles (Supplementary Fig. S6B). NST values exceeded the 50% boundary point in all four seasons, suggesting that stochastic rather than deterministic process played a more important role in shaping ARG compositions.
Multiple predictors for soil ARGs at a watershed scale Significant co-occurrence correlations were observed in the ARGmicrobial community network (Spearman's R 2 > 0.6, P < 0.001) ( Supplementary Fig. S7). Actinobacteria, Firmicutes, and Proteobacteria were highly connected with a variety of ARG subtypes ( Supplementary Fig. S7), which covered almost all major classes of antibiotics and were thought to be possible ARG hosts. For example, lunB-02 is strongly connected with Turicibacter which is affiliated with Firmicutes, while tetT is strongly connected with Lysobacter which belongs to Proteobacteria. In addition, we found most single bacterial taxa to be significantly correlated with multiple ARG subtypes (Spearman's R 2 > 0.6, P < 0.001). Season played a role in determining the abundance of certain phyla ( Supplementary Fig. S7) as annual mean temperature and annual mean precipitation were significantly correlated with Actinobacteria, Firmicutes, and Proteobacteria in soil (P < 0.01) (Supplementary Fig. S8).
Finally, we used RF modeling to compare the relative importance of spatial and temporal variability and microbial community composition in shaping the distribution of ARG profiles at a watershed scale (Fig. 4). Ten of the most abundant bacterial phyla were also incorporated in this model, because the Mantel test indicated that the ARG profiles were significantly associated with changes in bacterial community composition (spearman' r = 0.1698, P < 0.001). Our RF ranks showed that seasonal change was the most important predictor of ARG profiles (Fig. 4). Although, the importance of spatial variability in predicting of soil ARGs was weaker than temporal variability, its importance was still stronger than Nitrospirae, which was the most important microbial predictor for soil ARGs. Proteobacteria, as the most dominant phylum across time and space (Fig. 1D), exhibited a weaker predictive power regarding soil ARGs.
DISCUSSION
Temporal and spatial decay relationships of soil resistomes In the present study, a significant and negative correlation was observed between soil ARG profiles and seasonal dissimilarity (Eucliden distance), highlighting the importance of seasonal variability in predicting the soil antibiotic resistomes. It should be noted that seasonal variation is a latent variable, which is related to human activities and shifting biotic and abiotic factors (e.g., soil antibiotic concentrations). For example, prior work investigating the seasonal variation of soil antibiotics at the same site found a higher concentration of soil antibiotics in winter than summer. 42 Consequently, antibiotic residues derived from organic fertilizers applied in winter may pose a selection pressure on the soil microbes, increasing resident antibiotic resistance and accelerating new ARG occurrence and subsequent dispersal. This was supported by our observation of a more diverse and abundant soil ARG profile in winter and spring compared to soil ARG profiles in either summer or autumn. As ARGs can be present as both extracellular DNA (eDNA) and intracellular DNA, given the persistence of eDNA in livestock waste and soil, extracellular ARGbased transformation could be playing an important role in the proliferation of ARGs in the spring and winter soil samples after the application of organic fertilizers. 43,44 Given the importance of microbial communities in shaping soil ARG patterns 45,46 and the prevalence of co-occurrence patterns between ARG subtypes and microbial taxa, 47,48 one reasonable explanation of temporal ARG distribution patterns may be shifts in soil microbial communities caused by seasonal change. In addition, we found that different types of ARG displayed a different seasonal distribution pattern. For example, the abundance of genes conferring resistance to Aminoglycosides was significantly enriched in spring, while those resistant to Vancomycin were more enriched in autumn. These differences might be due to the seasonal preference of soil microbes that contain these genes. OLS analysis showed that Actinobacteria, Firmicutes and Proteobacteria, each of which contained ARGs encoding resistance to different classes of antibiotics, were significantly correlated with annual mean temperature and annual mean precipitation at a watershed scale. To give a more comprehensive understanding of soil microbial seasonal preference, further studies are needed.
Correlations between geographic distance and ARG similarity were season-dependent with significant distance-decay relationships identified in spring, summer and winter but not in autumn (P = 0.18). Moreover, from previous studies in the same watershed, 28-51 ARGs were detected only in farmland soils, with seasonal variation affecting farmland-related ARGs. For instance, the greatest number of farmland-related ARGs in the resistomes were detected in spring (ploughing season), while the fewest were detected in autumn (the fallow period). 45 Based on neutral community modeling (fitness > 0.8) and NST index( Supplementary Fig. S6), stochastic forces rather than deterministic selection dominated soil resistomes at a watershed scale. According to Hubbell's neutral theory (Hubbell, 2001), community similarity would be predicted to decrease with increasing spatial distance due to dispersal limitations. Thus, a distance-decay relationship would emerge even without differences in environmental conditions or niche requirements. In this study, nMDS visualization showed limited spatial differentiation in ARGs at the watershed scale. Moreover, NCM analysis estimated that the dispersal ability of most AGRs in spring was higher than other seasons, which may have partially explained the high diversity of soil ARGs in spring. Temporal variation driving the pattern of soil resistome We report that seasonal variability was more important than geographical distance in shaping soil ARG profiles at a watershed scale. Previous studies have suggested that anthropogenic factors (e.g., pesticide usage) were the main forces shaping ARG patterns in cropland and sediments at a continental scale, 17,18 while ARG composition in forest soil was regulated by the diversity of herbaceous plants and bacterial communities. 21 However, it should be noted that both anthropogenic impacts and vegetation turnover are seasonally dependent. Furthermore, agronomic activities can also induce significant shifts in microbial community composition 49 and in turn influence ARG patterns. Thus, season is a comprehensive proxy that encapsulates many changes to soil physiochemical conditions that may be driving soil ARG patterns.
Our results contrast with recent studies which found that spatial variability is more important than temporal variability when shaping soil microbial communities. 50,51 These studies were conducted at either a much smaller or larger spatial scale than this study. Thus, this inconsistency is potentially due to microbial systems being studied at different spatial, temporal, and phylogenetic scales, i.e., different processes may dominate at different scales in microbial systems. 52 Notwithstanding this, it may not be possible to completely separate the impacts of temporal and spatial variation on microbial patterns, since seasonal changes could affect microbial dispersal and environmental heterogeneity, which in turn could lead to stronger or weaker impacts of spatial variation. In the present study, significant though weak distance-decay effects were observed in soil bacterial communities. Moreover, the slopes of the distancedecay curves were steeper in the spring and autumn than those of summer and winter, indicating that seasonal variability may influence the balance between deterministic and stochastic processes in governing the assembly of soil microbial communities. Accordingly, it would be good to include more parameters that involved in seasonal change and spatial variability and quantifying the spatial and temporal variability, when designing future similar studies.
CONCLUSIONS
To the best of our knowledge the present work evaluates, for the first time, the relative importance of spatial and temporal variability on ARG composition and abundance at a watershed scale. Our results suggested that distance-decay patterns in soil 4 Random Forest modeling indicating the importance of different predictors on soil resistome profiles. The accuracy importance measure was computed for each tree and averaged over the forest (5000 trees). Percentage increases in the MSE (mean squared error) of variables were used to estimate the importance of these predictors, with higher MSE% values implying a greater importance of the predictors. Significance levels are as follows: ***P < 0.001. MSE mean squared error.
ARGs were seasonally dependent. The significant distance-decay relationships identified in spring, summer and winter had low fitness indicating weak distance-decay patterns of soil resistomes. The significant negative correlation between soil ARG profiles and seasonal dissimilarity together with RF model, indicated that seasonal variability played a major role in shaping soil ARG profiles. This study increases our understanding of soil ARGs predictors, which is crucial for predicting changes in soil ARGs due to environmental change and human disturbance.
|
2021-07-27T00:06:05.653Z
|
2021-05-20T00:00:00.000
|
{
"year": 2021,
"sha1": "a8bbe17db2be91874c1f7dd7d5ae271e5b229afb",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s43705-021-00018-y.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8acaa834fd1e9c8fbb9af0522aca9e27648d9fff",
"s2fieldsofstudy": [
"Agricultural And Food Sciences",
"Environmental Science"
],
"extfieldsofstudy": [
"Environmental Science"
]
}
|
119209084
|
pes2o/s2orc
|
v3-fos-license
|
Sequential Regeneration of Charmonia in Heavy-Ion Collisions
We investigate the production of psi(2S) in nuclear collisions at RHIC and LHC energies. We first address charmonium production in 200 GeV d-Au collisions at RHIC; the strong suppression of psi' mesons observed in these reactions indicates mechanisms beyond initial cold nuclear matter effects. We find that a more complete treatment of hadronic dissociation reactions leads to appreciable psi' suppression in the hadronic medium of an expanding fireball background for d-Au collisions. When implementing the updated hadronic reaction rates into a fireball for 2.76 TeV Pb-Pb collisions at LHC, the regeneration of psi' mesons occurs significantly later than for J/psi's. Despite a smaller total number of regenerated psi', the stronger radial flow at their time of production induces a marked enhancement of their R_{AA} relative to J/psi's in a momentum range p_t \simeq 3-6 GeV. We explore the consequences and uncertainties of this"sequential regeneration"mechanism on the R_{AA} double ratio and find that it can reproduce the trends observed in recent CMS data.
I. INTRODUCTION
Charmonium production in ultra-relativistic heavy-ion collisions (URHICs) has been studied for over 30 years. The originally proposed J/ψ suppression signature of Quark-Gluon Plasma (QGP) formation [1] has evolved into a more complex problem where both suppression and so-called regeneration (or statistical hadronization) mechanisms need to be considered. Their interplay and relevance depend on collision energy, system size and the 3-momentum of the measured charmonia, see, e.g., Refs. [2][3][4] for recent reviews. The phenomenological modeling of these mechanisms, and their relation to the underlying in-medium properties, has progressed significantly in recent years. In particular, kinetic transport approaches, when calibrated with existing data from SPS and RHIC, have shown predictive power in describing novel features of the J/ψ production systematics observed in the new energy regime of the LHC [5][6][7][8] (although significant uncertainties due to, e.g., the opencharm cross section persist [9]). This includes the marked increase of the nuclear suppression factor compared to RHIC energies, along with its pronounced dependence in transverse momentum.
Much less is known about the 2S excited state, ψ ′ (3686). Its small "binding" energy of about 45 MeV (relative to the DD threshold) renders controlled theoretical calculations of its in-medium properties (binding energy and inelastic reaction rates) challenging. Experimentally, the ψ ′ over J/ψ ratio has been measured at the SPS [10], where it was found to drop by up to a factor of 3 in central 17.3 GeV Pb-Pb collisions. This is consistent with the statistical hadronization approach [11], but it can also be explained by transport approaches with large inelastic reaction rates of the ψ ′ in the hadronic phase [12,13]. More recently, ψ ′ data have become available for 0.2 TeV d-Au collisions at RHIC [14] and 5.02 TeV p-Pb collisions at LHC [15]. ψ ′ mesons were found to be significantly more suppressed than J/ψ mesons, which is diffcult to reconcile with initial coldnuclear-matter (CNM) effects since the passing time of the highly Lorentz-contracted incoming nuclei is much smaller than the formation time scale of the charmonia. Consequently, final-state effects have been put forward to explain these data, e.g., using the comover suppression model [16], which achieves a good description of the collision energy and rapidity dependence of the ψ ′ and J/ψ including the expected CNM effects on the parton distribution functions (see also Ref. [17]).
However, rather unexpected results have emerged from recent measurements by the CMS collaboration [18] for the double-ratio of the nuclear modification factor, R AA , of ψ ′ over J/ψ in 2.76 TeV Pb-Pb collisions at the LHC (preliminary results are also available from AL-ICE [19]). At slightly forward rapidities, 1.6<|y|<2.4, and for transverse momenta 3<p t <30 GeV, this double ratio is around 0.9±0.45±0.3 for peripheral and semicentral collisions, but significantly exceeds one for central collisions, 2.3±0.5±0.35. Especially the latter has evaded any model explanations thus far, see, e.g., the detailed studies in Ref. [20]. On the other hand, around midrapidity, and for momenta 6.5<p t <30 GeV, a double ratio of around ∼0.5 is found, which is much more in line with common expectations of a stronger suppression of ψ ′ due to its much weaker binding.
In the present paper we put forward a potential mechanism to (partially) resolve the above "puzzle". Based on the rather large inelastic reaction rates for the ψ ′ in hadronic matter that we deduce from its suppression in d-Au (also in line with the aforementioned SPS data), we argue that the inverse reactions of ψ ′ formation in Pb-Pb collisions must also happen in the later, hadronic stages of the fireball evolution. In particular, the ψ ′ regeneration processes happen later than those for the J/ψ whose much larger binding energy leads to an earlier "freezeout" than for the ψ ′ . The consequence of such a "sequential freezeout" is that the collective expansion velocity of the medium leads to harder p t spectra for the ψ ′ , which, in terms of the R AA , can outshine the J/ψ in a momentum range of p t M ψ , which happens to coincide with the CMS p t cut. On the other hand, at higher p t , the regeneration contribution ceases giving way to a more sequential-like suppression pattern of primordially produced charmonia.
Our paper is organized as follow: In Sec. II we update our hadronic reaction rates for both ψ ′ and J/ψ, apply them within a schematic fireball for d-Au collsions at RHIC and compare our results to PHENIX data. In Sec. III we implement the updated hadronic rates in our earlier constructed transport approached. In particular, we estimate for the time windows for sequential charmonium regeneration, and elaborate the uncertainty for the underlying charmonium p t spectra in the context of the CMS data for the ψ ′ /J/ψ R AA double ratio. We conclude in Sec. IV.
II. HADRONIC DISSOCIATION OF CHARMONIA
Dissociation rates of charmonia in hadronic matter are usually considered to be much smaller than in the QGP (see, e.g., the discussion in Ref. [2]). However, for the ψ ′ this is not so obvious, since the proximity of its mass to the DD threshold provides a large phase space for break-up reactions. In the following, we revisit hadronic reactions rates for J/ψ and ψ ′ mesons based on effective meson Lagrangians (Sec. II A) and evaluate their consequences for final-state effects in d-Au reactions at RHIC (Sec. II B).
A. Update on Dissociation Rates
Our starting point is the previously employed hadronic reactions rates [13] based on flavor-SU (4) meson exchange models [21,22] for the processes J/ψ + ρ → D +D, D ⋆ +D ⋆ (exothermic for m J/ψ + m ρ > m D + mD and endothermic for m J/ψ + m ρ < m D ⋆ + mD⋆ ) and J/ψ+π → D ⋆ +D, D+D ⋆ (endothermic). For those reactions, the J/ψ dissociation rate at T =170 MeV amounts to 1-2 MeV, corresponding to a lifetime of 100-200 fm/c. Even with the typical uncertainties (factor ∼2-3) associated with the hadronic formfactor cutoff values, this is too small to affect the J/ψ abundance during the 5-10 fm/c lifetime of the hadronic phase in URHICs. For the ψ ′ , geometric scaling by the vacuum radius has been assumed, increasing its rates by a factor (r ψ ′ /r Jψ ) 2 , which is approximately compatible with constitutentquark model calculations [23].
A hadron resonance gas (HRG), however, contains many more species than π and ρ. To estimate their Previous results for a πρ gas with Λ=1 GeV (solid red lines) are compared to our updated results for a meson resonance gas using Λ=1 GeV and 2 GeV (solid green and blue lines, respectively). The blue-dotted lines additionally account for finite meson chemical potential that build up for temperatures below the chemical freezeout in URHICs [24].
impact on the charmonium dissociation rates, we simply adopt the existing ρ-and π-induced matrix elements and shift their kinematics according to the pertinent 2particle threshold, i.e., 2 and s X thr = (m J/ψ + m X ) 2 for exothermic and s X thr = (2m D ) 2 for endothermic ss (s X thr = (m D + m Ds ) 2 if X contains a strange quark). We include a total of 52 non-strange and single-strange meson species, up to a mass of m X = 2 GeV. As before, we apply geometric scaling to obtain the reaction rates for the χ c and ψ ′ . Our results for a meson gas in chemical equilibrium are summarized by the solid lines in additional resonances enhance the J/ψ dissociation rate by factor of ∼2.5, and another factor of 2.5 when increasing the hadronic formfactor cutoff from Λ=1 GeV to 2 GeV, reaching a maximal rate of 10.5 MeV. For the ψ ′ , geometric scaling leads a maximun rate of up to 35 MeV, translating into a lifetime of 7 fm/c which is now comparable to the duration of the hadronic phase in URHICs. This becomes even more significant if chemcial freezeout is accounted for, which implies the build-up of meson chemical potentials which lead to a slower decrease of the meson densities as temperature decreases, cf. dotted lines in Fig. 1.
B. Charmonium Prodution in 0.2 TeV d-Au Collisions
We proceed by implementing our updated hadronic reaction rates into the thermal-rate equation framework developed in Ref. [25]. Another new aspect relative to our previous work [26] is that we allow for final-state effects in small collision systems, specifically for d-Au at RHIC, where a rather large ψ ′ suppression has been observed. (while J/ψ's are much less suppressed).
Toward this end, we construct a thermal fireball using the same methods as before for AA collisions. We determine the total entropy of the fireball by matching the observed final-state hadron abundancies. Based on a Glauber model for the centrality dependent nculear overlap function, we initialize the fireball with a transverse radius of R 0 ≃2.5fm/c. The expansion is then modeled with the same acceleration as in 0.2 TeV Au-Au before [26]. For simplicity, we stick with our first-order equation of state, transitioning from a quasi-particle QGP into a HRG through a mixed phase at T =180 MeV (we do not expect large changes when utilizing a modern cross-over EoS, which has been checked for dilepton [27] and bottomonium observables [28]). Kinetic freezeout is also constructed as before, with a freezeout temperature mildy decreasing with centrality, from T fo ≃155 MeV in peripheral to ∼142 MeV in central collisions, resulting in fireball lifetimes of 0.5-3 fm we have assumed the same initial spectra for J/ and from pp collisions,/c, cf. Fig. 2.
For the most central collisions, a short QGP phase with initial temperature T 0 =190 MeV is followed by a 1 fm/c mixed phase and a 1.5 fm/c hadronic phase. The initial CNM effects are assumed to be identical for all charmonia, essentially given by a shadowing suppression [16] which we mimic with an "effective" nuclear absorption cross section of σ abs =2.4 mb. Our results for the centrality dependence of the nuclear modification factor, are summarized for both Ψ = J/ψ, ψ ′ in Fig. 3, in comparison to PHENIX data [14] (N coll : number of primordial N N collisions). For the J/ψ, there is little suppression beyond CNM effects; the additional final-state effects are due to a small QGP suppression on the direct J/ψ's, as well as a suppression of the feeddown from χ c 's and ψ ′ , with little room for additional hadronic suppression (thus, in principle, preferring a small value for the hadronic cutoff parameter). On the other hand, for the ψ ′ , our baseline QGP suppression is not enough to account for the marked suppression beyond CNM effects.
Here, a large formfactor cutoff of Λ=2 GeV is preferred to augment the hadronic suppression. An additional increase of QGP suppression rate of the ψ ′ by a factor of 2 could also be helpful (such an increase may arise, e.g., from nonperturbative heavy-quark interactions with light partons). Further increasing the hadronic rate by a factor of 2 1 goes in a similar direction, i.e., reducing the discrepancy with the most central data point. We finally note that a hadronic ψ ′ dissociation rate well beyond the one from the πρ gas (by a factor of 5 or more) was previously deduced within our setup [13] to be able to account for the ψ ′ suppression observed in S-U and Pb-Pb collisions at the SPS [10]. Our newly calculated rates in the present ms. are in line with this notion.
III. SEQUENTIAL REGENERATION OF CHARMONIA
In this section, we investigate the consequences of the updated hadronic reaction rates within our previously constructed thermal fireball expansion (Sec. III A), followed by a more generic evaluation of the associated uncertainties specifically in the context of the ψ ′ /J/ψ R AA double ratio (Sec. III B). We focus on 2.76 TeV Pb-Pb collisions at the LHC. Our earlier predictions for inclusive J/ψ production in these reactions [5] resulted in a fair agreement with the centrality, transverse-momentum and rapidiy dependecies observed by the ALICE and CMS collaborations, and thus serves as our framework to evaluate ψ ′ observables. Since the pertinent CMS data are for "prompt" J/ψ and ψ ′ production, we do not include contributions from B feeddown.
A. Fireball Model
For our fireball results we focus on the so-called "strong-binding scenario", where the in-medium charmonium properties are taken with guidance from a Tmatrix approach with the internal energy from lattice-QCD as underlying potential. This assumption gives a better agreement both with correlators from lattice-QCD [31] and the overall charmonium phenomenology at SPS and RHIC [26]. For definiteness, we employ the hadronic rates with Λ=0.75(2) GeV for the J/ψ (ψ ′ ), and the factor-of-2 increased QGP rate of the ψ ′ .
Let us first inspect the time evolution of direct J/ψ and ψ ′ mesons in 0-20% central Pb-Pb at midrapidity (without shadowing), as following from the solution of the kinetic rate equation, see Fig. 4. Compared to our previous results (cf. lower panel in Fig. 2 of Ref. [5]), the J/ψ now picks up a regeneration contribution in the hadronic phase, by about 0.15 units in R AA , which, despite a small rate, is due to the large equilibrium limit. Most of the production, however, occurs prior to the onset of the mixed phase at τ ≃ 5.5 fm/c. On the contrary, ψ ′ production only starts to set in a that point, leveling off around τ ≃9-10 fm/c, when the temperature of the fireball has dropped to about 150-160 MeV. The main qualitative and robust feature here is that the lower dissociation temperature of the ψ ′ , relative to the J/ψ, implies a later production through regeneration in the time evolution of the fireball in URHICs. The sequential regeneration of J/ψ and ψ ′ has rather dramatic consequences on their transverse-momentum (p t ) spectra. Following our previous work [32], we approximate the p t spectra of the regeneration components with the standard blast-wave expression, (4) implying thermalized charm-quark distributions. As our default, we evaluate this expression when most of the pertinent charmonium yield has built up, i.e., at T c for the J/ψ (τ =5.5 fm/c in Fig. 4) and at T =155 MeV for the ψ ′ (τ =9.7 fm/c in Fig. 4). The resulting nuclear modification factors for both regeneration and surviving primordial components are displayed in Fig. 5 (we have assumed the same initial spectra for J/ψ and ψ ′ from pp collisions, figuring into the denominator of R AA ). Clearly, there is a significant uncertainty associated with this procedure which we address in Sec. III B below (we recall, however, that our pertinent predictions for the J/ψ gave fair agreement with the observed p t spectra at LHC). Three main qualitative features can be gleaned from comparing the R AA 's for J/ψ and ψ ′ in Fig. 5: (a) At low p t 3 GeV, the regeneration yield of the J/ψ dominates over the one from ψ ′ , as a consequence of the approach toward equilibrium which favors the smaller J/ψ mass; (b) At "intermediate" p t ≃ 3 − 6 GeV, the significantly harder blast wave for the ψ ′ generates a shift of the "flow bump" which can exceed the regeneration contribution in the J/ψ R AA ; (c) at still higher momenta, p t 6 − 8 GeV, regeneration gives way to (suppressed) primordial production, where the J/ψ R AA exceeds again the one from the more weakly bound ψ ′ . Items (b) and (c) are in qualitative agreement with the trends observed in the pertinent R ψ ′ AA /R J/ψ AA double ratio observed by CMS [18]. In the next section we explore some of the uncertainties in our calculations.
B. Schematic Model
To better quantify variations in the interplay of the different production components of both J/ψ and ψ ′ , let us first formulate a baseline scenario motivated by existing experimental data for the J/ψ and the thermal fireabll calculations in the previous sections. For simplicity, we assume the primordial parts to be constant in R AA (p t ). For 0-20% Pb-Pb collision we take 0.15-0.25 for the J/ψ (compatible with high-p t CMS data for prompt J/ψ [33]) and 0-0.075 for the ψ ′ , to reflect its stronger absoprtion as a loosely bound states. The constancy in p t reflects our currently limited knowledge about the p t dependence of the dissociation rates (see, e.g., Ref. [32]), formation time effects, etc. For the regeneration components, we choose total yields such that the total momentum-integrated R AA amounts to 0.55-0.65 for the J/ψ (compatible with ALICE data [34]) and 0.4 for the ψ ′ (as suggested by our fireball results). For the sequential freezeout, which determines the temperature and flow strength in the blast-wave spectra of the regeneration components, we stick with the correlation given by the expanding fireball, i.e., τ ≃ 5.5(9.7) fm/c for the J/ψ (ψ ′ ) corresponding to T fo =180(155) MeV and v=0.32(0.46)c. Recall that the effective slope parameter for the blast-wave spectra is approximately given by T eff ≈ T + m Ψv 2 , which is mostly driven by the flow term due to the large mass of the charmonia. In Fig. 6 we display the pertinent p t spectra for central Pb-Pb(2.76 TeV), normalized by the N coll -scaled yields in pp, to mimic the relative magnitude of ψ ′ and J/ψ contributions in their respective R AA 's. The plot highlights again the main effect proposed in this paper: due to the stronger flow for the regenerated ψ ′ , its R AA can rise above the one for the J/ψ in a limited p t window around 3-6 GeV, before primordial production takes over again.
For semi-central Pb-Pb (20-40%) we construct the baseline following the same reasoning as for central collisions. We increase the primordial R AA 's for J/ψ to 0.35-0.45 [33] and to 0.1-0.2 for ψ ′ , and evaluate the blastwave p t spectra at the same freezeout temperatures as for central collisions (but with flow velocities given by the fireball expansion for 20-40% centrality).
The resulting double ratios covering the abovespecified ranges in the primordial components are shown GeV, respectively, and are compared to CMS data [18]. In the upper (lower) panel, the boxes indicate theoretical uncertainties in the modeling of the primordial (regeneration) component of the charmonium yields and spectra. The vertical purple lines represent our "best fit" in the upper panel, and a "realistic" baseline for the blast-wave variations in the lower panel.
in the upper panel of Fig. 7 and compared to CMS data [18] (we did not include the ALICE data [19] as they contain feeddown contributions from B-meson decays, which could becomne rather significant at high p t , see, e.g., Ref. [20] for a calculation including those). The basic trends of the data can be reproduced within our approach. For defintieness, we quote (approximate) "best fit" value of 0.15 and ∼0 (0.35 and 0.2) for the primordial J/ψ and ψ' values, respectively, in (semi-) central Pb-Pb collisions, resulting in the purple horizontal bars in the upper panels of Fig. 7. Finally, we illustrate the sensitivity of the double ratios to the blast-wave parameters, by varying the freezeout temperatures over the ranges T J/ψ fo =180-200 MeV and T ψ ′ fo =150-165 MeV, along with the pertinent flow velocities from the fireball model, and fixing the primordial components at R J/ψ AA =0.2(0.35) and R ψ ′ AA =0.05(0.15) for (semi-) central Pb-Pb. The larger regeneration components and larger flow velocities in central collisions render the double ratios more sensitive to the details of the sequential regeneration mechanism.
IV. CONCLUSIONS
In the present work, we have investigated the production systematics of ψ ′ mesons in URHICs. We first revisited the problem of hadronic ψ ′ dissociation and found that a more complete inclusion of hadronic states in a resonance gas suggests a marked increase of its inelastic reaction rates. When implementing these rates into an expanding fireball for d-Au collisions at RHIC, we found a much improved description of the rather strong suppression of ψ ′ mesons observed in these reactions. This is similar in spirit to, and thus supports, the recently suggested comover suppression effects [16] in dA and pA reactions at RHIC and LHC. We then evaluated ψ ′ transport in Pb-Pb collisions at the LHC using our existing fireball approach which previously provided fair agreement with J/ψ data. The key features in our approach are the lower dissociation temperature of the ψ ′ relative to the J/ψ and its sizable hadronic reaction rates. This implies a sequential freezeout of those two mesons, with most of the ψ ′ regeneration occurring later in the fireball evolution. The larger collective medium flow then leads to a marked enhancement of the ψ ′ regeneration yield in a p t region around 3-6 GeV, transitioning to (suppressed) primordial production at higher p t . While quantitative predictions of this mechanism are beyond current theoretical control, we have shown that variations in the ingredients to the sequential regeneration scenario produce trends in the ψ ′ over J/ψ R AA double ratio which agree with recent CMS data. We therefore believe that the qualtitative features of this mechanism are robust and provide a promising candidate to contribute to the understanding of these data. In fact, if corroborated, sequential regeneration may serve as a tool to extract in-medium properties of the ψ ′ from URHIC data.
|
2015-07-28T19:19:08.000Z
|
2015-04-02T00:00:00.000
|
{
"year": 2015,
"sha1": "238de588133e05ec3e511a07c2e50dfc6fae9d4f",
"oa_license": "elsevier-specific: oa user license",
"oa_url": "http://manuscript.elsevier.com/S0375947415002055/pdf/S0375947415002055.pdf",
"oa_status": "BRONZE",
"pdf_src": "Arxiv",
"pdf_hash": "238de588133e05ec3e511a07c2e50dfc6fae9d4f",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
56059971
|
pes2o/s2orc
|
v3-fos-license
|
A Novel Fuzzy Sliding-Mode Control for Discrete-Time Uncertain System
This paper considers the sliding-mode control problem for discrete-time uncertain systems. It begins by presenting a discrete variable speed reaching law and a discrete-time sliding-mode controller (DSMC) designed using the proposed reaching law, followed by an analysis of their stability and dynamic performance. A sliding-mode controller with simple fuzzy logic is then proposed to further strengthen the dynamic performance of the proposed sliding-mode controller. Finally, the presented DSMC and the DSMCwith fuzzy control for adjusting the parameters in this paper are compared with one of the previous proposed classic DSMC systems. The results of this simulation show that the DSMC presented here can suppress chatter and ensure good dynamic performances when fuzzy logic is used to tune the parameters.
Introduction
A sliding-mode control (SMC) system has proved to be an effective control strategy for incompletely modeled or uncertain systems, so much so that many relevant results can be found in various publications, for example, [1][2][3][4][5].SMC is a popular robust strategy which can offer good robustness against the parameter uncertainties and external disturbance by limiting the effect in a bounded area.The main advantage of SMC is that it can guarantee the stability of the system by switching the control law.Movement in general SMC systems can be divided into a reaching phase and a sliding mode [6] because the basic idea of SMC is to drive system states onto a predefined switching surface and move on it to the point of origin after the system state reaches the switching surface.The need to research SMC in discrete time (DSMC) is due to the widespread use of computers for the control problem.In recent years, DSMC has received a great deal of research interests.A new method of DSMC is proposed in [7].The algorithm comprises not only functions in the sliding variable but also functions in other variables.It provided a general method to design reaching law.The ideal reaching law is proposed in [8] which used LMI to list sufficient conditions of switching surface existence in Delta operator system, and the result was compared with the index reaching law.Fast convergence and the attenuation of disturbance are studied in [9][10][11].In [9], a new controller is present which is able to handle the effect of interconnection for large-scale systems and unmatched uncertainty.It also shows a fast convergence to the desired value.In [10,11], quasi-sliding-mode control was proposed to the system which has eternal disturbance.Fast convergence and small chatting are ensured by using these two algorithms.The applications of DSMC in practical systems are also studied in [12,13] which study the SMC on servo system with external disturbance.
However, its major drawback in practical applications is that the system trajectories of DSMC systems cannot reach their origin.They only tend to reach a chattering stage surrounding the origin [14][15][16].Numerous methods have been developed to eliminate this phenomenon in DSMC [17][18][19]; for instance, in [18], robustness is ensured and chattering is eliminated by using an optimal integral sliding surface.An output-feedback quasi-sliding-mode control scheme is proposed in [19] which eliminates the chattering phenomenon and also ensures the uncertainties and disturbances are robust.Bartoszewicz designed a quasi-sliding-mode controller for a discrete-time system in [20] that can reduce chattering caused by high frequency.Qu et al. proposed a DSMC with a disturbance compensator in [10] that obviously reduced chattering.A lot of effort has been made to overcome the chattering problem, such as a state observer, adaptive techniques, and an intelligent control method [21][22][23][24][25].Some other methods also can be applied into DSMC to overcome chattering, like the neural network in [26] which designed the adaptive adjusting parameters in scalar form to reduce the online computation burden and can adjust the sliding surface dynamically and the genetic algorithm in [27] which can optimize the parameters in sliding-mode controller.
Fuzzy control (FC) has been widely used in many practical systems [28][29][30][31][32][33].It is a technique that requires expert knowledge to design a controller which is a model-free design method that is insensitive to variations in parameters and external disturbances.In [29], an adaptive fuzzy slidingmode control is used for nonlinear systems to estimate the unknown gain of the switching control.A fuzzy basis function network is used to approximate the unknown dynamics of a robot with two arms [30] and a fuzzy adaptive state observer is designed to estimate the unmeasured state in [32].Fuzzy control has been widely used to adaptive parameters and optimal control [34][35][36][37][38][39][40][41].An adaptive fuzzy decentralized output-feedback controller was designed for a class of switched nonlinear large-scale systems which contain the unknown nonlinearities and dead zone in [34].Chen et al. [35] proposed an adaptive tracking control for a class of nonlinear stochastic systems with unknown functions.The fuzzy-neural networks were used to approximate unknown functions.In [36], Chen et al. considered the problem of observer-based adaptive fuzzy control for a class of nonlinear time-delay systems in nonstrict-feedback form.The fuzzy logic systems were used to approximate the unknown nonlinear functions.In [37], a hybrid task-space trajectory and force tracking based on fuzzy system and adaptive mechanism were proposed to compensate for the external perturbation, kinematics, and dynamic uncertainties.Liu et al. [38] proposed an adaptive fuzzy optimal control for a class of unknown nonlinear discrete-time systems which contained unknown functions and nonsymmetric dead zone.The fuzzy logic systems were employed to approximate the functions in the systems.In [39], an adaptive fuzzy inverse compensation control method was proposed for uncertain nonlinear system, and an adaptive fuzzy controller was developed to establish the close-loop system stability.For the Henon Mapping chaotic system, Gao and Liu [40] proposed a direct heuristic dynamic programming to handle the optimal tracking control, and the fuzzy logic system is applied to measure the long-term utility function.In [41], an adaptive fuzzy controller was proposed for uncertain nonlinear systems with backlash.The fuzzy logic systems were used to approximate the unknown functions, unknown backlash, and backlash inversion.One feature of fuzzy control is that "if-then" rules are based on expert knowledge [42,43].Fuzzy sliding-mode control (FSMC) combines two theories of FC and SMC, which means the fuzzy control method can optimize SMC dynamically in more detail [44][45][46].For example, when a system trajectory is leaving the switching surface, it can provide a large control force to drive the trajectory back to the switching surface according to fuzzy rules.
In order to reduce the chattering and accelerating convergent speed in discrete-time uncertain systems, a new reaching law is proposed for discrete-time systems with external disturbances.It begins by introducing a discretetime system with external disturbances, the proposed SMC, and then a FMSC is proposed to improve the reaching phase.This method guarantees the robust behavior of the system, whereas FSMC improves system performance based on a performance analysis.Simulated examples are given to verify the proposed method, followed by the conclusions.
System Description
The general description of a single-input uncertain system is as follows: where where is a positive and known constant.It should be noted that the uncertainty means the system contains external disturbance in our system.In order to alleviate chattering, the sample hold effects should be taken into account, so by applying zero-order-hold (ZOH) sampling [47], the discrete-time description of system (1) can be written as where and is the sampling period.
The goal here is to design a suitable FSMC controller for system (3).The switching function () is defined as where = [ 1 2 ⋅ ⋅ ⋅ ] is designed to drive the system to the origin when travelling along the switching surface: The control object is to drive the system to the switching surface defined in (5) and travel to the origin point along the switching surface .
Fuzzy Sliding-Mode Control Design
The variable speed reaching law is a kind of common reaching law [48] that can effectively overcome the chattering problem; its continuous form is and its discrete form is where > 0, ‖()‖ 1 is the state norm, and is the sampling period.
From this reaching law, we get the following controller for system (3): The variable speed reaching law (7) can overcome the chattering problem, but there are two shortages: one is that the chattering will be serious at the beginning of the reaching phase, and the other is that the reaching speed is too slow near the switching surface.These two shortages will seriously affect the system performance.where , > 0.
From this reaching law, we get the following controller for system (3): To ensure the system can meet its performance requirements, we analyzed the system as follows.
Stability Analysis.
To design a sliding-mode controller for discrete-time uncertain system, the following two conditions should be met to guarantee the system trajectory reaching the switching surface and converge to zero in a limited time [49]: Substituting the reaching law (9) to condition ( 11 When () is very small, by using the equivalent infinitesimal, we can get To guarantee (( + 1) + ())() > 0, must be less than 2. When () is big, we get the following formula: So if < 2, then (( + 1) + ())() > 0 is guaranteed.So when < 2 and < 2, the system trajectory can reach the switching surface and converge to zero in a limited time.Specifically, when () = 0, we can get ( + 1) = 0 from (9).So in the next time, () will be zero all the time if there is no disturbance.Therefore, the proposed reaching law (9) can meet conditions (11), which means that the switching surface is existent and the system trajectory will reach the surface in a limited time.
According to the abovementioned analysis, the system trajectory will enter a region near the switching surface and will not go away.So we can conclude that system (3) with controller ( 10) is stable and it will reach the surface in a limited time.
Dynamic Performance Analysis.
From reaching law (9), we can obtain the following two formulas: We define the switching region as where Δ = log 2 (|()| + 1).The width of the switching region is Meanwhile, the width in reaching law (7) by comparing the width between reaching laws ( 9) and ( 7), we find that the width we proposed is smaller than (7), which means the system trajectory will be in a smaller region when it crosses the switching surface.So the system with controller (10) has a smaller chatting than controller (8).
To analyze the parameters of the proposed reaching law, the derived reaching rate is given here.The derivative of a continuous system is defined as follows [50]: Assuming if the sampling period is small enough, the following equation can be obtained: For the first-order differential equation ṡ = (), we give a numerical solution for (20) with different values of and .Here, we assumed (0) = 5. Figure 1 shows how and affect the reaching rate.
Figure 1 indicates that when and are increasing, the reaching rate is faster, but when is larger, the reaching rate near the sliding surface is smaller.The results of simulating sliding dynamics indicate that when and are too big, the chattering problem will be terrible, so based on the experience of and , the reaching law (9) we presented here has further improvements in DSMC.
Further Improvement.
Based on shortages of the reaching law (9), we used fuzzy control (FC) to adjust the parameters and by referring to Figure 1.FC has replaced conventional technologies in many fields.One major feature of FC is its ability to express ambiguous thinking, so when the mathematical model of a system exists accurately or exists but with uncertainties, FC can deal with the unknown process efficiently, although it will sometimes complicate the analysis due to the huge number of fuzzy rules for high-order systems.To eliminate the chattering problem and accelerate the reaching rate in this study, a fuzzy slidingmode control method is proposed.The membership function of our FSMC method is shown in Figures 2 and 3; it is a oneinput-two-output FSMC, and because there is only one input, the number of fuzzy rules is greatly reduced.Fuzzy control rules are designed to map the input linguistic variables to the output linguistic variables and as follows: where FSMC() means that and are the function of switching function .
Here, the membership functions of and are normalized in the interval [0.6, 1.8], such that , > 0 and satisfy condition (11).
In our article, the Mamdani fuzzy inference method [51] is used to determine the fuzzy rule such that if is , then is and is . ( The FC we proposed has one input () and two outputs (, ).Linguistic variables which imply inputs are classified as NB, NS, ZO, PS, and PB, and outputs are classified as PZ, PS, PM, and PB.Inputs are normalized in an interval of [−4, 4], and outputs are normalized in [0.6, 1.8], as shown in Figures 2 and 3.It is possible to assign a set of decision rules as shown in the following based on Figure 1.These rules contain the relationships between input and output that define the control strategy.Each control input has five fuzzy sets so there are only 5 fuzzy rules; this overcomes the problem of complex fuzzy rules.
The fuzzy rule set designed for FSMC is Finally, defuzzification should be designed to determine the numerical value of and based on the fuzzy rule set.Here, the centre average method is used for defuzzification as follows: where is the domain of linguistic variables and .V is the point at which the membership function of ( ) of () achieves its maximum and V (V) is the degree of membership of to .From what has been discussed above, a fuzzy slidingmode controller can be obtained; however, the experience of experts should be referred to if the membership function and fuzzy rule set need to be adjusted for different problems.Here, experience was gained from Figure 1, which shows how and affect system performance.
Numerical Simulation
Consider a second-order system with time-varying uncertain [10] as follows: When the sampling period = 0.1, the discrete-time description of system (25) can be obtained as Here, the initial state was assumed to be (0) = [2.1 1] and = [1 1] .It should be pointed out that a long sampling period was chosen to exhibit the system dynamics.
In the following, we used different controllers to compare the controlling performance of system (26).
Case 1.Consider system (26), when controller (14) in [10] is implemented with the expression that is where = 5 and = 1.System performances are shown in Figures 4-6 and indicate that when the system trajectory is close to the sliding surface, chattering is small, but this chattering cannot reach the origin; it can only surround the origin.
Case 2. When controller (8) with parameters = 1, = 1 is implemented for system (26), then system performances are shown in Figures 7-9, respectively.Unlike controller (28), the reaching rate accelerates at the beginning and chattering decreases, but the reaching rate at the beginning has evidently not improved; this can also be gleaned from Figure 1 where = = 1.
Case 3. When controller ( 8) is implemented for system (26), then by using FLC (21) to adjust parameters and , the switching function, system states, and controller are shown in Figures 10-12.Figure 10 shows that the dynamics of FSMC have improved the reaching rate in all regions and overcome the chattering problem.
From what has been discussed above, the SMC and FSMC methods we proposed for discrete-time system with uncertainty are very capable, and the proposed reaching law performed better than (28).By improving the reaching law with fuzzy logic, the system presents much better dynamic performances at every stage.
Conclusion
This paper has presented a new discrete reaching law, and a discrete sliding-mode controller with fuzzy logical control has been designed by using the proposed reaching law and fuzzy set.A comparison of three cases, that is, (1) the previous DSMC presented by [10], (2) the DSMC presented in the paper, and (3) a DSMC presented with fuzzy control, has been carried out.The experimental results show the obvious effectiveness of the proposed SMC, especially with fuzzy control.
In the method presented, the chattering phenomenon that frequently appears in a conventional SMC was overcome, and the reaching rate has also been accelerated.
|
2018-12-08T18:24:29.571Z
|
2016-08-30T00:00:00.000
|
{
"year": 2016,
"sha1": "eb2a845e259e7e2ccc588e24a9655c2eecf8efca",
"oa_license": "CCBY",
"oa_url": "http://downloads.hindawi.com/journals/mpe/2016/1530760.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "0532404262557ec1818e185b3667943714289f86",
"s2fieldsofstudy": [
"Engineering",
"Computer Science"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
234032160
|
pes2o/s2orc
|
v3-fos-license
|
The phytomeliorating effect of spring rape in rice crop rotation in the Republic of Kalmykia
Mid-crops accelerate rice field soil cultivation and increase rice yields. When spring rape is introduced into the rice crop rotation, the soil acquires an increased ability to restore the soil layer. The bulk density of soil in the rice–spring rape crop rotation decreases by 8 … 10%. When cultivating rice on a reclamation field, this indicator increases by 4 … 6%. It has been established that rapeseed crops have phytomeliorative properties, improve the ameliorative state of rice fields, have a positive effect on the fertility of brown semi-desert soil, and increase rice productivity.
Introduction
To increase the yield of fodder products, in addition to perennial grasses in rice sieve rotations, it is possible to use mid crops, which also serve as the grandfather's reserve for increasing the crop production from a unit of area. One of such crops in rice crop rotations can be spring rape. During the growing season, mid crops make full use of climatic factors such as heat, moisture and fertility of irrigated areas. Such crops allow the soil to be enriched. Cultivation of intermediate phytomeliorant crops allows rapeseed to use heat, silty moisture and fertility of irrigated lands; the additional volume of rice production is obtained in the form of green artemic fodder. The soil becomes enriched with organic matter due to the receipt of root and mowing plant residues behind the plant. This contributes to an increase in rice productivity and accelerates the processes of soil rape cultivation [1].
As a fodder crop, spring rape is a necessary source of fodder protein. The fodder and nutritional value of spring rape is much higher than in many other agricultural inundated crops. The green mass of rape, which is a valuable type of fodder, has a higher protein content than legumes [2].
The aim of the study is to increase the productivity of rice fields by cultivating spring rape as a phytomeliorant. It also increases the amount of moisture after rice cultivation.
Materials and methods
In the conditions of the rice irrigation system, located on the territory of the Federal State Unitary Enterprise "Kharada" of Oktyabrsky District of the Republic of Kalmykia, experiments were conducted to study the formation of the aboveground mass and plant residues of spring rape. The territory of the experimental site is represented by brown semi-desert medium and heavy loamy soils 2 with the following indicators: bulk density of the arable layer is 1.27 ... 1.32 t/m 3 , down the profile, the bulk density increases and in the 1-meter layer it is 1.68 t/m 3 , the nitrogen content in the active layer is low -35.2 ... 49.1 mg/kg, the content of mobile phosphorus is increased -36.3 ... 41.2 mg/kg, the content of exchangeable potassium is high -431 ... 464 mg / kg, the lowest moisture capacity in the layer of 0 ... 0.4 m is 1.12 ... 1.26%. The soils are weakly and moderately saline, the type of salinity is chloride-sulfate, the total of r soluble salts in the meter layer is 0.103 to 0.256%. Ground waters are found at a depth of 1.4 ... 2.0 m, chloride-sulphate-sodium-calcium with a salinity of 2.9 ... 4.2 g / l [3].
Results
The important agrotechnical value of spring rape as a catch crop has been established when it is cultivated in a rice crop rotation as a reclamation crop. Inaccessible soil phosphates are converted into forms that are assimilable, which significantly improves the phosphorus nutrition of plants. The aboveground mass of spring rape is rich in protein (up to 32% for absolutely dry matter), vitamins, minerals (phosphorus, calcium, sulfur, etc.). As a green fertilizer with systematic plowing for seven years, the aboveground mass of catch crops increases the content of humus and total carbon by 0.17 and 0.12%, respectively, even with long-term permanent cultivation of rice.
Thus, an increase in the productivity of the main crop in the rice crop rotation, as well as the creation of optimal conditions for the growth and development of rice are possible with the cultivation of catch crops as green fertilizers [4].
It was found that with permanent sowing of rice, the density of soil in the active root layer (0 ... 40 cm) increases by 2 ... 3%. The ability of soil to restore the optimal structure can be improved by including spring rapeseed in the rice-rice link of the rice crop rotation. When cultivating the main crop after reclamation and inclusion of spring rape, soil density reduced by 8 ... 10%. An analysis of the data showed that in the rice-spring rape crop rotation link, the total porosity and porosity of aeration increase by 4 ... 6% and 8 ... 11%, respectively. They become optimal in the cultivation of spring rapeseed in the ameliorative field of rice crop rotation. When spring rape is included in the rice crop rotation, the cultivation has a positive effect on the structural condition of soil. During the growing season, spring rape forms a dense herbage, with a significant increase in the aboveground mass and root system. At the same time, the amount of the most valuable soil aggregates with a particle size of 0.25 ... 10 mm increases by 9.96 ... 16.06%, respectively, and the density of soil composition decreases, which favorably affects the development of agricultural crops.
The predominance of aerobic processes in the soil causing the redistribution of soil fractions was observed during the cultivation of a dry land crop. Such processes occur due to a decrease in dusty particles and an increase in the share of valuable aggregates (Figure 3). The structural coefficient increases by 0.55 ... 0.78. The bulk of rice roots is located at a depth of 20-30 cm. The value of the arable horizon for rice is more important than for many other crops. Improving the nutrition of rice crops and maximizing the preservation of fertility are prerequisites for soil restoration. These conditions are possible only if they are provided with an energy material. An increase in the biological activity of the soil of rice fields and its fertility is possible due to the additional supply of plant residues of spring rape [5]. According to the data obtained, the total amount of plant residues of spring rape varied from 1.96 to 4.14 t / ha (Table 1). The largest amount of root and crop residues is observed in the variant of nitrogen fertilization at a dose of N120 kg / ha a.i.
When included in rice crop rotations, catch crops contribute to the more active humus formation. Chemical and microbiological soil processes during the decomposition of organic matter change with constant flooding of rice fields. In flooded old arable fields, the decomposition of organic matter occurs at a slower pace due to a weak oxygen supply. [6].
The upper arable horizon of the rice paddy soil is usually more fertile. Plowing rapeseed residues into the surface soil horizon of rice fields creates an additional supply of organic matter to the soil, which significantly enhances its biological activity. At the same time, the main nutrients become more accessible for rice plants, the conditions for their absorption are improved. It structures the arable horizon.
Conclusion
It was found that rape crops have a phytomeliorative effect on the fertility of brown semi-desert soil, the reclamation state of rice fields improves, and the yield of rice increases. The soil and climatic conditions of the Sarpinskaya lowland meet the biological requirements of spring rape cultivated in a rice crop rotation. The phytomeliorating effect of spring rapeseed in rice crop rotation is as follows: improvement of water-physical properties (porosity of aeration and total porosity increase by 8 ... 11% and 4 ... 6%, respectively); bulk density is reduced by 8 ... 10%; the number of valuable soil aggregates (0.25 ... 10 mm) increases by 9.96 ... 16.06%, the structural coefficient increases by 0.55 ... 0.78; the risk of flooding the territory and the level of groundwater are reduced by 34%. The decomposition of rapeseed residues contributes to the intake of organic matter up to 4.3 t/ha, which increases the humus content by 14 ... 17%, the phytosanitary state of rice checks improves (by 43 ... 76%), and the productivity of rice culture increases by 0.43 ... 0.52 t / ha ...
|
2021-05-10T00:03:30.426Z
|
2021-02-01T00:00:00.000
|
{
"year": 2021,
"sha1": "9b122a71aa04ed88b0a7363261ace61fb5a68878",
"oa_license": null,
"oa_url": "https://doi.org/10.1088/1755-1315/659/1/012130",
"oa_status": "GOLD",
"pdf_src": "IOP",
"pdf_hash": "b34e29cfc90ad89881b3c98b95eedf1d56bae634",
"s2fieldsofstudy": [
"Agricultural And Food Sciences"
],
"extfieldsofstudy": [
"Physics",
"Geography"
]
}
|
254444202
|
pes2o/s2orc
|
v3-fos-license
|
Hearing Minority Voices: Institutional Discrimination Towards LGBTQ in Disaster and Recovery
Within the themes of CASCADE NET, this paper focusses on less heard voices and the need to develop new social spaces. Disaster vulnerability identi fi es diversity in society through a lens of constraints to solutions, on such bases as demography, socio-economic status, cultural, ethnic and gendered minorities within society, and marginalized groups as well as physical proximity to a hazard. The focus of disaster risk reduction is on building resilience through the strengths and capacities in society, but it has a tendency to ho-mogenize characteristics of resilience to the community level, thereby fl attening and hiding diversity. LGBTQ people are largely ignored as minority groups with speci fi c information needs. Speci fi c response and recovery processes and actors exacerbate the vulnerability of the LGBTQ minority, especially in evacuation, support, counselling, and rehousing. The role of faith-based organizations (FBO) in providing these services during disaster relief and recovery is examined in this paper. This paper identi fi es and critiques the attitudes and practices of some FBO towards LGBTQ groups in their provision of disaster relief services.
during extreme weather events. The vulnerability of LGBTQ people is exacerbated during disaster events as a consequence of homophobic discrimination and insensitivity towards their needs. Within this identification of enhanced vulnerability, a problematic issue that is identified in the literature concerns a trend towards privatization of disaster relief and recovery support services to faith-based organizations (FBO), especially conservative Christian agencies that are hostile towards LGBTQ people. The danger of increasing discrimination towards LGBTQ people by such FBO during and following a disaster is a primary focus of this paper. The aim is to identify the risk in order to enable policy and practice strategies to reduce that risk and enhance the resilience of this minority group in the face of disaster.
First, this paper briefly examines the broader context of Disaster Risk Reduction before examining a range of themes around LGBTQ vulnerability that are identified in the literature, and especially the increasing roles of FBO in relief and recovery. Since the 1990s, International Decade for Natural Disaster Reduction (IDNDR) emergency management has shifted its emphasis from vulnerability to resilience, with international strategies aimed at building and enhancing resilience. Diversity is acknowledged in our understanding of vulnerability, but less so in resilience enhancement. Vulnerability is defined in relation to different demographic, social economic, cultural, and special needs groups (Anderson-Berry and King 2005). While vulnerability recognizes the susceptibility to risk in individuals and groups, resilience builds on the strengths of people and communities. Consequently, resilience tends to focus on the whole community and is less sensitive to diversity. Within a strategy of building resilient communities, communication, warnings, education and messages are aimed at an amorphous public (King 2010). There is a tendency to homogenize society. The complication of these two concepts of vulnerability and resilience is that people are both vulnerable and resilient at the same time (Handmer 2003;Gurtner and King 2021). These are not at opposite ends of a single scale but are interacting parallel conditions of being.
One response amongst disaster researchers and emergency managers has been to attempt to diversify messages (Young and Jones 2022). There is a recognition of differences in demographics, class and social economic status, culture and ethnicity, language, geography, and gender. Thus, as an example, the Centre for Disaster Studies, James Cook University, Australia pursued work on awareness and education for children and subsequently for Australian indigenous people as examples of diversity as much as targeting specific needs.
In disaster risk reduction, the emergence of the recognition of LGBTQ people and communities has mostly taken place in the last couple of decades, although it should be noted that it is within this same timescale that emergency management has attempted to diversify its messages. There is some linkage through gender and feminism, but increasingly, the emphasis is specific to diverse sexualities and the difficulties of LGBTQ people in facing social acceptance. Whereas the need to recognize diversity in relation to emergency management messages, guidelines, and warnings was focused on appropriate language, terminology, and understanding, the social marginalization issues in relation to LGBTQ members of society are less in language (although heteronormative terms, attitudes and assumptions distort messages and awareness) than overt exclusion in some instances from disaster relief, services and facilities, and recovery support. Social homophobia that exists in day-to-day attitudes to LGBTQ people is sharpened and in some situations has emerged as overt discrimination during the phases of disaster relief and recovery.
Method
A literature review was carried out through a number of internet search engines using the variety of abbreviations around the terms lesbian, gay, bisexual, trans, transgender, transsexual and queer, intersex and plus in their various combinations that are defined and discussed below. These were combined with the terms disaster, crisis, climate change, extreme events and sub sets of response, recovery and emergency management. The range and breadth of experiences and issues is indicative rather than exhaustive. The review in turn identified a limited number of related reviews which bring together findings that are summarized here. Rather than repeating a similar summary of each paper identified in these reviews, this paper identifies the issues that are reported in each paper. In particular, this paper seeks to identify specific issues and finds the role of FBOs in disaster relief service provision a problem that requires both further research and ameliorative action.
The literature of LGBTQ experiences that are summarized in this paper identifies loss of privacy, discrimination, rejection, condemnation and in some societies, vulnerability to arrest and imprisonment. This paper therefore reviews the literature of experiences from the perspectives of: (1) Emergency management messages and diversity; (2) LGBTQ experiences in disasters; (3) FBO in the LGBTQ experience; (4) LGBTQ resilience and leadership in disaster.
As LGBTQ people have gained acceptance and recognition in some societies, research and analysis of LGBTQ as minorities has gained traction with focus in areas of feminist and gender geography where gay and lesbian spaces and queer theory have generated an extensive literature that is not the focus of this paper (Hubbard 2016;Gaillard et al. 2017). Against this background of scholarship, a consideration of LGBTQ experience in disasters has emerged in recent decades (Larkin 2019). Larkin (2019) carried out a global review of LGBTQ experiences. She identified 172 papers and reports, narrowing these down to 38 which were ultimately selected for analysis of themes and issues. The issues she identifies are cited in this paper as a summary of her review. Wisner andGaillard (2009), Larkin (2019) and Dominey-Howes et al. (2014) refer to a lack of research "specific to gender and sex minorities in disasters" (Larkin 2019: 61), but Larkin's (2019) literature review captures a range of issues, shortcomings, and policy responses across a number of nations -USA, India, Nepal, Samoa, Indonesia, Haiti, Australia, New Zealand, Japan, and Philippines (also Knight and Sollom 2012). Larkin's (2019) review lists discrimination and vulnerability, policy gaps, lack of LGBTQ access to relief combined with lack of trust of agencies by members of the LGBTQ community, along with the destruction and loss of safe places and problems associated with relief provision by FBO. She also summarizes the resilience of LGBTQ groups, the contributions of many LGBTQ people to DRR leadership and relief, and the emergence of inclusive policy. A more specialist coverage in the Forced Migration Review journal (Couldrey and Herson 2013) focused on the experiences of LGBTQ refugees in 26 case studies. While 10 of the articles are generically global, the plight of LGBTQ refugees is identified in Eastern Europe, the EU, Myanmar/Burmese migrants in Thailand, Canada, Brazil, UK, Korea, Kosovo, Bolivia, USA, Israel, Kenya, and Nepal. Dominey-Howes et al. (2014) approached the experience through five case study disasters in the USA, Haiti, India, Philippines, and Indonesia to identify issues and service needs, concluding by calling for more extensive baseline research. By 2021, Goldsmith and Raditz reviewed an extensive literature of the impact of disasters on LGBTQ communities, focusing on the USA, that identified issues of limited service provision, recognition of LGBTQ, the problems associated with discriminatory practices of FBOs, and entrenchment of existing inequalities. Their review identifies issues of bias and discrimination in service provision to LGBTQ people, a lack of recognition of LGBTQ individuals and couples, problems with services provided by FBOs, as well as poverty, homelessness, mental health, health needs amongst the LGBTQ community members and outright blame being placed on LGBTQ people by FBOs, especially those of the fundamentalist Christian right. Recommendations from their review focus on anti-discrimination training, community-based response and health care. However, they also point to a lack of empirical studies (Goldsmith and Raditz 2021), a lack earlier identified by Gaillard et al. (2017). This lack of empirical studies is noted by Goldsmith and Raditz as contributing to an absence of LGBTQ recognition in DRR policy.
Examples of surveys of LGBTQ respondents are recent and most are quite small (Carman et al. 2020;Parkinson et al. 2018;Haworth 2021) but COVID-19 has prompted fresh research such as Grant et al. (2021) which elicited 231 responses from LGBTQ respondents. Privacy and trust are disincentives to participation in surveys, despite ethical assurances, but greater constraints to web-based surveys are researcher identification of networks and permission to share requests to ensure the safety of respondents. Despite these research constraints, surveys and interviews illustrate and reinforce themes and issues of LGBTIQ experiences. Thus, there are already significant research studies and case studies that underscore the difficulties faced by LGBTQ people during and after disaster.
Emergency Management Messages
Following the UN IDNDR, the top down command and control model of emergency Management has steadily transformed (King 2004a) towards inclusion, cooperation, and recognition of diversity. The relative recency of this change in emphasis partially explains why many minority groups in society have been neglected. Furthermore, the heteronormative language of emergency management information (Gorman-Murray et al. 2019) extends across society in media, advertising, public information and general assumptions on what is normal (Hubbard 2016). Minorities and subgroups are recognized and often addressed, but under an umbrella of heteronormative images and assumptions. The information needs of indigenous people in Australia, for example, have for a long time been ignored because of a cultural blindness (Wensing 2012(Wensing , 2016. This blindness is pervasive and extends beyond a failure to recognize indigenous people (Gaillard et al. 2017) LGBTQ people are confronted with information and hazard awareness campaigns that often portray nuclear families or heterosexual couples doing the right thing. Gaillard (2022) suggests that the language of emergency management and DRR is western in origin, thereby assuming a cultural norm that exacerbates blindness or fails to account for cultural complexity, especially the needs of minorities. Gaillard asserts that gender is presented as binarynot diverse.
At the very least such heteronormative portrayals are irritatingmany LGBTQ people have long grown accustomed to such images and informationbut more pervasively, it goes over the heads of individuals, failing to engage them because it is not about them. This is how multiple minorities are excluded from safety messages (King 2004b). Education and awareness raising are important issues for hazard preparation and personal and community safety, where greater diversity can and needs to be incorporated into the safety messaging. However, most of the literature reviewed is focused upon LGBTQ experiences during and after disasters during the relief and recovery phases.
Themes in the Literature Concerning LGBTQ Experiences in Disasters
The review that was carried out in preparing this paper identified five groups of themes and issues. These are diversity of and within the LGBTQ minority, discrimination, issues around service provision and health needs, inequality risk and exclusion, privacy and trust, and the primary focus of this paper, the part played by FBO in LGBTQ Disaster Experiences. It is because some Christian FBO are increasingly part of a trend towards privatization of relief and recovery services that each of these issues and themes needs to be explained. The five issues that are identified below are not separate from the role of FBO, but rather are exacerbated by some of those agencies. This paper summarizes these issues and themes and then focusses on the problem of discriminatory FBOs in provision of disaster relief and recovery services.
Diversity
A crucial starting point to a review of problems and issues is the recognition of a lack of homogeneity of the LGBTQ community (Dominey-Howes et al. 2014).
The papers and reports that have been reviewed for this paper use a range of terms where LGBT is common to all, but variants include I for intersex, Q for queer or questioning and þ as an umbrella for several additional positions of sexuality. Ng (2015) uses LGBT in reference to Chinese activism on the grounds that the Q is not used in China, although she recognizes the increased popularity of LGBTQ. It would be correct to cite the exact abbreviation as it is cited in each paper, but this can be awkward in terms of changing use throughout this paper, and many experiences and issues are cited by multiple scholars who often use slightly different abbreviations. This paper uses LGBTQ for convenience and inclusivity, but stresses that it represents considerable diversity within a population of diverse sexuality.
LGBTQ, is used by the Q Christian Fellowship, formerly the Gay Christian Movement, whose organizational name change and choice, with a þ as well, is illustrative of its attempt to extend inclusivity among diverse people who are victims of active discrimination but who are exercising their need to come to terms with a discriminatory organizationthe church. Dwyer's (2022) tool for evaluating inclusion uses the term Sexual Orientations, Gender Identities and Expressions and Sex Characteristics (SOGIESC) as a substitute for LGBTIQþ. The complexity of the variance of the abbreviations for LGBTQ symbolizes the diversity of the minorities within a minority. There is no single experience, socially, and specifically in disaster (Dominey-Howes et al. 2014;Haworth 2021). Each disaster is unique as is the impact on individuals Gaillard et al. (2017) adding that while there is a general recognition of minorities in social service provision, this is largely absent in disaster risk reduction.
Service provision and health needs
Discrimination is practiced in the refusal of agencies, such as NGOs that are tasked with disaster relief, including and especially some FBOs, in their provision of health and service needs, as well as access to services (Parkinson et al. 2018;Larkin 2019). A specific issue identified by Gaillard et al. (2017) and Stukes (2014) is the provision of appropriate bathroom facilities for transpeople in evacuation centers. There are jurisdictions that do not recognize LGBTQ partners (Haskell 2014) and even where legal rights exist, partners are excluded from some health provisions such as recognition of same sex partners as next of kin during disaster response (Haskell 2014). Inadequacy of health needs provision and specifically mental health care to LGBTQ are recognized by United Nations Human Rights (2020), Grant et al. (2021), and Gorman-Murray et al. (2019).
Inequality risk and exclusion
Despite public leadership and articulation by many middle-class gay and lesbian individuals, there is a general pattern of lower social economic status, homelessness, unemployment, and economic disadvantage in the diverse LGBTQ community (Goldsmith and Raditz 2021;Carman et al. 2020;UN 2020;Haskell 2014;Stukes 2014). Inequalities across the diversity of LGBTQ people and communities are exemplified in racial and gender divisions (Haskell 2014;Stukes 2014). Transindividuals and partners are especially vulnerable, both in terms of general poverty and intensification of discrimination and exclusion in a disaster (Carman et al. 2020). Consequently many LGBTQ people are marginalized, living on the edges in terms of housing and income (Goldsmith and Raditz 2021). Recently, the COVID-19 pandemic has intensified pre-existing social and economic inequalities (Haworth 2021(Haworth , 2022 adding significantly to mental health impacts (Grant et al. 2021). During disaster response, evacuation and relief, risk has been enhanced for many LGBTQ people, including incidents of violence in evacuation centers (Parkinson et al. 2018), as well as increased domestic violence (United Nations Human Rights 2020) and access to appropriate facilities (Nicholson 2022).
Privacy and Trust
Haskell (2014) and Stukes (2014) wrote against a background case study of LGBTQ people's experiences in Hurricane Katrina. They point out that for many people, this disaster was the first time of being out in public. Haskell (2014) argues that many LGBTQ people are privately out but publicly closeted. The idea of a closet can be somewhat divisive amongst LGBTQ communities as it is associated with being hidden from sight in contrast to Gay Pride and the experience of being 'out' in the mainstream community. Being out is not simple for many LGBTQ people. It may invite discrimination, loss of employment and career, and homophobic violence. In complex roles of service provision, LGBTQ individuals may be faced with subjugating their own needs and rights in front of the albeit misguided conservative attitudes of the clients or the community they serve. This is typified in the dilemma of LGBTQ religious leaders and workers whose mission is to serve people who would otherwise reject and vilify them. Haskell refers to "stories of closeted LGBT unheard" (2014:10). People seek privacy for security and safety (Carman et al. 2020;Grant et al. 2021;Larkin 2019;Goh 2018). Safe D King 2241005-8 spaces are lost in a disaster, in some cases physically destroyed (Larkin 2019;Haskell 2014), but in others interrupted or temporarily lost, including online places (Grant et al. 2021;Gaillard et al. 2017). The reciprocity of trust between institutions and LGBTQ individuals is a repeated theme aimed at enhancing resilience and communication and reducing vulnerability (Haworth 2022;Gaillard 2022;Dwyer 2022;Duckworth 2022).
LGBTQ people seek privacy and safe places. Being out, LGBTQ may invite danger or discrimination or may restrict careers and community roles of service to others. Surveys and reviews report a lack of trust of mainstream service providers (Carman et al. 2021;Larkin 2019). Stukes (2014) identifies the loss of specific safe places in New Orleans following Hurricane Katrina. The loss of a local Metropolitan Community Church (MCCa diverse affirming church organization) and subsequent involvement of other MCCs played a significant role in terms of initial loss and then the recovery of safety and support. Even without a lack of trust, LGBTQ respondents identified heteronormative information, and specifically language and advice that is unfriendly towards transpeople (Haskell 2014;Larkin 2019).
All of the issues that are summarized above may be targeted and their negative impacts reduced, by strategies and policies that encourage best practice responsibilities on the part of individual service providers and their organizations. Disaster relief and recovery programs may be evaluated in terms of either their effectiveness or bias. For example Dwyer (2022) presents a risk assessment matrix as a rapid assessment tool of inclusion of LGBTQ (SOGIESC) people in disaster assistance programs. This identifies a continuum of hostile, unaware, inactive, inclusive, through to transformative. This matrix is subdivided by impact, cause and examples. The basic risk evaluation matrix is summarized in Table 1. Each cell in Dwyer's original table (Dwyer 2022, p. 35) identifies detailed examples of situations and strategies. These are not reproduced in Table 1 which is a summary of the matrix categories Under hostile practices, the impact is identified as 'norms-based marginalization and exclusion of people with diverse SOGIESC is exacerbated' (Dwyer 2022, p. 35). In the next cell of the matrix, the cause is identified as 'The organization is aware of likely negative impact on people with diverse SOGIESC but goes ahead anyway because either it chooses not to address diverse SOGIESC issues or actively discriminates against people with diverse SOGIESC' (Dwyer 2022, p. 35). It is especially significant that for a tool that was developed in the Pacific Island nation of Vanuatu, the hostile example that is cited is 'A faith-based organization is contracted to deliver relief, however its theology commitments or those of its in-country partners cast people with diverse SOGIESC as sinners; OR a secular organization puts aside SOGIESC concerns because they prefer to use the funds elsewhere or do not want to deal with the complexities of this work.' (Dwyer 2022, p. 35).
However the matrix cells of impact, cause and examples become progressively less harmful in each of unaware and inactive, and are positive and thus desirable project goals in the inclusive and transformative columns. The impact of a transformative goal is 'norms-based marginalization and exclusion of people with diverse SOGIESC is ameliorated and challenged', the cause is that 'the organization has developed competency to challenge norms-based discrimination that excludes people with diverse SOGIESC. It has revised its ways of working and has programs and partnerships that positively include people with diverse SOGIESC in mainstream programs while offering targeted alternative programs where safety requires' and an example is 'A cash-based social protection program designed in partnership with diverse SOGIESC CSOs and accounts for the impact of diverse SOGIESC marginalization on family and community relationships. The program provides holistic support that addresses longer-term livelihood challenges and counters community stigma' (Dwyer 2022, p. 35). The value of this tool is in applying it across a range of cultures, locations, and extreme events. In particular, it turns the focus round from the bad and harmful, while clearly identifying what these harmful actions are, to the positive inclusive and transformative approaches that enable resilience, capacity and trust to be built with LGBTQ people (here abbreviated to SOGIESC).
While homosexuality remains a criminal offence in some countries (69 countries mostly in Africa and the Middle East, Reality Check team 2021), there have been many positive trends towards acceptance, tolerance and inclusion. Government agency processes, policies and legislation, especially in western countries, acknowledge diversity, even though in Disaster Risk Reduction there is some way to go. Where direct government services are involved in disaster relief and recovery it is relatively straightforward to evaluate projects and train employees to do the right thing. However, services which are provided by FBO may involve some agencies and individuals who filter their responses and treatment of LGBTQ people in particular, through a lens of scriptural dogma.
Faith-Based Organizations in LGBTQ Disaster Experiences
It is in the areas of relief and services provided by FBO that researchers have identified the most blatant prejudice and discrimination (Stukes 2014;Richards 2010;Haskell 2014;Goldsmith and Raditz 2021). Governments in many parts of the world have either devolved welfare services inadvertently to FBO, or have left such services to organizations that have long provided relief, evacuation centers and temporary shelter (Wisner 2010). On the face of it, this is a rational sharing of the workload. Throughout the world, faith-based agencies operate at the smaller scales of local involvement, with buildings, personnel, and ideologies that urge the provision of assistance and care to people in need and suffering (Ng 2015;Hackworth and Akers 2010).
Religious faith provides motivation and purpose for citizens to want to help people who are in need or are suffering (Ng 2015;Hackworth and Akers 2010;Ager et al. 2015). Not only from the perspective of the giver, the recipients of services from FBO may experience a greater spiritual support and meaning that goes beyond the purely physical support and assistance (Ager et al. 2015). At times of disaster, many people who are not normally religious, turn to faith institutions for comfort and meaning. This transcends the world religions. Islamic, Buddhist, and Judaistic as well as Christian FBO respond to community needs in disaster (Ferris 2005;Baidhawy 2015;Wirtz and Ecke 2014;Ng 2015;Kawai 2014;Wisner 2010). The term FBO includes everything from local churches, mosques, temples or synagogues, etc. through to the hierarchy of the religion and service provision organizations that they have established to meet specific needs. Involvement of FBO is global and cross cultural, and scales from the local to international. In many instances, probably for most people, the response and relief experience is positive and supportive. However, minorities are vulnerable to pre-existing prejudices and social vilification attitudes, which in the context of this paper focusses on the activities of some conservative Christian agencies. LGBTQ individuals may have avoided conflict or confrontation through privacy and a low profile, but in a disaster people are forced into the sphere of influence of both government and non-government organizations including FBO.
At the local level, immediate response and recovery is driven by local individuals and community groups. LGBTQ people may be limited in their access to a range of services in small communities, but in such places, the providers may already be known. Post-disaster services and relief are provided by government departments, from central to local and NGOs as well as FBOs. In a small community where people may be members of multiple clubs and organizations, relief services are provided by known community members rather than a faceless
2241005-11
organization, where levels of trust and acceptance already exist (Wisner 2010). Unfortunately, a simple local trusted relationship in small communities is not the experience of most people, primarily because we live in large urban centers. During a disaster, local service providers are overwhelmed anyway. Community self-help breaks down in more severe disasters where local capacity may be destroyed and relief is brought in from outside by strangers. In a government-based hierarchy, this ramps up to calls for assistance from larger jurisdictions.
At the same time, throughout the world, governments have downsized and privatized services; even authoritarian regimes such as China (Ng 2015). In the developing world, governments are frequently weak and under resourced. Alongside such limitations, two processes have driven the shift from government provided services to increasing reliance on non-government organizations (local, international, and faith-based). In the developing world, because governments are under resourced and often weak in terms of governance, there is a reliance on nongovernment organizations (King 2004b). In the developed world, the trend has been towards downsizing of government services generally, but also as part of the neoliberal agenda. This has resulted in devolution of services to the private sector, which includes FBOs (Hackworth and Akers 2010;Ng 2015;Riera and Poirier 2014). In some cases, the devolution to FBOs was probably unintended but in societies such as USA, many Christian organizations are strongly committed to neoliberalism and supportive of conservative governments (Hackworth and Akers 2010). Concurrently, these FBOs are frequently anti-big government, answerable to their faith base and independent of government processes in terms of training, practice and inclusion. FBOs enter communities and countries with a set organizational agenda that is often not modified to meet local needs (King 2004b). They provide services in which they are skilled and committed with minimal reference to local needs, while many at the same time proselytize. Riera and Poirier (2014) describe UNHCR reliance on faith-based agencies in the running of refugee camps, and identify discrimination practices against minorities and people of other faiths as well as LGBTQ. Couldrey and Herson's (2013) edited collection lists 26 case studies of discrimination and poor practice in relation to forced migrants with specific focus on LGBTQ. Hackworth and Akers (2010) cite the example of NORM, the New Orleans Rescue Mission. NORM was positively supported by a United States government that was not only set on a purposeful neoliberal devolution of services, but which had also run down Federal Emergency Management Agency (FEMA) to the point where it was not able to provide proper response to the catastrophe of Hurricane Katrina. NORM required relief recipients to attend church, were they were proselytized by a fundamentalist interpretation of faith and judgment. Many other FBO involved in relief in New D King
2241005-12
Orleans after Katrina were also very evangelical. These state supported FBO had little or no association with pre-existing churches, especially African-American. Thus when they emerged as major providers of relief services after Hurricane Katrina they had no local relationship on which to build in many of the poorer parts of New Orleans. Stukes (2014) additionally cites the initial incapacity of the LGBTQ Metropolitan churches (MCC) to service its own community because of alienation of LGBTQ people by the broader evangelical churches that had generated a distrust of all FBO.
Within Christianity (as well as Islam and other world religions which are not the focus of this paper), a major rift has long existed between the evangelical, fundamentalist Christian right, and the liberal traditions which are most often identified in the more established denominations. It is against this trend of service devolution to FBO that increasing relief and care falls into the hands of some intolerant organizations, as well as the more tolerant liberal FBO. FBO are extremely diverse. This critique of negative and intolerant practices is specific to some FBOs.
LGBTQ experiences in disaster identify homophobia, stigmatization, blame and prejudice (Parkinson et al. 2018;Goldsmith and Raditz 2021;Richards 2010;Haskell 2014;Stukes 2014). The same researchers, as well as those who focus on FBO (Ng 2015;Hackworth and Akers 2010;Ager et al. 2015;Stukes 2014;Ng 2015), cite examples of evangelists placing blame for disaster on LGBTQ people as God's punishment for sins. Consequently some FBO providing relief services have practiced exclusion, refusing to assist LGBTQ people and especially those who are trans (Stukes 2014;Riera and Poirier 2014;Richards 2010;Couldrey and Herson 2013;Ng 2015). Ng's (2015) case study of Uganda explains how U.S. based evangelists whipped up homophobia in a predominantly Christian population in the period leading up to 2014 when the death penalty was introduced (and later annulled) for some homosexual acts. However, she also shows how FBO specifically Christian International Non-Government Organizations worked with the Chinese government to provide welfare services which included LGBTQ recipients. Similar inferences may be drawn from the Hurricane Katrina experience (Stukes 2014;Hackworth and Akers 2010) that suggest that government leadership and control of FBO is critical either in tacitly condoning, or actively supporting discrimination by FBO, or alternatively in framing inclusionary non-discriminatory provision of aid. Dominey-Howes et al. (2016) specify sound recommendations to direct FBO in providing relief, and the staffing of evacuation and recovery centers in such a way to include LGBTQ people within the broader legislative framework that outlaws discrimination. There are contradictory positions where a government (such as Australia's legislative proposal on religious discrimination) would permit discrimination by FBO on the grounds of theological interpretation and freedom of religion, while simultaneously outlawing discrimination in terms of human rights, specifically as practiced against LGBTQ individuals.
Dominey-Howes et al. (2016) make a specific recommendation calling for LGBTQ people of faith to examine how they manage relations between faith and secular society and especially in the disaster situation. This is very much easier said than done. In this review paper, it is this statement that generates a research question that is significant to the experience of LGBTQ people receiving disaster relief from FBOs. Stukes (2014: 73-74) in her thesis explains her personal involvement as an African American Lesbian emergency manager. The experiences identified in many of the articles cited in this paper are already evident from qualitative research of the stories and accounts of rejection and discrimination in the Christian church in its day to day, pre-disaster dealings with its LGBTQ members. The experience of LGBTQ Christians in terms of their general selfconflict and direct conflict with their churches and communities is a focus of background research that is compiling faith discussions and written explanation (QCF 2021). The diverse LGBTQ Christian community struggles in bringing change, or at least acceptance and safety, within their churches. These tensions and conservative attitudes then extend into the disaster event when FBOs provide relief and recovery services, intentionally or inadvertently bringing discriminatory attitudes to strangers in the wider community. FBOs are ill suited or unable to accept their own LGBTQ members, they are by extension liable to disengage and marginalize LGBTQ people in the broader community.
Many LGBTQ people are distrustful of religious organizations, and like a significant proportion of the population in the Western world, do not attend a religious institution. They distrust FBO because of their stated condemnatory attitudes, their vilification and blame (Goh 2018). These attitudes can be avoided most of the time for most LGBTQ people by staying away from religion and having nothing to do with it. However, for many LGBTQ people, the most direct confrontation with the attitudes of FBO will come when there is no alternative relief provider in a disaster. At such a time, the vulnerable individual and the intolerant organization come into contact and confrontation.
LGBTQ people of faith on the inside of the religious organization can experience far worse. The Q Christian Fellowship (formerly Gay Christian Movement) published a series of personal accounts by members of their struggle to come to terms with sexuality, faith, rejection, and acceptance (QCF 2021). Experiences of LGBTQ people inside the church and especially those in evangelical fundamentalist denominations extend through rejection by family and community, self-D King
2241005-14
doubt, self-harm and suicide attempts, depression and substance abuse, the abusiveness of gay conversion, dismissal, expulsion, verbal abuse, condemnation and threats.
LGBTQ people who remain in FBO do so despite the hatred and vilification of them either as individuals or as members of a minority. In most FBO, the task of negotiating with or changing the organization is extremely difficult. Some Christians move to tolerant or affirming, or even specifically LGBTQ churches, such as Metropolitan Community Church, (MCC) but these are small in number and location. Most LGBTQ people of faith find support networks and maintain work and service in organizations that reject and vilify them. To effect a change from within is a mammoth taskit is hard enough to survive and maintain faith.
In some denominations, progress has been made in achieving inclusiveness through affirmation. Churches have declared themselves affirming towards LGBTQ members and are put forward as appropriate FBO to offer nondiscriminatory fellowship and service to the broader community. The Uniting Church in Australia is such a broadly affirming church that theoretically welcomes LGBTQ people. However, this extends neither to all individuals nor to whole states, such as Queensland, which rejects gay leadership and same-sex marriage. Especially in the north of Queensland, churches tend to reject and vilify LGBTQ people, despite secular laws which prohibit this, and despite an overall affirmatory declaration from the whole church.
Leadership and resilience
The CASCADE NET will focus on community agency in adaptation to extreme weather events seeks to identify capacity for resilience in the community, in this case, a minority vulnerable community. Disasters are mostly short duration events that must be structured by emergency managers to lose minority and cultural blindness in order to encompass and serve all members of the community according to their needs. This requirement for emergency management as a state organization is relatively straightforward and is backed up by legislation, but it will still take some time to achieve effective inclusivity. All of the scholarly articles that have been reviewed in this paper contain sound recommendations concerning training, minority and specifically LGBTQ awareness, inclusive language and imagery, cooperation and resilience building. Many countries and jurisdictions are acknowledged as possessing appropriate policies, but these are often not practiced fully or even at all by some operational agencies (Duckworth 2022). To tackle this gap between intention and practice, Duckworth outlines a mechanism for a radical transparency to build trust between minority communities and relief agencies.
2241005-15
Apart from being victims of discrimination, LGBTQ people, coming from a very diverse minority are also leaders, careers, and innovators. Gorman-Murray et al. (2019: 4) state the "remarkable resilience, social capital, and adaptive capacity within LGBTQ communities". Pre-existing networks and support structures exist within LGBTQ communities. These must be sought and activated during times of crisis and disaster. Stukes (2014), Dominey-Howes et al. (2014), and Grant et al. (2021) draw attention to the strength of leadership and resilience that exists among LGBTQ people and their communities. This also exists within strong affirming and supportive, predominantly straight, extended families in their love and protection of their own LGBTQ members. Equally important are the leadership roles played by many LGBTQ people in mainstream society, who will be part of the response and recovery of the wider communities in which they serve. There are, for example, lesbian and gay emergency managers who primarily serve the whole community, and whose attitudes influence the agenda of tolerance. Dominey-Howes et al. (2014) and Gaillard et al. (2017) refer to specific minority groups within LGBTQ in Asian countries, who play key roles of leadership in disaster recovery. The leadership and resilience exist with the LGBTQ minority to advance change in DRR policy and practice from within the agencies, even in tackling the conservatism of FBOs.
The much greater problem is the entrusting of leadership of disaster relief, the operation of evacuation centers and provision of welfare support services to FBO that are blatantly anti-LGBTQ, and who consider it acceptable to deny the services with which they have been entrusted by the government. Clearly, there are many FBO who do good work, as cited by Ng (2015) in China, but there are fundamentalist organizations which condemn LGBTQ people and refuse services and support. If these services and goods are provided by government through these organizations, such actions are illegal, but in many instances goods and services are provided by the charity of organizations that are only loosely answerable to the government. Some of these organizations may be controlled and structured by good governance, but in times of disaster, services are often provided by volunteers whose training and awareness may be very limited, and some of whose anti-LGBTQ opinions are entrenched. LGBTQ people on the inside of these organizations have direct experience of the prejudice and intolerance that is directed towards them. Many are active in refusing to walk away from that prejudice and discrimination (QCF 2021), and understand fully the depth of the problems involved in dealing with these FBO. Reversing the intolerance of evangelical FBO towards LGBTQ people may not be achievable even in the long term, given its centuries' old history. Training in tolerance and inclusivity is not a solution to entrenched prejudice. An immediate focus must be the unsuitability of such organizations to be permitted to provide aid and relief after a disaster.
Conclusion
New and emerging areas of research from a variety of disciplines such as Anthropology, Sociology, Gender and Feminist Studies, and Disaster Studies argue for the importance of LGBTQ peoples' needs and engagement in disaster response and recovery. Researchers should identify discrimination and call out those organizations, or sections within them, which practice discrimination and speak publicly against LGBTQ individuals and communities both outside the period of a disaster event but especially during a disaster impact and recovery (Stukes 2014). More research is needed to fully understand how those inside these FBO can work within disaster relief operations and recovery in a way that does not impinge on the rights of LGBTQ members of the community during times of disaster and recovery. Participatory action research informed by queer theory is a research strategy for working inside organizations and communities (Castan Broto 2021; Parkinson et al. 2018). The question to be addressed will be whether FBO which advocate or practice discrimination should be identified and removed from disaster relief operations even if they have a long history of successful and effective relief and recovery strategies.
LGBTQ researchers within such organizations have a critical role and responsibility in enacting transformation that may harness the goodwill and capacity of FBO to care for all members of the impacted community.
|
2022-12-09T16:06:35.524Z
|
2022-12-05T00:00:00.000
|
{
"year": 2022,
"sha1": "13c235f757ff09f7654589e3a402a9a5abbeec1f",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1142/s2345737622410056",
"oa_status": "HYBRID",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "318683c2d120b90b33844585cc9bf2818aa716f5",
"s2fieldsofstudy": [
"Environmental Science",
"Sociology"
],
"extfieldsofstudy": []
}
|
18637461
|
pes2o/s2orc
|
v3-fos-license
|
Ly$\alpha$ Leaks in the Absorption Spectra of High Redshift QSOs
Spectra of high redshift QSOs show deep Gunn-Peterson absorptions on the blue sides of the \Lya emissions lines. They can be decomposed into components called \Lya leaks, defined to be emissive regions in complementary to otherwise zero-fluxed absorption gaps. Just like \Lya absorption forests at low redshifts, \Lya leaks are both easy to find in observations and containing rich sets of statistical properties that can be used to study the early evolution of the IGM. Among all properties of a leak profile, we investigate its equivalent width in this paper, since it is weakly affected by instrumental resolution and noise. Using 10 Keck QSO spectra at $z\sim6$, we have measured the number density distribution function $n(W,z)$, defined to be the number of leaks per equivalent width $W$ and per redshift $z$, in the redshift range $5.4 - 6.0$. These new observational statistics, in both the differential and cumulative forms, fit well to hydro numerical simulations of uniform ionizing background in the $\Lambda$CDM cosmology. In this model, Ly $\alpha$ leaks are mainly due to low density voids. It supports the early studies that the IGM at $z\simeq6$ would still be in a highly ionized state with neutral hydrogen fraction $\simeq 10^{-4}$. Measurements of $n(W,z)$ at $z>6$ would be effective to probe the reionization of the IGM.
ionizing background. At redshift z > 5, however, they are no longer to show forests features, but consist of complete absorption troughs separated by spikes of the transmitted flux (e.g. Becker et al. 2001;Fan et al. 2006). That is, although the cosmic hydrogen gas at z > 5, in average, is opaque for Lyα photons, there are many tiny regions, which are Gunn-Peterson transparent and lead to Lyα photon leaking.
The nature of the leaking is crucial to understand the physics of reionization. According to commonly accepted scenario of reionization, at early stage, only isolated patches around ionizing sources are highly ionized. The subsequent growing and overlapping of the ionizing patches lead to a uniform ionizing background and the end of reionization (e.g., Ciardi et al. 2003;Sokasian et al. 2003;Gnedin 2004;Mellema et al. 2006) The ionization fraction of the IGM and the ionizing radiation underwent an evolution from highly non-uniform patches to a quasi-homogeneous field. Before the patch-to-uniform transition, only ionized patches would be transparent to Lyα photons. After the transition, the low density voids will also be Gunn-Peterson transparent. Therefore, the origin of Lyα leaks will constrain the epoch of the patch-to-uniform transition. In this Letter, we study the origin of Lyα leaks in the observed spectra of QSOs at z ≃ 6.
Several statistics have been introduced to describe the transmitted flux of Lyα absorption at high redshifts, including the probability distribution function (PDF) of the flux (Fan et al. 2002;Becker et al. 2007), the distribution of the size of dark gaps (Songaila & Cowie 2002;Fan et al. 2006), and the largest peak width distribution (Gallerani et al. 2007). We will focus on the profile of the leaking features in the transmitted flux. We fit these statistical features with samples of hydrodynamic simulation with a uniform ionizing background, and analyze the possibility of explaining the leaks by ionized patches embedded in neural IGM background.
Samples
1. Observational spectra of high redshift QSOs. The observational spectra used here are 10 of the 12 Keck spectra of QSOs at redshift z > 5.8 compiled in Fan et al. (2006). We excluded 2 BAL QSOs. The data have a uniform resolution of R ∼ 4000 and are re-binned to a resolution R = 2600. To avoid the mixing of Lyβ absorption and the effect of QSO's H II region, only the rest frame wavelength between 1050 to 1170Å are used. To study the evolution of Lyα leaks, we divide the spectra into two redshift bins, 5.4 < z ≤ 5.7 and 5.7 < z ≤ 6.0. The observed flux, f obs , is normalized with a power-law continuum f con ∝ ν −0.5 . The noise level of transmitted flux F ≡ f obs /f con is about 0.018 ± 0.012 and 0.014 ± 0.008 for above two redshift bins. For more details, we refer to Fan et al. (2006).
It has been shown that the observed dramatic decrease and abnormally large scatter of Gunn-Peterson optical depth at z ≃ 6 (Fan et al. 2006) can be well fitted by models of a uniform ionizing background (Lidz et al. 2006;Liu et al. 2006). That is, the large scattering of Gunn-Peterson optical depth may still be mainly due to the inhomogeneity of the IGM density field. Therefore, we investigate whether such a uniform ionizing background can explain the leaks. In this context, the uniform photoionization rate is adjusted to yield the same mean optical depth as observational data at the redshifts considered. A thermal energy of T = 3 × 10 4 K is added at z = 10, and only adiabatic cooling and shock heating are followed. With this method, the photoionization rate are found to be equal to 0.63 and 0.33 in units of 10 −12 s −1 at redshift z =5.55 and 5.85, and the corresponding neutral hydrogen fraction are 3 × 10 −5 and 7 × 10 −5 . The simulated spectra were smoothed with a Gaussian window of a FWHM corresponding to R ∼ 4000, and re-binned to pixels of the size of R = 2600. We add Gaussian noises to the re-binned fluxes with variances equal to the observational noise level.
Identification of Lyα leaks.
Lyα leaks are identified as contiguous pixels where the fluxes have a maximum larger than 2 or 3 σ of the local noise level. The boundaries of a leak are defined as positions where the fluxes are smaller than a threshold F th , or are the minimum between the neighboring leaks. That is, if there are two local maximums above 2 or 3 σ of the noise level, each one is identified as a leak. We take the threshold F th = 0.02 in this letter. Note the identification of a leak depends mostly on the condition of the maximum flux (see discussion on Figure 1 below). With this method we decompose the transmitted fluxes between Gunn-Peterson troughs into Lyα leaks of different profiles. The Lyα leaks contain information different from the size of dark gaps and largest peak width, both of which measure only length scales.
To test the identification condition, we count the number of fake leaks due to noise in 100 simulation samples. The percentage of fake leaks are 2.3%(13%) and 0.4%(1.4%) for the 2 σ and 3σ identification in redshifts range 5.4 − 5.7(5.7 − 6.0). Similarly, we also count the number of missed leaks due to noise. The percentage of such missing leaks are 5%(8%) and 15%(22%) for the 2(3) σ in redshifts range 5.4 − 5.7(5.7 − 6.0), respectively. The fluxes of missing leaks generally are around F th . Therefore, the leak identification with F th = 0.02 is statistically reliable. In the 10 Keck spectra, there are totally 173 and 147 leaks in the redshift range 5.4 ≤ z ≤ 5.7, and 39 and 32 leaks in 5.7 < z ≤ 6.0 for the 2 and 3 σ identification, respectively. The fluxes of smallest leaks are a little higher than F = 0.02, while big leaks can have F ≃ 0.3.
Equivalent width functions.
Similar to emission and absorption lines, we can measure the profile of Lyα leak with equivalent width, which is defined as the area under its flux profile, W = F dλ, where the integral is over the range between the boundaries. For our observed samples, W lasts in the range from 0.06Å to about 5Å. In general, the equivalent width W measures the strength of the leaking, or the Gunn-Peterson optical depth within the leaking regions. The statistical description we used is the equivalent width function n(W, z), which is the number of leaks of W at redshift z per unit W per unit z. The equivalent width function reflects the distribution of the strength of leaking.
We count the observed W into 15 bins with logarithm size ∆ ln W = (1/15) ln(10/0.01). The results are shown in Figure 1, which is for leaks at redshifts 5.4 − 5.7 (top) and 5.7 − 6.0 (bottom), and the 2 σ (left) and 3 σ (right) identification. The error bar is of Poisson fluctuation. The functions n(W, z) are weakly dependent on the identification. Although the total numbers of leaks of 2σ and 3σ samples are different, the shape of n(W, z) for both samples are about the same. As expected, for large leaks of W > 0.5Å, the functions n(W, z) are independent of the 2 or 3σ condition, while for small leaks of W < 0.5Å, the function n(W, z) of 3σ is a little lower than that of 2σ. Figure 1 also shows the results given by 100 simulation samples. The solid curves are the mean of the samples, and the dot lines give the jackknife error estimator, which is to divide the 100 samples into 10 subsamples and compute the variance over the 10 subsamples. We see that the distributions of n(W, z) of simulation samples generally are good to fit the observed samples. To test the effect of noise, we also calculated the function n(W, z) of simulation samples without noise addition, and the results are shown in Figure 1 as the dashed lines. Without noise addition, the leaks are identified as local maximums above F th =0.02. Figure 1 shows that the noise has no effect on big leaks (W > 0.5Å). While for small leaks (W < 0.5Å), samples without noise addition give higher number of leaks than samples with noise addition. This is because the identification condition of 2 and 3 σ is more rigorous than the condition of F th =0.02. Therefore, the effect of noise on n(W, z) does not change the consistence between observed and modeled n(W, z). The measurement of W is insensitive to the noise, as it is on the area of profile.
We see from Figure 1 that a few data points at small W show fluctuation around simulation result. It is probably caused by the binning. To solve this problem, we calculate the cumulative equivalent width function defined as n(> W, z) = ∞ W n(W, z)dW , which is less dependent on the binning. Since the distributions of leaks of 2 and 3 σ identification are similar, only 2 σ identification condition is applied. The results are presented in Figure 2. The solid curves are the mean of simulated samples, and the dot lines give the jackknife error estimator as in Figure 1. It shows clearly that the cumulative width functions of observed leaks are smooth, and gives a better fitting with simulation samples. Figure 3 presents n(> W, z) vs. z for leaks of W = 0.4, 1, and 1.6Å . The redshiftevolution of leaks with larger W is more significant than smaller W leaks. This is natural in low density voids scenario. The larger voids have lower probability, or are the events given by the tail of the PDF of voids. The PDF tail underwent a strong evolution at high redshifts. At redshift z > 6, there are only very few leaks identified from observation data, and therefore, we will not extend the analysis to z > 6.
Since all leaks in simulated spectra are due to low density voids, the results show that the distribution of observed leaks are consistent with low density voids assuming the uniform ionizing background. It is interesting to point out that the tail of the cumulative width function shown in Figure 2 is close to Gaussian distribution with respect to logarithm W . Therefore, n(> W, z) approximately has a lognormal tail of W .
Ionized patches.
We now estimate the Lyα leaking due to the ionized patches around ionizing sources. Considering a simple model, ionizing sources embed in a fully neutral IGM at high redshift. The scale of ionizing patches can be estimated with radius R = R s [1 − exp(−t/t rec )] 1/3 , where R s is the Stromgren sphere radius, t rec and t are, respectively, the recombination time and the active age of the ionizing source. It has been shown that, due to the retardation effect of photon propagation, the scale R actually is an upper limit to the ionized volume Qiu et al. 2007a). The retardation effect is more apparent for clustered sources (Qiu et al. 2007b). Moreover, it is also shown that the fraction of H I within ionized sphere generally is larger than 10 −6 unless the intensity of sourcesṄ > 10 55 sec −1 (Qiu et al. 2007a).
It has been shown that the damping wing of the neutral IGM absorption make ionized patches opaque to Lyα photons if the size is too small (Miralda-Escude 1998). This effect is more significant if a small fraction of H I is remained in patches. For instance, an ionized patch with neutral fraction of 5 × 10 −6 around a galaxy at z = 6 can yield a flux F = 0.02 only if the comoving radius R ≥ 3.5 h −1 Mpc, orṄ ≥ 9 × 10 53 sec −1 , which requires a luminosity L ≥ 1.6 × 10 10 L ⊙ if assuming a spectra of L ν ∝ ν −3 . Here we also assume all the ionizing radiation of a galaxy is capable to contribute to the ionizing sphere, and the luminosity L ≥ 1.6 × 10 10 L ⊙ gives a lower limit to the required luminosity to produce a leak with F = 0.02.
With these results, one can estimate the number of leaks with F ≥ 0.02 due to galaxies by using the luminosity function of galaxies at z = 6 (Bouwens et al. 2006). The probability that a line intercepts patches at an comoving impact parameter r = 1.5 h −1 Mpc (since the cross radius should be larger than 3.5h −1 Mpc, we should use a smaller impact radius) for galaxies with luminosity > 1.6 × 10 10 L ⊙ is (e.g., Peebles 1993) here φ(> 1.6 × 10 10 L ⊙ ) is the comoving number density of galaxies with luminosity > 1.6 × 10 10 L ⊙ . On the other hand, Figures 1 and 2 show that the number density of leaks with F ≥ 0.02 at 5.7 < z < 6 is ≃ 35. Therefore, if the IGM z = 6 is mostly neutral, and the only ionized regions are the patches around galaxies, the leaks of F ≥ 0.02 given by the ionized patches of galaxies would be no more than 20% of the observed result.
Discussions and conclusions
The transmitted fluxes between Gunn-Peterson troughs of high redshift QSO's absorption spectra are not only one peak of the flux, but contain rich structures, which can be decomposed into Lyα leaks. The Lyα leaks have profiles similar to emission lines, and can be measured by equivalent width W . The equivalent width functions, n(W, z) are effective statistical measurement of the process of reionization. We show that the equivalent width functions of the observed spectra at redshifts 5.4 < z < 6.0 can be well fitted by hydro simulation of the ΛCDM cosmology assuming the ionizing background to be uniform. In this model, all the Lyα leaks are leaking through low density voids.
The mean transmitted flux at z is given byF ∝ n(W, z)W dW . Therefore, by adjusting the photoionization rate to match the observed Gunn-Peterson optical depth, the mean of W for simulation samples should be the same with observation. Thus, Figure 1 actually is to show that once we adjusted the mean flux to be the same as observation, the simulation yields the same distribution of the observed W . In other word, the scattering of W is caused by the fluctuations of mass density field of HI. Therefore, a small inhomogeneity of the ionizing background would be allowed. That is, the distribution of W would still be able to be fitted with a fluctuated ionizing background if its variance is much less than that of HI.
In addition to the distribution of W , the evolution of W is also helpful when differentiating models. For example, in voids scenario, the evolution of W reflects the evolution of low density voids; while for ionized patches, it reflects the evolution of the UV luminosity function of ionizing sources.
We show that the ionized patches of galaxies embedded in a fully neutral IGM at redshift z ≃ 6 are not enough to produce the observed leaks. We also show that leaks can only be produced by patches around strong ionizing UV photon sources, but not weak sources. Especially, big leaks (W > 0.5Å, or F > 0.1) have to come from very strong sources. Therefore, at higher redshift, Lyα leaks only probe strong ionizing sources. Thus, from the existence of many big leaks at z ≤ 6 we can conclude that the patch-to-uniform transition of the ionizing background would occur at z > 6, and most of the IGM at z ≃ 6 are still in a highly ionized state of neutral fraction f HI ≃ 10 −4 . This result is consistent with the analysis of the transmitted flux PDF (Becker et al. 2007), the QSO proximity zones (Lidz et al. 2007) and the luminosity function of Lyα emitting galaxies (Dijkstra et al. 2007).
It should be pointed out that the resolution of the observed data is low, ∼ 3Å, which corresponds to a comoving size ≃ 0.7 h −1 Mpc. In contrast, most of the simulated leaks possess a intrinsic width < 3Å. Thus the low resolution data provide only a test of smoothed leaking features. One cannot see whether the smoothed features is due to individual or clustered leaks. Higher resolution spectra would be able to test not only the width functions, but also the spatial correlations of the leaks. They can also provide other measurements of Lyα leaks, like the FWHM, which would be effective for a confrontation between real data and models.
The statistics of Lyα leaks at redshifts ≤ 6 actually is the statistics of voids formed in the early universe. The equivalent width functions, n(W, z) of Lyα leaks are similar to the mass function of galactic clusters. Thus, one can expect that the width functions of voids are sensitively dependent on cosmological parameters, and play a similar role as the mass function of clusters. For instance, the formation of large voids is found to be sensitively dependent on the mass parameter Ω m (Miranda & de Araujo 2001). With data of leaks, we may set constraint on cosmological parameters at high redshifts. This approach will be reported separately. Fig. 1.-Equivalent width function n(W, z) for leaks of 2σ (left) and 3 σ (right) identification at redshift ranges z = 5.4−5.7 (top) and 5.7−6 (bottom). The data points are from 10 Keck QSO spectra. The error bars are from Poisson fluctuation. The solid lines are calculated with 100 simulated spectra; dot lines are the range of variance over the 10 subsets, each of which contains 10 spectra; dash lines are for samples without noise. The noise has little effects on equivalent width W . Fig. 2.-Cumulative width function n(> W, z) for leaks at redshift ranges z = 5.4 − 5.7 (top) and 5.7 − 6 (bottom). The data points are from 10 Keck QSO spectra. The solid lines are from 100 simulated spectra; dot lines are the range of variance over the 10 subsets, each of which contains 10 spectra.
|
2007-11-05T23:01:32.000Z
|
2007-11-05T00:00:00.000
|
{
"year": 2007,
"sha1": "67ecf88cb3e5ad259bf21f762e11d617b8400bf3",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/0711.0773",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "67ecf88cb3e5ad259bf21f762e11d617b8400bf3",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
235286181
|
pes2o/s2orc
|
v3-fos-license
|
Calculating control area and reserves of single well by triangulation method
The arithmetic average method is a conventional method to calculate the geological reserves. However, it ignores the influence of heterogeneity in the block, especially the faults. The calculated single well controlled reserves are often quite different from the actual status, which can’t guide the development actions accurately. Therefore, it is necessary to introduce a more accurate calculation method. The triangular network method introduced in this article can improve the accuracy when calculating the control area of single well in fault area, support to evaluate the control reserves of single well, and reflect the changes of geological reserves near each well point accurately.
Introduction
By calculating the control area of a single well near the fault, and then calculating the control reserves, the triangular network method can quantitatively refine the reserves distribution, which can lay the foundation for the remaining oil research and lower potential tapping.
Calculation method of control area of single well
In order to clarify the controlled reserves of single well around the fault and the controlled reserves of fault area, the breakpoints on the fault are treated and optimized in this article, and each fault is regarded as a virtual well point. When the triangulation network is constructed, the virtual well points also participate in the formation of the triangulation network, which can effectively analyze the controlled reserves of the breakpoints and intuitively describe the remaining oil situation at the edge of the fault.
The trigonometric network method is used to calculate the control area of single well in fault area. In this method, each injection production well and breakpoint are connected in turn through the point by point insertion algorithm, and the triangular network structure is constructed through the basic principle of triangular network, and then the midpoint of the connecting line between the well point and the adjacent well is taken, and then the outer center of the triangle connected with the well point (the obtuse angle triangle can take its center of gravity) is obtained, and the outer center and the midpoint are connected into a closed area, which is the well scope of control [1][2][3] . Refer to Figure 1. Figure 1.Triangle network structure and single well control boundary diagram When calculating the control reserves, there are still some geological reserves in the control range of the water injection well. Even if the water injection well does not produce oil, these reserves will flow in the process of water injection development, and can be recovered from the surrounding production wells. Therefore, when the central well is a water injection well, the control area of a single well of the water injection well needs to be split proportionally to each production well around the well, so as to Ensure that all areas around oil and water wells participate in production. Accordingly, the following principle of well control area splitting is formulated: a). When two adjacent wells are injection well and production well respectively, the control area of the wells on both sides of the oil-water well connection line needs to be split to the wells on the connection line. Refer to Figure 2. Figure 2 is a schematic diagram of well control area splitting in a well control unit, in which L1 is the oil well, L2 is the well connected with L1 well, and the blue area is the control area of L1 well. According to the principle of area splitting, the brown area is the control area split by surrounding wells to L1 well, forming a single well control area. b). When two adjacent wells both are water injection wells, it is necessary to divide the control area of wells on both sides of the connection line between wells to the adjacent wells. c). When the water injection well in the area is a transfer well, if the well is still injected water in the production stage, it is necessary to regard the transfer well as a water well, and divide the geological reserves and production controlled by the transfer well to the adjacent oil wells. If the well stops injection in the production stage, its control area is divided to the adjacent oil wells, but it does not participate in the later reserves division.
For each breakpoint on the fault line, i.e. virtual well location, the calculation method of control area is the same as that of well point, and a triangular network is constructed together with the surrounding well points and breakpoints. In the early stage of development, there are few wells in the fault area and the well location is far away from the fault. Therefore, the control area of single well in the fault edge area should be adjusted. Scaling the breakpoint and the control range of single well is more in line with the actual situation, and the calculation is more reasonable [4][5][6][7][8][9] .
After the construction of triangulation network, the control area of well point is divided according to the attributes of the triangle well point, so as to determine the area of each small triangle in each well point and breakpoint control area, then the control area of a single well can be calculated by stacking all the small triangle areas [10] . Refer to Formula (1).
A ∑ A ( 1 ) Ai -area of triangles around well point.
Calculation method of single well controlled reserves [10]
After the control area of single well in fault area is calculated by triangle network method, the control reserves of each single well and breakpoint in the area are determined by volume method. Refer to Formula (2). N = A h ω ( 2 ) N-Geological reserves of crude oil，10 4 t; A-Oil bearing area, km 2 ; h -effective thickness, m; ω -Single storage coefficient,10 4 t/(km 2 ꞏm). After calculating the control area of single well in each sub layer in fault area by triangle network method, according to the reserve calculation parameters of main control well, the attribute parameters are interpolated and weighted average with the attribute values at triangle vertex, and then the control reserves of single well in each sub layer are calculated. The total reserves are the sum of the single well controlled reserves of each sub layer. Refer to Formulas (3), (4).
( 4 ) N -Geological reserves of crude oil, 10 4 t; M -Single storage coefficient;(With the change of well point, the relationship between break point and oil-water transition zone, oil-gas transition zone and pure oil area, the thickness of oil layer changes, and the single reservoir value is also different) ; A -Control area of single well and small layer，km 2 ; h -Average effective thickness of single well，m.
Calculation method of single well controlled remaining reserves [10]
Based on the controlled reserves and production of a single well in the fault area, the remaining reserves in the controlled area of a single well can be calculated, and then the remaining reserves can be subdivided according to the triangulation method, and finally the distribution of the remaining reserves can be obtained. N 1 =N-Q N 1 -Remaining reserves, 10 4 t; N -Geological reserves of crude oil, 10 4 t; Q -Cumulative oil production, 10 4 t.
In the process of calculating the remaining reserves of a single well, it is necessary to split the controlled geological reserves of wells and the cumulative oil production of oil wells to ensure the accuracy of the calculation of geological reserves. Considering the effective thickness of oil layers, effective permeability, interlayer interference, distance between injection and production wells, connectivity and measures, a specific split formula is established. Refer to Formulas (5), (6), (7). R j -Split coefficient of geological reserves; K j h j -Formation coefficient, Product of effective thickness and effective permeability; C j -Connectivity coefficient of oil and water wells, Let 1 be connected, 0 if not connected; M j -Oil well measure coefficient, 0 for water shutoff, Take 1 without fracturing, Take 2 fracturing measures; d j -Distance between oil and water wells, m; I j -Interlayer interference coefficient; Y j -Yield split coefficient, 10 4 t. According to the partition coefficient of well reserves, based on the established triangle network, the well reserves are split into the surrounding oil wells, and then the oil well production is split into the triangles of each small layer, so as to refine the distribution of the remaining reserves of a single well, and then guide the measures for the lower part of a single well.
Conclusion
Compared with the arithmetic average method, the triangulation method considers the heterogeneity of strata in the block, especially the influence of faults, when calculating the reserves of a single well. Its calculation results have been proved to be more accurate in practice, and it has more guiding significance for the formulation of measures for the lower part of a single well.
|
2021-06-03T01:12:49.278Z
|
2021-01-01T00:00:00.000
|
{
"year": 2021,
"sha1": "65fc45d28600fd5e1ce56e7dfc07aadeca4ba50d",
"oa_license": null,
"oa_url": "https://doi.org/10.1088/1755-1315/770/1/012021",
"oa_status": "GOLD",
"pdf_src": "IOP",
"pdf_hash": "65fc45d28600fd5e1ce56e7dfc07aadeca4ba50d",
"s2fieldsofstudy": [
"Geology"
],
"extfieldsofstudy": [
"Physics"
]
}
|
240493167
|
pes2o/s2orc
|
v3-fos-license
|
Quality of life and health-related utility after trans-oral surgery for head and neck cancers
Purpose The purpose of this study was to assess utility coefficients of health states following two minimally invasive surgical approaches for head and neck cancer, namely trans-oral robotic surgery and trans-oral laser microsurgery. Those utility coefficients will be later exploited in an economic evaluation study comparing the two approaches. Methods The above cited economic evaluation will be done from the Swiss healthcare system perspective and, as such, Swiss healthcare professionals were interviewed to elicit utility coefficients. Health states, ranging from remission to palliative care, were described using clinical vignettes. A computerized tool (UceWeb) implementing standard gamble and rating scale methods was used. Results Utility coefficients for 18 different health states were elicited with the two methods from 47 individuals, for a total of 1692 values. Elicited values varied from 0.980 to 0.213. Comparison with values elicited in previous studies show the need for population-specific elicitation, mainly for the worst health states. Conclusion Herein we report health utility coefficients for the Swiss population for health states following minimally invasive trans-oral surgery. This study provides utility values that can be used not only for a specific cost-utility analysis, but also for future studies involving the same health states. Supplementary Information The online version contains supplementary material available at 10.1186/s12955-021-01836-3.
Background
In head and neck surgery minimally-invasive techniques have recently been introduced to reduce access-related morbidity and their associated functional deficits. Such novel approaches are mostly used to access cancers in the oropharynx [1,2]. Trans-oral approaches such as transoral laser microsurgery (TLM) and trans-oral robotic surgery (TORS) have provided improved visualization for tumors previously amenable only to transmandibular or other 'open' surgical approaches [3]. While the oncologic principles for treating head and neck cancers remain the same regardless of the approach, the costs related to various approaches are different. As such, an economic evaluation is needed to inform and guide decision making about adoption of techniques.
Our group has developed a model for comparing those two surgical approaches through an economic evaluation, and precisely a cost-utility analysis (CUA) [4][5][6]. The model has been developed as a decision tree [7] embedding Markov processes [8] to represent a patient's transitions among different health states following transoral surgery [9]. Direct costs associated with every intervention and every health state treatment have been included in the model, while quality-adjusted life years (QALYs) are used as primary health outcome [10]. To calculate QALYs, a patient's expected survival is split in time intervals, each one (Ti) spent in a specific condition Ci. Assuming that Ci is associated with a utility coefficient (UC) UCi, QALYs are computed as the weighted sum of Ti × UCi. UCs range from 0 to 1, where 0 is for the worst possible health state (usually death) and 1 is for the best possible one (perfect health). Thus, a UC quantifies a patient's preference for a given health state.
Different methods have been proposed in the literature for eliciting UCs for sub-optimal healh states. A complete review of the utility elicitation methods is beyond the scope of this paper and can be found in [11]. In the following we introduce the most popular ones, namely the standard gamble (SG) method, the Time Trade-Off (TTO) and the rating scale (RS). In the SG, based on the axioms of utility theory [12], the interviewee must envision a hypothetical scenario in which he must choose between living the remainder of his life in the sub-optimal S state or accepting a gamble in which the outcome may be perfect health or alternatively sudden death. The gamble may be presented as a hypothetical treatment (e.g., a surgical intervention or a drug) that may result in perfect health, but presents a risk p of death, due to complications or toxicity for example. The first risk value proposed can be arbitrary (e.g., 0.5), and then varied (increased or decreased according to the answers) until the interviewee can't decide about the gamble. At that point, the UC is computed as (1-p). In the TTO method, the patient is asked to choose between living his entire remaining life (t1) in the state S or to live shorter (t2 < t1) but in a perfect health state. Like the probability p in SG, the amount of time a patient is proposed to give up in order to heal completely (i.e., t1-t2) is varied until the patient is indifferent between the two choices. The UC is then calculated as t2/t1. The RS method, conversely, is simply calculated by asking a respondent to rate S on a scale between 0 and 100, then normalizing to the range 0-1. Values derived from RS are not true UCs as they fail to meet the utility theory requirement of "decision under uncertainty" [13]. However, they may help to validate utilities derived by other methodologies and are often easier to comprehend. Finally, also indirect utility elicitation methods, based on questionnaires, are widely used. The most popular example is through the administration of the EQ-5D questionnaire and its conversion in a UC (note that this conversion may lead to values less than zero, i.e., states considered worse than death).
For what concerns utility elicitation in head and neck cancer, a systematic literature review was previously published [14]. Meregaglia et al. state that "there is currently a lack of research for some disease phases including recurrent and metastatic cancer, and treatment-related complications", which is specifically what we target in our article, i.e., how quality of life is affected by early and late consequences of trans-oral surgery. Indeed, only one study by de Almeida et al. [15] reported UCs related to a number of health states following TORS or radiotherapy in oropharyngeal cancer patients. In that study, UCs were elicited using the SG and RS methods. UCs for 21 health states, presented through scenarios, were elicited from 50 healthy subjects belonging to the North-American population. UCs for head and neck cancer have been elicited also by Noel et al. [16] in Canadian population. However, their aim was to investigate face and construct validity of different elicitation methods, and no specific post-surgery states were addressed. Finally, Liao et al. [17] compared utility values ascertained from Taiwan and Sweden, and suggested the need for understanding population differences, which is corroborated by Caulley et al. [18]. As a matter of fact, UCs are subjective measures, and may vary among different populations, depending on cultural, geographical, and economical factors [19,20]. Thus, in principle, UCs should be elicited from the same population for which the economic analysis is intended. Unfortunately, utility elicitation is a very time consuming and challenging task, and often researchers rely on values found in the literature, even if these values have been collected from different populations. To avoid this methodological pitfall in our economic analysis, we aimed to ascertain UCs directly from members of the Swiss population. Notably, no study in the literature targeted Swiss population, and no study other than [8] reported UCs for the health states after minimally invasive surgery.
Thus, this paper presents the methodology adopted for eliciting UCs, and the actual UCs associated to the health states represented in the above mentioned decision tree for CUA.
Study population
We performed a cross-sectional study where we interviewed a set of 47 healthcare professionals on a voluntary basis from the Otolaryngology division of Centre Hospitalier Universitaire Vaudois (CHUV) located in Lausanne, Switzerland. We prepared a series of factsheets describing all the health states considered in the study (see also additional material). The preparation of the scenario sheets involved feedback-refinement iterations among the clinical experts from CHUV. The final version of the sheets has been used during the interviews with all the volunteers.
To calculate the interviewees' sample size, we considered estimating the mean value of UCs with a precision (margin of error) of 0.05, assuming a standard deviation of 0.15. To achieve a confidence level of 95%, the minimum number of persons needed is 38 [21]. To be conservative, and to prevent loss of statistical power due to drop-out or data issues, we enrolled 47 volunteers.
The computerized tool for utility elicitation
Utility elicitation is a challenging task. In order to obtain consistent values from all the individuals, it is important that everybody is interviewed following the same procedure. We used a computer based tool, UceWeb, a system developed by our group [4,22,23] and that has been previously validated in other medical contexts [24,25]. The main features of UceWeb are: (1) a graphical interface implementing several methods for utility elicitation; (2) a decision support facility suggesting the best elicitation method according to the interviewee's and health state characteristics; (3) a common terminology for the health states (SNOMED) for which UCs are elicited (additional states, not covered by SNOMED, may be added if necessary); (4) a collaborative environment where different researchers can feed the UCs repository, while providing a basic profile for every interviewed individual (age, gender, country, if he is a patient or not, and other features known to affect an individual's preferences). This will allow future studies to have larger and larger sets of UCs for specific target populations. The study described in this paper contributed to this repository indeed; (5) a direct link to TreeAge Pro [26], allowing to run a decision tree just after having elicited UCs from a single patient or having retrieved a UC set for a target population from the repository; (6) UceWeb is a multiuser (and multilingual) system and this allowed the interviewers to work in parallel and save time. For the current study, French language has been used.
Utility elicitation sessions
The elicitations were done in 9 sessions over a 101 days period. As mentioned, all the interviews were done using UceWeb and supported by the same set of factsheets describing the clinical context, the treatments and their possible consequences. In this way, all the volunteers received the same information in the same format. Two physicians (FS and LD) performed all the interviews and answered all the volunteers' questions. First of all, the clinical problem was described, relying on available literature [27][28][29]. In the following textbox we report the exact words used for informing the volunteers at the beginning of each interview, which is useful for the reader as well, to understand the context, while omitting some details for sake of space.
After describing the clinical problem, the interview is introduced through a hypothetical shared decision-making framework: Imagine being a patient who consults an otolaryngologist following the diagnosis of cancer of the oropharynx. Your doctor will describe different treatment options and you will finally choose the option that best suits your wishes. After sufficient discussion with him, you will have enough information about the risks and benefits to be able to make a decision.
After this introduction, the elicitation exercise begins. First of all, the oncologist describes the scenario #1, depicting the first month consequences of the surgical intervention, which are the same for TORS and TLM, and then he describes all the complications and further treatments that are considered in the decision model, for a total of 18 health states, which are reported in Table 1: As an example, the following textbox reports the factsheet for scenario #4 (the other ones are available in the additional material, all of them have been proposed to the volunteers in the French version): After explaining each scenario and addressing questions of the interviewee, the following questions are asked, implementing the RS and SG methods, respectively: 1. On a scale of 0 to 100, where 0 represents death and 100 represents perfect health, where would you place the above scenario? (Let UCrs be the UC elicited with this first question).
2. Now imagine that you can choose between an adjuvant CRT and a pill that, taken at home, has the same effectiveness as the adjuvant CRT, but without side effects. However, taking the pill carries a risk of sudden death (we will go shortly in the value of the risk). Would you consider taking the pill if the risk is low enough?
According to the standard gamble procedure, if the person does not accept even a very small risk, the interview is stopped, and the UC of the described state turns out to 1. On the contrary, if the person accepts the gamble, the first risk value proposed is (1-UCrs). As described in [22], this initialization shortens the SG elicitation procedure.
The above two questions are the same for all the 18 scenarios, resulting in the RS and SG values for each health state.
Methods of analysis
Descriptive statistics as mean, standard deviation, and quantiles (min, q1, median, q2, max) have been calculated for each set of collected UCs, grouped by health state. Furthermore, to compare the results obtained with the two elicitation methods employed, health states have been ordered in a ranked list, and correlation between SG-and RS-derived UCs has been analyzed using non-parametric statistical tests. Statistical analyses were performed to investigate correlations of elicited UCs and patient demographics and profile characteristics. All statistical analyses have been performed with R (https:// www.R-proje ct. org/).
Oropharyngeal carcinomas (OPCs) represent a major health problem. The oropharynx is the region of the throat behind the mouth. It contains, as main structures, the two palatine tonsils on each side and the base of the tongue. These areas play a major role in swallowing and breathing. Unfortunately, this region can be the site of cancer. The most common cancer found in the oropharynx is squamous cell carcinoma. … Patients with OPC associated with HPV tend to be young, non-smoking men.
The possible treatments.
The type of treatment that can be offered to patients suffering from OPC is of three types: a surgical intervention which removes the cancerous tissue, radiotherapy (RT), which induces the death of cancer cells, and finally chemotherapy (CT) in the form of a medicine (taken intravenously) that attacks cancer cells all over the body through the bloodstream. How these treatment modalities are chosen and combined depends on the stage of the cancer, the availability of treatment, and the patient's preferences. The most common methods and combinations are: -Surgery only.
-Surgery followed by radiotherapy.
-Surgery followed by CCRT. When radiotherapy or chemo-radiation are added after surgery, this treatment is called "adjuvant". In the case of adjuvant radiation therapy, the radiation dose is reduced compared to primary radiation therapy alone.
In order to improve functional recovery after surgery, i.e. your ability to eat, breathe and speak normally, new techniques have been developed. They consist of fully endoscopic approaches to the tumor by mouth, thus avoiding access through the neck and reducing the morbidity associated with access. This leads to a much faster recovery after surgery and a better functional result. The tumor is visualized with endoscopes and microscopes.
The two main endoscopic techniques practiced today are trans-oral robotic surgery (TORS) and trans-oral laser microsurgery (TLM). For TORS, a retractor is used to open the mouth to create space for the robotic camera and surgical instruments. Then, the surgeon uses a surgical robot to view and access the structures of the pharynx (back of the throat). The tumor is removed with an electric knife. In TLM, retractors are also used to open the mouth and access the throat, but with this technique, the surgeon uses a microscope and a laser to remove the tissue.
Both techniques show pros and cons. When TLM is performed, the visual field is usually quite small …this can compromise the correct assessment of surgical margins and trigger unnecessary adjuvant therapy. However, the precision of TLM is exceptionally high ….
TORS, on the contrary, allows for resection in one piece based on better visualization with the available endoscopes and cutting instruments…. Therefore, the analysis of surgical margins is more precise. On the contrary, the accuracy of the dissection, particularly at the deep margins, may not be as satisfactory as with the laser and the microscope.
Surgical approaches to the oropharynx are generally associated with dissection of the neck lymph nodes in order to remove the nodes eventually carrying tumor cells. An incision in the neck is made to allow the surgeon to remove the lymph nodes potentially contaminated with cancer cells.
Scenario # 4: Chemoradiotherapy (CRT) after surgery.
Imagine you have carcinoma of the oropharynx and your specialist recommends that you have a CRT after surgery. For this treatment, you will be treated every day (around 45 min) during the week, except on weekends. The treatment will last an average of 6 weeks. Side effects can occur at any time during or after radiation therapy. You may develop side effects months or years after radiotherapy. Most side effects go away on their own or can be treated, but some side effects may last longer or become permanent. During treatment, you will receive, usually three times, in addition to radiotherapy, chemotherapy. Chemotherapy may create additional side effects. Temporary side effects during treatment are: -Irritation of the pharynx -Swallowing problems (you may need a feeding tube in 25% of cases) -Dermatitis (inflammation of the skin, shown in figure a) of the neck ( 9 6 % ) -Nausea / vomiting (27%). -
Asthenia during and immediately after treatment -Tingling of the arms or legs (25%) -Hearing loss (25%) -Fall in the number of white blood cells (50%) which may require hospitalization
The long-term side effects that you may experience, and that are more common than with radiotherapy only, are: -Permanent swallowing difficulties and pulmonary infections which may require frequent hospitalizations With respect to profession, unemployed people show significantly lower UCs for 4 health states (1,2,8 and 11). While people employed in commercial activities show significantly lower UCs for 3 health states (1,8 and 12). No other significant correlations were found between profession and SG utilities. Regarding marital status, the only nearly significant difference was found for state 17 with lower SG UCs elicited from married people. No differences were found when considering RS values. No significant differences were found when comparing elicited values in people with different education level.
Comparing our results with other, similar studies, Table 5 summarizes the relationship with UCs elicited in the present study and the study by de Almeida [15] on a North American population that, as mentioned in "Background" section, is the only study reporting UCs directly comparable with the ones in our study.
Discussion
Our study covers UCs of all relevant complications, treatments or health state change (e.g., remission/relapse) after Head and Neck trans-oral surgey, in agreement with the literature review by Adelstein et al. [32]. Moreover, our study addresses the lack of UCs for treatment-related complications highlighted by Meregaglia et al. [14].
There is some debate as to whether patients or healthy individuals of the general population should be interviewed to derive UCs. On one hand, patients are more aware about their health state and how it affects their life, and on the other hand, the general population is probably more suitable to decide on the relative value of that health state in comparison to other comparable states as these individuals are not personally biased by their existing state. Moreover, patients tend to develop adaptive behaviors to cope with a certain heath state, thus overestimating the true utility of that state [18]. Several advisory bodies including the United States Public Health Services panel recommend using members of the general public to elicit utilities [11,33,34] [35]. For these reasons, we elicited UCs by interviewing healthcare professionals, on a voluntary basis. Although these individuals are not patients, they have a more nuanced understanding of the patients' health states following treatment.
Another discussion topic is represented by the elicitation methods. We used direct methods (SG and RS). As mentioned in [16] indirect measures like EQ-5D may be more capable of discriminating different subsets of patients. However, there are limitations in using EQ-5D for our specific population of Swiss healthcare professionals. As a matter of fact, as reported on the EQ-5D website, the questionnaire is mainly for self-administration, and "proxy versions were developed for use in special cases where patients are mentally or physically not (shown in the figure). This device will serve the same purpose as the feeding tube in the nose, except that it will be surgically inserted directly through the skin into your stomach. It is much more comfortable to carry than the feeding tube in the nose. Insertion is a simple procedure, which can be performed under general or local anesthesia.
Results
As mentioned, 47 Swiss individuals, mean age 40.8 years (range 21-68 years, standard deviation 14.79), were enrolled for the study and interviewed in the second trimester of 2019. A complete interview took from 18 to 69 min (43.5 ± 12.5 min). Table 2 summarizes the characteristics of the respondents. Table 3 shows the summary statistics of the SG-elicited UCs and RS values for the different scenarios. We report mean and standard deviation to facilitate comparison with other possible values reported in the literature, and median and quartiles to reflect the non-normal distribution very skewed towards 1 for the SG method.
Considering all the elicited values, RS and SG values showed a good correlation (Fig. 1). The smoothing line has been fitted using LOWESS [31] which is an algorithm for robust locally weighted regression (Fig. 2).
Considering the mean values, the 18 health states were ranked in the same order by the two methods 9 out of 18 times. Table 4 reports the ordering of elicited UCs from lowest (e.g. palliative care) to highest (remission).
In order to assess the need for tailoring decision models according to some population characteristics, we investigated the correlation of the elicited values with the respondent profile variables (gender, age, profession, marital status, education level). No significant difference was found between males and females (Wilcoxon rank sum test p value: 0.1824). With respect to age, no significant correlation was found with UCs elicited with SG (unless a slight negative correlation for the state 11-post-operative hemorrhages, p value = 0.08338, rho − 0.255219). Considering RS values, significant direct correlations were found with age for states 1 (p value = 0.002363, rho 0.4330367) and 2 (p value = 0.002985, rho 0.4282678), while significant negative correlation was found for state 11 (p value = 0.0456, rho − 0.2930534), giving support to the above finding with SG. Table 3 The values elicited with standard gamble and rating scale methods 19:250 capable reporting on their health-related quality of life, for instance because of severe intellectual disability or mental health problems". Moreover, for our specific aims, an appropriate value set of EQ-5D for Switzerland is not currently available for conversion of EQ-5D scores to UCs [36]. For these reasons, we did not administer EQ-5D in addition to SG and RS in our UC elicitation study.
As we reported in Table 4, RS and SG rank in the same order the 5 worst and the 2 best states. Thus showing higher agreement for poorest states. Also, a substantial agreement exists in ranking of the worse health states between SG and RS, suggesting a better reliability of the elicited values than what was previously reported [16].
Our study has some limitations. Confidence intervals for elicited UCs are wider than the ones reported in de Almeida [15]. This might suggest incrementing the number of interviewees, which would also allow performing sub-group analyses that may reveal more representative preferences. Finally, our interviewees had, in large part (36/47), a high level of instruction. Thus, a non-negligible selection bias has to be taken into account. Other selection biases in the present study are represented by the mean age of the interviewees (41 y.o.) and the sex (female 70%): indeed, the typical population of Head and Neck cancer patients are male and 65 y.o. [37]. Finally, our population was not profiled on tobacco use, which is a well-known behavioral risk factor for head and neck cancer [38].
Conclusion
To the best of our knowledge this is the first study collecting UCs after head&neck trans-oral surgery in Switzerland. Moreover, our study collected, for the first time, UCs of health states associated to both TORS and TLM, and their treatment-related complications.
The elicited values will be used first of all in a decision model for the cost/utility analysis of TORS vs TLM from the perspective of the Swiss Health National System. Beyond the specific application, the elicited values represent useful data for further economic evaluations that will consider the same health states, some of which are general enough to be applied to different cancer-related scenarios.
Additional file 1. Scenario descriptions for health-state utility elicitation.
|
2021-11-03T13:58:50.541Z
|
2021-11-03T00:00:00.000
|
{
"year": 2021,
"sha1": "7dd52b907271669834d45ed7f80a6557dda79a41",
"oa_license": "CCBY",
"oa_url": "https://hqlo.biomedcentral.com/track/pdf/10.1186/s12955-021-01836-3",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "7dd52b907271669834d45ed7f80a6557dda79a41",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
230796794
|
pes2o/s2orc
|
v3-fos-license
|
ReportFlow: an application for EEG visualization and reporting using cloud platform
Background The cloud is a promising resource for data sharing and computing. It can optimize several legacy processes involving different units of a company or more companies. Recently, cloud technology applications are spreading out in the healthcare setting as well, allowing to cut down costs for physical infrastructures and staff movements. In a public environment the main challenge is to guarantee the patients’ data protection. We describe a cloud-based system, named ReportFlow, developed with the aim to improve the process of reporting and delivering electroencephalograms. Methods We illustrate the functioning of this application through a use-case scenario occurring in an Italian hospital, and describe the corresponding key encryption and key management used for data security guarantee. We used the X2 test or the unpaired Student t test to perform pre-post comparisons of some indexes, in order to evaluate significant changes after the application of ReportFlow. Results The results obtained through the use of ReportFlow show a reduction of the time for exam reporting (t = 19.94; p < 0.001) and for its delivering (t = 14.95; p < 0.001), as well as an increase of the number of neurophysiologic examinations performed (about 20%), guaranteeing data integrity and security. Moreover, 68% of exam reports were delivered completely digitally. Conclusions The application resulted to be an optimal solution to optimize the legacy process adopted in this scenario. The comparative pre-post analysis showed promising preliminary results of performance. Future directions will be the creation and release of certificates automatically.
Background
Electroencephalography (EEG) is an electrophysiological monitoring method used to record the electrical activity of the brain in millisecond-range temporal resolution [1]. EEG is used to diagnose neurological disorders such as sleep disorders, epilepsy, dementia, to name a few [2][3][4][5]. Differently from other techniques such as computed tomography, positron emission tomography or magnetic resonance imaging, EEG is non-invasive, cheap and fast of using, and this makes it a valuable screening tool although its lower spatial resolution [6]. As several prediction data mining techniques used to support healthcare decision-making [7], EEG signal processing integrated with computational algorithms based on machine learning methods may contribute to a deeper comprehension of the disease, as well as to support physicians in early diagnosis [8]. However, a visual inspection performed by trained experts remains the most reliable manner to diagnose a disease. The recent advancements of Information and Communication Technologies (ICT) allow accessing a large amount of information remotely [9], and assessing missing value from data as well [10]. In this scenario, the cloud represents a practical solution to the problems of storing and sharing a large amount of electronic health records (EHR) or other types of health data, providing several benefits to the user and organization. Indeed, cloud systems can be used to share information in realtime and to optimize economic and temporal resources [11], besides to be a tool to solve the problem of prediction from a huge healthcare database [12], cutting down costs required by management and administration of physical infrastructures. It allows for higher productivity compared to previous manual exchange of data. However, some security requirements for data sharing in cloud computing systems have to be guaranteed [13]. Thus, the provider must ensure data security and privacy of sensitive information, especially in complex domains like healthcare [14].
Open Access
According to their accessibility, the cloud systems can be classified into: (1) public (external), when it is accessible from anywhere over the internet such as Google's Drive online storage; (2) private (internal), when its infrastructure is developed to manage private cloud environment; (3) hybrid, as a combination between public and private cloud systems [15].
The use of a security system based on Public Key Infrastructure (PKI), symmetric and asymmetric key encryption and digital signature may guarantee a fine-grained access control scheme able to maintain the security and integrity of data on public cloud infrastructures, preserving patient privacy, and preventing the policy enforcers from comprehending the access control policies transmitted with the data [16]. Symmetric-key encryption is based on cryptography algorithms using only one key for both encryptions of plaintext (human-readable data) and decryption of the unintelligible ciphertext [17]. On the contrary, asymmetrical encryption uses a pair of public and private keys to encrypt and decrypt the data, according to the RSA algorithm developed by Rivest, Shamir, and Adleman [18]. Thus, the data can be encrypted using the public key and decrypted through the private key, which is known only to the owner. Similarly, digital signature employs asymmetric cryptography to provide data authentication and integrity [19]. Therefore, the private key is used to encrypt owner information data, whereas the public key to check the owner of the data.
In this paper we report a portable PC application, called ReportFlow, that we developed to share sensitive data over public cloud, and to speed and simplify the medical report process of EEGs reporting and delivering. The paper is structured as follows: in the methods section, we introduce the case study that inspired our work and describe the functioning of ReportFlow, proving cloud specifics and the key management adopted. In the next sections, we report and discuss some preliminary results. Finally, we expose the limitations, advantages, and disadvantages of such an application, comparing it with the most recent related works.
Case study
The IRCCS Centro Neurolesi Bonino Pulejo of Messina (Italy) includes a neurological main center and three external facilities where different units take place. EEGs and evoked potentials of children are performed at the neurophysiology diagnostic unit (NDU) placed at the main center, but to report the examination it is needed to wait for a child neuropsychiatrist who comes from the peripheral unit (which is distant about 25 km). Sometimes, this issue can also require a couple of days because of the duties at the peripheral unit.
To fulfil the request of EEGs reporting by optimizing times and costs, we developed, in Python programming language (version 3.7), ad-hoc PC application called ReportFlow for sharing instrumental examinations among members of a clinical team including staff from different units (i.e. child neuropsychiatry, neurophysiology laboratory, and administrative office). ReportFlow exploits the public cloud platform and a PKI system. We tested the application to improve the procedure of EEGs reporting adopted in our hospital setting, which encompasses the four steps summarized in Table 1.
The ReportFlow application functioning
The functioning of ReportFlow uses a role-based access control (RBAC) [20], an approach to restricting system access to authorized users. Thus, some operations and features are differentiated to the user according to his role, identified by the certificate stored in a USB Write Once Read Many pen drive. Notably, to protect the data stored on the Cloud, ReportFlow takes care of all cryptographic activities, besides to manage all certificates and check their validity by using OpenSSL, an open source general-purpose cryptography library. Public keys and other information are directly stored on the public cloud into specific folders. ReportFlow encrypts the data through a Triple Data Encryption Symmetric Algorithm (Triple DES or 3DES) [21], using a 128 characters randomly generated password before uploading it on the Cloud. This password will be used to decrypt the file when the user (i.e. physician or the medical clerk) will need to use it. ReportFlow encrypts the random password using the public keys of the user according to his role. This ensures that only authorized operators can decode files. The encryption process takes place locally on the PC of the user, whereas the results are uploaded on the Cloud. The cloud syncing activity is not performed directly by ReportFlow but is demanded to the cloud application installed on the operator PC. Figures 1 and 2 show the ReportFlow functioning, whereas an additional text file shows the pseudo-code (see Additional file 1).
The cloud platform
The cloud platform used in this study is Google Drive online storage of G-Suite [22]. The provider's account (i.e. the IRCCS), which has unlimited storage space, is used to set and manage a shared drive with employees. The shared drive contains three different folders, which are accessible to employees according to their responsibilities (roles) within the company, e.g. all team members Table 1 The EEGs recording and reporting process pre-post the development of ReportFlow
Before ReportFlow After ReportFlow
Step 1 At the main centre, the neurophysiology technician performs the EEGs. The recordings are stored on the lab's PC At the main centre, the neurophysiology technician performs the EEGs. The recordings are encoded and stored on the Cloud by ReportFlow, i.e. the application updates the synchronized folder containing the EEG records. Then, ReportFlow notifies to the physician that a new examination has been recorded Step 2 The physician (i.e. the child neuropsychiatrist) bi-weekly goes to the main centre to report the EEGs performed. Then the reports are printed and personally signed by the physician The physician uses ReportFlow from a PC of any IRCCS facility, selects the EEG record and reports it. The report is encoded and digitally signed by the physician. ReportFlow takes care to notify and update the synchronized folder containing the EEG reports. ReportFlow notifies to the medical clerk that a new examination has been reported Step 3 EEG reports are archived by a medical clerk The medical clerk uses ReportFlow to validate the correct archiving of the EEG reports. ReportFlow sends to the patient the EEG report by email Step 4 The medical clerk calls the patient to warn that the EEG report is ready and can be either personally picked up If the patient does not provide an email, the medical clerk calls him to warn that the EEG report is ready and can be either personally picked up Fig. 1 The EEG reporting and delivering process of the NDU. The first folder, shared in read-only mode, includes all certificates; the second folder, shared in read/ write mode, includes the EEGs recorded (i.e. XML files containing both the EHR and the diagnostic examination, for each patient); and the third folder shared only to physicians and administrative staff, includes the EEGs reported. To keep in sync, the local folder with Cloud folder the Google Drive File Stream application was used, running on Windows and Mac OSX, while the Google Drive Ocamlfuse was used on Linux OS.
The cryptography and key generation
For the management and administration of keys and certificates, the software XCA is used. A private key is created for the provider, and with this key, in turn, a Certification Authority (CA) certificate is associated with the NDU involved. Thus, for each subject in NDU, an RSA private key and an X.509 certificate, signed with NDU-CA, are generated. To make the release of multiple certificates easier, the certificate signing request (CSR) is used. CSR is a special message for a CA used to apply for a digital identity certificate [23]. Indeed, CSR contains the public keys of the NDU and some identifying information, i.e. name, working unit, role. The same CSR can be used for different CAs to get several certificates, each of them associated with a different unit, in order to correctly identify the staff member and his associated privileges. Each member of the NDU is provided with a USB Key containing the duple {private key; X.509 certificate}, as well as the ReportFlow application with needed libraries, and the Cloud tools.
Data collection
Data referred to two consecutive semesters of 2019 were extracted from the administrative hospital database. To identify the examinations, we selected healthcare services with code 89.14.x (electroencephalogram) and 84.15.x (evoked potential) of the Italian national nomenclature of outpatient specialist care, adopted following the Ministerial Decree of January 2017. The only exclusion criterion applied was to be over 18 years of age.
Statistical analysis
We defined a list of indexes and we compared them before and after the ReportFlow development, i.e. we compared the first semester of ReportFlow use (T1) with the previous semester (T0). Statistical analysis was performed by using the 3.5.0 version of the open-source software R. A p < 0.05 was considered as statistically Fig. 2 Flow chart of the EEG reporting and delivering process significant. Results for continuous variables were expressed in mean ± standard deviation, whereas categorical variables in frequencies and percentages. The X 2 test with continuity correction was used to assess for statistical differences in proportions, whereas the unpaired Student t-test was used to compare continuous variables.
Results
As reported in Table 2, we found a significant reduction of average times in both EEG exam reporting (t = 19.94; p < 0.001) and delivering (t = 14.95; p < 0.001) when the ReportFlow application was used. Figure 3 shows the magnitude of these time reductions, more evident in EEG report delivery times. Moreover, the rate of phone calls to patients was significantly lower (χ 2 = 94.87; p < 0.001), the number of EEG/EP exams performed increase of 20%, and the child neuropsychiatrist was able to visit about 30% of outpatients more than before. Finally, with the introduction of ReportFlow, about 68% of exam reports were delivered completely digitally.
Discussion
The results confirm that the use of ReportFlow can improve the process of EEG reporting and delivering, reducing times and increasing performance. Indeed, we observed an increase in the number of neurophysiologic Fig. 3 Boxplots of reporting times and delivering times before and after the use of ReportFlow examinations performed, as well as the number of outpatients visits at the child neuropsychiatric unit. The service with the greater benefit was the report delivery, thanks to the speedup of the administrative procedure. Therefore, the use of ReportFlow supported the hospital in cost-saving (e.g. for paper, stationery, phone calls) and facilitated the patients as well.
Expanding the use of ReportFlow also to other repositories, in future it will be possible to manage different kinds of data, and offer a wide range of services through a process completely dematerialized, i.e. reporting and delivering of Magnetic Resonance Imaging evaluations, genetic testing and so on.
All crypto functions (coding and decoding) were implemented on the client-side by ReportFlow, without exposing sensitive data over the public cloud. Indeed, ReportFlow uses the OpenSSL PKI system to store data over the public cloud, ensuring the security level needed by healthcare institutions, and also making the EEG reporting process faster and simpler than the process originally used in our structure. Healthcare companies often keep legacy systems, i.e. old method, technology and application programs. Even if these systems work satisfactorily, they could be replaced by more secure, accurate and modern systems [24]. In our case study, the legacy process forced personnel to be physically present at the same location where the neurophysiological exam was carried out, besides requiring printing the EEG reports, which now are directly delivered to the patients digitally by ReportFlow. Thus, the hospital provides a more efficient service by reducing costs and waiting times for both personnel and patients. Indeed, physicians can report an evaluation everywhere and at any free time, e.g. during a break between two visits; the medical clerk can monitor in real-time the state of the reporting, soliciting the physician to report urgent cases; the patient has to wait less time for obtaining the exam report. On the whole, all involved staff can monitor the process through the ReportFlow interface, which includes a notification system.
Limitations, advantages and disadvantages
The main limitation of this study concerns the key management, since the key and certificate preparation is not automated and needs trained SysAdmins. Thus, in future we plan to develop an internal tool for USB key preparation with the generation of private key and CSR, allowing the staff members to request the certificate automatically as well. Moreover, the application does not check the quality control of EEGs and the management of report format. However, in future it will also include a semiautomated data quality control script before data sharing, in order to ensure that planned acquisition parameters and data annotation are followed and data files are not corrupted. A disadvantage of this method could be the possibility of occurring in conflicting activities, for example when two or more physicians simultaneously report the same outcome. In order to minimize such a conflictual eventuality, Report Flow application applies some techniques i.e. using lock file (with double check) and deleting from the cloud all decoding passwords except for the current user. Unfortunately, this problem could not be totally solved since the cloud infrastructure is here used as a global notification mechanism.
Working in a public environment, the most important disadvantage of ReportFlow concerns data security. Although the protection and privacy of data hosted by cloud providers are not so high as in a private cloud, ReportFlow encode data using a random generated key in turn encoded using a PKI system and, with this cryptography process, the security of the patients' sensitive data can be protected from third parties. Moreover, even if we have used Google Drive, ReportFlow is not linked to any specific cloud services and it is economically convenient. Indeed, ReportFlow did not constitute an additional cost for the company because Google is the actual Cloud Service, which is one of the most stable and scalable cloud solutions. Finally, ReportFlow does not need specific customization, but it is only based on syncing features of the cloud service. This avoids limits and restrictions in customization due to the public infrastructure of the cloud.
Related work
The advent of the cloud has created a new paradigm in providing and using infrastructures, leading to new challenges concerning security and privacy of sensitive data, especially in the healthcare environment [25]. Several contributions in method development for healthcare data sharing we can find in literature. In this section we summarized some previous methods of data sharing on cloud in the healthcare environment, and compared them with ReportFlow according to the access control, the encryption technique, and the key management (Table 3).
Recently, Pugazhenthi and Chitra [26] described a method called IDHKE able to securely share the secret keys to the receiver in the stage of decryption, since an encryption is used. The key is safely generated using one random prime number, a master secret key and parameter value. In their work, Ahmed et al. [27] described a method providing patient privacy and accountability in the health information sharing environment by using the open-source CONNECT software to enable eHealth Exchange specifications, although their research lacked a thorough representation of dynamic access-control policy solutions. In another study, Basu et al. presented a cloud-based platform for securely managing and sharing healthcare information at a large scale, but without an access structure to clarify data-sharing management [28]. Similarly, Marwan et al. [29] proposed a novel methodology based on a multi-cloud concept, to meet the security level required by health care, avoiding loss of data, unauthorized access, and privacy disclosure. Sneha and Asha [30] proposed to use k-anonymity for privacy-preserving EHR. Ibrahim et al. [31] provided a comprehensive solution for EHR by using a cryptographic role-based technique and the Kerberos protocol to carry session keys. However, these studies require neither a privacy-aware cloud infrastructure or specific protocols, with the disadvantage of having to sustain hardware acquisition costs, besides requiring skilled human resources to manage, monitor and support the cloud infrastructure. This can be a problem for healthcare companies, which are lacking software developers and computer scientists, and they usually pay for cloud services such as Google or Microsoft Azure. Therefore, ReportFlow was aimed to offer a working model simple and efficient, without using special protocols or middleware applications over the cloud infrastructure.
Works in public key encryption environments are very limited. For example, Hwang et al. [32] constructed a knapsack encryption scheme for the problem of public key encryption based on permutation combination algorithm. However, this permutation method appeared to be not efficient to the security of the scheme [33]. Shao et al. proposed a public key encryption protocol supporting multiple receivers for medical information sharing based on bilinear maps. Data owner stored only one copy of his encrypted file and its corresponding encrypted keywords on the cloud for multiple designated receivers [34]. A data-sharing system was proposed by Wei et al. [35], in which the data holder encrypts data with the public key and then uploads it to the cloud servers, regardless of various access requirements. Chu et al. [36] presented a public-key encryption scheme that produces constantsize cypher-texts for efficient delegation of decryption rights in the cloud data sharing in a hierarchical structure. However, these works in public key encryption have long ciphertext related to the number of receivers, do not support receiver revocation without re-encrypting, and do not preserve the membership of receivers. All crypto functions (encryption and decryption) of ReportFlow were implemented on the client-side, without exposing sensitive data over the public cloud. Most papers about EHR access control in public key environments are based on attribute-based encryption (ABE) to encrypt data and to provide the hierarchical access structure for fine-grained data sharing. However, in a practical application, EHR data could be stored in multiple clouds due to the need for scalability and privacy. Rezaeibagha and Mu [37] proposed an EHR data sharing system, based on several cryptographic building blocks and secret sharing, with RBAC to protect patient's privacy stored in different types of clouds (i.e. private and public clouds), whereas ReportFlow is based on Ciphertext-policy attribute-based encryption (CP-ABE) scheme, one of the most suitable in a public environment since it can guarantee data owners' direct control over their data and provide a fine-grained access control service. Indeed, users obtain their private keys only after data encryption, without knowledge of the actual set of users that will be able to decrypt only by specifying the actual policy [38].
Concerning electrophysiological data sharing, Thilakanathan et al. [39] presented a system using a sensor connected to a mobile phone via Bluetooth to stream encrypted electrocardiographic (ECG) data to the cloud. According to Tran et al. [40], the system is based on proxy re-encryption where keys are partitioned and shared with other physicians. Revoking a user would simply involve removing the corresponding physician's key partition in the cloud. Our system is similar but all encryption processes are handled directly by ReportFlow and not require other external devices, except for the one that contains the certificates. This device can be a USB key or a smartcard or also a Bluetooth connected mobile phone.
Conclusions
The ReportFlow application developed for sharing, visualizing, reporting and delivering EEG records, resulted to be an optimal solution to optimize the legacy process adopted in our hospital. Report Flow provides a userfriendly graphical interface in order to have a good learning curve for the hospital staff. Using ReportFlow the reporting process becomes independent by the location: technicians can take the diagnostic examination everywhere, also in patients' homes using a portable EEG recorder, and the physician can visualize and evaluate the EEG tracing at any time, even from a remote location. Moreover, the EEG report is instantly available, and the administrative staff can archive it in real-time, while the application automatically delivers it to the patient. The comparative pre-post analysis showed promising preliminary results of performance, although the application is still in the testing phase. Notably, the report delivering service was sensitively speeded up due to the improvement of the whole process.
Health data security is the main purpose of Report Flow. Such purpose is achieved executing all encoding and decoding processes locally exploiting the OpenSSL PKI system. Thus, the public cloud is used only as a storage of encrypted data.
Future directions will be the creation and release of certificates automatically, as well as the implementation of an automatic process for revocations management. A future challenge will be to improve the current locking mechanism, to avoid simultaneity in data transcription, and notify early the progress of the reporting process.
|
2021-01-07T09:01:31.057Z
|
2021-01-06T00:00:00.000
|
{
"year": 2021,
"sha1": "d590f09df95d447f8c62c5265c47e65e94c609e3",
"oa_license": "CCBY",
"oa_url": "https://bmcmedinformdecismak.biomedcentral.com/track/pdf/10.1186/s12911-020-01369-7",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "c3ff26f2ee86c45a5b0fb1d84441e486718f6603",
"s2fieldsofstudy": [
"Medicine",
"Computer Science"
],
"extfieldsofstudy": [
"Computer Science",
"Medicine"
]
}
|
258213637
|
pes2o/s2orc
|
v3-fos-license
|
rAAV-PHP.B escapes the mouse eye and causes lethality whereas rAAV9 can transduce aniridic corneal limbal stem cells without lethality
Recently safety concerns have been raised in connection with high doses of recombinant adeno-associated viruses (rAAV). Therefore, we undertook a series of experiments to test viral capsid (rAAV9 and rAAV-PHP.B), dose, and route of administration (intrastromal, intravitreal, and intravenous) focused on aniridia, a congenital blindness that currently has no cure. The success of gene therapy for aniridia may depend on the presence of functional limbal stem cells (LSCs) in the damaged aniridic corneas and whether rAAV can transduce them. Both these concerns were unknown, and thus were also addressed by our studies. For the first time, we report ataxia and lethality after intravitreal or intrastromal rAAV-PHP.B virus injections. We demonstrated virus escape from the eye and transduction of non-ocular tissues by rAAV9 and rAAV-PHP.B capsids. We have also shown that intrastromal and intravitreal delivery of rAAV9 can transduce functional LSCs, as well as all four PAX6-expressing retinal cell types in aniridic eye, respectively. Overall, lack of adverse events and successful transduction of LSCs and retinal cells makes it clear that rAAV9 is the capsid of choice for future aniridia gene therapy. Our finding of rAAV lethality after intraocular injections will be impactful for other researchers developing rAAV-based gene therapies.
INTRODUCTION
Gene therapy is undergoing a successful revival for disorders with unmet therapeutic needs.To date, recombinant adeno-associated viruses (rAAVs) are one of the most frequently used vectors for in vivo delivery of gene therapies [1].rAAV-based gene therapies have shown acceptable long-term safety and efficacy profiles in several clinical trials, leading to FDA-approval of Luxturna in 2017 and Zolgensma in 2019 [2].However, recent studies have raised safety concerns associated with high doses of rAAV in both preclinical studies [3][4][5] and clinical gene therapy trials [6,7].Therefore, when developing a new rAAV gene therapy, elements such as viral capsid, route of administration, dose, and promoter need to be carefully considered.Among these elements, we will explore the choice of capsid and route of administration in our study to pave the way for developing a new rAAV-based therapy for the eye disorder aniridia.
Aniridia is a rare congenital eye disorder, which is caused by autosomal dominant variants in the paired box 6 (PAX6) gene.It clinically manifests as reduced visual acuity and photophobia in childhood due to hypoplasia of the retinal fovea and iris, and gradually leads to blindness by the second to third decade of life due to glaucoma and/or slowly progressing corneal vascularization and clouding [8,9].This time course means there is a therapeutic window available to undertake treatment.To date, various symptomatic interventions are available for aniridia patients to slow the disease progress, but none prevent the eventual blindness.Limbal stem cell (LSC) deficiency has been proposed as the underlying mechanism responsible for corneal clouding in aniridia [10][11][12] and as an important contributing factor to the corneal epithelial thinning observed in this disease [13][14][15][16].In a healthy cornea, the slow-dividing LSCs are present throughout life and continuously replenish the epithelium with new cells to maintain the transparency of the cornea [17].This makes LSCs of the cornea important targets for intervention in patients with aniridia.
The rAAV capsid controls tropism and together with the route of administration has the ability to greatly influence the safety and efficacy of gene therapy [18].To date, 13 naturally occurring AAV serotypes have been identified, and numerous variants within these serotypes have been isolated from various species including humans and non-human primates (NHPs) [19].rAAVs were generated on the foundation of these naturally occurring capsids and have been studied by many groups to determine their tissue tropism [20].rAAV9 is among the most-commonly used capsids for systemic delivery to the central nervous system (CNS), owning to its natural ability to cross the blood-brain barrier (BBB) after intravenous injection leading to neuronal tropism in neonate pups, and astrocytic tropism in adult mice [21] as well as in NHPs [22].In the last decade, it has become increasingly popular to apply capsid engineering approaches such as directed evolution to further improve the transduction efficiency of rAAVs.In 2016, capsid engineering studies led to the development of rAAV-PHP.B (hereafter referred to as PHP.B) with superior BBB-crossing ability after intravenous injection in adult mice compared to its parental rAAV9 capsid [23].The following year, a directed evolution strategy led to the discovery of a second-generation capsid, rAAV-PHP.eB,with further enhanced CNS transduction efficiency after systemic delivery in adult mice [24], creating new opportunities for capsid selection when developing gene therapy.However, shortly after, researchers discovered that the excellent properties of PHP.B capsids for systemic delivery to the CNS depends on interaction with the LY6A receptor and therefore are mouse strain [4,[25][26][27] and species-specific [4,28,29].In spite of this, the PHP.B family of capsids is still widely used in preclinical gene therapy studies [30][31][32][33][34][35] and are valuable in demonstrating efficacy in a mouse model prior to NHP studies, recognizing that the final capsid selection must be optimal for humans.
Intravitreal injections of rAAV9 have previously proven effective for the transduction of various PAX6-expressing retinal cells in wildtype (WT) mouse eyes [36][37][38].These studies used a "current" marker (e.g., green fluorescent protein) showing where the virus is currently, which may be suitable for modeling traditional geneaugmentation therapies.In contrast, much less work has been done with rAAV carrying a "historical" marker (e.g., Cre, rearranging the genome), which aims to show all cells transduced over time and thus is better suited for modeling newer gene editing therapies such as CRISPR.Cre has been previously used in the cornea to show the ability of intrastromal injection of rAAV8[Y733F] to reach PAX6expressing corneal epithelial cells, as well as the therapeutically significant corneal LSCs in WT mouse eyes [39].Now it is important to build on these works and explore the transduction patterns of rAAV9 and the novel PHP.B capsids in the aniridic retina and cornea using a historical marker to make the findings translatable, especially to gene editing strategies for this disease.
There is a good mouse model of aniridia known as Small eye (Sey), which is also caused by a dominant variant in the Pax6 gene [16,40].The corneal abnormalities seen in these mice mimic those of aniridia patients [16] and have been similarly attributed to defects in LSCs [41].The Sey mice used in this study have been extensively characterized [14] and bred to additionally carry a loxP-3xStop-loxP-tdTomato Cre-reporter allele [42].The resulting mice have previously been used for testing and optimizing gene editing delivery methods for aniridia [43].Injecting these reporter mice with rAAV expressing improved Cre (iCre) removes the transcriptional stop signal upstream of tdTomato gene, resulting in tdTomato epifluorescence expression that can be visualized at the single cell level in histological analysis, allowing characterization of vector target-cell specificity.This has been demonstrated for corneal LSCs in WT mice when transduction of LSCs led to tdTomato expression, the cells divided and gave rise to tdTomato positive transient amplifying cells, that became terminally differentiated epithelial cells of the central cornea [39].This cell movement in the corneal epithelial layer generated a red stripelike banding pattern and indicated the successful targeting of functional LSCs [39].
Here we present our findings from testing two capsids (rAAV9 and PHP.B), in two mouse strains (WT and Sey), by three different administration routes (intrastromal, intravitreal, and intravenous).We recommend the best capsid and administration routes for future gene therapy for aniridia.In addition, these findings deliver an important message of rAAV lethality for all rAAV ocular gene therapies in development.
Mice were maintained in the pathogen-free Centre for Molecular Medicine and Therapeutics (CMMT) mouse facility on a 7 a.m.-7 p.m. light cycle, 20 ± 2 °C with 50 ± 5% relative humidity, and had food and water ad libitum.All procedures involving mice were in accordance with the Canadian Council on Animal Care and UBC Animal Care Committee (Protocols A17-0204, A17-0205, A21-0140, and A21-0184).
Intravitreal and intrastromal injections
Sample sizes (number of mice per group) were based on a previously published similar experiment [45].Mice were genotyped and grouped by one person and injections were done on a separate day by a different person who did not re-examine the eyes.Adult mice (2 months old for intravitreal injections and 3 months old for intrastromal injections) were weighed, anesthetized with isoflurane inhalation, and injected with 5 mg/kg Metacam (DIN: 02240463) for pain management.All procedures were performed while mice were at surgical plane of anesthesia.Alcaine eye drops (DIN: 00035076) were applied to numb the eye receiving the injection, and Optixcare (Aventix, Burlington, Canada) lubricating eye gel was applied to the contralateral eye to prevent dryness.Custom-made 33and 34-gauge removable Small Hub RN needles (length: 9.53 mm; point style: 4; bevel angle: 30°) (Hamilton Company, Reno, NV) were used for intravitreal and intrastromal injections, respectively.For rAAV injections, 2 μl of undiluted virus at a titre of either 4.2 × 10 13 vg/ml of PHP.B iCre or 3.8 × 10 13 vg/ml of rAAV9 ssAAV-smCBA-iCre-WPRE (hereafter referred to as rAAV9 iCre) in PBS + P was loaded into a 2.5 μl syringe (Cat #7632-01, Hamilton Company).For vehicle-only injections, 2 μl of PBS + P was loaded into the syringe.Final injection doses were: 8.4 × 10 10 vg/mouse for PHP.B iCre and 7.6 × 10 10 vg/mouse for rAAV9 iCre.Intravitreal injection was performed by inserting the needle perpendicular to the sclera close to the limbus region and slowly advancing within sclera to vitreous chamber before dispensing the virus.Intrastromal injection was performed following a previously described method [39,43].Briefly, the needle was inserted into the stroma near the corneoscleral junction and was carefully advanced to the center of the stroma where the payload was dispensed.The central injection method in combination with the injection volume were chosen with the goal of reaching as much of the cornea as possible after a single injection.Injection success was a predetermined inclusion criterion for intrastromal injections, as this is a challenging technique to perform on the small mouse eye.Mice that did not receive a successful intrastromal injection were immediately euthanized.Mice were weighed and eyes closely examined daily for five days following injection.After that, mice were monitored weekly until the first lethality was observed, leading to careful examination of all the remaining mice twice a week until they were euthanized and tissues were harvested from a subset.
Slit lamp imaging
Slit lamp imaging was performed according to a previously described method [49].Briefly, 3-month-old mice were placed under general anesthesia.The Phoenix Micron Anterior Segment Imaging system (Phoenix Research Labs, Pleasanton, CA) was used to visualize the corneal phenotype of the mouse left eye.
Histology
Intraocularly-injected mice were initially planned to be harvested at 1 month and 7 months post-injection, but due to the unforeseen lethality in PHP.B iCre mice, the experimental mice were euthanized and tissues were harvested once they reached a humane endpoint and the nonreporter control mice were euthanized without harvesting.Despite the absence of phenotype in rAAV9 iCre mice, harvest time for these mice was also adjusted to match the PHP.B iCre mice for comparison purposes.However, in the case of intrastromal rAAV9 iCre injections, a subset of injected mice was aged and harvested at 7 months post-injection to allow for the formation of LSC stripes.At the time of harvest, mice were euthanized by anesthetic avertin (Cat #T48402, Sigma-Aldrich) overdose, and transcardially perfused according to a previously described method [50].Tissues were then harvested and post-fixed overnight in 4% paraformaldehyde in PBS at 4 °C.On the following day, for cryosections, tissues were transferred to 25% sucrose in PBS for a minimum of 24 h at 4 °C or until they were processed.Tissues were embedded in Tissue-Tek optimal cutting temperature compound (Cat #4583; Sakura FineTek, Torrance, CA) in cryomolds on dry ice.Next, 20 μm thick cryosections were cut and directly mounted onto Superfrost Plus slides (Cat #1255015, Thermo Fisher).Slides were stored at −80 °C until staining was performed.For staining, slides were warmed to room temperature for 5 min and pressed underweight for 30-60 min.Cryosections were then washed twice in PBS and incubated with Hoechst nuclear stain (1:1000) in PBS for 10 min.Following incubation, cryosections were washed twice more with PBS, before being mounted with ProLong Gold Antifade reagent (Cat #P10144, Invitrogen).Slides were kept in dark at 4 °C until visualization.For corneal flat mounts, fixed eyes were stored in PBS until they were processed.Eyes were placed in a petri dish filled with PBS.Corneas were dissected using surgical scissors (Cat #15000-03, Fine Science Tools, Foster City, CA) and forceps (Cat #11251-33, Fine Science Tools) under a Leica MZ APO dissecting microscope (Leica Microsystems, Wetzlar, DE).Four radial cuts were made into the corneas before transferring them onto slides (one cornea per slide).Corneas were then mounted with ProLong Gold Antifade reagent with DAPI (Cat #P36935, Invitrogen) and flattened with a cover slip.Slides were kept in dark at 4 °C overnight before imaging.All injected eyes were processed and analyzed, whereas non-ocular tissues from a minimum of 2 mice per genotype were processed and analyzed for each group at each harvest time point.Results were confirmed by an individual blinded to treatment groups.
Microscopy and image processing
tdTomato epifluorescence was visualized in cryosections and corneal flat mounts using an Olympus BX61 fluorescent microscope (Olympus, Tokyo, Japan) at either 4× magnification (cerebellum, spinal cord, liver, heart, and pancreas) or 10× magnification (eye cryosections and corneal flat mounts).The cerebellum and eye images were taken as Tile scans.Corneal flat mounts were re-imaged as Tile scans using a Leica SP8 confocal microscope (Leica Microsystems) at 20× magnification by z-projection through the epithelium that was defined at the center of each cornea.The resulting images were processed using Image J (http://rsbweb.nih.gov/ij/), and Illustrator (Adobe, Mountain View, CA).In some images, Hoechst/DAPI was enhanced consistently throughout the entire image for better visualization.
Statistical analysis
GraphPad Prism version 9.1.1 for Windows (GraphPad Software, San Diego, CA, www.graphpad.com)was used for all statistical analysis and plotting Kaplan-Meier survival curves with 95% confidence intervals (based on the asymmetrical method) [51].Log-rank test was performed for statistical comparison of the survival curves and p < 0.05 was accepted as statistically significant.
PHP.B iCre injections caused lethality in mice, regardless of administration route
Our previous work with rAAV9 shows its favorable tropism for the retina and cornea after intraocular injections in WT mice [36][37][38].However, rAAV9 is known to be inefficient in transducing the retina after intravenous injection in adult mice [52].At the time this study began, PHP.B was chosen as the most advanced capsid for non-invasive systemic delivery to the retina in adult mice [23].Therefore, the PHP.B capsid was used for all three administration routes (intravenous, intravitreal, and intrastromal), whereas the use of rAAV9 was limited to intraocular injections.
Intravenous injections were carried out at five doses of PHP.B iCre virus, with mice injected at the three highest doses showing a progressive phenotype leading to lethality (Fig. 1A).Uninjected mice served as controls.Mice injected with the highest dose of 5 × 10 11 vg/mouse were first observed to have a mild "clasping" phenotype for their hind legs 33 days post-injection.The clasping phenotype progressed to hind leg paralysis and these animals were euthanized 43 days post-injection.Mice injected with the second highest dose of 1.5 × 10 11 vg/mouse were first observed to have a mild clasping phenotype for their hind legs 58 days postinjection, with the phenotype progressing quickly, the remaining mice were euthanized 65 days post-injection.Mice injected with the third highest dose of 7.5 × 10 10 vg/mouse were first observed to have a very mild clasping phenotype in their hind legs at 69 days post-injection, with the phenotype progressing to severe and mice beginning to have difficulty moving around the cage due to hind leg paralysis at 153 days post-injection, at which point the mice were euthanized.Mice injected with the fourth highest dose (second lowest) of 3 × 10 10 vg/mouse showed a mild clasping phenotype seven months (215 days) post-injection.Mice injected with the lowest dose of 1.5 × 10 10 vg/mouse showed no phenotype at seven months (215 days) post-injection, which was the experimental end point.The Kaplan-Meier survival curves were significantly different (p < 0.0001) among the five groups according to the log-rank test.Although the lowest intravenous dose (1.5 × 10 10 vg/mouse) did not result in any observable phenotype, we concluded that this dose was probably too low to have a therapeutic impact.Therefore, to continue to compare both capsids we chose to focus on intraocular (intravitreal or intrastromal) injections.
To maximize transduction, and because we did not anticipate any deleterious impact of intraocular injections, undiluted PHP.B and rAAV9 iCre viruses were used for intravitreal injections.Throughout these experiments, non-reporter virus injected mice and vehicleonly injected mice served as controls to account for the potential impact of transgene expression or delivery method on the mice wellbeing.Experimental mice and non-reporter control mice were intravitreally injected with 8.4 × 10 10 vg/mouse of PHP.B iCre.Surprisingly, we found a dead mouse among the experimental mice 13 days post-injection.Upon careful examination of the remaining injected animals, mice in both the experimental and non-reporter control groups, displayed ataxia and weight loss and were subsequently harvested between 17 and 62 days post-injection when they reached a humane endpoint.The ataxia phenotype observed was characterized by motor coordination and balance deficits, as well as head and body tremors.Overall, this phenotype had a more rapid onset, was faster progressing, and more severe than what we observed after the intravenous injections.It is worth noting that ataxia is not part of the aniridia phenotype in either patients with heterozygous PAX6 loss-of-function variants [53] or Sey mice [14].Importantly, in a parallel study, experimental mice and non-reporter control mice were intravitreally injected with 7.6 × 10 10 vg/mouse of, not PHP.B, but rAAV9 iCre.At the time of harvest (62 days post-injection), these mice did not display any adverse events.As expected, no phenotype was observed in mice that received vehicle-only injections.
Similar to intravitreal injections, to maximize transduction, PHP.B and rAAV9 viruses were used at their highest possible dose for intrastromal injections.These injections were initiated at the time when we did not anticipate any deleterious impact of intrastromal injection.Non-reporter virus injected mice and vehicle-only injected mice served as controls.Experimental mice and non-reporter control mice were intrastromally injected with 8.4 ×10 10 vg/mouse of PHP.B iCre.Surprisingly, a dead mouse was observed at 16 days post-injection.Except for one mouse in the experimental group, all remaining mice in both experimental and non-reporter control injected groups displayed varying degrees of ataxia and weight loss after the injections and were subsequently Fig. 1 Mice injected with PHP.B or rAAV9 iCre viruses had different survival curves.Cages were closely examined for phenotype once or twice a week.Animals were euthanized once they reached a humane endpoint and tissues were harvested from a subset.Kaplan-Meier survival curves (solid or dashed lines) were generated with 95% confidence intervals (dotted lines).A Intravenous injections of high doses of PHP.B iCre (5 × 10 11 , 1.5 × 10 11 , and 7.5 × 10 10 vg) caused severe adverse events leading to euthanasia of the entire dose group on the same day.Intravenous injections of high doses of PHP.B iCre significantly reduced the probability of survival compared to the groups that received lower doses (3 × 10 10 and 1.5 × 10 10 vg).B(i), C(i) Intravitreal and intrastromal injections of PHP.B iCre significantly reduced the probability of survival compared to rAAV9 iCre injected mice.B(ii), C(ii) The probability of survival was not significantly different between the PHP.B iCre injected WT and Sey mice.B(iii), C(iii) The probability of survival was not significantly different between the PHP.B iCre injected tdTomato reporter and non-reporter mice.Log-rank test was used to determine the significance of the data.iCre improved Cre, N Sig non-significant, Sey Pax6 small eye, Sig significant, vg viral genome, WT wild type.Fig. 2 Intravitreally-delivered PHP.B and rAAV9 iCre viruses escaped the eye.Fluorescence images of mouse tissues (cerebellum, spinal cord, liver, heart, and pancreas).Injection doses were: 8.4 × 10 10 vg/mouse for PHP.B iCre and 7.6 × 10 10 vg/mouse for rAAV9 iCre.Mice were perfused 17-30 days after intravitreal injection of iCre viruses (see Methods for more detail).Tissues from a minimum of 2 mice per genotype were analyzed.The sectioned tissues were evaluated for tdTomato epifluorescence.The tdTomato signal was observed in the central nervous system cerebellum, and spinal cord of PHP.B iCre injected WT and Sey mice, but not in rAAV9 iCre injected mice.In contrast, tdTomato signal was detected in the peripheral tissues liver and heart, but was almost absent in pancreas, in both PHP.B iCre and rAAV9 iCre injected WT and Sey mice.tdTomato expression was not observed in age-matched vehicle-only injected mice (negative controls).The number in brackets refers to the mouse ID.Scale bar = 500 μm.Blue Hoechst nuclear stain, iCre improved Cre, red tdTomato epifluorescence, Sey Pax6 small eye, WT wild type.Fig. 3 Intrastromally-delivered PHP.B and rAAV9 iCre viruses escaped the eye.Fluorescence images of mouse tissues (cerebellum, spinal cord, liver, heart, and pancreas).Injection doses were: 8.4 × 10 10 vg/mouse for PHP.B iCre and 7.6 × 10 10 vg/mouse for rAAV9 iCre.Mice were perfused 16-30 days after the intrastromal injection of iCre viruses (see Methods for more detail).Tissues from a minimum of 2 mice per genotype were analyzed.The sectioned tissues were evaluated for tdTomato epifluorescence.The tdTomato signal was observed in the central nervous system cerebellum, and spinal cord of PHP.B iCre injected WT and Sey mice, but not in rAAV9 iCre injected mice.In contrast, tdTomato signal was detected in the peripheral tissues liver and heart, but was almost absent in pancreas, in both PHP.B iCre and rAAV9 iCre injected WT and Sey mice.tdTomato expression was not observed in age-matched vehicle-only injected mice (negative controls).The number in brackets refers to the mouse ID.Scale bar = 500 μm.Blue Hoechst nuclear stain, iCre improved Cre, red tdTomato epifluorescence, Sey Pax6 small eye, WT wild type.Fig. 4 In PHP.B ocular-injected mice tdTomato signal was higher in ataxic versus healthy mice.Fluorescence images of tissues from earlyharvested mice (17 days post-intravitreal injection and 16 days post-intrastromal injection) with severe ataxia compared to late-harvested mice (62 days post-intravitreal injection and 68 days post-intrastromal injection).A Intravitreal PHP.B iCre injections: Tissues from WT and Sey mice with severe ataxia exhibited a higher tdTomato intensity and spread in comparison to healthy mice.B Intrastromal PHP.B iCre injections: Tissues from WT and Sey mice with severe ataxia exhibited a higher tdTomato intensity and spread in comparison to healthier mice.The number in brackets refers to the mouse ID.A subset of mice presented in Figs. 2 and 3 was re-analyzed for this comparison.Scale bar = 500 μm.Blue Hoechst nuclear stain, iCre improved Cre, red tdTomato epifluorescence, Sey Pax6 small eye, WT wild type.
harvested between 16-68 days post-injection when they reached a humane endpoint.The fast-progressing ataxia in these mice was similar to our observations after intravitreal injections.Importantly, in a parallel study, experimental mice and non-reporter control mice were intrastromally injected with 7.6 ×10 10 vg/mouse of, not PHP.B, but rAAV9 iCre.At the time of harvest (68 days post-injection), these mice did not display any adverse events.As expected, no phenotype was observed in mice that received vehicle-only injections.
Kaplan-Meier survival analysis showed significantly reduced survival in mice intraocularly injected (intravitreal or intrastromal) with PHP.B iCre compared to those injected with rAAV9 iCre (logrank test, p = 0.0005 for intravitreal injections and p < 0.0001 for intrastromal injections) (Fig. 1Bi and Ci).Survival curves were not significantly different between WT and Sey mice that received intraocular injections of PHP.B iCre (log-rank test, p = 0.97 for intravitreal injections and p = 0.42 for intrastromal injections) (Fig. 1Bii and Cii).This demonstrates that survival of these mice was not influenced by Pax6 genotype.Lastly, survival curves were not significantly different between experimental and non-reporter control groups that received intraocular PHP.B iCre injections (logrank test, p = 0.15 for intravitreal injections and p = 0.10 for intrastromal injections) (Fig. 1Biii and Ciii).Hence, the observed lethality was attributed to the PHP.B iCre virus, rather than tdTomato or lacZ expression.
Intraocular injected rAAVs escaped from the eye
To determine the rAAV transduction in non-target cells after intraocular injections (intravitreal or intrastromal), we evaluated cerebellum, spinal cord, liver, heart, and pancreas for the presence of tdTomato expression (Figs. 2 and 3).Histological analysis of early-harvested (16-30 days post-injection) tissues revealed tdTomato expression in the liver and heart, but was almost absent in pancreas, in both PHP.B and rAAV9 iCre injected WT and Sey mice.This demonstrates the ability of both these capsids to escape from the eye, regardless of the route of administration.Nevertheless, no tdTomato expression was observed in the cerebellum and spinal cord of mice intraocularly injected with rAAV9 iCre, which were the mice that did not display any adverse events.Whereas the PHP.B iCre injected mice demonstrated strong tdTomato expression in these tissues, regardless of the route of administration, and demonstrated severe ataxia and lethality.Histological analysis of tissues from vehicle-only control mice did not show any tdTomato expression.
In PHP.B ocular-injected mice tdTomato signal directly correlated with severity of ataxia
We compared tissues from early-harvested (16-17 days postinjection) mice that had severe ataxia with late-harvested (62-68 days post-injection) mice to better understand the correlation between tdTomato expression and the severity of phenotype.For the intravitreal PHP.B iCre injections of experimental mice, we had two healthy WT and three healthy Sey mice at the end of the study.For the intrastromal injections we only had one healthy Sey mouse at the end of the study, as the only WT mouse that made it to 68 days had moderate ataxia, it was included in this analysis.In order to do a head-to-head comparison, tissues from early-harvested PHP.B iCre experimental mice (Figs. 2 and 3) were re-analyzed together with lateharvested tissues under the same histology and imaging conditions.We found that for both WT and Sey mice, tdTomato signal intensity and spread was higher in the cerebellum, spinal cord, liver, and heart of mice with severe ataxia compared to healthy mice after intravitreal PHP.B iCre injections (Fig. 4A).Similarly, for the intrastromal PHP.B iCre injections, tdTomato signal intensity and spread was higher in the cerebellum, spinal cord, liver, and heart of mice with severe ataxia compared to healthier mice (Fig. 4B).
Intravitreal PHP.B and rAAV9 iCre injections resulted in transduction of a variety of retinal cells Gene therapy for aniridia will be focused on delivering virus to the PAX6-expressing cells of the retina (ganglion, amacrine, horizontal, and Müller glia) and cornea (LSCs and epithelial).Intravitreal PHP.B iCre injections resulted in tdTomato expression in all four PAX6expressing cells of the retina (identified by location and morphology), in both WT and Sey mice (Figs. 5 and S1).Whereas, intravitreal injection of rAAV9 iCre into WT mice led to tdTomato expression in all but the Müller glia.In contrast, in Sey mice tdTomato expression was observed in all four PAX6-expressing cells.Interestingly, tdTomato expression was not restricted to the retina despite the intravitreal injection site being close to the retina (Fig. S1).Expression was also observed in the corneal stromal and endothelial cells, but was almost absent from the epithelial layer in both WT and Sey mice, regardless of rAAV capsid.tdTomato expression was absent in the vehicle-only injected eyes (negative controls).
Intrastromal PHP.B and rAAV9 iCre injections resulted in transduction of all three corneal cell layers Intrastromal PHP.B and rAAV9 iCre injections resulted in tdTomato expression in all three corneal cell layers, encompassing corneal epithelial cells (basal cells, wing cells, and squamous cells), stromal cells (keratocytes), and endothelial cells, identified by location and morphology in both WT and Sey mice (Fig. 6 and S2).Additionally, sectioning demonstrated that the central intrastromal injection method used resulted in even transduction throughout the cornea (Fig. S2).tdTomato was only expressed in the retinal blood vessels of intrastromally-injected mice (Fig. S2).tdTomato expression was absent in the vehicleonly injected eyes (negative controls).
Intrastromal rAAV9 iCre injections successfully transduced corneal LSCs in adult aniridic mouse eyes
In Sey eyes, the exact age at which the LSC deficiency manifests is unknown.In this study, all intrastromal injections were performed when mice were 3 months old.By 3 months, the cloudy cornea phenotype of aniridia is already apparent in the Sey eyes (Fig. 7).The success of an rAAV gene therapy targeted towards LSC deficiency may depend on whether any functional LSCs are still present in the damaged Sey corneas at the time of injection.To investigate this, we looked at WT and Sey corneas 7 months after intrastromal rAAV9 iCre injections.It must be noted that tdTomato expression at 7 months post-injection is not a reflection of continuing virus presence, but is due to the permanent genomic rearrangement by the iCre/loxP recombination system that makes the transduced cells, and any cells descending from them, to persistently express tdTomato.Fluorescence microscopy demonstrated widespread Fig. 5 Intravitreally-delivered iCre viruses successfully transduced a variety of retinal cells and some corneal cells.Fluorescence images of injected mouse retinas and corneas.Injection doses were: 8.4 × 10 10 vg/mouse for PHP.B iCre and 7.6 × 10 10 vg/mouse for rAAV9 iCre.Intravitreal injection of PHP.B iCre into WT mice resulted in tdTomato expression in all four PAX6-expressing cells of the retina, ganglion, amacrine, horizontal, and Müller glia.rAAV9 iCre injection resulted in tdTomato expression in three PAX6-expressing cells, excluding Müller glia.But, intravitreal injections of PHP.B and rAAV9 iCre into Sey mice resulted in the expression of tdTomato in all PAX6-expressing cells of the retina.Intravitreal injections of PHP.B and rAAV9 iCre viruses also resulted in tdTomato expression in the stromal and endothelial layers of the cornea, but tdTomato expression was almost absent from the epithelial layer in both WT and Sey mice.tdTomato expression was not observed in age-matched vehicle-only injected eyes (negative controls).Retinal cells were identified by their location and morphology.Images with the same Roman numeral belong to one mouse, and correspond to overview images in Fig. S1.For the PHP.B iCre WT and Sey groups, each fluorescence image is representative of 11 eyes.For the rAAV9 iCre WT and Sey groups, each fluorescence image is representative of 4 and 8 eyes, respectively.Scale bar = 100 μm.A amacrine, blue Hoechst nuclear stain, End endothelium, Epi epithelium, G ganglion, GCL ganglion cell layer, H horizontal, iCre improved Cre, INL inner nuclear layer, IPL inner plexiform layer, M Müller glia, ONL outer nuclear layer, OPL outer plexiform layer, red tdTomato epifluorescence, Sey Pax6 small eye, Str stroma, WT wild type.
tdTomato expression in the cornea of both WT and Sey mice (Fig. 8).Confocal microscopy confirmed that the tdTomato stripes are located in the corneal epithelium.In general, in the Sey eyes, the tdTomato stripes that originated from the limbus, as expected did not reach the center of the cornea [41].Overall, two out of three WT corneas and five out of eight Sey corneas showed tdTomato stripes in the epithelium, indicating that functional LSCs are present in 3month-old Sey corneas, which can be successfully transduced with rAAV9.tdTomato expression was absent in the cornea of vehicleonly injected mice (negative controls).
DISCUSSION
For ocular gene therapy, rAAV has been a popular choice [1].Given the deteriorating state of the aniridic eye [54], a nonocular delivery system such as intravenous was quite attractive.However, recent news of several deaths associated with intravenous delivery of high-dose rAAV therapies [6,7] represents a major setback for the field and emphasizes the importance of establishing maximally-tolerated viral dose during preclinical gene therapy studies.Thus, the primary goal of our intravenous rAAV dose-response studies was to find the maximally-tolerated viral dose.Unfortunately, we concluded that the only dose of PHP.B iCre that did not cause toxicity in our mice was probably too low to be a viable option for our gene therapy.Thus, to continue to compare both capsids we shifted to the option of intraocular delivery.
Luxturna, an rAAV2-based augmentation gene therapy delivered intraocularly was approved in 2017 by the FDA for treating patients with a rare form of inherited vision loss [55].Given this, and the intravenous toxicity, we shifted our focus to intraocular delivery with an aim to concentrate the rAAV in the eye and improve the safety and efficacy profiles.We were surprised to see severe ataxia and early lethality in our intravitreally-injected mice when the PHP.B capsid was used.It was even more shocking to see severe ataxia and early lethality after intrastromal PHP.B iCre injections, considering that the injection was done into the cornea, an avascular tissue in WT eyes.These observations were especially unexpected since in the literature ataxia has not been reported with the use of PHP.B in mice, and currently the only report of PHP.B-associated ataxia comes from the study in piglets that received high dose of intravenous injection of this virus [3].Thus, this study is the first report of ataxia and lethality after intraocular injections of PHP.B.Furthermore, the adverse events that we observed in our study after intravitreal and intrastromal injections of PHP.B were substantially more severe than the local ocular toxicity (defined as morphological changes to the retina and retinal pigment epithelium) that another group observed in mice after subretinal injections of multiple rAAVs (although not PHP.B) [56].Interestingly, the PHP.B capsid was used by an independent group for successful intravitreal delivery of CRISPR/ Cas9 components to the retina of mouse pups (1 × 10 9 vg/eye) [57].The absence of toxicity could be due to lower virus dose, different promoter and regulatory elements, but is most likely due to the very short harvest time of 2 weeks.In our work, we are confident that the observed lethality is not due to promoter, viral genome dose, or vector preparation, as both PHP.B (lethal) and rAAV9 (non-lethal) iCre viruses had the exact same viral genome; were used at comparable doses; and were prepared in the same facility.Our findings are consistent with a group that showed intravenous injection of PHP.B virus at a high dose in an adult NHP resulted in severe acute toxicity and ultimately euthanasia, while the animal that received the same dose of rAAV9 showed no clinical symptoms [4].Together, these results highlight the inherent differences in tropism in even closely related capsids (PHP.B and rAAV9) that can dramatically impact safety.
The eye is an easily accessible and contained structure, and therefore is widely considered to be a safe target for gene therapy [58].Typically, tissue transduction after escape from the eye has not been reported, even for the PHP.B capsid [57].However, several studies have performed biodistribution analysis after intraocular rAAV2 [59], rAAV8 [60][61][62][63], or rAAV8G9 [64] injections and detected vector genome in extraocular tissues.A limitation of these studies is that they do not make it clear whether or not extraocular cells had been transduced.In the current study, histological analysis demonstrated that both PHP.B and rAAV9 viruses escaped from the eye after intravitreal and intrastromal injections, and then transduced peripheral tissues.The "historical" reporter used in this study (a genomic rearrangement turning on tdTomato), may have more effectively captured the true extent of the viral escape since transduced cells are permanently and clearly marked.This is compared to the more commonly used "current" reporters (e.g., green fluorescent protein), for which signal is determined by viral copy number, and there is no signal if the virus is lost from the cell.Applying Occam's razor to explain both intravitreal and intrastromal escapes, we suggest that PHP.B and rAAV9 capsids have crossed the adult blood-ocular barrier from the eye side into the bloodstream, resulting in extensive transduction of non-ocular tissues.In the future, we would recommend replacing a ubiquitous promoter as used here, with a target-cell-specific promoter to minimize transgene expression in off-target tissues [36][37][38]45].
In this study, we must consider the mechanism whereby PHP.B and rAAV9, which both reached the bloodstream regardless of route of administration, differentially resulted in ataxia and death (PHP.B lethal and rAAV9 non-lethal).It is important to note that in adult mice, PHP.B capsids outperform rAAV9 in crossing the BBB and transducing the CNS [23,25].Our results support this in that we saw transduction of cerebellum and spinal cord only with the lethal PHP.B capsid.Another candidate region from NHP studies would be the dorsal root ganglia (DRG) [3,5,65], however, DRG are not protected by the BBB and both PHP.B and rAAV9 capsids have been shown in NHP to have similar tropism for DRG [4].Thus, in our study, we conclude that the ataxia phenotype observed with PHP.B is most likely due to differential transduction of the cerebellum and/or spinal cord.
LSC deficiency has been proposed as the underlying mechanism responsible for corneal clouding, a major contributor to blindness in aniridia [10][11][12].Thus, LSC are a critical therapeutic target.Intrastromal injection of rAAV8[Y733F] has been shown to transduce LSCs in adult WT mice [39], but similar data was not available for Sey mice.know from other work that PAX6 haploinsufficiency results in fewer functional LSCs compared to WT corneas [41].Nevertheless, the exact age of onset for LSC deficiency in the Sey cornea is unclear and therefore there was uncertainty whether rAAVs would be able to transduce these therapeutically important cells in an already cloudy Sey cornea.Here we have shown for the first time that 3-month-old cloudy Sey corneas still have functional LSCs and that they can be transduced with rAAV9.As expected, the tdTomato stripes from these LSCs fail to reach the center of the Sey corneas, consistent with the previously observed defective migration [41].Regarding route of administration, we were surprised to see that intravitreal injection of rAAV9 was not only excellently suited for transducing the therapeutically important retinal cells, but it also reached the cornea.Since it would be exciting if one injection could transduce both the therapeutic retinal and corneal LSC targets, this observation deserves further study.Overall, our findings are most promising in the context of a future gene therapy for aniridia, which can transduce LSCs and may restore the normal migration pattern from these cells, thereby eliminating corneal clouding and the resulting blindness.
In this study, we conclude that PHP.B iCre can cause lethality in mice regardless of the administration route (intravenous, intravitreal, or intrastromal).Furthermore, we hypothesize the mechanism of escape from the eye by both capsids lies in their ability to cross adult blood-ocular barrier from the eye to the bloodstream, leading to the transduction of non-ocular tissues.Their differential tropism for CNS is the likely explanation for the observed lethality phenotype with PHP.B only.Most importantly, we have demonstrated that functional LSCs are present in the cloudy cornea of 3month-old Sey mice.Lack of adverse events and successful transduction of the therapeutically important corneal LSCs and retinal cells with rAAV9 in the Sey eyes makes this capsid a good candidate for use in future aniridia gene therapy development.Finally, our findings regarding rAAV lethality will be impactful for other researchers working on developing rAAV-based gene therapies for ocular disease.
Fig. 6
Fig. 6 Intrastromally-delivered iCre viruses successfully transduced all three corneal cell layers.Fluorescence images of injected mouse corneas.Injection doses were: 8.4 × 10 10 vg/mouse for PHP.B iCre and 7.6 × 10 10 vg/mouse for rAAV9 iCre.Intrastromal injections of PHP.B and rAAV9 iCre showed successful transduction of cells in all three layers of the cornea (Epi, Str, and End) in WT and Sey mice.In the corneal epithelial layer, tdTomato expression was seen in basal cells, wing cells, and squamous cells.Sections were chosen to show all the transduced cell types in one image.Overall, the results did not show a bias in transduction of corneal cell types between PHP.B and rAAV9.tdTomato expression was not observed in age-matched vehicle-only injected eyes (negative controls).Corneal cells were identified by their location and morphology.Images with the same Roman numerals belong to one mouse, and correspond to overview images in Fig. S2.For the PHP.B iCre WT and Sey groups, each fluorescence image is representative of 6 and 10 eyes, respectively.For the rAAV9 iCre WT and Sey groups, each fluorescence image is representative of three eyes.Scale bar = 100 μm.B basal cell, blue Hoechst nuclear stain, E endothelial cell, End endothelium, Epi epithelium, iCre improved Cre, K keratocyte, red tdTomato epifluorescence, S squamous cell, Sey Pax6 small eye, Str stroma, W wing cell, WT wild type.
Fig. 7
Fig. 7 Corneal clouding was present in 3-month-old Sey corneas.Representative slit-lamp images of uninjected Sey and WT eyes at 3 months old.Sey eyes were imaged at two light settings to better capture the cloudiness of the corneas.All Sey eyes displayed severe corneal clouding, whereas age-matched WT eyes had transparent corneas.Sey Pax6 small eye, WT wild type.
Fig. 8
Fig. 8 Intrastromal injections of rAAV9 iCre transduced corneal LSCs in Sey mice.Representative fluorescence and confocal images of WT and Sey corneal flat mounts 7 months after injection.Corneal flat mounts were first examined under the fluorescence microscope to assess tdTomato expression in all three layers of the cornea, and then subjected to confocal microscopy to focus specifically on the entire corneal epithelial layer.Intrastromal injections of rAAV9 iCre resulted in tdTomato stripes indicative of LSC transduction in both WT and Sey corneas.tdTomato expression was not observed in age-matched vehicle-only injected corneal flat mounts (negative controls).Blue DAPI nuclear stain, red tdTomato epifluorescence, Sey Pax6 small eye, WT wild type.
|
2023-04-20T06:16:19.849Z
|
2023-04-19T00:00:00.000
|
{
"year": 2023,
"sha1": "69b28196710272804b15a521a526a5855d966628",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s41434-023-00400-6.pdf",
"oa_status": "HYBRID",
"pdf_src": "Springer",
"pdf_hash": "398c5a81943b996a608286cecd3ada2655431125",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
10988545
|
pes2o/s2orc
|
v3-fos-license
|
Nonholomorphic N=2 terms in N=4 SYM: 1-Loop Calculation in N=2 superspace
The effective action of N=2 gauge multiplets in general includes higher-dimension UV finite nonholomorphic corrections integrated with the full N=2 superspace measure. By adding a hypermultiplet in the adjoint representation we study the effective action of N=4 SYM. The nonanomalous SU(4) R-symmetry of the classical N=4 theory must be also present in the on-shell effective action, and therefore we expect to find similar nonholomorphic terms for each of the scalars in the hypermultiplet. The N=2 path integral quantization formalism developed in projective superspace allows us to compute these hypermultiplet nonholomorphic terms directly in N=2 superspace. The corresponding gauge multiplet expression can be successfully compared with the result inferred from a N=1 calculation in the abelian subsector.
Introduction
The nonholomorphic N = 2 potential H(W,W ) in the effective action of SYM theories has been an object of study for some time now [1]- [3]. This potential is integrated with the full N = 2 superspace superspace measure and therefore is a dimensionless real function of the N = 2 gauge field strengths W andW . In the nonabelian sector it contributes to the N = 1 Kähler potential K(φ,φ) and in the abelian sector it can only contribute to N = 1 higher derivative terms [2]. Scale invariance and U(1) R invariance restricts the form of the N = 2 potential to be where H o depends on gauge invariant, scale independent combinations of the nonabelian N = 2 field strengths. The pure abelian piece is contained only in the second term. In N = 4 SYM theories the abelian nonholomorphic potential is believed to be generated only at 1-loop [3] since higher loop and nonperturbative contributions would break the scale and U(1) R invariance of H. Nonperturbative contributions have been studied in [4] and they give vanishing results. It is therefore possible to determine the exact form of H Abelian by performing a 1-loop calculation. Recently this type of calculation has been done in N = 1 superspace [5] by considering the higher derivative operator This is one of the N = 1 components of (1) [2]. Its contribution to the 1-loop abelian effective action was computed using N = 1 superspace quantization. The resulting coefficient c was found to be nonvanishing. This has interesting implications for 3-branes in ten dimensions because they are believed to be effectively described by N = 4 SYM at low energies: the presence of a nonvanishing N = 2 nonholomorphic potential introduces acceleration dependent terms in the scattering of 3-branes in addition to the standard velocity dependent terms [5].
In this article we compute the nonholomorphic corrections to the N = 4 SYM effective action directly in N = 2 superspace using the N = 2 path integral quantization that we developed in [6], [7]. This quantization involves N = 2 superfields that contain the familiar N = 1 hypermultiplet and gauge degrees of freedom 1 .
First we calculate all finite 1-loop corrections to the N = 2 hypermultiplet effective action dropping terms with spinor or space-time derivatives on the external fields. The calculation is greatly simplified using N = 2 gauge propagators in the Landau gauge.
We then isolate the dependence of the effective action on the N = 2 hypermultiplet superfield whose N = 1 projection is part of the chiral hypermultiplet isodoubletῩ 0 | θ 2 α =0 = Q. This contribution to the N = 2 effective action is a nonholomorphic potential H(Ῡ 0 , Υ 0 ) whose N = 1 projection can be rotated by a Z 2 subgroup of the global SU(4) R of N = 4 SYM into the N = 1 projection of a corresponding pure gauge piece H(W,W ). Symmetry arguments therefore determine the form of the 1-loop N = 2 nonholomorphic potential in the low energy gauge effective action of N = 4 SYM.
The N = 2 potential we find for the abelian sector is of the form (1). The coefficient c is exactly the same as that calculated in N = 1 superspace [5]. Due to the nonlinearity of the nonabelian superfield strengths it is not clear if the 1-loop nonabelian piece H o can be reproduced from the knowledge of the hypermultiplet effective action and we cannot test the proposal in [2].
N=formalism
In this section we briefly review the superfield content of the N = 4 SYM in N = 2 superspace and we give the Feynman rules for quantization of these N = 2 multiplets. For a more detailed explanation we refer the reader to [6]- [7] and references therein. The conventions we follow are those of ref. [8].
Gauge multiplets and hypermultiplets can be described by off-shell representations of N = 2 supersymmetry using superfields that live in projective superspace. This is a subspace of N = 2 superspace whose anticommuting coordinates are the following linear combinations of the N = 2 Grassmann coordinates: Θ α = θ 2α − ζθ 1α andΘα =θα 1 + ζθα 2 parameterized by a complex projective coordinate ζ. Accordingly, projective superfields Ω obey the constraint Charged hypermultiplets can be described by an infinite power series in the projective coordinate 2 . We refer to this multiplet as the polar multiplet As a consequence of the constraints (3) the highest order coefficient is a chiral superfield in N = 1 superspaceDαῩ 0 = 0 and the next order is a complex linear superfieldD 2Ῡ 1 = 0. These two superfields contain the physical degrees of freedom of the hypermultiplet. The other coefficients are auxiliary superfields in N = 1 superspace.
Gauge vector multiplets are described by an infinite series with negative and positive powers of the projective complex coordinate that we call the tropical multiplet. This multiplet is real under conjugation and since there are no lowest order or highest order coefficients, all of them are unconstrained in N = 1 superspace The coefficients v n , |n| > 1 are gauge degrees of freedom, v 0 is related to the usual N = 1 real gauge prepotential v = v 0 + nonlinear corrections of the covariantly chiral spinor field strength and v −1 is related to the prepotentialψ = iv −1 + nonlinear corrections of the ordinary chiral scalar φ that appears in the N = 1 components of the classical gauge action The manifestly N = 2 supersymmetric action describing the coupling of a polar hypermultiplet in the adjoint representation of the gauge group to the tropical gauge multiplet is the following This action can be dualized to the give the usual description of the hypermultiplet in terms of two chiral fields [7]: the complex antilinear superfield Υ 1 is traded for a chiral Lagrange multiplier Q, the chiral coefficient superfieldῩ 0 is identified with its N = 2 partnerQ and the auxiliary superfields decouple. The resulting interacting action is the usual one The convention we adopt to define the path integral is such that the kinetic term of scalars is convergent in euclidean space For the gauge multiplet, the kinetic piece of (6) has a simple expression in terms of tropical multiplets while the interaction vertices are more complicated. Since we are only going to compute 1-loop amplitudes with external hypermultiplets coupling to internal gauge multiplets, all we need is the gauge propagator in projective superspace. The kinetic action (10) is therefore enough to use the path integral quantization of the model. We recall examine the Feynman rules that we use to compute the set of diagrams proposed. The polar hypermultiplet propagator is [6] where C 2 (A) is the second Casimir in the adjoint representation of the gauge group T rT a T b = C 2 (A) δ ab and ∇ 4 1 = ∇ 2 (ζ 1 )∇ 2 (ζ 1 ). The tropical multiplet propagator is [ where α denotes the gauge fixing parameter (the factor g 2 in the propagators is the one consistent with the holomorphic normalization of the kinetic term as opposed to canonical normalization in the sense of [11]). The interaction vertices that contribute to the relevant graphs are obtained from the first two orders in the expansion of (7) Now we can use all the powerful tools of path integral quantization to calculate the 1loop effective action of the polar multiplet. The only peculiarity of the N = 2 formalism is that we must complete the Grassmann measure of each vertex to have a full N = 2 superspace measure. This procedure eliminates four projective spinor derivatives in one of the propagators stemming from the vertex. For example in the vertex with an external arctic multiplet Once the measure has been completed in all the vertices, we do the "D"-algebra to reduce all propagators but one to bare Grassmann delta functions. These are the basic Feynman rules that we use in the next section to construct the 1-loop effective action of the polar multiplet.
1-loop nonholomorphic terms in the hypermultiplet effective action
Now that we have presented the rules to calculate Feynman diagrams in N = 2 superspace, we focus our attention on those amplitudes of interest to us. We want to consider graphs with any number of external polar multiplets at zero momentum. The calculation is greatly simplified working with the gauge propagator in the Landau gauge α = 0. To illustrate the techniques used in this novel N = 2 quantization we present the simplest graphs in some detail. The more complicated ones only involve a larger amount of algebra. In the N = 2 formalism tadpoles and seagulls vanish automatically [6], and at one loop we always have the same number of external arctic and antarctic polar fields. Therefore the first graph we study is the two point function. After completing the Grassmann measure on both vertices we find the graph on Fig. 3 and a similar graph in which the external hypermultiplets are exchanged.
The "D"-algebra on this graph is trivial: we already have one bare propagator and the other one is acted upon by eight spinor derivatives. We just have to reduce The resulting contribution to the effective action −iΓ(2) is UV finite (this is the well known nonrenormalization of hypermultiplets) and local in the N = 2 Grassmann coordinates (16) Next we consider the graphs with four external polar multiplets. After completing the superspace measure in the vertices we obtain the graphs in Fig. 2 and similar ones in which we exchange the external hypermultiplet of each cubic vertex by the internal hypermultiplet.
The "D" algebra is trivial only on the upper graph. In the other two we integrate by parts the spinorial derivatives of all propagators but one. As usual, it is easiest to do so on the graph. Since we are interested on nonholomorphic terms without external derivatives we keep only the contribution where all spinor derivatives end up acting on the same propagator.
Finally we can integrate the bare Grassmann delta functions and reduce the spinor derivatives acting on the last one. For example in the middle graph As a result all the graphs with four external hypermultiplets and no external derivatives have the same euclidean loop momentum integral id 4 p/(−2πp 2 ) 4 . The complex coordinate dependence is slightly different though. The first graph gives a contribution to −iΓ(4) the second is and the last one This simple result illustrates a few features that will be reproduced by higher n-point functions: i) integrating the complex coordinates we can see that the coefficient superfields enter quadratically Υ iῩi ; i) graphs containing one or more internal hypermultiplet propagators do not contribute terms that depend purely on powers of Υ 0Ῡ0 .
This simplifies our calculation considerably because we are interested in selecting terms that depend only on the N = 2 superfield containingQ = Υ 0 | θ 2 α =0 . Terms mixing auxiliary superfields Υ i , i > 1 and Υ 0 do not modify the pureῩ 0 Υ 0 piece because they enter at least quadratically. We may set the auxiliary fields to zero using their algebraic field equations.
We focus our attention on Υ 0 because the N = 1 superfieldQ is rotated by Z 2 subgroup of SU(4) R into the N = 1 gauge scalarφ. Since this symmetry is nonanomalous the nonholomorphic potential must be accompanied by a corresponding nonholomorphic function of N = 2 superfields whose N = 1 projection is preciselyφ, φ. These are the N = 2 chiral gauge field strengths W,W . Now that we know what kind of amplitudes to look for, we collect all the relevant graphs with any number of external hypermultiplets but no internal hypermultiplets and find This is almost the Taylor expansion of a logarithm but it is missing the first order term. To find this term let us go back for a moment to the result of the first graph (16). It does not seem to contain a piece depending on the chiral superfield we are interested inῩ 0 Υ 0 . Notice however that it is possible to rewrite the contour integrals in (16) as follows The first term is a projective quantity and therefore it vanishes if we integrate it with the full N = 2 superspace measure. Thus we obtain a contribution to the effective action that we can write in two equivalent ways To help us decide which form we use let us recall that the physical superfield Υ 1 | θ 2 α =0 is mapped by duality to one of the chiral fields Q of the hypermultiplet on-shell description and the superfieldῩ 0 | θ 2 α =0 is identified with its partnerQ. Since we expect the global SU(2) R symmetry of these two chiral fields to be realized in the effective action 3 it seems natural to choose an equally weighted combination This choice will prove to be correct when we compare the corresponding nonholomorphic gauge effective action with the result inferred from its N = 1 components [5].
Adding (22) and the Υ 0Ῡ0 piece in (25) we find the nonholomorphic contribution to the effective action The corresponding nonholomorphic potential for the N = 2 gauge field strength W is therefore H(W,W ) = 1 2 dp 2 (4π) 2 p 2 T r ln 1 + To simplify our analysis let us consider the case of SU(2) SYM. The gauge operator in the argument of the logarithm can be diagonalized [2] U (WW +W W ) The first term is just a numerical constant and the integration over Grassmann coordinates cancels it. Since the splitting point is arbitrary we can choose ξ ≫ 1 and the second term gives to an arbitrary degree of accuracy. In this form it is easy to see that the IR scale is irrelevant, since all the terms that depend on it are killed by the N = 2 superspace integral [2], [3]. Our calculation gives the correct answer for the 1-loop nonholomorphic abelian potential with a coefficient We can alternatively regularize the momentum integral using dimensional regularization H Abel (W,W ) = 1 (4π) 2 µ 2 −ǫ dp 2 (p 2 ) 1−ǫ ln 1 + The divergence of the integral can be isolated by standard manipulation The integral in (39) gives a finite constant and the upper limit of the last term contains the regulated divergence as we let ǫ → 0 . The resulting nonholomorphic potential is
|
2014-10-01T00:00:00.000Z
|
1998-04-01T00:00:00.000
|
{
"year": 1998,
"sha1": "08ef2119dfeeb71a110f6ee08668375118e7cd36",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/hep-th/9804010",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "08ef2119dfeeb71a110f6ee08668375118e7cd36",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
225387774
|
pes2o/s2orc
|
v3-fos-license
|
Advanced oxidation process based water disinfection- the microbiology beyond bacterial inactivation
Different water treatment regiments are revealed to have potential in enriching antibiotic resistant bacteria (ARB). Advanced oxidation processes (AOPs) based disinfection techniques have been studied widely in the recent times due to their advantages over conventional treatment methods. However, bacterial response and adaptations against the hostile environments of AOPs is not clearly understood yet. Based on the existing knowledge on the ways in which bacteria surpass the antibiotic treatment, here we propose few important aspects of bacterial adaptation which could be true for AOPs as well since both antibiotics and AOPs generate reactive oxygen species (ROS) during their modes of action. We discuss the plausible role of ROS in the selection of ARB and bacterial heterogeneity as a strategy to bypass the lethal action of AOPs. Understanding bacterial adaptation during disinfection plays a vital role in devising strategies to outclass the bacterial survival. Hence, more importance should be given to such studies in the near future for the successful implementation of AOPs.
Introduction
The quality of water holds an imperative role in shaping the economy of the world. The consequences of water contaminated with microbial pathogens are often daunting. Around 1.7 billion cases of diarrhoea, mostly due to the gastrointestinal infections caused by the consumption of contaminated water, are reported in children every year, thus resulting in the death of more than 2.2 million people worldwide ("Diarrhoeal disease"., "WHO | Waterrelated diseases," 2016). The health aspects of microbial contaminants present in drinking water is well documented in the literature (Ashbolt, 2015;Pandey et al., 2014). The adaptability of microbes, bacteria in particular, against the action of antibiotics at a staggering pace has made the situation more inimical. On the other hand, various biological water treatment methods are now being blamed for facilitating the evolution of antibiotic resistant bacteria (ARB) and the dissemination of antibiotic resistant genes (ARG) (Guo et al., 2017;Zheng et al., 2018). Disinfection strategies are of high asset in treating water because of their performance and efficacy. Among all the disinfection processes, chlorination and UV treatment are well explored for bacterial inactivation and are still practiced worldwide despite their disadvantages like formation of toxic by-products and the possibility of bacterial reactivation (Nyangaresi et al., 2018;Pichel et al., 2019). Moreover, the promotion of ARG holds true for chlorination according to some recent reports (Jia et al., 2019;. Advanced oxidation process (AOP) is another water disinfection strategy which includes processes like ozonation, photocatalysis, photo-Fenton process, electrocatalysis, sonophotocatalysis etc. (Deng and Zhao, 2015).
The commonality in the action of antibiotics and AOPs lies in the production of reactive oxygen species (ROS). While chlorination and UV have direct effects on the DNA, the ROS produced during AOPs target multiple sites in the bacteria viz. membrane, DNA and protein, thereby creating a hostile environment similar to that of antibiotics (Pichel et al., 2019;Van Acker and Coenye, 2017). It is to be noted that in AOPs, ROS is produced in the reactor system containing bacteria and it can further induce oxidative stress inside the cells by forming intracellular ROS, creating an environment akin to that of antibiotics (Giannakis et al., 2016). However, the fate of these effects depends on the bacterial stress response to the pressure caused by the process. This can result in bacterial inactivation, killing or survival.
Bacterial stress response to antibiotics is a well-studied subject whereas the studies related to the bacterial reaction during disinfection is mostly limited to the role of commercial disinfectants and disinfectant by-products in triggering antibacterial resistance (ABR) (Li and Gu, 2019;Poole, 2012;Tkachenko, 2018).
It is the ability of bacteria to attune to the external pressure that enables them to resist the action of antibacterial agents or processes including water disinfection. However, most of the studies on AOP based disinfection focusses on designing the process and synthesis of catalysts, which are indeed important, and neglects the response of bacteria during the process, which ultimately determines the fate of the process on a longer run. In this article, we aim to provide a comprehensive knowledge on the bacterial phenotypic responses and their genotypic determinants that are possible during water disinfection, focussing on AOPs, based on the existing information on bacterial response against stress inducing factors like antibiotics. We expect to bridge this gap by delivering the know-hows of water disinfection which are beyond bacterial inactivation and thus serving few essential strategies which can be considered during the development of water disinfection regiments in the future.
Stress response and ABR-the extended role of ROS in disinfection
The history of ABR dates back to early 1940's when bacterial penicillinase was discovered (Abraham and Chain, 1940). Interestingly, this was detected way before penicillin was instilled as a therapeutic agent, pointing at the ability of microbes to survive unfavourable conditions. Decades later, due to the misuse of antibiotics, we are now in an era with the arsenal of active antibiotics diminishing day by day due to the staggering pace of ABR evolution (Alekshun and Levy, 2007). It is of a huge concern especially when biological water treatment plants and some disinfection techniques select ABR and enrich ARGs during the treatment. AOPs are generally not attributed for the selection of ABR, mostly because of the lack of information, and only few data are available in the literature on this aspect. For example, a recent study on the continuous application of ozone resulted in the decrease of the targeted bacteria and ARGs although higher doses of ozone were required to decrease the regrowth potential of the bacteria. (Iakovides et al., 2019). On the contrary, a study conducted on the effect of ozonation on the overall bacterial community and ARGs alluded the selection of bacteria which are GC rich and hence more resistant to the treatment along with the enrichment of few resistant genes (Alexander et al., 2016). Therefore, it can be inferred that the selection of ABR during ozonation depends on the microbial communities present in the water and further enhancement is required to curtail the enrichment of ARGs.
The situation of ARG enrichment and thus water treatment facilities becoming a reservoir for the evolution of drug resistance is very intriguing as we ask what makes a microbe to upregulate its ARGs which are not directly related to the treatment. Now given the fact that majority of bactericidal antibiotics induce ROS formation, especially HO · , and it leads to ABR as a result of the oxidative stress response, it is justly possible that similar reaction might be happening inside bacterial cells when they are treated using AOPs (Cirz et al., 2005;Kohanski et al., 2007). As a consequence of bacterial SOS response due to antibiotic induced ROS formation inside the cells, mutation in the DNA occurs via the activation of error-prone polymerases. A comprehensive review on the oxidative stress response associated drug resistance can be found elsewhere (Dwyer et al., 2009). The situation becomes more complex when ROS produced during antibiotic treatment not only helps in resistance acquisition against that antibiotic but also partially protects the bacteria from antibiotics belonging to other classes (Hoeksema et al., 2018). Horizontal gene transfer (HGT) is regarded as one of the key driver of ARG dissemination. It is already known that disinfectants like chlorine and H 2 O 2 at sublethal concentration can promote inter and intra genera HGT (Zhang et al., 2017).
Similar experimental analysis on the impact of nanomaterials which are generally used in AOPs indicates a positive correlation with HGT Lu et al., 2020;. Strikingly, here also ROS plays a major role via the activation of SOS response followed by alteration in the cell membrane permeability and upregulation of the membrane associated HGT genes. Hence, not surprisingly, the bacteria exposed to photocatalysis are also reported to be on the edge of promoting HGT (Guo and Tian, 2019). However, it can be comprehended that all of these convoluted pathways are very unlikely to happen if the production of ROS is harsh enough to foster cell death. AOPs in general work via 3 ways. i) Exogenous production of ROS which attack the bacterial membrane, ii) Diffusion of ROS into the cytoplasm, iii) Endogenous production of ROS by the bacteria itself due to the oxidative stress triggered by exogenous ROS. Indeed, the question of the evolution of ARB will remain relevant only if the ROS produced during AOP is not sufficient enough to kill the bacteria. A graphical representation of ROS induced effects on bacterial survival/death is described in Figure 1.
The use of sublethal concentration of antibiotics has been an essential tool to study the evolution of ARB. Here, a set of bacteria is allowed to survive the treatment either by (Giannakis et al., 2018). In another study, the perineal exposure of E. coli to photocatalysis was found to have an accelerated impact on the ARB evolution . They further extended their work in understanding the stress response of the bacteria and how ARGs are regulated during the events of photocatalytic exposure. This experimental evidence hints at the role of ROS in eliciting the AR. In contrast, similar studies conducted by our group using sublethal sonophotocatalysis on Salmonella Typhimurium resulted in an overall decrease in the MIC of several antibiotics belonging to different class (Rahman et al., 2020). This could be due to the severe pressure offered by the process although similar responses should be verified on other bacterial species before we generalise the effect.
Figure 1. Illustration of the possible effects of ROS produced during AOPs against bacteria
Resistance acquisition is an energy intensive process and is mostly accompanied by a fitness cost. Briefly, the relative growth of ARB in the absence of antibiotics or stressor molecules will be lesser than its isogenic counterpart in order to balance out the expenditure of acquiring resistance (Andersson and Hughes, 2010;Melnyk et al., 2015). While addressing how the oxidative stress rendered by AOPs regulate the fitness cost in ARB, Yin et al cautioned about the accentuating effect of sublethal photocatalysis on reducing the fitness cost of ARB and thereby promoting the proliferation of resistance (Yin et al., 2019). More importantly, if the disinfection process is acting as factor behind compensatory evolution, a phenomenon by which ARB neutralizes the fitness cost, than a mere selective pressure should not be overlooked. Hence, it is very much desirable to use clinically relevant ARB alongside antibiotics for studying the influence of disinfection on the evolutionary traits of bacteria in selecting ARGs.
Disinfection processes were historically designed to kill or inactivate microbes but not to tackle the issue of the emerging ARB and ARGs. Current data also supports the view of bacteria getting resistant to disinfectants like antibiotics (Mc Carlie et al., 2020). The aforementioned data verily opens up the possibility of ROS induced cross resistance to antibiotics in bacteria during AOPs. Signifying the adaptive capability of bacteria against adverse conditions, they indeed pose high risk in causing process resistance unless significant enhancement in the process design and use of proper techniques to validate the cell death after the treatment is not put in place.
ROS induced heterogeneity in bacteria-a survival strategy
Heterogeneity is one among the many strategies devised by bacteria while responding to the cues of environmental stress. It can be explained as the tactics deployed by a subset of a clonal population to ward off a particular stress. Interestingly, this state can be achieved either via genetic modifications (genetic heterogeneity) or phenotypic modifications (phenotypic heterogeneity) (Davis, 2020). Phenotypic heterogeneity is a complex trait within an isogenic population in which a subpopulation having special phenotypic features emerge as a result of stress without having much alterations in the genomic content. Bacterial persistence against antibiotics is a good example of phenotypic heterogeneity. Since persistence is often used alongside resistance and tolerance in the scientific literature, a clear distinction in using these terms was proposed recently (Brauner et al., 2016). Briefly, resistance is the ability of bacteria to increase the MIC of an antibiotic, achieved by mutation or HGT, whereas tolerance is defined as the potential to withstand high concentration of antibiotics without changing the MIC. While the feature of tolerance displayed by the whole clonal population is similar to that of resistance, persistence is a non-hereditary form of tolerance exerted only by a subset of bacteria, resulting in a biphasic response to the treatment. Tolerance/persistence can be distinguished using minimum duration for killing Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 13 August 2020 doi:10.20944/preprints202008.0291.v1 (MDK) assay as it estimates the duration taken by the antibiotic at a concentration exceeding its MIC to kill the bacterial population (Brauner et al., 2016).
Slowing down the metabolic activity and thereby arresting the growth is a key factor which helps in the generation of persisters (Brauner et al., 2016). Among the multiple factors that have been identified as the key regulators of persistence, repression in SOS response appears to be the most important driver of persistence. The SOS response genes which are responsible for the inhibition of cell division were found to have upregulated in the persister population enriched by ampicillin treatment (Gefen and Balaban, 2009). Toxin-antitoxin (TA) modules have also been reported to have important function in bacterial persistence as they are closely associated with bacterial growth arrest (Gerdes and Maisonneuve, 2012). During stress, the antitoxin which inactivates the toxin gets destroyed and eventually the release of toxin results in shutting down the growth promoting factors in bacteria. When MqsR toxin in E. coli was made to overexpress, it resulted in the suppression of RpoS mediated stress response and triggered the formation of persisters. Interestingly, the same study indicated the repression of RpoS when the cells were treated with hydrogen peroxide and in turn a spike in persister colonies was observed (Hong et al., 2012). This suggests that the probability of the selection of bacterial persistence in ROS generating environments like AOPs should not be disregarded. This idea was supported by a very recent study where exposing bacteria to several cycles of sublethal photocatalysis improved the bacterial tolerance towards several antibiotics . Nevertheless, the final outcome of AOPs is likely to be dependent on the level of stress it incurred on the bacteria as understood from the TA modules which promoted persistence and thereby their survival during nominal stress as opposed to the activation of killing pathway when the stress level exceeded the limit of damage repair (Wu et al., 2011) Bet-hedging is another strategy exercised by a subset of bacteria to evade environmental stress where they preadapt before the onset of a particular stress (Schröter and Dersch, 2019).
Although this heterogeneous population carries a significant fitness cost in the environment without having the stress they adapted for, they tend to survive when they are exposed to the stress while the non-adapted get killed in the meantime. This is particularly advantageous for the bacteria which are already exposed to several fluctuating environments. For example, the T3SS (type 3 secretion system) expressing Salmonella subset injects effector proteins to the host cells to counteract the lethality of ROS during phagocytosis albeit the heavy fitness cost they carry due to this secretion which results in reduced metabolic activity and thereby increase in antibiotic tolerance (Diard et al., 2014). Bet-hedging becomes a very important mechanism of bacterial escape from the oxidative stress especially when they are pre-exposed to various scales of oxidative stress. Hence, it is very likely that the reactivated bacterial colonies post disinfection (if any) may show features of heterogeneity. As AOPs are generally considered in the tertiary steps of water treatment, it may have to encounter microbial communities which are pre-adapted to stress. Situation becomes more serious while considering hospital wastewater/ effluents as they foster the diversification of differentially evolved microbes to bypass adversities (Lamba et al., 2017;Petrovich et al., 2020). Hence, we urge the use of clinically and environmentally relevant bacterial strains expressing different phenotypes/genotypes not only for testing the efficacy of disinfection process but also to ensure that it is not selecting characteristics like persistence. Stress derived protein aggregation in a clonal population also appears to have functions in driving heterogeneity (Mortier et al., 2019). Protein aggregates formed during proteolytic stress are distributed stochastically to one of the cell poles via nucleoid occlusion and the ensuing asymmetric segregation of these proteins during the division of the surviving cells results in a heterogeneous population (Lindner et al., 2008). In a recent study, Govers et al reported that the protein aggregates inherited in E. coli, formed as a result of sublethal heat shock, elevated the heat resistance compared to its isogenic counterpart devoid of protein aggregates (Govers et al., 2018). Interestingly, it also improved the resistance against the proteolytic stress inflicted by ROS. A report by Shi et al proposed protein aggregation as the reason behind the bactericidal activity of photocatalysis using silver nanoparticles (AgNPs), refuting the existing evidences on the ROS generation and its associated lethality . They suggested the possibility of AgNPs acting as a mediator to transfer the light energy to the proteins, resulting in protein aggregation followed by cell death. Based on the proteomics data, they also claimed the damage to be irreversible and therefore it is very unlikely to have bacterial resistance against AgNPs in the presence of light. However, to validate the protein aggregation mediated cell death and to rule out the possibility of persistence associated, microscopic tools and MDK assay should have been performed respectively especially on the grounds of reports on the bacterial resistance to AgNPs and persistence linked with protein aggregation (Govers et al., 2018;Panáček et al., 2018).
Keeping the complexities associated with bacterial adaptation in mind, studies related to bacterial stress response during the disinfection would help to tweak the system in order not to get selected for traits like persistence.
Conclusions
The common feature shared by most of the antibiotics and AOP based disinfection processes is the production of ROS. The fate of bacterial survival under the action of ROS depends on the magnitude of the ROS production. Bacteria have evolved different strategies to outmanoeuvre the stress caused during antibiotic treatment. Like so, it is justly possible that they can bypass the lethal actions of AOPs unless effective measures are taken to curtail the stress response. Based on the existing knowledge on the bacterial stress response against antibiotics, we propose to extend this information for the betterment of AOP based disinfection ( Figure 2) as follows -Inclusion of sublethal AOPs to study the bacterial response and adaptation to the process via the plausible development of resistance and persistence to different antibiotics.
-Use of clinically and environmentally adapted bacteria for the disinfection process along with their wildtypes. -Exploration of disinfection efficacy against bacterial consortia in their natural form.
-Use of microscopic and flow cytometry based tools to validate the bacterial death.
-Above all, increase the process efficiency by increasing the ROS production, for example: development of nano-catalysts with superior ROS producing capacity.
Conflicts of interest
There are no conflicts to declare.
|
2020-08-20T10:09:32.627Z
|
2020-08-13T00:00:00.000
|
{
"year": 2020,
"sha1": "7cf3290dce81c2d9559aed5c74dc099270f63983",
"oa_license": "CCBY",
"oa_url": "https://www.preprints.org/manuscript/202008.0291/v1/download",
"oa_status": "GREEN",
"pdf_src": "Anansi",
"pdf_hash": "97ff001c440d60d8ca42414fd2d23a6b39c0f6ca",
"s2fieldsofstudy": [
"Environmental Science",
"Chemistry"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
56182825
|
pes2o/s2orc
|
v3-fos-license
|
Alien terrestrial crustaceans ( Isopods and Amphipods ) Chapter 7 . 1
A total of 17 terrestrial crustacean species aliens to Europe of which 13 isopods (woodlice) and 4 amphipods (lawn shrimps) have established on the continent. In addition, 21 species native to Europe were introduced in a European region to which they are not native. Th e establishment of alien crustacean species in Europe slowly increased during the 20th century without any marked changes during the recent decades. Almost all species alien to Europe originate from sub-tropical or tropical areas. Most of the initial introductions were recorded in greenhouses, botanical gardens and urban parks, probably associated with passive transport of soil, plants or compost. Alien woodlice are still confi ned to urban habitats. Natural habitats have only been colonized by three amphipod species in the family Talitridae.
Introduction
Th e orders in the arthropod subphylum Crustacea are mainly composed of aquaticliving species, at least during part of their life-cycle.Most alien terrestrial crustaceans belong to the order Isopoda, suborder Oniscidea, commonly named woodlice.But several species recorded in Europe belong to the order Amphipoda, and are commonly known as "lawn shrimps" or "landhoppers".
In 2004, the total number of valid Isopod species worldwide was 3637 (Schmalfuss 2003).Woodlice are adapted to various terrestrial environments from sea shores to deserts and have established on all continents.As decomposers of organic plant material, isopods play an important role in ecosystems (Holthuis et al. 1987, Zimmer 2002).Most European species prefer humid and moist micro-habitats (Vandel 1960) like soil, leaf litter, mosses and decaying wood.Several species are known for their myrmecophylic nature.
Amphipods are generally marine or limnicolous, and only a few species can live permanently on land (mainly in the family Talitridae).Some live near the sea, on beaches where they hide under logs and dead algae and vegetation.Th e true terrestrial amphipods live on the surface of mulch and moist ground (Fasulo 2008).Many of the habitat features of terrestrial amphipods are similar to those of isopods.Th ese little animals are most commonly noticed by their strong, rapid jumps upon being disturbed.
Taxonomy of alien terrestrial crustaceans
Th irty-eight species belonging to ten diff erent families were recorded during this study.Th e four most commonly represented families (all belonging to Isopoda) are Trichoniscidae (seven species), Porcellionidae (fi ve species), Philosciidae and Armadillidiidae, both with fi ve species (Figure 7.1.1.).Two main categories were considered: • Aliens to Europe, including 17 crustacean species originating from other continents ( Some of the species above have proved to be very successful colonizers and are currently considered as part of the native fauna in parts of Europe, e.g. in Hungary.However, their synanthropic nature and their extremely wide distribution range suggest a long colonisation history as it is the case for Armadillidium vulgare.
In the remainder of this chapter, we will focus mainly on the species alien to Europe.
Temporal trends of introduction in Europe of alien terrestrial crustaceans
Th e total number of crustaceans alien to Europe has slowly increased during the 20 th and the early 21 st centuries, but without any acceleration in the rate of arrival.Two alien species were fi rst discovered in Europe in the 19 th century, about nine species in the fi rst half of the 20 th century and only fi ve species since then.Th e majority of these alien species have been found in several other countries after their discovery in Europe.However, the number of occupied countries over time has grown steadily rather than exhibiting exponential growth.A similar pattern is apparent for woodlice species alien to Europe.However, because of sparcer information on this group, the date for the fi rst introduction is roughly known for only approximately 50% of species.To our knowledge, at least six species of woodlice classifi ed as aliens of Europe were noticed in the fi rst half of the 20 th century and only fi ve more species since then.
Th us, unlike many other invertebrate phyla, the temporal trend in alien crustaceans (both intra-European and alien) has shown no marked changes during recent decades.As "silent invaders" (Hornung et al. 2007) no terrestrial crustaceans are classifi ed as pests in Europe; they are elusive animals.We suspect frequently a large gap between the date of introduction and "discovery" of alien woodlice species.For example, during an intense eight year survey of the isopod fauna in a large region representing 15% of Hungary, three new alien species for this country were found (Farkas 2007).
To conclude, the atypically gradual trend in the number of alien terrestrial Crustacea in Europe could be an artefact of incomplete knowledge.Because of both the increasing worldwide trade in ornamental plants and the general ecology of terrestrial crustaceans (i.e.often hidden in soils), it is more realistic to expect a future exponential increase in the number of alien species (especially intra-European aliens).
Origin of the alien species
Species alien to Europe almost all originate from sub-tropical or tropical areas (Table 7.1.1.).Only one species -Protracheoniscus major (Dollfus, 1903)-is likely to be native from Central Asia.For several species, their ranges are poorly known (they are also often introduced in other tropical areas).However, several species do have a precise origin.Th e most widely distributed alien woodlouse in Europe is the tropical American Trichorhina tomentosa (Budde-Lund, 1893), while the most widely distributed amphipod is Talitroides alluaudi Chrevreux, 1901.It should be noted that a least six of the seventeen alien species were originally described from Europe (Great Britain, France and Germany) after their introduction.
Th e crustaceans alien in Europe generally originate from the Mediterranean basin (seven species), from western and south-western Europe (fi ve species).
Distribution of the alien species in Europe
Within Europe, Crustaceans of alien origin have mainly been recorded in western countries, where they appeared fi rst.Th e four countries with most species are Great Britain (11 species), the Netherlands (10 species) and Germany (nine species) (Figure 7.1.2).Comparatively few alien species have been recorded in central and eastern Europe to date (e.g.only four species in Hungary).In this part of Europe, the Central-Asian P. major is one of the most widespread alien crustaceans.Th e high number of aliens in western European countries may be linked to the high number of scientists and the intensity of soil research (Hornung 2009).
Th ere are only very few records of alien crustaceans on European islands.Trichoniscus pusillus has been reported from the Azores and Madeira, T. provisorius and A. assimile from the Azores but these species are native of Continental Europe.To our knowledge, the only alien aliens recorded on islands are talitrids, Arcitalitrus dorrieni (Hunt, 1925) in Scilly and Guernsey, Talitroides topitotum (Burt, 1934) in the Azores and Madeira, and T. alluaudi in the Azores and the Canaries.All these species occur outdoors and are therefore considered as naturalised.Th e rarity of alien terrestrial crustaceans on European islands is likely to be due to the primarily introduction route being major greenhouses in large metropolitan cities (see below).
Crustaceans classifi ed as aliens of Europe are typically species which have expanded their range approximately northwards and eastwards.Th e eastern and central countries have a higher number of these species than more westerly countries of Europe.For example, Germany and the Czech Republic, taken together, have nine species of alien woodlice of European origin, about 45% of the total in this category.A striking example of successful colonization and establishment of such species is given by A. nasatum.Th is woodlouse is believed to be native to Italy, southern France and Spain (Vandel 1962).Since the start of the 20 th century, it has been introduced into greenhouses in a number of additional countries of Northern and Central Europe (e.g.Denmark, Finland, Germany, Hungary, Poland, Slovakia, Sweden), making this species one of the most widely distributed alien woodlice of Europe.Moreover, numerous reports highlight the successful establishment of outdoor populations in several western and central European countries (e.g. the Netherlands, Czech Republic, Romania, Slovenia) (Berg et al. 2008, Giurginca 2006, Navrátil 2007, Vilisics and Lapanje 2005).
Some of the aliens of Europe have also invaded other continents and can be considered as very successful invaders.Th e most notable ones are A. vulgare, P. scaber and P. pruinosus.Armadillidium vulgare and P. pruinosus are probably native from Mediterranean regions.In northern temperate parts of Europe, these species are restricted to synanthropic habitats (e.g.gardens, cellars, compost heaps).P. pruinosus is one of the woodlice that has been spread most by man across the world (Vandel 1962) and can now be considered as "synanthropically cosmopolitan" (Schmalfuss 2003).
A consequence of the dominance of Mediterranean origin for species classifi ed as aliens of Europe is their decreasing number towards the north of the continent (Vilisics et al. 2007).In the northernmost countries of Europe (e.g.Finland (Vilisics and Terhivuo 2009)) only the most tolerant habitat-generalists, as well as intra-European aliens, are able to become successfully established.
. Pathways of introduction of alien terrestrial Crustaceans
Because a great majority of the fi rst isopod introductions were recorded in greenhouses, botanical gardens or urban parks, it is clear that many were associated with passive transport of soil, plants or compost.With few visible eff ects in such biotopes, terrestrial crustaceans colonize and spread as undetected "silent invaders" (Hornung et al. 2007).Th us, most introductions were unintentional.Th e one known exception is the spreading of T. tomentosa, commonly sold as pet food, triggered by trading activity in Europe.Th is probably explains why, among all the alien crustaceans, T. tomentosa is the most widespread species in Europe.
Another interesting case is the Mediterranean species P. schoblii.Th is myrmecophylous woodlouse is a commensal of the ant Lasius neglectus Van Loon, Boomsma & Andrásfalvy, 1990 and was fi rst recorded in Hungary in 2001, a few years after the introduction of the ant.P. schoblii was probably introduced at the same time as its ant host (Tartally et al. 2004).It has since been found regularly (Hornung et al. 2005, Tartally et al. 2004, Vilisics 2007, Vilisics et al. 2007) and is now considered established, as is L. neglectus.
Ecosystems and habitats invaded in Europe by alien terrestrial Crustaceans
To our knowledge, the only alien crustaceans invading natural habitats are three talitrid species.Arcitalitrus dorrieni has invaded leaf litter understoreys of deciduous woodlands in Great Britain and Ireland (Cowling et al. 2003, Vader 1972).Talitroides alluaudi is known outdoors in the Canary Islands, and T. topitotum in the Madeira Islands, both species in the Azores (Vader 1972).All other species are generally limited to highly artifi cial habitats and artifi cial ecosystems: mostly greenhouses, urban parks and houses (especially cellars).Th e proportion of introduced isopods can be very high in urban areas.A study in Budapest revealed that 35% of the total species (n = 28) were introduced (Vilisics and Hornung 2009).Th e major settlements of Hungary were characterised as "hotspot for non-native species" (Hornung et al. 2008).Th is could certainly be applied to many major cities in other European countries.
For the tropical species, especially those recorded only once or twice in Europe, they may not be considered as established (Table 7.1.1.)since their survival is completely dependent on warm greenhouses.Among all alien woodlice, none have spread to more natural habitats.However, the situation is diff erent for intra-European woodlice native to southern or Mediterranean Europe.Th ese established aliens can successfully expand by dispersal from very disturbed areas (where they were originally introduced) to more semi-natural habitats in rural-suburban zones (Vilisics and Hornung 2009).With global warming and the large-scale disturbance of biomes in Europe, that trend could increase, especially for the species with large ecological spectra.
Ecological and economic impact of alien terrestrial Crustaceans
Alien crustaceans in Europe are not known to carry diseases or to have an impact on native species and natural habitats.Further, they have no economical impact.Based on existing literature, the occurrence of alien woodlice is strictly bound to the urban environment (e.g.greenhouses, botanical and private gardens); alien terrestrial isopods do not yet seem able to survive or to expand to more natural ecosystems.
Th e case of the alien amphipod A. dorrieni is quite diff erent.Terrestrial amphipods are known to have many eff ects on the soil and leaf litter (Friend and Richardson 1986).Arcitalitrus dorrieni has invaded deciduous and coniferous woodlands in western parts of Great Britain.In Ireland, a study showed that 24.7% of annual litter fall in a coniferous woodland was ingested by this species.It is suggested that "this introduced species plays a more important role than native macrofaunal species in nutrient turnover in this particular woodland habitat" (O'Hanlon and Bolger 1999).It is possible that other, as yet undetected, ecological impacts are likely.
Terrestrial crustaceans can represent a large percentage of biomass and abundance in the soil macrofauna (Gongalsky et al. 2005).Th us any successful invasion by a terrestrial alien crustacean could induce some disturbance if it established in relatively natural habitats.For example, in a forested area of Florida, a study on the introduced European woodlouse A. vulgare showed that this species' activity "had a strong eff ect on the chemistry of the mineral layer" (Frouz et al. 2008) and concluded that in some cases it may signifi cantly alter soil conditions".
Figure 7
Figure 7.1.1.Taxonomic overview of the Isopoda and Amphipoda species alien to and Alien in Europe.
Figure 7
Figure 7.1.2.Colonization of continental European countries and main European islands by myriapod species alien to Europe.Archipelago: 1 Azores 2 Madeira 3 Canary islands.
Table 7 .1.1.
List and main characteristics of the Crustacean species alien to Europe.Country codes abbreviations refer to ISO 3166 (see appendix I).Habitat abbreviations refer to EUNIS (see appendix II).Only selected references are given.Last update 16/10/2009.
|
2018-12-07T18:51:07.567Z
|
2010-06-07T00:00:00.000
|
{
"year": 2010,
"sha1": "d090a1114ae75d8e907ecf56043149d5197b3812",
"oa_license": "CCBY",
"oa_url": "https://biorisk.pensoft.net/article/1848/download/pdf/",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "d090a1114ae75d8e907ecf56043149d5197b3812",
"s2fieldsofstudy": [
"Environmental Science",
"Biology"
],
"extfieldsofstudy": [
"Biology"
]
}
|
10865402
|
pes2o/s2orc
|
v3-fos-license
|
Targeted delivery of Tet1 peptide functionalized polymersomes to the rat cochlear nerve
Polymersomes are nanosized vesicles formed from amphiphilic block copolymers, and have been identified as potential drug delivery vehicles to the inner ear. The aim of this study was to provide targeting to specific cells within the inner ear by functionalizing the polymersome surface with Tet1 peptide sequence. Tet1 peptide specifically binds to the trisialoganglioside clostridial toxin receptor on neurons and was expected to target the polymersomes toward the cochlear nerve. The Tet1 functionalized PEG-b-PCL polymersomes were administered using routine drug delivery routes: transtympanic injection and cochleostomy. Delivery via cochleostomy of Tet1 functionalized polymersomes resulted in cochlear nerve targeting; in contrast this was not seen after transtympanic injection.
Introduction
Targeted drug delivery aims to enhance the biological availability of a chosen therapeutic agent while reducing adverse effects. The ability to target drugs to the spiral ganglion cells (SGCs) and cochlear nerve (CN) would create an advanced therapeutic strategy for the treatment of sensorineural hearing loss. Composed of US Food and Drug Administration (FDA)-approved medical materials, poly(ε-caprolactone)block-poly(ethylene glycol) (PEG-b-PCL) polymersomes (PMs) are efficiently taken up by the SGCs and spiral ganglion Schwann cells. This occurs without overt toxicity in vitro spiral ganglion cultures indicating that this type of nanoparticle holds promise for drug delivery to treat sensorineural hearing loss. 1 Targetability of PMs to the neural element of the cochlea was achieved through ligand mediated multivalent binding to tyrosine kinase receptors and to p75 neurotrophin receptors. 2 However, no in vivo studies on the targetability of functionalized PMs were reported and there is a risk that targeting tyrosine kinase receptors and p75 neurotrophin receptors may result in altered cell signalling, which in turn may affect the neural activity and viability of the SGCs.
An alternative approach to targeting SGCs and their processes might be achieved by using the Tet1 peptide (sequence H-L-N-I-L-S-T-L-W-K-Y-R), which was identified by phage display and showed a strong affinity to differentiated PC12 cells, primary motor neurons, and dorsal root ganglion cells. Tet1 peptide binds to the trisialoganglioside clostridial toxin (GT1b) receptors, 3 which are also expressed in the cochlea. 4,5 Any disruption in the neural activity and viability of SGCs might be avoided because GT1b is not involved in SGC cell signalling. Modification of nanoparticles with the Tet1 peptide has resulted in improved targeted gene delivery to neurons both in vitro and in vivo. 6,7 We hypothesized that SGCs or CN targeted drug delivery could be achieved by using PEG-b-PCL PMs functionalized with the Tet1 peptide.
In our previous study, unlabelled PMs that were delivered onto the round window membrane (RWM) did not appear in SGCs and spiral ganglion satellite cells. 8 In the present study, PMs were functionalized with the Tet1 peptide. The targetability was evaluated using two therapeutic approaches in which the PMs were delivered via either transtympanic injection or cochleostomy. The inner ear distribution of Tet1 functionalized PMS (Tet1-PMS) was then assayed using confocal microscopy. Two different control samples of PMs, unlabelled PMs and PMs functionalized with a scrambled Tet1 sequence (ScrTet1), were used in this investigation.
Methods
Preparation of peptide functionalized polymer NH 2 -PEG5.8K-b-PCL19K (60 mg) was dissolved in DMF (2 mL). 4-nitrophenyl-iodoacetate (10 mg) was added, and the reaction mixture stirred for 4 hours. Diethyl ether (50 mL) was added and the solution left overnight. The resulting precipitate was filtered and washed with diethyl ether to give iodoacetate-PEG5.8K-b-PCL19K functionalized polymer (45 mg, 75% yield). A total of 12 mg of the solid was dissolved into DMF (1 mL) to give a pale yellow solution that turned clear when cysteine terminated peptide (1 mg) was added. The reaction mixture was stirred overnight, concentrated in vacuo, and used crude in PM preparation. The peptide sequences used were Tet1 (NH2-HLNILSTLWKYRC-COOH) and a ScrTet1 (NH2-LHNYTWLSLRKIC-COOH).
PM preparation
The carboxy cyanine dye DiI was dissolved in DMF at a concentration of 0.1 mg/mL. Tet1-PEG5.8K-b-PCL19K (6.0 mg) was added to the DiI/DMF solution (0.4 mL), and this was placed in an ultrasonic bath for 10 minutes to aid dissolution. The polymer solution was then added dropwise (∼1 drop every 8 seconds) to rapidly stirred PBS (1.60 mL). The sample was then dialyzed against PBS (400 mL), and the buffer solution was replaced four times over the course of 48 hours, after which the sample was removed from the dialysis tubing. Control PMs with the scrambled ScrTet1 peptide (ScrTet1-PMs) and unlabelled PMs were made in an identical fashion using ScrTet1-PEG5.8K-b-PCL19K and iodoacetate-PEG5.8K-b-PCL19K respectively ( Figure 1A). Analysis by dynamic light scattering showed the typical hydrodynamic diameter of the Tet1-PMs to be 105.0 ± 20.0 nm. Both the ScrTet1-PMs and unlabelled PMs had hydrodynamic diameters similar to the Tet1-PMs. The zeta potential measurements for the Tet1-PMs and ScrTet1-PMs were negative, with typical measurements for the Tet1/ScrTet1-PMs and unlabelled PMs being -0.764 and -0.578 mV, respectively. Before use in in vivo experiments, the PMs were sterile-filtered through a 0.2 µm cellulose acetate syringe filter.
Animals
Twenty-seven 5-month-old male SD rats with normal Pryer's reflexes and weighing 350-450 g (supplied by the Experimental Animal Unit, University of Tampere, FL) were used in this study. All rats were treated in accordance with the directives of the regional ethics committee of the University of Tampere. All experimental procedures were performed under general anesthesia after intraperitoneal injection of a mixture of 0.8 mg/kg of medetomidine hydrochloride (Domitor, Orion, Espoo, Finland) and 80 mg/kg of ketamine hydrochloride (Ketalar; Pfizer, Helsinki, Finland).
The animals were assigned to two groups: in group 1, Tet1-PMs, ScrTet1-PMs, and unlabelled PMs were delivered by transtympanic injection bilaterally; in group 2, Tet1-PMs, ScrTet1-PMs, and unlabelled PMs were delivered by cochleostomy unilaterally. Three days after administration of the PMs was begun, the animals were sacrificed and their cochleae were harvested. The cochleae were processed to either whole mounts or cryosectioning. submit your manuscript | www.dovepress.com
Dovepress
Protocols for transtympanic injection and cochleostomy of PMs have previously been reported. 8 Briefly, for transtympanic injection, 40 µL of each PM type was injected into the middle ear cavity; for cochleostomy, an osmotic pump filled with 100 µL of PMs (the PMs were diluted 1:1 with artificial perilymph) was connected to the scala tympani (ST) (via a hole in the bony labyrinth that was made with a 0.5 mm diameter burr) by a polyethylene tube prefilled with the same solution ( Figure 1B). The analysis was carried out blind, the PMs were coded prior to administration, and evaluation of the tissue by confocal microscopy was completed before decoding.
During the procedure, the animals' eyes were protected with Terramycin eye cream (Pfizer). Atipamezole hydrochloride (2 mg/kg) was injected intraperitoneally immediately after the operation to accelerate the animals' recovery from anesthesia. Saline (2 mL) was administered via subcutaneous injection in the neck. Rimadyl (1 mg/kg; Pfizer) was injected to relieve pain. Baytril (10 mg/kg; Orion, Germany) was injected intraperitoneally once a day to prevent middle ear infections.
Histology analysis
The rats were anesthetized using the procedure described above, and this was followed by fixation through cardiac perfusion with 0.01 M PBS containing 0.6% (v/v) heparin (pH 7.4) and then 4% paraformaldehyde (Merck, Espoo, Finland). The isolated cochleae were rinsed with water for 1 minute to remove any PMs that may have attached to the middle ear mucosa and the round window membrane (RWM), and fixative was added for 2 hours (for whole mount samples) or overnight (for cryosection samples).
The methods for perfusion, sample preparation, and confocal microscopy after PM delivery have previously been reported. 8
Fluorescence signal analysis
Colocalization of the Tet1-PMs, ScrTet1-PMs, and unlabelled PMs with NF-200 immunostaining in confocal micrographs was analyzed using ImageJ 1.42q software (National Institutes of Health, Bethesda, MD).
Tet1-PMs delivered via cochleostomy
Distribution of PMs in the spiral ganglion and the cochlea nerve All of the PMs (Tet1-PMs, ScrTet1-PMs, and unlabelled PMs) were detected in the cochlea nerve (CN) within the spiral lamina canal, within the modiolus, and in the spiral ganglion satellite cells in Rosenthal's canal (Figure 2). Only the Tet1-PMs were detected in nerve fibers in the tractus spiralis foraminosus region (Figure 2 and Figure 4A), which are multiple tiny osseous canals that allow the axons to pass from the Rosenthal canal to the cochlea nerve in the modiolus.
A fluorescence signal intensity gradient was visualized from the distal to the proximal parts of the peripheral processes of the SGCs ( Figure 3A-C). Nevertheless, the distribution patterns of the PMs in the peripheral processes of the SGCs were different: the Tet1-PMs were evenly distributed in the nerve fibers, while the majority of ScrTet1-PMs appeared within the epineurium surrounding the nerve bundle (a few of the PMs were found in nerve fibers), and the unlabelled PMs were detected mainly within the epineurium with only sparse uptake in the nerve fibers of peripheral processes (Figure 2; Figure 3D-F).
The neural structures showing PM distribution were further identified by immunostaining using NF-200, S100, and peripherin. As can be seen, the Tet1-PMs were localized within or adjacent to NF-200-positive nerve fibers of SGCs ( Figure 4B and F). In contrast, the scrambled and the unlabelled PMs were remote from the NF-200-positive nerve fibers ( Figure 4C and D). Co-localization with S100-positive structures that represent the spiral ganglion satellite cells occurred with all PM types ( Figure 5C, F, and I). None of the PMs were observed in neuronal soma of the SGCs identified by NF-200 immunostaining ( Figure 5B, E, and H) and peripherin immunostaining (not shown).
Distribution of PMs in rWM, cochlea scala, and spiral ligament
All the PMs (Tet1-PMs, ScrTet1-PMs, and unlabelled PMs) were observed in the RWM, the mesothelial cells of the ST, and SV (Figure 2; Figure 3A-C), and the anterior and posterior spiral modiolar vein in the modiolus. In the scala media, PMs occasionally appeared in the inner and outer pillar cells of Corti's organ. In addition, the scrambled and the unlabelled PMs were occasionally found in Hensen cells and Claudius cells (Figure 2; Figure 3B and C), while the Tet1-PMs were not detected in these cells. Tet1-PMs and ScrTet1-PMs were detected in the entire spiral ligament. Unlabelled PMs were only detected in the type IV and V spiral ligament fibrocyte districts of the SL (Figure 2; Figure 6). 9,10 In contrast, all the PMs were occasionally found in the strial vascularis.
Targeted axonal distribution of Tet1-PMs Limited distribution of Tet1-PMs in cochlea after transtympanic injection
The Tet1-PMs administered via transtympanic injection showed no difference in distribution pattern in the cochlea compared to ScrTet1-PMs and unlabelled PMs, and no targeting to the SGCs and CN was observed. However, PMs were observed in spiral ligament fibrocytes that lack GT1b receptors, indicating non-specific uptake. Furthermore, no PMs were detected in the SGCs, a phenomenon that might be due to limited passage of the PMs through the RWM to the perilymph (Figure 2). The RWM is the major barrier between the middle ear and the cochlea, and consists in rats of three tractus spiralis foraminosus region. Tet1 peptide is reported to bind to the GT1b receptor, which exists abundantly in neuronal membranes in the central nervous system 14 and peripheral nerve axolemma. 15 All the PMs (whether with or without the Tet1 peptide) were observed in spiral ganglion satellite cells of the spiral ganglion and the CN in the modiolus, as identified by S100 immunostaining, but only the Tet1-PMs were detected in nerve fibers identified by NF-200 immunostaining.
Neurons have active axonal transport machinery that moves substances from their distal terminals to the neuronal soma. 16 We hypothesized that the distribution of Tet1-PMs in the tractus spiralis foraminosus region ( Figure 4A) was a consequence of active axonal transport from the CN in the modiolus to the neural soma of SGCs. However, the Tet1-PMs were not observed in the neuronal soma. This may be because the SGCs are enclosed by spiral ganglion satellite cells and there is no communication between these cells -the Tet1-PMs were located in distal axon compartments of SGCs that have little or no communication with the cell body compartment. 17,18 In addition, gene expression in the lateral ventricle was reported after administration of Tet1-PEG-b-polyethylenimine containing plasmid DNA. 6 This indicated that the PMs might release their cargos in the distal axon compartment, which could then penetrate separately into the cell body compartment of the neuron. Thus, Tet1-PMs could target the neurofilament of the CN; however, whether they effectively release their cargo into the cytoplasm needs further investigation.
Affinity of PMs to spiral ganglion satellite cells and spiral ligament fibrocytes
The Tet1-PMs, ScrTet1-PMs, and unlabelled PMs administrated via cochleostomy were detected in spiral ganglion satellite cells. This suggested nonspecific affinity between these PMs and the spiral ganglion satellite cells. This is in line with previous in vitro studies with cochlear explants 2 and spiral ganglion mixed cultures. 1 We suggest that the reason more Tet1-PMs and ScrTet1-PMs than unlabelled PMs were detected in the SL may be because the surface modification of PMs using the Tet1 and ScrTet1 peptides changes their affinity with spiral ligament fibrocytes. The PMs observed in spiral ligament fibrocytes were there due to non-specific uptake because the spiral ligament fibrocytes lack GT1b receptors.
Publish your work in this journal
Submit your manuscript here: http://www.dovepress.com/international-journal-of-nanomedicine-journal The International Journal of Nanomedicine is an international, peerreviewed journal focusing on the application of nanotechnology in diagnostics, therapeutics, and drug delivery systems throughout the biomedical field. This journal is indexed on PubMed Central, MedLine, CAS, SciSearch®, Current Contents®/Clinical Medicine, Journal Citation Reports/Science Edition, EMBase, Scopus and the Elsevier Bibliographic databases. The manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. Visit http://www.dovepress.com/ testimonials.php to read real quotes from published authors.
|
2014-10-01T00:00:00.000Z
|
2012-02-23T00:00:00.000
|
{
"year": 2012,
"sha1": "870d8d21f31ee71a439cf3bbeab57ba8def14bc2",
"oa_license": "CCBYNC",
"oa_url": "https://www.dovepress.com/getfile.php?fileID=12120",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "f6abab4941da8334a7c8fb38f2f1658cd42b7206",
"s2fieldsofstudy": [
"Engineering",
"Medicine"
],
"extfieldsofstudy": [
"Medicine",
"Materials Science"
]
}
|
53967903
|
pes2o/s2orc
|
v3-fos-license
|
oligoMask : A Framework for Assessing and Removing the Effect of Genetic Variants on Microarray Probes
As expression microarrays are typically designed relative to a reference genome, any individual genetic variant that overlaps a probe’s genomic position can possibly cause a reduction in hybridization due to the probe no longer being a perfect match to a given sample’s mRNA at that locus. If the samples or groups used in a microarray study differ in terms of genetic variants, the results of the microarray experiment can be negatively impacted. The oligoMask package is an R/SQLite framework which can utilize publicly available genetic variants and works in conjunction with the oligo package to read in the expression data and remove microarray probes which are likely to impact a given microarray experiment prior to analysis. Tools are provided for creating an SQLite database containing the probe and variant annotations and for performing the commonly used RMA preprocessing procedure for Affymetrix microarrays. The oligoMask package is freely available at https://github.com/dbottomly/oligoMask.
Introduction
It has been observed that for mRNA microarrays from a given sample, genetic differences of that sample relative to the probe sequences can affect hybridization to short oligonucleotide probes resulting in false positives or negatives depending on the experimental and array design (Walter et al., 2007;Alberts et al., 2007).Several approaches currently exist to identify and flag/remove probes that have hybridization artifacts due to genetic variants.Removal of probes based on pre-defined genetic variant databases is one such approach (Benovoy et al., 2008).Software (Kumari et al., 2007) and databases (Duan et al., 2008) allowing the interrogation of the relationships between microarray probes and single nucleotide variants have been described.The R package CustomCDF (Dai et al., 2005) is an example of this approach that removes probes from an environment formed from the Affymetrix chip description file (CDF) prior to analysis.One potential limitation of the CDF filtering approach is that for some of the more recent arrays platforms such as the Affymetrix Gene or Exon arrays the use of such environments has been superseded (e.g. the SQLite databases in the oligo (Carvalho and Irizarry, 2010) package or ROOT scheme files as in the xps (Stratowa et al., 2013) package).
In addition, the actual expression data itself can be interrogated to identify and mask out variants.R packages exist to effectively deal with a two group comparison between several strains or species through procedures based mainly on the expression data such as maskBAD (Dannemann et al., 2012) and SNEP (Fujisawa et al., 2009).However, models based on two (genetic) groups have limited utility when analyzing more complicated experimental designs such as those found in expression-based analyses using more genetically diverse mouse lines such as Diversity Outbred (Svenson et al., 2012), Collaborative Cross (Collaborative Cross Consortium, 2012) or other Heterogenous Stock (Chia et al., 2005) mice.
In order to facilitate eQTL mapping and other expression analyses in complex mouse crosses we devised an R package oligoMask based on the use of high quality publicly available genetic variant databases to screen microarray probes identifying probes impacted by variants.The key to this is the relatively recent availability of genome-wide variant databases in the variant call format (VCF) such as those from the Sanger Mouse Genomes Project (Keane et al., 2011) and the 1000 Genomes for humans (1000Genomes Project Consortium, 2012) as well as the ability to query and parse these files via the VariantAnnotation (Obenchain et al., 2014) package.The oligoMask package is designed to work in conjunction with the oligo Bioconductor package to facilitate removal of aberrant probe expression prior to the commonly used robust multi-array average (RMA) (Irizarry et al., 2003) pre-processing procedure for Affymetrix arrays.Our package works by removing potentially impacted probes from the overall expression matrix prior to the call to the RMA processing functions.This can be done before the background correction step or after the normalization step.The annotation for these impacted probes are most easily derived from VCF files with the parsed data stored in an SQLite database.This database can be optionally wrapped in an R package with appropriate metadata to facilitate sharing and reproducibility.High-level S4 classes and methods provide a convenient interface with oligoMask and oligo.In addition, users can define new database schemas, add custom data as well as create their own functionality.Below we give an overview as well as demonstrate using publicly available data the steps involved for the use of oligoMask.
Example data
The example data presented in this article and in the vignette was downloaded from the gene expression omnibus (GEO) with accession number GSE33822 (Sun et al., 2012).For demonstration purposes we use a subset (n=8) of the dataset including only those samples derived from whole brain which received the vehicle treatment and that were run on version 1 of the Mouse Gene ST array.In our example, we are looking for expression differences between the NOD/ShiLtJ (NOD) inbred strain and the C57BL/6J (B6) inbred strain, the genome of which serves as the mouse reference genome.As we can expect the microarray probe sequences to be heavily biased towards the reference genome, looking for differential expression between these two strains may be problematic as differences in expression may be due to hybridization artifacts or true gene expression differences.First we create a NOD-specific database and then filter out those probes that are impacted by at least one variant in the NOD strain but not in the B6 strain and then carry out the differential expression analysis as per standard statistical workflows.
Workflow Creating a variant database
The first step in the use of oligoMask is the creation of an SQLite database containing the probe annotation (including alignments to a given genome), variant annotation and the overlap,if any, between probes and variants.In a general sense, the probes sequences are first realigned to the given genome using the BSgenome and Biostrings Bioconductor packages (Pages, 2013;Pages et al., 2013) with the probe location and mappability of the probes being recorded.The locations of the uniquely mapping probes are then used to compute overlap with the variants in the specified VCF file using import and overlap functionality in the VariantAnnotation package.The locations of the variants in the genome, type of variant and the individual/population it was observed in is also recorded along with the overlap between probe alignments and variants.A convenience function for database creation is provided (create.sanger.mouse.vcf.db) for use with the case of variants derived from VCF files from the Sanger Mouse Genomes Project and variants of the Affymetrix Mouse Gene ST arrays.The oligoMask Vignette demonstrates in the section 'Data preparation' how the NOD variant database package (om.NOD.mogene.1.0.st) can be created using this function.
Additional array platforms and variant genotype file types can be supported through a modification of create.sanger.mouse.vcf.db as well as specifying the database schema as a "TableSchemaList" object as returned in the pre-defined SangerTableSchemaList function.The "TableSchemaList" S4 class serves a similar role as an object-relational mapping approach (ORM) in other languages and allows the R code to interact with a given database in a general way.
Masking procedure
The masking procedure first requires the installation of the oligo package along with the appropriate platform design databases that can be downloaded from Bioconductor.In our use case of Affymetrix Gene ST arrays, the CEL files are first read in using the read.celfilesfunction of oligo resulting in a "GeneFeatureSet" object.Next, the oligoMask database package is loaded, the parameters for the masking procedure are defined and finally the RMA summarization is performed as is shown below starting from the "GeneFeatureSet" object distributed with oligoMaskData.library(oligoMask) library(oligoMaskData) library(om.NOD.mogene.1.0.st) library(pd.mogene.1.0.st.v1) library(limma) data(oligoMaskData) var.parms <-VariantMaskParams(om.NOD.mogene.1.0.st, geno.filter= FALSE, rm.unmap = FALSE, rm.mult = FALSE) sun.gfs.mask<-maskRMA(oligoMaskData, target = "core", apply.mask= TRUE, mask.params= var.parms) The result of these commands is a summarized "GeneFeatureSet" object with all probes overlapping variants from the NOD inbred strain of mouse removed prior to the background correction step of RMA.Users can control several aspects of the masking procedure through creation of a parameter The R Journal Vol.6/1, June 2014 ISSN 2073-4859 object.For instance users can additionally remove probes that map to multiple locations as well as those that do not map at all to the reference genome by supplying TRUE to rm.multi and/or rm.unmap.
Similarly, masking can be performed using only those variants that passed quality filters encoded in the VCF file by setting geno.filter to TRUE.
The maskRMA method carries out the RMA procedure and provides a similar interface to the rma method from oligo.In addition it requires specification of a "VariantMaskParams" object and whether the masking procedure should be performed before the background correction function or after background correction and normalization but before summarization by setting the mask.typeargument to before.rmaor before.summaryrespectively.
Assessment of masking procedure
As a demonstration of oligoMask next we perform a basic linear-model based differential expression analysis with the Sun et al. 2012 data comparing results with and without the NOD mask applied.Below we illustrate the basic approach using the masked data.
Conclusion
oligoMask is a flexible R/SQLite framework for pre-processing and QA/QC of hybridization based expression data.Not only can it remove the effect of spurious probe intensities due to genetic variants but it can additionally correct for design artifacts (probes mapping to multiple places or not mapping at all).It utilizes SQLite and works in conjunction with the oligo Bioconductor package.Currently it supports the Affymetrix Mouse Gene ST array though support could be easily added for other array types or species as described above.Approaches for removal of probes based off of the variant and mapping information in the SQLite database are already implemented.More sophisticated algorithms could be built on top of the database to provide masking additionally based off of the The R Journal Vol.6/1, June 2014 ISSN 2073-4859
Figure 1 :
Figure 1: Concordance of differentially expressed genes between the masked and unmasked versions of the analysis
|
2018-12-01T00:28:13.926Z
|
2014-01-01T00:00:00.000
|
{
"year": 2014,
"sha1": "9fe4a54080e724adbbd5aefe634ac1159d2e2005",
"oa_license": "CCBY",
"oa_url": "https://journal.r-project.org/archive/2014/RJ-2014-018/RJ-2014-018.pdf",
"oa_status": "HYBRID",
"pdf_src": "Anansi",
"pdf_hash": "9fe4a54080e724adbbd5aefe634ac1159d2e2005",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Computer Science"
]
}
|
255841304
|
pes2o/s2orc
|
v3-fos-license
|
Il-6 signaling between ductal carcinoma in situ cells and carcinoma-associated fibroblasts mediates tumor cell growth and migration
Ductal carcinoma in situ (DCIS) is a non-obligate precursor lesion of invasive breast cancer in which approximately half the patients will progress to invasive cancer. Gaining a better understanding of DCIS progression may reduce overtreatment of patients. Expression of the pro-inflammatory cytokine interleukin-6 increases with pathological stage and grade, and is associated with poorer prognosis in breast cancer patients. Carcinoma associated fibroblasts (CAFs), which are present in the stroma of DCIS patients are known to secrete pro-inflammatory cytokines and promote tumor progression. We hypothesized that IL-6 paracrine signaling between DCIS cells and CAFs mediates DCIS proliferation and migration. To test this hypothesis, we utilized the mammary architecture and microenvironment engineering or MAME model to study the interactions between human breast CAFs and human DCIS cells in 3D over time. We specifically inhibited autocrine and paracrine IL-6 signaling to determine its contribution to early stage tumor progression. Here, DCIS cells formed multicellular structures that exhibited increased proliferation and migration when cultured with CAFs. Treatment with an IL-6 neutralizing antibody inhibited growth and migration of the multicellular structures. Moreover, selective knockdown of IL-6 in CAFs, but not in DCIS cells, abrogated the migratory phenotype. Our results suggest that paracrine IL-6 signaling between preinvasive DCIS cells and stromal CAFs represent an important factor in the initiation of DCIS progression to invasive breast carcinoma.
Background
Ductal carcinoma in situ (DCIS) of the breast is a preinvasive lesion and a risk factor for subsequent invasive ductal carcinoma (IDC) [1]. DCIS represents about 20 % of newly diagnosed breast cancers in the United States [2]. If left untreated, approximately half of DCIS tumors will progress to IDC while half will remain indolent [3,4].
Although there are many subtypes of DCIS, it is not currently possible to identify which will progress. This has led to aggressive treatments, specifically radiation with either lumpectomy or mastectomy [5].
Components of the tumor microenvironment are increasingly implicated in the progression of many cancers. Early morphological and physiological changes in breast epithelium are minimal, and compounding factors such as tumor-suppressive paracrine signaling from myoepithelial cells [6] or the extracellular matrix [7] may hide early indicators of ductal cell aberration. Such changes may include, but are not limited to: gene expression modulation, epigenetic alterations, and loss of genomic stability in both the epithelial and stromal compartments. In the tumor microenvironment, carcinoma-associated fibroblasts (CAFs) represent a fibroblast population or mixture of sub-populations that can promote tumor progression [8][9][10][11][12][13]. Although this mechanism is not fully understood, it is known that CAFs secrete numerous cytokines and growth factors [14].
Interleukin 6 (IL-6) is a pro-inflammatory cytokine shown to alter cell morphology, modulate cell migration and the epithelial to mesenchymal transition [15][16][17]. Many of these processes occur upon IL-6 activation of the transmembrane IL-6 receptor (IL-6R), which heterodimerizes with the ubiquitously expressed cell surface receptor glycoprotein 130 (gp130). Downstream activation of the Janus kinase/signal transducers and activators of transcription (JAK/STAT) pathway initiates IL-6 target gene transcription [18,19]. Alternatively, secreted IL-6 can bind to the soluble IL-6 receptor (sIL-6R), which then binds gp130 at the cell surface and initiates intracellular signaling. This form of IL-6 signaling has been coined "IL-6 trans-signaling" (IL-6TS) [20]. IL-6 has been linked to the upregulation of proteases such as cysteine cathepsins and matrix metalloproteinases that are known to play a role in cancer progression [21,22]. The IL-6 signaling axis is commonly upregulated in invasive cancers, suggesting that IL-6 may be an important mediator of events involved in tumor cell invasion [23][24][25].
The effects of IL-6 signaling inhibitors on breast cancer cell morphology and proliferation have been evaluated in monotypic cultures. Two studies have shown that inhibiting autocrine IL-6 signaling in either triple negative breast cancer cell lines [26], or the ER-positive MCF7 cell line [27] significantly inhibited cell growth. Additionally, Leslie et al. show that knockdown of IL-6 in an invasive variant of the MCF10A cells, MCF10A-H-RasV12, inhibited cell migration in a transwell assay, and inhibited growth in a xenograft mouse model [28]. Although these studies evaluated paracrine signaling, cells were treated with exogenous recombinant protein rather than co-culturing different cell types. Therefore the authors were unable to evaluate in 3D the dynamic cell:cell interactions between two separate human cell types or the cell:microenvironment interactions.
Our 3D mammary architecture and microenvironment engineering (MAME) culture model mimics in vivo architecture, providing a suitable setting to study cell:cell interactions and, notably, physiologically relevant cell:cell signaling over time. Additionally, 3D in vitro cell culture models better represent in vivo tumor drug response, which would facilitate efficacious therapy development at the preclinical stage [29]. Here we examine the role of IL-6 in progression of preinvasive breast DCIS to an invasive phenotype, and show how co-culture of DCIS cells with CAFs changes DCIS growth and invasive potential.
Cell lines
MCF10A human breast non-transformed epithelial cells, MCF10.DCIS and SUM102 human breast DCIS cells, and WS-12Ti human breast carcinoma-associated fibroblasts were provided by Dr. Bonnie F. Sloane. All primary fibroblasts were derived from human breast tissue. CAF40T were derived from biopsy tissue diagnosed as invasive carcinoma. NAF98 were derived from benign tissue. Both CAF40T and NAF98 fibroblast cell lines were provided by Dr. Simon W. Hayward. These fibroblasts were immortalized in Dr. Sloane's lab and are designated CAF40TKi and NAF98i, respectively. The FB-NF, NAF-FB, and FB-CAF primary fibroblasts were derived from patient biopsies diagnosed as: benign (FB-NF, NAF-FB), or invasive carcinoma with accompanying DCIS (FB-CAF) and provided by Dr. Fariba Behbod. The FB-NF fibroblasts were immortalized in Dr. Sloane's lab and designated FB-NF-i. The FB-CAF and NAF-FB fibroblasts were not immortalized. All patient derived cells were received de-identified and therefore are exempt from IRB oversight.
Cell culture
In this study we utilized non-tumor forming MCF10A human breast epithelial cells [30] and the human DCIS cell lines MCF10.DCIS and SUM102, which were maintained as previously described [31]. See supplemental methods for more detailed information (Additional file 1). All 3D MAME cultures were performed using Cultrex (3433-005-01, Trevigen, Gaithersburg, MD) similar to previously described [32]. Briefly, cell culture dishes were coated with 100 % Cultrex. Cells were added on top of solidified Cultrex and allowed to adhere for 30-45 min before being overlaid with 2 % Cultrex in phenol red-free DMEM F12 media containing 2.5 % fetal bovine serum (Additional file 2: Figure S1). In co-culture experiments, fibroblasts were added first and allowed to adhere before adding tumor cells. Once tumor cells had adhered to the matrix, an overlay of 2 % Cultrex was added. Media were changed every 4 days.
Measurement of multicellular structures
Differential interference contrast (DIC) images of three random fields at 20X magnification were used to measure multicellular structures. Three individuals of whom two were study-blinded measured the diameter and perimeter of structures and the number and length of interconnections between structures. All three data sets were used in the quantification. This analysis was performed using Zen imaging software (Zeiss, Thornwood, NY). Volume measurements were obtained using 3D fluorescent images and quantified using Volocity software (Perkin-Elmer, Waltham, Mass).
Gene expression
RNA was isolated from cells grown in either 2D monolayer or 3D MAME cultures. For 2D culture, cells were washed in phosphate buffered saline and harvested using 0.5 % Trypsin/EDTA (Life Technologies, Foster City, CA) and pelleted. The cell pellets were resuspended in TRIzol® Reagent (Life Technologies, Foster City, CA) for RNA extraction. All qRT-PCR reactions were performed using Taqman Assays (Life Technologies, Foster City, CA). See supplemental list of Taqman Assays (Additional file 3: Table S1).
ELISA
ELISA kits (human IL6R-ab46029, human IL6-ab46044, and human GAPDH-ab119627) were purchased from Abcam® (San Francisco, CA). Aliquots of lysates were collected for ELISA assays and measurement of total DNA.
Immunofluorescence
Nuclei were labeled with Hoechst (33342, Thermo Scientific) or EDU (Life Technologies, Foster City, CA). Polyclonal antibodies to human IL-6 (AF-206-NA, R&D Systems, Minneapolis, MN) were used at a concentration of 1 μg/ml. Mono-specific antibodies to human cathepsin B have been previously isolated and characterized [33]. Cathepsin B immunostaining was performed as previously described with the exception that 1 % Tween 20 replaced the 0.01 % saponin [34]. For some studies, CAF40TKi were pre-labeled, prior to seeding in MAME co-cultures, utilizing CellTrace CFSE (carboxyfluorescein diacetate succinimidyl ester; Life Technologies, Foster City, CA) according to the manufacturer's protocol.
Drug treatments
For treatment of MAME cultures with IL-6 nAb (R&D Systems, AF-206-NA), we added 1 μg/ml of IL-6 nAb in the 2 % Cultrex overlay to 3D cultures on the first day of culture and refreshed with IL-6 nAb and 2 % overlay every 4 days. The antibody concentration was selected based on preliminary studies in which we determined the lowest concentration needed to inhibit growth of tumor structures. Oxymatrine (Sigma, St. Louis, MO) at 1-mg/ml (3.7 mM) was added 24 hours after cell seeding and was replaced with fresh drug every 4 days. The oxymatrine concentration was determined empirically based on the concentration at which proliferation was inhibited to 50 % of control. The protease inhibitors CA074Me and E64d (Sigma, St. Louis, MO) were used at a concentration of 10 μM [35].
Live cell proteolysis assay
Dye-quenched collagen IV (DQ-collagen IV, Life Technologies, Foster City, CA) was used admixed in Cultrex as previously described [32]. MAME cultures in optical glass bottom cell culture dishes were imaged live for a period of 10 to 60 min under 5 % CO 2 at 37°C.
Confocal microscopy
Confocal microscopy was performed on either a Zeiss LSM 510 or LSM 780 upright confocal microscope (Zeiss, Thornwood, NY). All cell cultures used for imaging were seeded on 40 mm optical glass bottom culture dishes at a density of 45 cells/mm 2 (~5000 cells/dish).
Statistics
Data were statistically analyzed using Student's t-test on GraphPad Prism 6.0 (GraphPad Software Inc., La Jolla, CA).
Ethics statement
All human subject materials and experiments in this study have been reviewed by the Wayne State University Institutional Review Board and deemed exempt according to the definition codified in the common rule at 45 CFR 46.102(d)(f ).
Human breast DCIS cells express pro-inflammatory cytokines
Increased levels of pro-inflammatory cytokines, including IL-6, in tumors and serum of breast cancer patients correlate with poor prognosis [26,[36][37][38]. Immunohistochemistry on breast tissue from female patients, age 30 to 81 with an average age of 49.5, confirmed IL-6 protein expression in 65 % of patient samples diagnosed with DCIS ( Fig. 1a and b, cf. 1c and d).
We utilized a 3D MAME culture model (Additional file 2: Figure S1), to study the role of IL-6 signaling between human breast DCIS cells and human breast CAFs. The MAME model allows for co-culture of multiple cell types in the context of a three dimensional microenvironment. Our model is advantageous over commercially available 3D platforms as it can be utilized for live cell microscopy, and imaging live cell functional assays (such as proteolysis) [32]. MAME cultures are tractable and can be altered to grow multiple cell types at various ratios or at various relative positions within the matrix. Additionally, the reduced density overlay allows for realtime collection and analysis of cytokine secretion without disrupting the longitudinal growth of the culture.
Here we examined the expression of IL-6 and the associated pro-inflammatory genes interleukin 1β (IL1β) and nuclear factor kappa light chain enhancer of activated B cells 1 (NFκB1) in human breast MCF10.DCIS and SUM102 cell lines. Our data show higher IL-6 mRNA levels in both MCF10.DCIS and SUM102 cell lines, as compared to the MCF10A breast epithelial cell line (Fig. 1e). Levels of IL-6 protein in lysates of the three cell lines were below the level of detection. When we measured secreted IL-6 in conditioned media from MCF10.DCIS or SUM102 cultures, we found that both DCIS cell lines secreted 10 to 30-fold more IL-6 than the MCF10A cells (Fig. 1f ).
Blocking IL-6 autocrine signaling inhibits MCF10.DCIS growth
To test whether blocking IL-6 autocrine signaling affected DCIS cell growth, we treated MCF10.DCIS MAME cultures for 8 days with either an IL-6 neutralizing antibody (nAb) or an equivalent concentration of . f Secretion of IL-6 protein from DCIS cell lines and non-transformed MCF10A cells as determined by ELISA. *P < 0.05, Student's t-test; mean ± SD a species and isotype matched antibody (Fig. 2). IL-6 nAb treatment resulted in a reduction in diameter of the multicellular 3D structures (Fig. 2b, cf. 2a, quantified in 2d). This was reversible as replacement of media on day 8, with media lacking IL-6 nAb, resulted in a significant increase in diameter of the MCF10.DCIS structures after an additional 48 h (Fig. 2c, cf. 2b, quantified in 2d).
To determine the effects of IL-6 nAb on MCF10.DCIS structures, we examined the expression of a panel of candidate genes that have been associated with tumor growth and invasion [39][40][41][42][43]. The expression of IL-6 was upregulated 2-fold in the treated cultures. The expression of TWIST1, vimentin, and Fibroblast Growth Factor 2 was downregulated greater than 2-fold, while minimal changes were observed in the expression of E-cadherin, N-cadherin, and NFκB1 (Fig. 2e). To test whether pharmacological suppression of IL-6 could reproduce IL-6 nAb mediated growth inhibition, we treated cells with oxymatrine, a naturally occurring inhibitor of IL-6 gene expression. Oxymatrine has been shown to prevent nuclear translocation of NFκB-p65 thereby inhibiting transcriptional activation of its target genes, which include IL-6 [44]. Oxymatrine treatment was able to replicate the growth inhibitory effects observed with IL-6 nAb (Additional file 4: Figure S2B, cf. S2A, quantified in S2C). Neither oxymatrine nor IL-6 nAb treatment resulted in marked cell death as cytotoxicity assays showed no difference in cell viability after 48-hour drug treatment (Additional file 4: Figure S2D).
Carcinoma-associated fibroblasts express IL-6 and promote DCIS cell proliferation and motility CAFs represent a population or group of populations of stromal cells that can promote tumor cell growth [14,[45][46][47]. The mechanism of supported tumor growth is likely through stromal-epithelial paracrine signaling. Therefore, we next evaluated human breast CAFs to determine their contribution of IL-6 in the tumor microenvironment. Additionally, we examined the role that CAFs play in MCF10.DCIS cell proliferation and motility in the 3D MAME model.
We examined the expression of IL-6 mRNA in normal human fibroblasts and CAFs grown in 3D. Here we found that CAFs exhibited elevated expression of IL-6 mRNA compared to normal fibroblasts (Fig. 3a). Protein levels of IL-6 in FB-NF-i normal fibroblast lysates were near the lower limit of detection and undetectable in NAF-FB or NAF98i lysates. IL-6 levels in CAF40TKi lysates were significantly higher than in FB-NF-i lysates (Fig. 3b). Levels of IL-6 in CAF-conditioned media were higher than in normal fibroblast-conditioned media (Fig. 3c).
We next co-cultured MCF10.DCIS cells with CAFs in 3D, at a seeding ratio of five tumor cells to one CAF [32], to evaluate the effect of CAF-secreted cytokines on DCIS cell proliferation and related changes in morphology of multicellular structures. We found that in MCF10.DCIS:C AF40TKi co-cultures there was an increase in the average diameter and volume of the multicellular structures and a prominent formation of branch-like interconnections between the structures (Fig. 3e, cf. 3d, quantified in Additional file 5: Figure S3A). Fluorescent imaging of MCF10.DCIS:CAF40TKi 3D co-cultures revealed that the branch-like multicellular connections between structures were primarily composed of tumor cells, yet contained some CAFs (Additional file 5: Figure S3B and Additional file 6: Video S1).
We also observed that CAFs induced an increased rate of proliferation in MCF10.DCIS cells. Using a thymidine analog to evaluate the rate of DNA synthesis, we observed that co-cultures had a consistently higher rate of DNA synthesis than CAFs alone or DCIS cells alone (Additional file 7: Figure S4A-D). CAF and DCIS co-culture using a slower growing DCIS cell line, i.e., SUM102, also resulted in changes in multicellular structure formation. We observed the presence of multiple invasive processes in SUM102:CAF co-cultures that were completely absent in cultures of the SUM102 cells alone (Additional file 8: Figure S5C and D, arrow, cf. S5A and B).
DCIS cells migrate with CAFs at the invasive front
CAFs express and secrete a number of proteases, which enhance their ability to migrate and remodel extracellular matrices [48][49][50][51]. In CAF:MCF10.DCIS co-cultures, we evaluated cell:cell interactions and motility. Here we found that MCF10.DCIS spheroids formed attachments to CAFs and remained associated with them throughout an 8-day culture. On day 1 (24 h after seeding) of a coculture we observed many single DCIS cells and a few CSFE labeled CAFs (Fig. 4a). After 3 days in culture DCIS cells had formed small spheroids that were in contact with one or more CAFs (Fig. 4b). On day 5 we observed CAFs at the invasive edges of DCIS tumor spheroids. A representative image shows a CAF (Fig. 4c, arrow) in contact with a DCIS spheroid. A time-lapse video shows the CAF leading the DCIS spheroid (Additional file 9: Video S2). High magnification imaging shows heterocellular contact between a tumor structure and a single CAF (Additional file 10: Figure S6). We did not observe these invasive characteristics in DCIS cell grown alone, as the tumor structures tended to roll in the extracellular matrix (Additional file 11: Video S3). By day 7 CAFs were seen at the invasive edge of most tumor structures (Fig. 4d).
Using confocal microscopy we examined the 3D spatial organization of the DCIS cells in relation to CAFs. A top view of a 3D reconstruction shows an MCF10.DCIS structure (only nuclei labeled) branching out toward CFSE labeled CAFs (pseudo-colored white) (Fig. 4e, Additional file 12: Video S4). The core of this structure (dashed circle) formed shortly after seeding and grew in size before the two branching outgrowths formed. These outgrowths are reminiscent of the "strand" multicellular migration pattern previously described [52]. A side view of the same structure shows that DCIS cells migrated downward from the core structure toward the CAFs (Fig. 4f). IL-6 protein expression in these co-cultures was detected using immunofluorescent staining (Additional file 13: Figure S7). A CFSE-labeled CAF showed high expression of IL-6, whereas the leading edge of the DCIS multicellular structure showed a gradient of IL-6 that was strongest near the attachment to the CAF (Additional file 13: Figure S7A, arrow). We also showed that MCF10.DCIS cells that do not migrate to serumfree media (Additional file 13: Figure S7D) migrated toward CAF-conditioned serum-free media (Additional file 13: Figure S7E, S7F).
CAFs and MCF10.DCIS cells utilize cathepsin B to degrade matrix
IL-6 has previously been shown to upregulate cathepsin B [21], a protease associated with breast cancer progression. Therefore, we examined the expression and localization of cathepsin B in relation to active proteolysis in our 3D MAME co-cultures. Our active proteolysis assays were performed during live cell imaging [33]. Fluorescent imaging of live cells in MAME co-culture revealed regions of proteolytic activity (green) around several multicellular structures (Fig. 5a). The cells were then fixed for cathepsin B immunofluorescent staining. Visualizing the same structures revealed cathepsin B expression at the interface between the multicellular structures and the matrix (Fig. 5b). Fluorescent overlay shows co-localization between DQ-collagen IV degradation and areas with high cathepsin B expression (Fig. 5c); however, matrix degradation also occurred in areas that did not stain for cathepsin B (arrow). Upon visualization using light microscopy, the cells in these areas had morphological features consistent with CAFs, which would be consistent with current knowledge of fibroblast migration and their secretion of many proteases in addition to cathepsin B [53,54].
We then used protease inhibitors to determine the contribution of all cysteine proteases and/or cathepsin B and L, to the overall proteolysis in the MAME culture. We treated cultures with E64d, a cell permeable pan-cysteine protease inhibitor, and found a 30 -40 % reduction of matrix degradation (Fig. 5d). We also treated cultures with the cell permeable cathepsin B/L inhibitor, CA074Me, and found a 60 -70 % reduction in matrix degradation (Fig. 5e). CA074Me has previously been shown to be a more efficacious inhibitor against cathepsins B and L than is E64d [35,55]. We did not observe complete inhibition of proteolysis with the cysteine protease inhibitors, a finding in accord with previous findings that DCIS cells and CAFs secrete many non-cysteine proteases [31,56,57].
Blocking IL-6 inhibits CAF-stimulated effects on human breast DCIS cells
We have shown that CAFs promote the proliferation and migration of human breast DCIS cells. To determine c Fluorescent overlay of dDQ-Col.IV (green) and cathepsin B immunostaining (red). Scale bar, 200 μm. Quantification of dDQ-Col.IV fluorescent intensity in MAME co-cultures treated with cell permeable inhibitors E64d (pan-cysteine protease inhibitor) (d) or the cathepsin B selective inhibitor CA074-Me (e). *P < 0.05, Student's t-test; mean ± SD whether CAF-stimulated migration and interaction with MCF10.DCIS cells could be inhibited by blocking IL-6, we co-cultured MCF10.DCIS cells with two human breast carcinoma-associated fibroblast lines WS-12Ti or CAF40 TKi, in the presence of IL-6 nAb or an isotype control antibody. IL-6 nAb reduced the size of MCF10.DCIS:CAF multicellular structures when grown with either CAF cell line (Fig. 6a-e). Multicellular structure volume measurements confirmed a significant reduction when treated with IL-6 nAb (Fig. 6f-h).
To confirm the role of IL-6 as a key player in the formation of the multicellular structure interconnections, we utilized shRNA targeting IL-6 expression. In CAF40T Ki and MCF10.DCIS cells, we achieved greater than 50 % reduction in secreted IL-6 (Additional file 14: Figure S8A). When we co-cultured CAF40TKi-shRNA control fibroblasts with MCF10.DCIS cells, we saw a phenotype similar to non-shRNA transduced cultures (Additional file 14: Figure S8B, cf. 3E). Knocking down CAF40TKi IL-6 expression in co-culture resulted in the formation of multicellular structures with uniform borders and few invasive processes (Additional file 14: Figure S8C). Co-culture with non-shRNA transduced CAF40TKi fibroblasts and shRNA-IL-6 transduced MCF10.DCIS cells showed greater MCF10.DCIS:CAF 40TKi interaction and multicellular structure branching (Additional file 14: Figure S8D). IL-6 signaling is propagated through either direct cell membrane receptor signaling or soluble receptor transsignaling (TS). In the DCIS cell lines, we found that IL-6R expression was low and sometimes undetectable by qRT-PCR. Additionally, we detected very low levels of sIL-6R secreted from MCF10.DCIS cells and none from SUM102 cells. In contrast, CAFs had a higher level of IL-6R gene expression and high secretion of sIL-6R as verified by ELISA. These findings suggest that IL-6TS is a likely mechanism of IL-6 action in these DCIS cells [58] (Additional file 15: Figure S9).
Discussion
A number of factors produced by CAFs have been shown to be involved in promoting malignant transformation in epithelial cells, these include TGFß [59][60][61] and CXCL12 (SDF-1) [59,61,62]. These factors are involved in eliciting a range of responses that are context-dependent and that benefit the tumor in various ways. The present study describes a novel addition to these known interactions and a new mechanism by which CAF can influence tumor progression. There is a direct correlation between serum IL-6 levels and poor prognosis in breast cancer patients [36,[63][64][65]. Studies have shown that CAFs [66], and various immune/inflammatory cells secrete pro-inflammatory cytokines including IL-6 and contribute to tumorigenicity [67][68][69][70][71][72][73].
In the current study, we show that IL-6 expression can be found in both the tumor and stromal compartments. In our IHC data we found approximately 65 % of patient samples had positive IL-6 staining; however, when we examined IL-6 expression in our DCIS tumor cell lines, we found the expression to be near the lower threshold of our assay. This discrepancy may be due to differences in gene expression between tissue and cells [74], differences in IL-6 expression with tumor grade/invasiveness [65], degree of "stemness" in cell lines vs. tissue [75], and differences between the assays and/or sample collection.
We next examined the interaction between human breast DCIS epithelial cells and human breast CAFs in the context of an in vitro 3D microenvironment. We hypothesized that CAFs promoted the migration of breast DCIS cells via paracrine signaling within the tumor microenvironment. Here we observed that DCIS cells migrated towards CAFs and upon attachment to CAFs, DCIS cells remained attached and migrated through the matrix following the lead of the CAFs. Similar findings of fibroblast-led migration have been reported for cocultures of invasive squamous carcinoma cells and CAFs [76]. It is also likely that CAFs migrated to DCIS cells, as some fibroblasts moved more easily and quickly through the matrix as seen in time-lapse videos; therefore it is possible that some fibroblasts migrated to and attached to DCIS cells thereupon leading migration of the DCIS cells.
An increase in DCIS cell proliferation and a change in multicellular structure morphology was observed in all of our co-culture experiments. DCIS multicellular structures showed invasive characteristics, having lost their uniform circular structure and had developed single or multiple protruding edges. An underlying mechanism for DCIS:CA F interaction, enhanced tumor cell proliferation, and migration was IL-6 signaling in the tumor microenvironment. We [31] as well as Sung et al. [77] have shown in 3D co-culture systems that paracrine HGF and its receptor cMet can drive the invasive potential of the MCF10.DCIS cells. We confirmed this in xenografts formed by orthotopic implantation of HGF-secreting fibroblasts and MCF10 .DCIS cells in SCID mice [31]. HGF and IL-6 have been shown to cooperatively enhance lung cancer cell invasion by upregulation of their corresponding cell surface receptors [78]. Also of interest is the stem cell-like properties of the proliferating DCIS cell population as Krishnamurthy et al. have shown that endothelial IL-6 enhances selfrenewal of cancer stem-like cells. Whether or not this is the case for CAF IL-6 in regard to proliferation of DCIS cells has yet to be determined [79].
Treatment of DCIS:CAF co-cultures with an IL-6 nAb abrogated the proliferation and migratory phenotype acquired by DCIS cells. This phenotype was primarily produced by the inhibition of CAF secreted IL-6 as shRNA knockdown of IL-6 in CAFs, but not in DCIS cells was able to replicate the phenotype. We also showed that MCF10.DCIS cells treated with IL-6 nAb had a down regulation of genes associated with EMT. A recent study suggests that tumor cell EMT is mediated through factors secreted from CAFs and that selective inhibition of TGFβ1 is sufficient to reverse EMT associated gene expression [80]. Other studies show cross talk between IL-6 and TGFβ signaling consistent with IL-6 and TGFβ both acting as drivers of EMT [81].
A Federal Drug Administration (FDA) approved humanized anti-IL-6R antibody, tocilizumab (Actemra), has shown promise in the treatment of inflammatory diseases particularly rheumatoid arthritis, Crohn's disease, and Castleman's disease. A major caveat of Actemra is that serum IL-6 levels are increased in patients after drug administration [82]. Since breast cancer patients with elevated serum IL-6 have poorer prognosis, Actemra may not be a practical therapy in these patients, however alternative approaches to reduce IL-6 signaling may prove efficacious.
Siltuximab, a monoclonal antibody against IL-6, is in clinical trials for therapies including; combinatorial treatment of metastatic renal cell carcinoma, multiple myeloma, and prostate cancer [83][84][85][86]. The FDA recently approved siltuximab for the treatment of multicentric Castleman's disease [87]. Studies have shown that siltuximab significantly inhibits the growth of nonsmall-cell lung cancer in primary xenografts [88], and ovarian cancer cell xenografts [89]. Therapeutic use of siltuximab for the treatment of breast cancer has not been fully evaluated, although preliminary studies with the estrogen receptor alpha positive MCF-7 breast cancer cell line suggests potential efficacy [90].
A limitation of the current study is that we did not have complete histories or demographic data to perform correlation studies to determine how IL-6 expression in DCIS relates to tumor prognosis. Another limitation is there is a limited number of commercially available DCIS cell lines. Here we used two cell lines that are commercially available. A third DCIS cell line that is commercially available is the SUM-225 DCIS cell line; however, this line comes from a metastatic recurrence so we chose not to use it. Another limitation is that our study only followed the progression and interaction of DCIS cells with CAFs, but not inflammatory cells associated with breast tumors. Additionally, we are investigating the probability of increasing the number of cell types grown together in our 3D MAME culture, for example: macrophages, adipocytes and lymphocytes.
Conclusion
In conclusion, we have shown that IL-6 paracrine signaling between DCIS cells and CAFs is a key mediator of early stage breast cancer cell proliferation and migration. Furthermore, IL-6 from CAFs facilitated the transition of DCIS cells from preinvasive to an invasive phenotype. This study highlights the necessity to explore paracrine signaling within the context of the tumor microenvironment and components therein. Table S1. Taqman Gene Expression Assays. Taqman Assays (primers) were selected for the human gene targets. Each Taqman Assay was selected based on suppliers "best coverage" criteria. (C) Incorporation of EDU in a representative structure from a MCF10.DCIS: CAF40TKi co-culture. Note that similar sized structures were chosen to compare levels of EDU incorporation in a 24-h period. Scale bars, 100 μm. (D) DNA was extracted from MAME cultures to quantify EDU concentration. Data shows a significantly higher concentration of EDU incorporation in DNA of co-culture as compared to monotypic cultures. ***P < 0.001, ****P < 0.0001, Student's t-test; mean ± SD. (TIFF 344 kb) Additional file 8: Figure S5. Carcinoma-associated fibroblasts enhance invasive growth of SUM102 cells in MAME co-culture.
|
2023-01-16T14:10:13.364Z
|
2015-08-13T00:00:00.000
|
{
"year": 2015,
"sha1": "5248605ddb1b988193ff1b4f22a8ce0f53be12b1",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1186/s12885-015-1576-3",
"oa_status": "GOLD",
"pdf_src": "SpringerNature",
"pdf_hash": "5248605ddb1b988193ff1b4f22a8ce0f53be12b1",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": []
}
|
229360482
|
pes2o/s2orc
|
v3-fos-license
|
Prevalence of malignant uterine pathology in utero-vaginal prolapse after vaginal hysterectomy
Materials and Methods: A descriptive observational cross-sectional study including 110 women with VH indication for hysterocele greater than or equal to POP-Q stage II with negative Pap smear still valid, and with normal transvaginal ultrasound in the past year. Patients with abnormal uterine bleeding or suspected premalignant or malignant uterine disease were excluded. All participants signed an informed consent document. The mean age was 64 year old (SD ±8.84), 85.5% of them being menopausal.
INTRODUCTION
Pelvic organ prolapse (POP) is a common health problem 1 affecting 30 to 50% of women, 2,3 and its prevalence increases with age and parity. 4 Its incidence continues to increase as a result of the aging of the world's populations and higher rates of obesity in some populations. 4 POP has a negative impact on quality of life of women and it is associated with physical, psychological and sexual problems. 5 Pelvic floor repair surgery is the main therapeutic tool for POP treatment. One of its surgical treatment options is vaginal hysterectomy (VH). The vaginal approach facilitates anatomical entry into the perineum and pelvis, resulting in easy access for the repair of vaginal relaxation. Once pelvic floor reconstruction surgery through the vagina has been defined, the surgical tactic should be selected considering the preservation or exeresis of the uterus. 6 This tactic should be planned with the aim of restoring normal anatomy and adequate support of the vaginal walls and apex.
Utero-vaginal prolapse (UV-POP) is the primary indication for most VHs in Uruguay, 6,7 and worldwide. 8,9 At present, VH is one of the most commonly performed surgical procedures to treat UV-POP and, generally, it is combined with other pelvic floor reconstruction procedures. 10 Hysterocele is usually accompanied by other conditions, such as cystocele, enterocele, rectocele, or relaxation of the pelvic floor. All these conditions must be corrected in the same surgical procedure. Tissue laxity facilitates the performance of hysterectomy through the vagina.
The vaginal removal of the uterus has a very long history, and references have been found, for instance, to conditions in which it had to be removed-in whole or in part-in Soranus of Ephesus, who reported that some authors indicated that it should be removed when the prolapsed uterus was of a blackish hue. 11 Since the first regulated VH was performed in the nineteenth century, 12 VH has been both revered and discredited, but it is still around today.
The reasons for preserving the uterus in women with UV-POP may be the patient's request or the gynaecologist's conviction about the advantages and disadvantages of conservative management. [13][14][15] Some authors argue that UV-POP is the result rather than the cause of POP, and hysterectomy associated with dissection of the pelvic floor could increase complications. 14,15 Other authors have voiced their concern about the possibility of undiagnosed pathologies or the risk of developing future diseases, such as cancer, if the uterus is not excised. 16,17 The dissenting opinions in this regard make it necessary to have open discussion with the patient about the pros and cons of hysterectomy or hysteropexy, so that the final decision would be informed, conscious and consensual.
The aim of this study was to establish the prevalence of malignant and premalignant uterine pathology in the histopathological findings in patients who have undergone VH due to UV-POP in a sample of women in our area.
MATERIALS AND METHODS
A cross-sectional descriptive observational and prospective study was carried out. Patients seeking consultation who were assisted by the Pelvic Floor Unit medical team between April 1, 2017, and March 31, 2019, were selected. The study was approved by the Ethics Committee of the Hospital de Clínicas of Medical School of Montevideo, University of the Republic. All the participants signed a consent document after being briefed about the study.
The sample included patients with asymptomatic prolapsed uterus greater than or equal to POP-Q stage II with an oncological cervical cytology study with negative Pap stain two years prior to VH and with normal transvaginal ultrasound (TVU) in the past year.
The Pelvic Organ Prolapse Quantification (POP-Q) system uses certain anatomical references such as the hymen, anterior vaginal wall, posterior vaginal wall, total vaginal length, urogenital hiatus, perineal body and cervix or vaginal apex as guiding points. All these references are expressed in cm and measured in proximal (above) or distal (below) in relation to the hymen and are designated as negative or positive respectively. Four prolapse stages from I to IV are established in which the fundamental considerations are the prolapse guiding points and their relation to the hymen expressed in cm and the total length of the vagina. 18 Patients were excluded if they had UV-POP with a history of abnormal uterine bleeding, a pathological Pap smear or a history of cervical dysplasias or premalignant endometrial lesions, a TVU with an endometrial line greater than 10 mm, a history of malignant ovarian tumours, use of tamoxifen or hormone replacement therapy, and those who did not give their consent for the study.
Out of a total of 133 patients assisted by the Unit team during the study period, 110 women (82.7%) were selected in accordance with the inclusion criteria, and 23 patients (17.3%) were excluded. Of the 23 cases excluded, 11 had a history of genital haemorrhage with premenopausal (three cases) or postmenopausal (eight cases) metrorrhagia, five patients had been diagnosed with cervical dysplasia or squamous intraepithelial lesion (SIL), four of whom were low-grade SIL (L-SIL) and 1 was high-grade SIL (H-SIL), five women had an endometrial line ultrasound measurement greater than 10 mm, and two cases had had clinical and ultrasound diagnoses of adnexial tumour.
The vaginal repair surgery was performed by the same team in all cases. All patients underwent a VH associated with vaginal repair surgery. Vaginal apex suspension was performed to the uterosacral ligaments in all patients in accordance with the McCall culdoplasty technique. 19 In cases with coexisting stress urinary incontinence, the middleurethral sling (MUS) technique was associated. 20,21 All extracted uteri were sent for histopathological study.
The following histopathological variables were analysed: presence of endometrial atrophy, endometrial hyperplasia with and without atypia, endometrial cancer, leiomyoma, focal and diffuse adenomyosis, chronic endocervicitis, L-SIL and H-SIL, as well as cervical carcinoma.
The statistical analysis was performed using the Excel 2013 Microsoft Office spreadsheet. For the qualitative variables in the cases where pathologies were detected, a frequency calculation was performed and expressed as absolute frequencies and relative percentage frequencies. For the quantitative variables, the mean, the standard deviation and range were calculated.
RESULTS
The mean age was 64 years old with a standard deviation of ±8.84, with an age range between 46 and 83 years. The mean body mass index (BMI) was 26.56 kg/m 2 with a standard deviation of ±3.23, with a range between 20.1 and 33.9 kg/m 2 . Most of the patients (85.5%) were postmenopausal ( Table 1). None of the patients had family history of endometrial cancer. Smoking was observed in 21.8% of cases. Hypertension was observed in 16,3% and diabetes in 11%. All women presented cystocele associated with hysterocele, while in 13.6% of them it was associated with high rectocele, and with enterocele in 1.8% (Table 1). 4.5% required a MUS technique.
As to the histopathological findings, 78.2% of cases did not have any pathology whatsoever. Amongst patients with endometrial pathology, 11 cases (10%) exhibited hyperplasia without atypia and 12 (10.9%) presented endometrial polyp without atypia. No cases of endometrial hyperplasia with atypia were observed. In some patients, an associated benign myometrial pathology was observed, such as uterine leiomyoma in 25.5%, and focal and diffuse adenomyosis in 23.6% of cases (Table 3). In regard to malignant pathology, only one (0.9%) of the patient exhibited endometrial cancer, and none of the patients showed a malignant or pre-malignant pathology of the cervix or ovaries ( Table 3). The ovaries were included in only two cases (1.8%) during the surgery with normal histological findings.
One case of endometrial cancer was found in a 71-year-old woman with stage IV hysterocele according to POP-Q, asymptomatic, with TVU showing a 1 mm endometrial line, who underwent a VH and bilateral adnexectomy. The histopathological result reports a moderately differentiated adenocarcinoma that extensively infiltrates the cervix and the myometrium as far as the serosa (stage II -2009 FIGO). Both annexes had a normal histopathological examination.
DISCUSSION
This study estimated the prevalence of malignant and premalignant uterine pathology in patients with UV-POP who However, the inclusion of all precancerous conditions had a prevalence of 21.8%. The premalignant lesions that were observed within the endometrium were hyperplasia and polyps without atypia. No case of atypia was observed. A lesional association with leiomyoma and adenomyosis of approximately 25% was found.
This makes it necessary to consider the importance of a clinical examination and a TVU at a time very close to the surgical procedure lest a detectable pathology is found only after surgery. Anyway, surgical behaviour would not vary in the cases analysed.
According to international literature, the incidence of malignant and premalignant endometrial pathology in asymptomatic women who undergo hysterectomy due to UV-POP ranges between 0.7 and 4.2%. 16,17,22,23 In a retrospective study 16 the authors included 644 women, 421 of whom were postmenopausal with asymptomatic POP, with 11 patients found to have an organic endometrial pathology (2.6%). As to these cases, only one case of endometrial cancer and 10 cases of hyperplasia were observed.
In another retrospective study 17 unsuspected endometrial cancer was found in four cases (0.8%) out of 517 asymptomatic patients who underwent VH, not to mention the high incidence of hyperplasia.
In another retrospective study, 22 three cases of premalignant and malignant endometrial pathology were observed in 456 asymptomatic patients (0.7%).
These three studies include women with asymptomatic UV POP with no preoperative endometrial assessment.
CONCLUSIONS
The conclusion is that it is necessary to assess the endometrium before POP surgery by means of TVU followed by an endometrial biopsy in cases of thickened endometrium.
In a retrospective study where they considered the preoperative assessment, Grigoriadis et al. 23 found 14 histopathological abnormalities in the uteri of 333 asymptomatic women who underwent VH (4.2%) and concluded that the incidence of premalignant and malignant gynaecological pathology is low but not negligible.
There is no agreement on the most appropriate course of action in the cases of asymptomatic patients with a thickened endometrium. 24 In a cohort study 25 the authors conclude that in asymptomatic postmenopausal women with an endometrium ≥11 mm, an endometrial biopsy should be performed.
Our research considered the value of the TVU, especially in order to assess the risk of unexpected premalignant and malignant endometrial pathology. One of the limitations of our study is that the number of patients analysed is relatively small, but it may be useful for future research in which other risk factors such as smoking and medical comorbidities are considered in addition to age and BMI. One of the strengths of this study is the prospective and cross-sectional design, in addition to the assessment of the risk of not diagnosing a premalignant and malignant pathology of the uterus through clinical assessment and Pap and TVU studies in the preoperative period.
It may be concluded that the prevalence of premalignant and malignant uterine pathology is low, at 0.9%. This leads to the reinforcement of the idea of the importance of a clinical history with no abnormal uterine bleeding, endometrial assessment by TVU close to the date of the surgery and cytological assessment of the cervix through a Pap smear which will allow adequate preoperative diagnosis in the case of a hysterocele. A briefing regarding the risk of premalignant and malignant pathology should be included in the information provided during the preoperative period in the form of a consent document. When conservative management, such as a surgical procedure for uterine fixation (hysteropexy) is chosen, it is imperative to make a careful preoperative assessment to rule out -with the exceptions shown by all studies-a precancerous or cancerous condition of the uterus.
Ethics
Ethics Committee Approval: The study was approved by the Ethics Committee of the Hospital de Clínicas of Medical School of Montevideo, University of the Republic (Date 2017, March 5).
Informed Consent: All the participants signed a consent document after being briefed about the study.
|
2020-12-22T15:08:04.935Z
|
2021-01-01T00:00:00.000
|
{
"year": 2021,
"sha1": "e7474e567a5542037c7af455e3d8077c286b2a5b",
"oa_license": null,
"oa_url": "https://doi.org/10.34057/ppj.2020.39.04.006",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "e7474e567a5542037c7af455e3d8077c286b2a5b",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": []
}
|
258864216
|
pes2o/s2orc
|
v3-fos-license
|
Reprogramming of glucose metabolism via PFKFB4 is critical in FGF16-driven invasion of breast cancer cells
Abstract Fibroblast growth factors (FGFs) are expressed in both developing and adult tissues and play important roles in embryogenesis, tissue homeostasis, angiogenesis, and neoplastic transformation. Here, we report the elevated expression of FGF16 in human breast tumor and investigate its potential involvement in breast cancer progression. The onset of epithelial–mesenchymal transition (EMT), a prerequisite for cancer metastasis, was observed in human mammary epithelial cell-line MCF10A by FGF16. Further study unveiled that FGF16 alters mRNA expression of a set of extracellular matrix genes to promote cellular invasion. Cancer cells undergoing EMT often show metabolic alteration to sustain their continuous proliferation and energy-intensive migration. Similarly, FGF16 induced a significant metabolic shift toward aerobic glycolysis. At the molecular level, FGF16 enhanced GLUT3 expression to facilitate glucose transport into cells, which through aerobic glycolysis generates lactate. The bi-functional protein, 6-phosphofructo-2-kinase/fructose-2, 6-bisphosphatase 4 (PFKFB4) was found to be a mediator in FGF16-driven glycolysis and subsequent invasion. Furthermore, PFKFB4 was found to play a critical role in promoting lactate-induced cell invasion since silencing PFKFB4 decreased lactate level and rendered the cells less invasive. These findings support potential clinical intervention of any of the members of FGF16-GLUT3-PFKFB4 axis to control the invasion of breast cancer cells.
Introduction
The conversion of a benign epithelial tissue toward a fully malignant carcinoma entails extensive remodeling in tissue architecture, driven by both genetic and micro-environmental factors [1]. Epithelial-mesenchymal transition (EMT), an embryonic trans-differentiation program, contributes largely in cancer progression by converting primary tumor cells into invasive in nature [2]. During EMT, notable cellular changes include loss of cell adhesions, change in cell polarity, interactions between new plasma membrane receptors and remodeled extracellular matrix (ECM) components; all finally lead to cell migration and generation of a more plastic mesenchymal cells [2,3]. This change in cellular phenotypes essentially involves the remodeling of the ECM components including collagens, laminin, fibronectin, proteoglycans etc., through a series of qualitative and quantitative changes. This ECM remodelling thus appears to be a key step in facilitating tumor growth and spread by interfering with cell-cell adhesion, polarity, and by augmenting growth factor signaling [4].
In normal cells, multiple metabolic cycles are always operative which provide constant supply of energy and biomass. However, this metabolic program is altered in cancer cells to sustain their continuous growth/proliferation and to adapt to new microenvironment at distant location [5]. A high rate of glycolytic flux, even in the presence of functional mitochondria and oxygen, is a typical metabolic hallmark of tumors. Although this 'aerobic glycolysis' is actually less energy-efficient pathway than oxidative phosphorylation, this reprogrammed metabolism confers several growth advantages on cancer cells including cell death evasion, cell proliferation, angiogenesis, and metastasis [6,7]. Thus, the phenomenon of metabolic alterations appears to facilitate successful spread and growth of metastatic tumour at distant sites [5]. The molecular mechanism underlying this metabolic reprogramming of cancer cells is complex and unique for each cancer type. Pinpointing the altered metabolic step can lead to more effectively targeting the metabolism of cancer patient with the consequence of reduced cancer invasiveness.
Gradual acquisition of cell invasiveness can take place due to sequential genetic and epigenetic changes, triggering induction of EMT in response to micro-environmental cues, for example secreted growth factors [8]. The fibroblast growth factor (FGF) family, comprising of 22 ligands, exerts their functions through four, highly conserved, cell-surface tyrosine kinase receptors (FGFR1, FGFR2, FGFR3 or FGFR4) [9]. FGF/FGFR signaling has established roles in the control of multiple biological processes, such as endocrine homeostasis, wound repair, cellular proliferation, differentiation, and survival [10]. However, FGF/FGFR pathway has been demonstrated to have a significant relevance in tumor growth, metastasis, and resistance to anticancer therapies [11][12][13][14]. Aberrant FGF/FGFR activation may occur either in ligand-dependent or -independent manner in several cancer types, including breast cancer. Overexpression of FGF ligands cause excessive mitogenic signaling through FGFR, promoting cancer progression. Example includes the elevated expression of FGF2 [15] and its involvement in driving basal-like breast cancer through the autocrine activation of FGFR signaling [16]. FGF2 could also promote BC growth in hormone-independent manner, leading to endocrine therapy resistance [17]. Fillmore et al. has already demonstrated the potential involvement of paracrine FGF9/FGFR signaling in the estrogen-mediated expansion of a breast cancer stem-cell-like subpopulation in vitro [18]. FGF16, a member of FGF9 sub-family, is a 207-amino acid protein with a core region of 120 amino acids that acts through binding to heparin and cell surface FGF receptor (FGFR) [19]. Recently, FGF16 is implicated in the progression of hepatocellular [20], lung [21], and embryonic carcinoma [22] with unique mechanistic action for each cancer type. In continuation to these findings, we attempted to investigate the potential oncogenicity of FGF16 and the underlying mechanism in breast cancer progression.
In the present study, the oncogenic potential of FGF16 in breast cancer progression has been demonstrated, where the activation of EMT was found to be critical. Further, it was unraveled that a set of genes involved in ECM remodeling is regulated by FGF16. Detail studies uncovered a hitherto unidentified role of FGF16 in promoting aerobic glycolysis where a robust expression of PFKFB4 was particularly observed. Overall, our findings provide a basis for developing a new metabolism-targeted therapeutic intervention to reduce aggressiveness of breast cancer.
FGF16 induces EMT in breast cancer cells
FGF signaling has been implicated in pathogenesis of many cancers [23]. To investigate the potential role of FGF16 in breast cancer development, we explored whether it can trigger EMT with immortalized human mammary epithelial cell-line MCF10A. In comparison with vehicle-treatment, cells treated with rhFGF16 exhibited morphological alteration, shifting from round cobblestone-like to fibroblast-like elongated shape ( Figure 1A). At the molecular level, reduced expression (by ∼50%, P<0.05) of epithelial markers like desmoplakin (DSP) and E-cadherin (CDH1), and significant increase in mesenchymal markers like vimentin (VIM) and N-cadherin (CDH2) was observed in FGF16-treated cells ( Figure 1B,C). Treatment of cells with FGFR-inhibitor, PD173074 (marked as PD), caused a marked reduction in mesenchymal markers and increase in epithelial genes ( Figure 1B,C). The same trend was observed in MCF7 cell line (Supplementary Figure S1A). Similarly, combined treatment of FGF16 with its neutralizing antibody reversed the change in expression of EMT markers ( Figure 1B,C), supporting a definite contribution of FGF16 in promoting EMT. This reduction in E-cadherin, concomitant with the increase in Vimentin, was reinforced at their protein levels by Western blot ( Figure 1D) and confocal imaging ( Figure 1E,F and Supplementary Figure S1B). In addition, the reorganization of actin cytoskeleton during morphological change in cells was quite evident from phalloidin staining (Supplementary Figure S1C). Here, pre-treatment of PD along with FGF16 exhibited partial morphological alterations. Further, morphometric analysis of nuclei in DAPI-stained cells revealed that FGF-treated cells had nuclei of significantly larger volume compared with those of control cells (Supplementary Figure 1D). This observation is in agreement with earlier report showing that the alteration in nuclear morphology and architecture are associated with transformation into cancer [24].
Growth factors, in general, confer their impact on any cellular events, majorly through the activation of signaling pathway(s). To verify this, the activation of P38, MAPK, JNK, and PI3K pathways were assessed and observed the and mesenchymal (C) markers were done by Q-PCR assay with RNA isolated from MCF10A. Relative gene expression is indicated as 'fold' change in the y-axis (mean + − SD) with statistical analysis by ANOVA followed by Tukey's Honest Significant Test (HSD). * denotes P<0.05, 'ns' stands for non-significant change. (D) Western blot was performed using specific antibodies with lysates of vehicle-, FGF16-and FGFRI-treated MCF10A cells. (E,F) Immunostaining with antibodies against E-cadherin and vimentin followed by Alexa Fluor-488 (green) and nuclei staining with DAPI was done in control (vehicle-treated) and FGF16-treated MCF10A cells. Magnification scale bar: 50 μm (E-cad), 25μm (Vim). (G) Western blot was performed using specific antibodies with cell lysates after respective treatment. (H) The change in gene expression was assessed by Q-PCR assay and relative gene expression is indicated as 'fold' change in the y-axis (mean + − SD) with statistical analysis by ANOVA followed by Tukey's Honest Significant Test (HSD) test. * denotes P<0.05, 'ns' stands for non-significant change. control cells at 24 h was calculated and represented with statistical significance calculated by ANOVA followed by Tukey's Honest Significant Test (HSD) test. * denotes P<0.05, 'ns' stands for non-significant change. (C,D) Transwell invasion assay was performed with MCF10A cells treated as mentioned. Cells at independent fields for each well were counted and plotted with error bar. The statistical significance was calculated by ANOVA followed by Tukey's HSD test. * denotes P<0.05, 'ns' stands for non-significant change.
MAPK pathway to be active by 40 ng/ml FGF16 at 1.5 h (Supplementary Figure S1 E,F and Figure 1G). Here, the working dose of PD and U0126 was also determined (Supplementary Figure S1G,H). Subsequent changes in the expression of CDH1 and CDH2 genes ( Figure 1H) convinced the implication of MAPK pathway in FGF16-induced onset of EMT. Activation of EMT facilitates increased cell motility and invasiveness, key events in cancer metastasis [2,25]. At first, the wound healing assay was performed which exhibited a faster wound closure upon FGF16 treatment compared with vehicle-treated (control) MCF10A cells after 24 h (Figure 2A, compare panels ii and iii), while PD-treatment significantly (P<0.05) inhibited FGF16-mediated cell movement (Figure 2A, compare panels iii and iv). The results are quantitatively displayed in Figure 2B. This finding was verified in the MCF7 cell line (Supplementary Figure 1I-J). Similarly, matrigel invasion assay demonstrated that rhFGF16 stimulated invasion of MCF10A cells by ∼1.5 fold ( Figure 2C,D, P<0.05), while PD-treatment remarkably prevented this invasion.
Since four isoforms make up FGFR, a natural question arises: which of the four isoforms is/are responsible for the early switch in FGF16-activated signaling? This question was addressed by estimating the relative binding affinities of the four isoforms to FGF16 using established in silico methods. The starting point was the crystal structure of FGFR1 bound to FGF9 (PDB ID: 5w59). Since FGF9 shares significant sequence homology with FGF16, homology models of FGF16 bound to all four isoforms were built using FGFR1 bound FGF9 as the template. Using the homology models as starting structures, molecular dynamics simulations were performed and the trajectories were analyzed using standard sampling techniques for estimating the relative binding free energies of the four isoforms with FGF16 (see Supplementary Methods, Supplementary Results and Supplementary Table S1) as: G = 0.0, 2.2, 3.2 and 7.1 kcal/mol for FGFR1, FGFR2, FGFR3, and FGFR4, respectively. The trend in relative binding affinities was then compared with minor sequence alterations in the four isoforms at the binding interface with FGF16 (Supplementary Figure S2). Binding affinities matched with the sequence alterations, especially FGFR4, with the least predicted affinity showed the maximum variation, suggesting that the relative binding affinities follow the trend: FGFR1 > FGFR2 > FGFR3 >> FGFR4.
The clinical relevance of FGF16 expression was investigated. Q-PCR assay (n = 5 paired and 8 orphan; Figure 3A; P<0.005) and immuno-flourescence-based IHC analysis (n=6 Paired; Figure 3B) demonstrated elevated expression of FGF16 in human breast cancer compared to normal breast tissues. Then, we extended our investigation on xenograft mouse model. Here 4T1-FLAG or 4T1-FGF16 (both stably expressed) was injected subcutaneously into female Balb/c mice and tumors appeared. The size of the tumor, as induced by FGF16, was significantly bigger than that of empty vector transfected set ( Figure 3C,D) supporting the contribution of FGF16 in the growth of primary tumor. Immunoflourescence-based IHC analysis demonstrated the reduction in E-cadherin ( Figure 3E) and the increase in vimentin ( Figure 3F) expression in FGF-induced tumor sections. Furthermore, a remarkable increase in the expression of laminin, a basement membrane protein involved in breast cancer invasion was also evidenced in tumor section ( Figure 3G). In addition, prominent staining of Ki67 supported the proliferation of cancer cells in both sets ( Figure 3H). Hematoxylin-Eosin staining of the respective tissue sections were performed (Supplementary Figure S3A). In summary, elevated expression in tumor samples and induction of cell invasiveness/migration supported the oncogenic potential of FGF16 in breast cancer.
FGF16 differentially alters the expression of ECM-associated genes
To identify the functional signatures underlying FGF16-induced oncogenicity, whole cell transcript analysis was performed with MCF10A after exposure with FGF16 for 24 h. The treatment of 24 h was selected since the changes in EMT markers were evidenced at this time point ( Figure 1). Notably, the expression of a large number of genes was found to be altered whereas several genes remained unchanged at this time point ( Figure 4A). Within this list of genes, while applying fold change cutoff of 1.5 (log2 FC >0.58 or < −0.58) and P-value <0.05, 956 genes were found to be differentially altered; among them 691 genes were up-and 265 genes were down-regulated (Supplementary Table S2 and 3). Next, DAVID bioinformatic tool was applied and it was found that differentially expressed (DE) genes are involved in a variety of biological processes including angiogenesis, nucleosome assembly, telomere organization, protein metabolic process, ECM disassembly, cell adhesion etc. (Figure 4B). At the same time, pathway analysis by PANTHER Classification System revealed that these DE genes are associated with several cancer-associated signaling pathways including angiogenesis, RAS-pathway(P04393), EGFR signaling pathway (P00018), PDGF (P00047)-, FGF (P00021)-, VEGF(P00056)-, cadherin (P00012)-and integrin-signaling pathway (P00034) and glycolysis (P00024, Supplementary Figure S3B).
Next, we undertook an in-depth analysis of the altered transcription profile. Interestingly, from the total DE gene list, we observed that 41 genes related to extracellular matrix (ECM) organization were deregulated by FGF16 ( Figure 4C). Moreover, the enrichment score plot from Gene Set Enrichment Analysis (GSEA) showed a differential distribution of ECM regulator genes ( Figure 4D) in FGF16-stimulated cells. Taken together, the bioinformatic analysis of the altered transcription profile revealed a possible role of FGF16 in ECM regulation.
Considering ECM remodeling to be a key determinant of cancer progression, we were motivated to verify the change in a subset of ECM-related genes by Q-PCR assay in the cell-line. Hence, we picked up representative genes of collagen, integrin, laminin, keratin families, all being structural components of ECM. FGF16 stimulation caused up-regulation of KRT14, MCAM, LAMA5, LAMC2, and COL12A1 genes and down-regulation of ITGB6, ITGB2, ITGA6, COL4A2, and COL8A1 genes in MCF10A whereas treatment of receptor inhibitor (RI) with FGF16 somewhat reversed the changes in their expressions ( Figure 4E). In continuation, immunofluorescence imaging confirmed the similar change in laminin and Integrin A6 in their protein levels (Supplementary Figure 3C,D) as well. On the contrary, the mRNA level of ITGA6 and ITGB6 were induced in the presence of FGF16 in MCF7 cell-line, which could be reverted with PD treatment (Supplementary Figure S3E). Altogether, the regulatory role of FGF16 over ECM related gene expression was established.
FGF16 induces reprogramming of glucose metabolism
Change in the metabolic program is perhaps an emerging driving force to support detachment from ECM to initiate cell migration [5]. Based on our cues on pro-migratory role of FGF16, the change in metabolism was investigated in this regard. Interestingly, RNA-sequencing data revealed a group of transporters i.e. solute carrier (SLC) genes to be differentially altered (Supplementary Figure S4A) by FGF16. Among them, a prominent up-regulation in GLUT3 (SLC2A3) was indicated in FGF-treated cells. Q-PCR assay also validated the increase in GLUT3 by ∼2.5 fold (P<0.05) whereas pre-treatment with PD reverted the FGF-induced expression ( Figure 5A). In contrast, such alteration was not evidenced in its isoform GLUT1 by FGF16 ( Figure 5A). A similar trend was noticed in the MCF7 cell line as well (Supplementary Figure S4B). Through the up-regulation of GLUT3 expression, cells enhance glucose uptake and subsequently glycolysis. Hence, 2-NBD-glucose uptake assay was performed in MCF10A. The presence of FGF16 increased the glucose uptake ( Figure 5B, P<0.05) that was compromised while GLUT3 was down-regulated by siRNA. Moreover, the combination of FGF16 and siGLUT3 showed the glucose uptake to be lower than FGF16 alone ( Figure 5B). In parallel, the uptake was measured in presence of Apigenin, a flavonoid inhibitor of GLUT1 mediated glucose transport [26], to verify the assay performance. Apigenin alone was found to be able to lower the glucose uptake but jointly with FGF16 could retrieve the same extent of glucose uptake as was observed with FGF16 alone ( Figure 5B). This finding could indicate the implication of GLUT3 over GLUT1 in FGF16-induced cellular glucose uptake. The knockdown efficiency of siRNA against GLUT3 was verified by QPCR assay (Supplementary Figure S4C).
Next, the influence of FGF16 on the expression of key glycolytic genes was assessed by Q-PCR assay. A noticeable increase was observed only for ENO1, GAPDH by FGF16 that was abrogated by PD pre-treatment ( Figure 5C). However, the change in expression of HK1 and HK2 was merely remarkable ( Figure 5C). Interestingly, between lactate dehydrogenase A and B (LDHA and LDHB), which govern the production of lactate, LDHA was also found to be elevated by ∼5 fold, not observed for LDHB, its other isoform ( Figure 5C). The same trend was observed in MCF7 as well (Supplementary Figure S4D). Next, the alteration in metabolic profile of MCF10A was examined by 1D proton NMR-spectroscopy. FGF16 enhanced the production of endogenous intermediates of glycolysis ( Figure 5D), namely glucose-6-phosphate (G6P), 2-phosphoglycerate (2PG) whereas pre-treatment with PD reversed this induction. Intriguingly, the intracellular level of lactate, the end-product of aerobic glycolysis, was found to be ∼2 fold higher despite a mere increase in pyruvate in FGF16-induced cells ( Figure 5D). A prominent reversal in the level of all these intermediates was evident in the presence of PD ( Figure 5D). Lactate, after being produced in cells, is mostly secreted in the medium. Hence the metabolic profile of extracellular portion was also examined. Partial least-square discriminant analysis (PLS-DA) of NMR spectral bins of extracellular metabolites demonstrated distinct separation among untreated control cells, cells treated with FGF16 alone or in combination with PD ( Figure 5E). Concomitant with the intracellular level, FGF16 increased the secreted lactate concentration by 1.8-fold (P<0.05) whereas pre-treatment with PD decreased it by ∼80% ( Figure 5F, P<0.05). This certainly indicates a FGF16-driven metabolic shift toward aerobic glycolysis concomitant with increase in cell number (Supplementary Figure S4E). Subsequently, we verified whether our earlier observation of induced wound closure by FGF16 is influenced by the increased cell proliferation. After finalizing the working dose of 5FU (Supplementary Figure S4F,G), wound healing assay demonstrated that the treatment of cells with 5-FU could merely affect the wound closure by FGF16 (Supplementary Figure S4H).
PFKFB4 is critical in FGF16-induced glycolysis and tumorigenesis
To establish a mechanistic link between FGF16 and glycolytic induction, the rate-limiting steps of this pathway were investigated. Since our RNA-sequencing data demonstrated a significant up-regulation of PFKFB4 expression by FGF16, we pinpointed on the members of PFK gene family to assess their potential role. Mammalian PFK-1 exists in three isoforms PFKP, PFKL and PFKM. Upon FGF16 over-expression, only PFKP showed a modest increase in its mRNA level, while the other two isoforms (PFKL and PFKM) were slightly decreased ( Figure 6A). Two isozymes of PFK2, i.e. PFKFB3 and PFKFB4, showed elevated expression in many cancer types and are involved in altered glycolysis [27]. Interestingly, the expression of PFKFB4 was significantly enhanced by FGF16 ( Figure 6A) whereas PD treatment abrogated the same. Here, the observation of PFKFB3 to be slightly decreased by FGF16 led us to hypothesize PFKFB4 to be a potential player in the control of FGF16-mediated glycolytic induction ( Figure 6A). This finding was verified with MCF7 cell line (Supplementary Figure S5A). Western blot assay demonstrated a similar trend in their protein levels (Supplementary Figure S5B). Detail study uncovered that GLUT3 and PFKFB4 genes were turned on by the activated MAPK-signaling pathway as U0126-tretament reversed the FGF16-induced observation ( Figure 6B). Subsequently, lactate assay was performed where an increase in extracellular lactate content was observed by FGF16 ( Figure 6C), typical to our earlier finding ( Figure 5F). Surprisingly, siRNA-mediated abrogation of PFKFB4 reduced this secretion by FGF16 by >2-fold ( Figure 6C, P<0.05). However, down-regulation of PFKP reduced lactate production albeit maintained a moderately higher level in combination with FGF16 ( Figure 6C, P<0.05).
Furthermore, we deciphered the influence of the FGF16-FGFR-PFKFB4 axis on facilitating the onset of ECM remodelling as well as EMT. For this, the changes in expression of two EMT regulators, ZEB1 and SNAIL, two mesenchymal markers, Vimentin (VIM) and N-Cadherin (CDH2), and one ECM degrading enzyme, Matrix Metalloproteinase-2 (MMP-2) were quantified at their mRNA level in both MDA-MB-231 ( Figure 6D) and MCF10A (Supplementary Figure S5C). Noticeable up-regulation of all the five genes was observed by FGF16 ( Figure 6D) that was compromised by PFKFB4-siRNA. In contrast, suppression of PFKP resulted in the up-regulation of all the five genes and maintained the very same trend in combination with FGF16 ( Figure 6D) as observed with FGF16 alone. Similar trend in the change in expression of these selected genes was found in MCF10A (Supplementary Figure S5C) as well. This signifies that FGF16 and PFKFB4 share an equivalent mode of action to augment cell motility at molecular level. The knockdown efficiencies of shRNA and siRNA, against PFKP and PFKFB4, respectively, were verified by QPCR assay and Western Blot (Supplementary Figure S5D,E). This investigation was extended at the phenotypic level by matrigel invasion assay in MDA-MB-231 because of its aggressive nature. Here, FGF16-driven cell invasiveness was reduced by ∼50% (P<0.05) when PFKFB4 expression was abrogated by siRNA ( Figure 6E). In contrast, shPFKP could not notably affect the cellular invasion alone or in combination with FGF16. Overall, our findings support the critical involvement of PFKFB4 isoform in FGF16-promoted metabolic shift followed by cell motility.
Discussion
The major cause of fatality in breast cancer is its aggressiveness, achieved through increased motility/invasion of cells, promoted by growth factors and cytokines, secreted from tumor micro-environmental cells. Concurrently, reactivation of EMT in adult cells triggers localized invasion and hence is considered to be the first step in metastatic dissemination [28] and is correlated, to some extent, to poor patient survival [29][30][31]. This transient up-regulation of EMT-inducing transcription factors occurs as a response to the activation of intracellular signaling pathways by various growth factors including FGF [32]. Therefore, FGF signaling has been implicated in promoting the oncogenic progression [10]. Apart from developmental role in cardiomyocytes [33], FGF16 has been identified to be potentially involved in hepatocellular carcinoma [20] as well as in lung cancer [21]. Our investigation was initiated with the finding of elevated expression of FGF16 in human breast tumor samples. Detail study uncovered its involvement in the activation of MAPK signaling cascade leading to onset of EMT followed by cell invasion (Figures 1 and 2). In conjunction, key components of ECM were found to be altered ( Figure 4) which may, in turn, support tumor growth and spread. Moreover, gradual growth in tumor size reinforced the oncogenic potential of FGF16. High expression of FGF8 has been implicated in promoting EMT and tumorigenesis in oral squamous carcinoma cells and in colorectal cancer [34,35]. Our finding further supports the emerging evidences demonstrating atypical expression of FGFs and aberrant FGF/FGFR signaling in cancer development.
Whole cell transcriptomic analysis unravelled that a group of genes of solute carrier (SLC) family is differentially regulated by FGF16, indicating a possibility of metabolic alteration, an oncogenic hallmark. The SLC superfamily encodes membrane transport proteins for exchanging molecules including glucose, amino acids, vitamins, nucleotides, inorganic ions, organic ions, neurotransmitters and drugs [36]. Altered expression of SLC genes has been demonstrated to affect various steps of tumorigenesis, including proliferation, apoptosis, invasion and metastasis, chemotherapy resistance [37]. Even SLCs have been considered to be a therapeutic target in the control of pancreatic and renal cell carcinoma [38,39]. Surprisingly, the elevation of mRNA level of GLUT3 (SLC2A3), a transporter with highest affinity for glucose, but not of its isoform GLUT1, was observed ( Figure 5) and eventually the increase in glucose uptake by FGF16. High expression of GLUT3 has been demonstrated in brain cancer [40] and is responsible to resist certain chemotherapy drugs as well [41]. The essential role of GLUT3 in metabolic alteration to promote breast cancer brain metastasis is also established [42]. In addition, it is known that elevation of GLUT3 promotes EMT, invasion, metastasis and inflammation in triple negative breast cancer [43].
Concomitant with the elevation of glucose uptake, the glycolytic flux was found to be increased with the secretion of lactate ( Figure 5). Whole cell metabolomics and expression profiling of glycolytic genes revealed that cells preferred glycolysis over TCA cycle under the influence of FGF16 ( Figure 5). The intermediates of the glycolytic pathway are utilized in multiple biosynthetic processes and NADPH in sequestering reactive oxygen species to resist cell death [44]. Aerobic glycolysis is known to be one of the most prominent metabolic changes associated with cellular transformation into cancer. Lactate is not only a mere waste product of aerobic glycolysis; it is actually exploited through the onset of oncogenesis and cancer progression. Higher lactate production has been documented in pancreatic ductal adenocarcinoma [45], renal cell carcinoma [46], breast cancer [47], and lung cancer cells [48]. Lactate promotes invasion by tumor acidification, continuous proliferation by activating 'metabolic symbiosis' , evasion of immune system and angiogenesis [49]. Further, lactate specifically assists EMT by reducing the pH of tumor microenvironment. It not only activates MMPs but also stimulates the expression of relevant transcription factors including VEGF, HIF-α, leading to cell migration [50].
In this report, PFKFB4 was unveiled to be specifically up-regulated by FGF16 ( Figure 6). Between two PFK families, PFK-1 catalyzes the second rate-limiting step of glycolysis, converting fructose-6-phosphate into fructose-1, 6-bis phosphate. Three isoforms of mammalian PFK-1 exist. While PFKL and PFKM are enriched in liver and muscle, respectively, the third isoform, PFKP, is mainly expressed in platelet and also elevated in most human cancer cells [51,52]. PFKFB4 is also an emerging contributor in many cancer types including lung adenocarcinoma [53], breast cancer [54][55][56][57] and gastric cancer [58]. Apart from glycolysis, PFKFB4 governs cancer progression through various mechanisms-regulating transcription through phosphorylating SRCs [53], affecting autophagy [27] or inducing the production of hyaluran, a major constituent of ECM [56]. Mechanistically, PFK-1 is allosterically inhibited by several products of glucose metabolism including ATP, citrate, and H + ions and is activated by fructose-2,6-bisphosphate (F26BP), a shunt product of glycolysis. F26BP has the ability to counteract the allosteric inhibition by ATP, thus stimulating PFK1, glucose uptake and flux through the entire glycolytic pathway as well [59]. The intracellular concentration of F26BP is controlled by a family of four bi-functional enzymes (PFKFB1-4), having both fructose-6-phosphate kinase and F26BP phosphatase activities [60]. The increased production of lactate by FGF16, a key highlight of our investigation, was mediated by PFKFB4 ( Figure 6). Even FGF16-induced EMT/invasion of cells was found to be lowered while abrogating the expression of PFKFB4 ( Figure 6). On contrary, these molecular changes remained unperturbed by PFKP knockdown (Figure 6). However, direct regulation between PFKFB4 and transcriptional machinery of EMT genes, if any, remained less ventured, thus providing a productive field for further investigation.
In conclusion, here we present a comprehensive mechanistic investigation on oncogenic potential of FGF16 in breast cancer through the initiation of EMT and invasion. The breast cancer cells were found to adopt the GLUT3-PFKFB4 axis under influence of FGF16 (Figure 7). Up-regulation of GLUT3 and PFKFB4 results in increased glucose fuelling in the cells coupled with elevated lactate production and secretion, indicating a metabolic shift towards glycolysis (Figure 7). This finding has the potential to open up a new avenue to design novel therapeutic agent(s) against FGF16-GLUT3-PFKFB4 axis that may reduce the invasiveness of breast cancer cells.
Cells were starved for 16 h prior to the treatment with recombinant human FGF16 (rhFGF16; Peprotech) at 40 ng/ml for 24 h, unless otherwise mentioned. Treatment with PD173074 (PD; Cayman, U.S.A.), an inhibitor of FGF-receptor, was done at 20 ng/ml for 30 min prior to FGF16 treatment. 0.1% BSA in PBS was used as vehicle for growth factor treatment and DMSO was used as vehicle in 'control' cells for PD-treatment. The treatment with the inhibitors was as follows: 5-Fluorouracil (5-FU), a potent cell proliferation blocker, at 50 μM; 2-deoxy glucose, glycolytic inhibitor at 50 mM [61] and U0126, a highly selective inhibitor of MAPK/ERK kinase at 10 μM [62,63]. The control set of each inhibitor was given equal volume of vehicle.
Cell imaging
MCF10A cells were starved for 16 h, treated with rhFGF16 for 48 h and the change in morphology was visualized and captured by Olympus 1X 83 inverted microscope.
Expression constructs and transfection
For transient/stable transfection, cell-lines were seeded in 1 × 10 5 cells/well in 6-well culture plate. After 24 h, 1 μg of construct was transfected using 2 μl Fugene-HD (Promega, Madison, U.S.A.) in Opti-MEM medium (Gibco) and kept for 5 h followed by replacement with fresh complete medium. The cells were incubated for 24 h before performing assays. For stable transfection, cells were selected in G418 (400 μg/ml). The shRNA mediated knockdown of PFKP expression was performed by transfection as described above.
siRNA transfection
The RNA interference was carried out by the siRNAs against PFKFB4 and non-targeting siRNA at 20 nM and GLUT3 at 10 nM using 10 μl INTERFERin reagent (Polyplus, USA) on cells seeded in 6-well culture plate. The cells were further incubated for 24 h before performing assays.
Quantitative real-time PCR (Q-PCR)
Total RNA was isolated from cells using RiboZol ™ RNA extraction reagent (Amresco) following the standard protocol. First-strand cDNA synthesis was done with RevertAid first strand cDNA synthesis kit (Thermo Fisher Scientific, U.S.A.) followed by Q-PCR assay by using Power SYBR green master mix (Invitrogen). The comparative C T method ( C T ) was used to measure relative gene expression where the fold enrichment was calculated as: 1/2 C T . Here, C T is the C T of the housekeeping gene subtracted from the C T of target gene.
C T is the C T of control sample subtracted from C T of treated sample. The fold change >1 means increased expression and <1 means decreased expression. The primer sequences are mentioned in Supplementary Table S4.
Western blot
Following proper treatment/transfection, whole cell extracts were prepared from cell pellets using RIPA buffer (50 mM Tris-HCl pH 8, 150 mM NaCl, 1% NP40 and 1 mM EDTA) by incubating in ice for 20 min with Protease Inhibitor Cocktail (Roche) followed by sonication. Then, lysed cells were centrifuged at 10,000 rpm for 10 min and supernatants were taken for measuring concentration. The samples were electrophoresed on 7% or 10% or 12% SDS-PAGE and transferred on nitrocellulose membrane (Merck Millipore) followed by blocking with non-fat skimmed milk (3%) for 1 h and probed with specific primary antibodies for overnight at 4 • C. For phospho-antibody, blocking was performed in BSA (5%). After incubating with HRP-tagged secondary antibody, blots were developed with Western ECL substrate (Bio Rad, Hercules, CA, U.S.A.). Primary antibodies used are enlisted in Supplementary Table S5.
Immunocytochemistry
Cells were seeded (8 × 10 4 cells/well) on coverslips in 6-well plate and grown to 70% confluency. After required treatment/transfection, the cells were fixed with 4% paraformaldehyde for 15 min, permeabilized with 1% Triton X-100 for 10 min and blocked with 3% BSA for 1 h. Cells were then incubated with anti-VIM, anti-E-cadherin, anti-laminin and anti-ITGA6 antibodies for 1.5 h. Following washes with 1× PBS-T buffer for 5 min for three times, the cells were incubated with Alexafluor 488-conjugated secondary antibody for 1 h at RT. For Phalloidin staining, cells were incubated with Alexafluor 488-conjugated phalloidin for 1.5 h. The coverslips were then washed with PBS-T and mounted with Prolong Diamond Anti-fade Mountant with DAPI (P36966, Thermo Fisher Scientific). The images were captured with Leica TCA SP8 confocal microscope with Leica Application Suite (LAS). All the imaging experiments were done in duplicate.
Nuclear morphometric analysis (NMA)
Nuclei of DAPI-stained (n=30) cells, captured by confocal microscope from two independent experiments, were selected for each condition of experiment. Nuclei were marked to assess morphometric data i.e. area, aspect, area box, radius ratio and roundness. ImageJ (NHI, Bethesda, MD, U.S.A., http://rsbweb.nih.gov/ij/) with NMA plug-in (http://www.ufrgs.br/labsinal/nma/) was used. Nuclear Irregularity Index (NII) was calculated for each condition as described in [64] and fold change values with respect to control set was plotted with Graphpad Prism 5.01.
RNA sequencing and analysis of differentially expressed genes
For RNA-seq analysis, total RNA was extracted from MCF10A cell line using RNAiso Plus (Takara Bio) following standard protocol. The sample was treated with DNase I and its quality was verified by Agilent Bioanalyzer ® RNA 6000 Nano/Pico Chip. Samples with an RNA integrity number (RIN) > 7 were further processed. One microgram of total RNA was used for library preparation using a NEBNext Ultra II DNA Library Prep Kit for Illumina (New England Biolabs). Sequencing was performed on NovoSeq 6000 at Sandor Speciality Diagnostics Pvt. Ltd. Raw Data Processing was performed by In-house perl scripts which removed low quality bases and adapter sequences. Cufflinks (Module: Cuffdiff) was used for differential expression analysis with parameters Log 2 FC = 0.58 and P-value cut-off <= 0.05. Three biological replicates were used for each sample.
The biological and functional classification of differentially expressed genes was executed by using DAVID biological resources and PANTHER Classification System (http://pantherdb.org/). To get the idea of enrichment of gene sets in a specific biological process in an unbiased manner, Gene set enrichment analysis (GSEA) was performed with GSEA tool (Broad Institute) using default setting.
NMR spectroscopy
For extracellular metabolomics, the media from transfected/treated cells were mixed with ice-cold methanol: chloroform (1:1) solution, vortexed for 30 s and then subjected to cold centrifugation at 12000 rpm for 20 min at 4 • C. Subsequently, the upper aqueous layer was collected, lyophilized and stored at −80 • C till further analysis. For intracellular metabolic profiling, cells were washed with ice cold PBS and then immediately quenched with 50% chilled methanol, counted under microscope using haemocytometer and harvested. Cells were then lysed by sonication at amplitude 50 for 1 cycle of 30 sec pulse followed by 20 sec gap followed by addition of chilled chloroform. Then, the sample was vortexed for 5 min and the whole suspension was then centrifuged at 12000 rpm for 20 min at 4 • C. The separated aqueous layer was collected and lyophilized properly. Dried samples were then stored at -80 • C till use.
NMR spectra were acquired using a Bruker Avance 700 spectrometer operating at 700 MHz and at 298 K. Regular one-dimensional proton NMR spectra were obtained using the pulse program 1D NOESYPR1D with a relaxation delay of 2 s. Spectra were acquired with 512 scans and the obtained FID were processed using a line-broadening factor of 1 Hz and Fourier-transformed. Spectra were manually phased, baseline-corrected, and chemical-shift-referenced to TSP (δ = 0.00).
Analysis of NMR spectra
1D NMR spectra were subjected to multivariate data analysis using the program Metaboanalyst 5.0. The data were normalized and scaled using Autoscaling. Principal Component Analysis (PCA) was utilized for data overview and outlier detection. The data was modeled with the supervised method of Partial Least Squares -Discriminant Analysis (PLS-DA) using the same scaling and centering parameters.
The metabolites were identified and quantified using an untargeted profiling approach using Chenomx NMR Suite 8.3 Software (Chenomx, Inc., Edmonton, Alberta Canada). Six spectra from each experimental group were considered for quantification of extracellular metabolites and four spectra were used for intracellular metabolite quantification. The concentrations obtained from the spectra were plotted with GraphPad Prism (Version 5.01).
Glucose uptake assay
Glucose uptake was measured using a cell-based assay kit (#600470, Glucose uptake cell-based assay kit (Cayman Chemical, Ann Arbor, MI)). Briefly, cells were seeded (10000 cells/well) in a black, flat-bottomed 96-well microplate in triplicate. After 24 h, cells were treated with FGF16 alone or in combination with siGLUT3 or Apigenin, an inhibitor of glucose transport by GLUT1 (1:500) in glucose-free medium. Following 24-h incubation, cells were incubated with 100 μg/ml 2-NBDG (a fluorescently-labeled deoxy-glucose analog) for 30 min at 37 • C in a humidified atmosphere of 5% CO 2. The cells were washed with PBS and the fluorescent signal was measured using Varioskan Flash Spectral Scanning Multimode Reader (Thermo Scientific). The experiment was done in three independent sets.
Lactate assay
Lactate excreted by cell-line upon respective treatment/transfection in culture media was measured using a plate-based colorimetric assay kit (MET-5012, Cell Biolabs). Briefly, cells seeded in 96-well microplate were transfected with FGF16 alone or in combination with siPFKFB4 or shPFKP. Post-incubation of 24 h, exhausted media was collected and processed according to the manufacturer's protocol. Lactate concentration was normalized with respect to cell number.
Trypan Blue exclusion assay
A total of 1 × 10 5 cells/well were seeded in a 12-well plate. After 24 h, cells were treated with FGF16 alone or in combination with 2-DG or 5-FU. Next day, cells were counted before and after Trypan blue staining using haemocytometer. The experiment was performed in triplicate. Cell viability was calculated using the following formula: % viable cells = [1.00 − (Number of blue cells ÷ Number of total cells)] × 100
Syngeneic mouse model
Female Balb/c mice (6-8 weeks old) were injected into mammary fat-pad with 1 × 10 6 4T1 cells (stably transfected with FLAG/FLAG-FGF16 constructs) suspended in 0.1 ml of PBS. After 10 days, primary tumors were visible and the tumor volume was monitored and measured (twice a week) for next 20 days followed by Avertin mediated anesthesia of mice and their sacrifice by cervical dislocation. The tumors were excised, measured and collected in formalin prior to prepare paraffin blocks. All animals were treated in accordance with the guidelines of the Committee on the Care and Use of Laboratory Animals of the Institute of Animal Resources and Institutional Animal Care and Use Committee (IACUC) and the ethical approval was obtained from central animal house and research facility, Bose Institute. For each group, 5 mice were used; among which tumor of three representative images were given. Entire animal experiment was done at animal house, Bose Institute. Tumor volume was calculated by the modified ellipsoidal formula: V = 1 / 2 (Length × Width 2 ).
Immunohistochemistry (IHC)
Immunohistochemical analysis was done using 5 μm thick paraffin-embedded sections of tumor xenograft. At first, the slides were de-waxed by heating at 70 • C for 30 min followed by 5 min wash in xylene and rehydration with 100%, 90%, 70%, 50% ethanol up to pure distilled water. For heat-induced antigen retrieval, the sections were then kept in 10 mM citrate buffer (65 • C) followed by wash with PBS and blocking in 3% FBS in 1× PBS for 1 h. Staining was done with anti-Laminin (1:400), anti-Ki67 (1:500), anti-E-cadherin (1:400) and anti-Vimentin(1:100) antibodies diluted in 1XPBS containing 1% FBS for 2h . After proper washing, the slides were incubated with the secondary antibody (Alexa Fluor-488 conjugated; 1:500) for 1.5 h followed by washing with PBS-T and staining with DAPI. The images were captured using Leica TCS SP8 microscope and LAS X software.Haematoxylin-Eosin staining of the sections was performed by following the standard protocol by professional pathologists.
Wound healing assay
Cells at 70% confluency were treated/transfected as mentioned above and prior to that, scratching was carried out with a 10 μl pipette tip (mentioned as t = 0 h at the figures). Cells were washed several times with PBS to remove the detached ones and supplied with new growth medium. Photographs of the scratches were taken at 0 h and 24 h with 10× objective lens of inverted microscope (Olympus 1X 83) equipped with in-built camera and software (CellSens). The images of 5FU-treated set were captured using 20× objective of the microscope equipped with camera LEICA DFC450C and in-built software (Leica Application Suite).The length of wound for all the sets was measured by ImageJ software. The experiment was performed in duplicate.
In vitro invasion assay
Corning ® BioCoat™ Matrigel ® Invasion Chamber (Catalog No. 354481) was used to assess in vitro invasion by FGF16. Briefly, 1 × 10 5 cells were added in serum-free medium in the upper chamber and allowed to invade in presence of either 10% FBS or rhFGF16 alone or in combination with PD173074 in the lower chamber.
To check the effect of PFKFB4 and PFKP, cells were first transfected with respective siRNA/shRNA and next day were trypsinized, counted and seeded in upper chamber to allow the cells to invade in presence of 10% FBS. After incubating for 22 h at 37 • C in 5% CO 2 , the invaded cells on the lower surface were fixed in methanol, stained with crystal violet and counted under microscope (Leica Microsystems). The images were captured with 20× objective of the microscope equipped with camera LEICA DFC450C and in-built software (Leica Application Suite). The experiment was performed in duplicate.
Clinical analysis with human patient tissue samples
Primary malignant breast tumor and adjacent normal tissue specimens were collected from affected individuals with informed written consent. The study was approved by the Institutional Ethics Committee of Saroj Gupta Cancer Centre and Research Institute, under regulation of the Govt. of India (Registration No. ECR/250/Inst/WB/2013/RR-20). All tissue samples were stored in −80 • C till further experiments. Whole RNA was extracted from five paired and eight orphan samples following TRIzol method followed by cDNA synthesis and Q-PCR analysis. Relative change in gene expression was normalized with 18SrRNA level. IHC was performed following the above-mentioned protocol with anti-FGF16 antibody (1: 50). All clinical and histo-pathological information were collected simultaneously. Collection/processing of tissue samples and follow-up experiments were done in SGCC&RI, Kolkata. The clinicopathological features of each of the clinical samples are mentioned in Supplementary Table S6.
Statistical analysis
All data were calculated and analyzed using Microsoft excel/GraphPad Prism (Version 5.01) and expressed as mean + − SD of independent replicates of each experiment where + − SD was represented by error bars. Differences between treatments and controls were compared using one-way analysis of variance (ANOVA), followed by the post-hoc multiple comparisons-Tukey's Honest Significant Test (HSD) whenever applicable (>2 experimental conditions), or using a Student's t-test, otherwise (=2 experimental conditions). P<0.05 was considered as significant.
Data Availability
All primary data and constructs are freely available upon request to the corresponding author. The RNA-seq data will be deposited to the public repository during the process of revision. thank Dr. Kuladip Jana, Bose Institute for the assistance in performing in vivo experiment and Prof. Jayanta Mukhopadhyay, Bose Institute for his support in this project.
|
2023-05-25T06:17:09.989Z
|
2023-05-24T00:00:00.000
|
{
"year": 2023,
"sha1": "9d8c67740cb559f7f6205cf0e41346b80a0cb6df",
"oa_license": "CCBY",
"oa_url": "https://portlandpress.com/bioscirep/article-pdf/doi/10.1042/BSR20230677/946692/bsr-2023-0677-t.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "f2a00da5632363480f5f2f5cce752b0715764253",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
235633470
|
pes2o/s2orc
|
v3-fos-license
|
Impact of model resolution on the representation of the wind field along Nares Strait
Nares Strait is a major pathway along which multi-year sea ice leaves the Arctic, an ice class that has seen a recent dramatic reduction in extent. The winds that blow along the strait play an important role in modulating this ice export as well as in establishing the Arctic’s largest and most productive polynya, the North Water, that forms at its southern terminus. However, its remote location has limited our knowledge of the winds along the strait. Here we use automatic weather station data from Hans Island, in the middle of the strait, to assess the ability of a set of atmospheric renalyses and analyses with a common lineage but with varying horizontal resolution to represent the variability in the wind field. We find that the flow is highly bidirectional, consistent with topographic channeling, with the highest wind speeds from the north and that a model resolution of ~ 9 km is required to capture the observed variability. The wind field at Hans Island is also found to be representative of variability in the flow along much of Nares Strait.
In the vicinity of Robeson Channel, meteorological observations were made at Fort Conger from August 1881 to August 1883 during the First International Polar Year by members of the United States Expedition to Lady Franklin Bay 20 . These observations indicated the preference for bidirectional wind flow, either in the NNE or SSW direction, along Robeson Channel. There were numerous events, with both directionalities, where the wind speeds were in excess of 15 ms −120 .
Hans Island (80 o 49′35'N, 66°27′35'W), jointly claimed by both Canada and Denmark, is situated in the middle of the Kennedy Channel, one of the narrower sections of Nares Strait (Figs. 1 and 2). Since 2008, an automatic weather station (AWS) has been operating on this island 21 with data currently available from 2016 onwards. This paper represents the first use of this data to characterize the wind field along Nares Strait.
In this paper, a sequence of reanalyses and an operational analysis with horizontal resolutions ranging from ~ 60 to ~ 9 km, that all share a common model framework and data assimilation system, are used to characterize the variability in the wind field as observed by this weather station. The approach used in this paper represents an improvement over previous studies of the impact of model resolution on the representation of topographic flow distortion that have used reanalyses with different data assimilation systems and model parameterizations [22][23][24] . Please refer to the Methods Section for more information of the AWS data and the model suite used in this paper. Figure 3 presents the wind rose for the Hans Island AWS data. The tendency for bidirectional flow, in the NNE or SSW directions, is apparent as is the preference for northerly flow. This directionality is consistent with the orientation of the Kennedy Channel in the vicinity of Hans Island (Fig. 2). The mean wind speed is 5.8 ms −1 with a 99th percentile wind speed of 18.7 ms −1 . For northerly flow, the maximum wind speed was 29.4 ms −1 with a 99th percentile wind speed of 19.7 ms −1 . The corresponding values for southerly flow were 22.4 ms −1 and 13.9 ms −1 respectively.
Results
The narrowness and steep topography of Nares Strait plays an important role in the region's wind climate. Before comparing the wind observations from Hans Island with those from the models, we show in Fig. 4 www.nature.com/scientificreports/ 15 arc-second, here subsampled to 1 km, digital elevation model 25 ; the eERA5 with a resolution of ~ 60 km; the ERA5 with a horizontal resolution of ~ 30 km and the ECOA with a horizontal resolution of ~ 9 km. The GEBCO cross-section (Fig, 4a) clearly represents the narrowness of Nares Strait in the vicinity of Hans Island as well as the steep topography along both sides of the strait. The eERA5 cross-section (Fig. 4b) has a very modest representation of Nares Strait as a broad minimum in topography. The ERA5 cross-section (Fig. 4c) is able to resolve more details of the strait but, for example, there is no open water present and the height of the topography along the sides of the strait is underestimated. The ECOA cross-section (Fig. 4d) approaches the level of detail in the GEBCO cross-section but is still unable to fully resolve the steepness or the narrowness of the strait. Figure 5 provides scatterplots of the observed surface wind speed at the AWS with the 10 m wind speeds from the three model datasets. Statistics that provide quantitative measures of the goodness of the fit are shown. These include the root-mean-square (rms) and bias errors as well as the least-squares slope and the correlation coefficient r. A least-squares slope of 1 implies a one-to-one relationship exists between the observations and model output.
The results show that both the eERA5 (Fig. 5a) and the ERA5 (Fig. 5b) have a marked tendency to underestimate the wind speed at the AWS site with the ERA5 having a slightly better fit. For example, the rms error for eERA5 is 4.56 ms −1 with a least-squares slope of 0.44; while for ERA5 the corresponding values are 4.09 ms -1 and 0.55. The bias error for both of these datasets was ~ −3 ms −1 . By comparison, the ECOA (Fig. 5c) has an improved fit to the data with a rms error of 2.18 ms −1 and a bias error of 0.17 ms −1 as well as a reduced tendency to underestimate the wind speed at the AWS site indicated by the least-squares slope of 0.8. The correlation coefficient increases from 0.67 for the eERA5 to 0.79 for the ERA5 to 0.87 for the ECOA. www.nature.com/scientificreports/ As noted above, the flow in the vicinity of Hans Island has a pronounced bidirectionality. Figure 6 shows the scatterplots of the observed meridional and zonal components at the AWS with the corresponding components of the 10 m wind from the three model datasets. In general, the characteristics noted above with respect to the 10 m wind speed hold for the components as well. In particular, there is a reduction in the rms and bias errors as well as an increase in the least-squares slopes as one transitions from the lower resolution to higher resolution datasets. In addition, the fit to the zonal component is generally poorer than that for the meridional component, a characteristic that is most noticeable for the eERA5 and ERA5 datasets.
In Fig. 7 the wind rose at Hans Island from the three model data sets are compared to observations for the period of overlap. The characteristics are consistent with the results from the scatterplots. In particular, there is an increase in frequency of occurrence of high wind speeds as one transitions from the eERA5 to the ECOA. The above noted quantitively poorer fit in the zonal component in the eERA5 (Fig. 7b) and ERA5 (Fig. 7c) results in an inability of these models to capture the preference for NNE to SSW bidirectional flow. As compared to the observed wind rose (Fig. 7a), the ECOA wind rose (Fig. 7d) has a tendency to underestimate the magnitude of extreme wind speeds. There was also a slight discrepancy in the orientation of the winds in the ECOA as compared to observations. It is of interest to characterize the representativeness of the model wind field at Hans Island for flow along the entirety of Nares Strait. To accomplish this, one-point correlation maps 26 of the 10 m wind speed at Hans Island with the wind speed at other grid points in the Nares Strait region were calculated for each of the model datasets ( Fig. 8). Also shown is the 0.5 and 0.7 correlation coefficient contours that encompasses the area in which the variability in the 10 m wind speed at Hans Island can explain ~ 25% and ~ 50% of the variability in the 10 m wind www.nature.com/scientificreports/ speeds 27 . The 0.7 correlation coefficient contours were also fit to ellipses using a least-squares approach 28 so as to allow for a determination of their eccentricity and orientation 29 .
The results for all three model data sets indicate a high degree of correlation in the wind speed along Nares Strait with that at Hans Island. For eERA5, the region with correlation coefficients greater then 0.5 and 0.7 extend over both Ellesmere Island and Greenland resulting, for example, in an elliptical fit to the r = 0.7 contour that has an eccentricity of 0.79 (Fig. 8a). This characteristic of the spatial correlation of the wind field, i.e. the correlation over the regions of high topography adjoining Nares Strait, is reduced in ERA5 resulting in an eccentricity of 0.95 for the r = 0.7 contour (Fig. 8b). In contrast, the 0.7 contour for the ECOA is, for the most part, restricted to Nares Strait resulting in an eccentricity of 0.98 (Fig. 8c). For all three datasets, there is an asymmetry in the extent of the region of elevated correlation with the Hans Island 10 m wind speed. In particular, the region of elevated correlation extends farther to the south as compared to the north. Indeed for the ECOA, the 10 m wind speeds as far south as Smith Sounds are correlated with that at Hans Island, at the r = 0.5 level. www.nature.com/scientificreports/ To test the hypothesis that this asymmetry is the result of the preference for northerly flow at Hans Island, the one-point correlation maps were recalculated for times in which either northerly or southerly flow were observed at Hans Island. Results are shown in Fig. 9 for the ECOA dataset. The one-point correlation map for northerly flow (Fig. 9a) is very similar to that when there is no preference to the wind direction (Fig. 8c). In contrast, the one-point correlation map for southerly flow (Fig. 9b) indicates that the region of elevated correlation extends over the southern Lincoln Sea as well as adjacent regions of north Greenland.
We conclude the presentation of results with composites of the environmental conditions associated with extreme northerly and southerly flow events at Hans Island. This was accomplished by identifying the times when the observed wind speed at Hans Island exceeded the 99th percentile thresholds for northerly, 19.7 ms −1 , and southerly, 13.9 ms −1 winds. For each event, only the time at which the wind speed was a maximum was included. As a result, there were 16 distinct northerly events and 15 distinct southerly events. For these times, the sea-level pressure, 10 m wind speed and 10 m wind fields from the 3 model datasets were averaged resulting in composites shown in Figs. 10 and 11.
The composite extreme northerly wind event at Hans Island (Fig. 10) is associated with higher sea-level pressures over the Lincoln Sea and lower sea-level pressures over northern Baffin Bay. The higher wind speeds are focused in the region to the south of Hans Island and generally increase from the eERA5 (Fig. 10a) to the ERA5 (Fig. 10b) to the ECOA (10c). In a similar vein, there is an increase in the localized gradients in sea-level pressure along the strait as well as over northern Baffin Bay as model resolution increases.
In contrast, the composite extreme southerly wind event at Hans Island (Fig. 11) is associated with a reversal in the meridional pressure gradient between the Lincoln Sea and northern Baffin Bay. In this case, the highest wind speeds tended to be to the north of Hans Island with a similar increase in wind speed and higher gradients in sea-level pressure with resolution. There was however also evidence of elevated winds over the southern Kane Basin in both the ERA5 (Fig. 11b) and the ECOA (Fig. 11c) that were absent in the eERA5 (Fig. 11a).
Discussion
The Nares Strait wind field plays an important role in modulating the export of thick multi-year sea ice from the Last Ice Area situated to its north as well as in maintaining the NOW located at its southern end 1,9 . There has been a significant loss of multi-year ice from the Arctic in recent years 14 as well as an increase in ice export along Nares Strait 7 and there is also evidence that the NOW is increasing in area 10 . The remoteness of the Nares Strait region has limited our ability to characterize the wind field thereby leading to uncertainty in its role in the www.nature.com/scientificreports/ changes to the ice that are now being observed in the region. The steep and narrow nature of the strait (Fig. 1) leads to high winds arising from interactions with the topography along its coast that remain a challenge for models to capture 1,18 . In this paper, we have used wind observations from an automatic weather station situated on Hans Island, in the middle of Nares Strait (Figs. 1, 2), to document the variability in wind field along the strait and the ability of numerical models to represent this variability. The Hans Island wind observations confirm that the flow along the strait is bidirectional, with preferred directions from the NNE or SSW and that northerly flow is most common (Fig. 3). There is evidence that extreme events with either directionality occur with wind speeds in excess of 20 ms −1 . One such northerly wind event destroyed an ice camp that had been established along the Kennedy Channel in April 2005 17 .
To assess the role that horizontal resolution plays in the ability of models to represent the variability in the wind observed at Hans Island, three different versions of the ECMWF's IFS were considered with resolutions of ~ 60 km (eERA5), ~ 30 km (ERA5) and ~ 9 km (ECOA). The common lineage of the models allows one to control for biases arising from difference in model architecture and parameterizations 22-24 . www.nature.com/scientificreports/ The comparison of cross-sections through the reference and model topographies (Fig. 4) clearly shows the improvement in representation of the narrowness and steepness of the strait in the vicinity of Hand Island in the ECOA as compared to the eERA5 and ERA5.
The comparison for the 10 m wind speed (Fig. 5) shows a measurable improvement in the representation of the observations with increasing model resolution with the largest jump occurring between the ERA5 (resolution ~ 30 km) and the ECOA (resolution ~ 9 km). For example, the rms error decreased from 4.09 ms −1 for the ERA5 to 2.18 ms −1 for the ECOA; while the slope of the least squares linear fit increased from 0.39 for the ERA5 to 0.8 for the ECOA. The latter characteristic implies that the eERA5 and the ERA5 have a tendency to underestimate the wind speed during high wind speed events to a greater extent than the ECOA. www.nature.com/scientificreports/ This may be related to the inability of these models to represent the open water in Nares Strait and its replacement with land that generally has higher surface roughness (Fig. 4). Nevertheless, all three versions of the IFS have relatively high correlation coefficients, varying from 0.67 for the eERA5 to 0.87 for the ECOA, suggesting that a degree of linearity exists between model and observations.
The comparison with the components of the wind (Fig. 6) shows the same improvement with increasing resolution with the fit to the meridional component being systematically better than the zonal component for the eERA5 and ERA5. To first order, the winds along Nares Strait are the result of the difference in sea-level pressure between the Lincoln Sea and northern Baffin Bay 1 . All the models, to a greater or lesser extent, represent the narrowness of the strait and hence the tendency for the flow to be ageostrophic and controlled by the pressure gradient in the along-strait, i.e. meridional, direction. The composite high speed events (Figs. 10 and 11) clearly show that the magnitude of these along-strait gradients are a function of resolution. In contrast, the zonal component of the wind along Nares Strait is constrained by a model's ability to represent the gradient in topography, that is largest in the zonal direction (Figs. 1 and 4). The poorer fit of this component of the wind for the eERA5 and ERA5 is associated with the inability of these models to represent this gradient. The comparisons of model wind roses with observations ( Fig. 7) are consistent with the above discussion and in particular highlight the tendency for the lower resolution eERA5 and ERA5 datasets (Fig. 7b, b) to have the winds aligned in the meridional direction as compared to the NNE to SSW direction in the ECOA dataset (Fig. 7d) and in the AWS data (Fig. 7a).
The one-point correlation maps of the 10 m wind field (Fig. 8) confirm this characteristic in that both the eERA5 and ERA5 have regions of high correlation that spill out over the topography along the strait. In contrast for the ECOA, the region of high correlation is restricted, for the most part, to Nares Strait. The eccentricities for the elliptical fits to the r = 0.7 contour, that range from 0.79 for the eERA5 to 0.98 for the ECOA provide a quantitative confirmation of this characteristic. The one-point correlation maps for the ECOA 10 m wind (Figs. 8c and 9) confirm that the Hans Island observations are representative of variability in the wind field along Nares Strait and, for southerly flow, over the southern Lincoln Sea as well.
The structure of the composite extreme wind speed events (Figs. 10 and 11) are consistent with the other results of this study and confirm the requirement of high horizontal resolution to resolve the impact that the topography of the Nares Strait region has on the wind field. In particular, high model resolution is required to resolve the localized large spatial gradients in the along strait sea-level pressure field that is responsible for the ageostrophic nature of the flow. Smith Sound as well as the Kennedy and Robeson Channels are regions where these large gradients occur. Both Smith Sound and the Kennedy Channel are regions where high speed northerly and southerly wind events occur. This confirms historical observations made during the nineteenth century 16 www.nature.com/scientificreports/ The results of this study suggest that a model resolution of at least 9 km is required to represent the wind field along Nares Strait providing a confirmation of earlier work 18 in which a 6 km limited area numerical weather prediction model was used to develop a 2-year climatology of the regional wind field as well as the study 30 that used the 9 km ECOA data to represent air-sea interaction over the NOW. It also suggests that climatologies 9,31,32 and ice/ocean models [33][34][35] of Nares Strait that are based on or forced by atmospheric datasets with horizontal resolutions coarser than 9 km may underestimate wind-driven processes active in the region. In particular, the coarser resolution datasets may not represent the narrowness and steepness of the strait leading to an underestimation of the along-strait pressure gradients responsible for the ageostrophic nature of the flow 2 . In addition, the coarser resolution datasets may not be able to resolve the open water of the strait replacing it with land ( Fig. 4) www.nature.com/scientificreports/ this study, we will use the wind speed and direction data for the period September 2 2014-September 30 2020 with the exception of the 3 month period with bad wind direction data during 2016. The data was subsampled to a 6-hourly interval providing data at 00, 06, 12 and 18 GMT. In this paper, we will make use of 3 model datasets based on the ECMWF's Integrated Forecast System (IFS). Included is the new fifth generation reanalysis from the ECMWF or ERA5 with a horizontal resolution of ~ 30 km, as well as its ensemble version, eERA5 with a horizontal resolution of ~ 60km 36 and the current version of their operational analysis or ECOA, with a horizontal resolution of ~ 9km 37 . The eERA5 ensemble consists of an unperturbed member and 9 perturbed members. For this work, we made use of the unperturbed member of the ensemble. For this study, we used the model data for the period 2016-2020. The eERA5 and ERA5 datasets are based on Cycle 41r2 of the IFS. Being an operational product, the ECOA data is based on a number of different cycles of the IFS from Cycle 41r2 up to Cycle47r1. No material changes to the IFS, that would impact the present study, occurred over the period 2016-2020. All three datasets use the same 4Dvar data assimilation system 38 , as well as same number of levels in the vertical, 137 up to 1 mb 36,39 .
For this study, the instantaneous 6-hourly 10 m winds and 100 m from these three datasets were interpolated to the location of Hans Island and the data was used at those times for which there are observations. A number of different interpolation schemes were used and the results were independent of scheme. For simplicity, the results shown are for a bilinear interpolation scheme. In total, ~ 550 days worth of data was included in this study. The AWS was situated at a height of approximately 170 m above sea level 21 .
|
2021-06-26T06:17:15.003Z
|
2021-06-24T00:00:00.000
|
{
"year": 2021,
"sha1": "b321ebeb8683e0f8af0457ae6e49ddac23e327c9",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s41598-021-92813-9.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "a8fec0284b34af1308b8709c6f9c08ad4d15370a",
"s2fieldsofstudy": [
"Environmental Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
255853597
|
pes2o/s2orc
|
v3-fos-license
|
Understanding cell signaling in cancer stem cells for targeted therapy – can phosphoproteomics help to reveal the secrets?
Cancer represents heterogeneous and aberrantly proliferative manifestations composed of (epi)genetically and phenotypically distinct cells with a common clonal origin. Cancer stem cells (CSC) make up a rare subpopulation with the remarkable capacity to initiate, propagate and spread a malignant disease. Furthermore, CSC show increased therapy resistance, thereby contributing to disease relapse. Elimination of CSC, therefore, is a crucial aim to design efficacious treatments for long-term survival of cancer patients. In this article, we highlight the nature of CSC and propose that phosphoproteomics based on unbiased high-performance liquid chromatography-mass spectrometry provides a powerful tool to decipher the molecular CSC programs. Detailed knowledge about the regulation of signaling processes in CSC is a prerequisite for the development of patient-tailored multi-modal treatments including the elimination of rare CSC. Phosphorylation is a crucial post-translational modification regulating a plethora of both intra- and intercellular communication processes in normal and malignant cells. Small-molecule targeting of kinases has proven successful in the therapy, but the high rates of relapse and failure to stem malignant spread suggest that these kinase inhibitors largely spare CSC. Studying the kinetics of global phosphorylation patterns in an unbiased manner is, therefore, required to improve strategies and successful treatments within multi-modal therapeutic regimens by targeting the malignant behavior of CSC. The phosphoproteome comprises all phosphoproteins within a cell population that can be analyzed by phosphoproteomics, allowing the investigation of thousands of phosphorylation events. One major aspect is the perception of events underlying the activation and deactivation of kinases and phosphatases in oncogenic signaling pathways. Thus, not only can this tool be harnessed to better understand cellular processes such as those controlling CSC, but also applied to identify novel drug targets for targeted anti-CSC therapy. State-of-the-art phosphoproteomics approaches focusing on single cell analysis have the potential to better understand oncogenic signaling in heterogeneous cell populations including rare, yet highly malignant CSC. By eliminating the influence of heterogeneity of populations, single-cell studies will reveal novel insights also into the inter- and intratumoral communication processes controlling malignant CSC and disease progression, laying the basis for improved rational combination treatments.
Background
Cancer is caused by the accumulation of genetic and epigenetic changes that eventually account for the unrestricted proliferative and metastatic capacity of malignant cells [1,2]. Despite of having a common cellular and genetic ancestor, deep genome sequencing of cancer cells together with histopathological and molecular marker analyses revealed a surprising heterogeneity of cancer cells within the tumor mass. Following a Darwinian selection scheme, clonal evolution results in dynamic changes of subclones, which can account for disease progression and drug resistance in response to therapy [3][4][5]. Notably, the malignant capacity of clonal cancer cells differs considerably in terms of tumor initiation, propagation, metastatic spread and therapy resistance. In most -if not all malignancies -these highly aggressive traits can be ascribed to the presence of rare and self-renewing cancer cells. Since this rare subpopulation displays several stem-like cell characteristics and is likely to derive from long-lived tissue stem cells, these cells are commonly -but not exclusively -referred to as cancer stem cells [6,7]. The terminology for selfrenewing cancer cells with tumor initiating and maintaining properties is diverse, controversial, contextdependent and research-field specific. Here, we will use the term cancer stem cells (CSC) for rare self-renewing malignant tumor cells that have the ability to initiate, maintain and propagate heterogeneous malignancies (for details about the terminology and nomenclature of CSC see [8]).
Cancer stem cells and tumor heterogeneity
The hierarchical CSC model of malignant development and growth is a result of numerous recent genetic, cellular and molecular analyses of cancer heterogeneity (see below). However, the first evidence pointing to the existence of stem-like tumorigenic cells dates back at least several decades. Kleinsmith and Pierse demonstrated in 1964 that single embryonal carcinoma cells within a teratocarcinoma can give rise to multiple cellular lineages [9]. By performing 1700 single cell grafts, of which 43 formed teratocarcinomas composed of at least 14 different somatic tissues, this study provided experimental support for the stem cell theory of cancer. The basic concept of this model, however, has already been hypothesized in 1907 by Max Askanazy, a Prussian pathologist, who speculated that based on histological similarities between tumors and embryonic tissues, cancer arises from cells with properties similar to those of the early embryo [10]. Much has changed since then from both a technical and mechanistic point of view, but the basic concept of tumors arising from undifferentiated stem-like cells has recently been supported for many cancer entities, using sophisticated and state-of-the-art transplantation and genetic tools. Together, these seminal studies (for elaborate reviews see [6,11,12]) have led to a hierarchical rather than stochastic model of malignant development and growth driven by self-renewing cancer stem cells (Fig. 1).
The first evidence for clonal and stem cell-derived development of malignancies in man came from a study with patients suffering from chronic myeloid leukemia (CML). In 1967, Fialkow et al. investigated females heterozygous for the X-linked glucose-6-phosphate dehydrogenase (G-6-PD), resulting in the expression of only one of the two enzyme types in a single cell. By analyzing the blood cells of three female heterozygous G-6-PD patients, the team found exclusive expression of only one allele of G-6-PD in all CML cells of a patient, suggesting that the malignancy arose from a single hematopoietic stem cell [13].
Nearly 20 years later, the existence and phenotypic characterization of leukemia initiating CSC was reported Fig. 1 Scheme of the hierarchical stem cell model in healthy and malignant tissue. a Asymmetric cell division of a stem cell (SC; depicted as dark blue cells) in normal tissue results in the generation of a daughter stem cell as well as committed and dividing progenitor cells that can give rise to terminally differentiated cells (shown as light blue cells) of the given tissue. b Genetic and/or epigenetic alterations can transform stem cells and/or progenitor cells, leading to the escape from intracellular and extracellular control mechanisms that restrain aberrant cell proliferation and uncontrolled tissue growth. Constant self-renewal and the production of heterogeneous malignant progeny is considered a hall mark of cancer stem cells (CSC). The CSC model in malignant tissue represents a hierarchical organization, where rare self-renewing and long-lived CSC give rise to the tumor mass consisting of heterogeneous cancer cells with variable degree of differentiation and proliferative capacity (orange cells). CSC are more resistant to radiation-and chemotherapy calling for targeted approaches that eliminate CSC in multi-modal treatment strategies [134] by Bonnet and Dick for acute myeloid leukemia (AML) [14]. The authors found that only the rare CD34 + CD38subpopulation of undifferentiated leukemic cells possesses self-renewing and leukemia initiating capacity. Since this study was based on engraftment experiments in immunocompromised NOD/SCID mice, the leukemia initiating cells were termed SCID leukemia-initiating cells (SL-IC). Although the first concepts of the hierarchical CSC model were based on studies of leukemic malignancies (reviewed in [15]), multiple evidence has been provided since for the existence of CSC in numerous solid tumors. The first report of CSC in a solid malignancy came from studies of primary breast cancer samples. Al-Haji et al identified rare, undifferentiated CD44 + /CD24 -/low cells as highly tumorigenic [16]. In this study, the authors demonstrated that as few as 100 CD44 + /CD24 -/low cells were sufficient to initiate the growth of tumors that could be serially passaged, each time giving rise to heterogeneous tumors comprising rare self-renewing CD44 + /CD24 -/low CSC and abundant non-tumorigenic cells.
During the past years, numerous reports have identified and confirmed the existence of rare CSC in the majority of human malignancies including cancers of the brain, the gastro-intestinal tract, skin and many other tissues [16][17][18][19][20][21]. Notably, CSC not only account for tumor initiation, growth and relapse in settings of minimal residual disease, dormancy and therapy resistance [22][23][24][25], but also are able to trans-differentiate for instance, into endothelial cells, thereby contributing to the tumor vasculature and malignant growth of glioblastoma [26]. As for the molecular determinants of CSC fate, it could be shown that the expression of a particular combination of transcription factors can reprogram non-CSC into CSC-like cells, analogous to the reprogramming and induction of pluripotent stem cells. In a glioblastoma model, expression of four factors, POU3F2, SOX2, SALL2 and OLIG2 in non-CSC is sufficient for the reprogramming of stem-like tumorpropagating cells (TPCs) with an epigenetic landscape comparable to the proper CSC population [27].
The notion that CSC are likely to derive from longlived tissue stem cells has been intensely studied in transgenic mouse models suitable for genetic labeling of stem cells and lineage tracing of stem cell progeny in a defined genetic setting including selected cancer driver mutations (for review see [28]). Such studies revealed, for instance, rare Lgr5-positive intestinal crypt stem cells with hyperactive Wnt signaling as those cells that fuel the growth of intestinal adenomas. Like wild-type intestinal stem cells, Lgr5 positive adenoma stem cells reside in the bottom of the crypt niche, where they generate aberrantly proliferating Lgr5-negative adenoma cells that build the tumor mass [29,30]. In line with a crucial role in fueling tumor growth, selective depletion of intestinal CSC resulted in rapid tumor regression, demonstrating the therapeutic potential of direct CSC targeting, although the relevance of these findings to human pathology and therapeutic relevance still remains to be addressed in detail [31] (for a general concept of CSC targeting see Fig. 2).
Deciphering the phosphoproteome of CSC for the development of anti-CSC therapies The highly malignant nature of CSC together with their pivotal role in disease relapse calls for a detailed and comprehensive understanding of the molecular processes regulating CSC behavior. Since kinases frequently represent the major effectors of oncogenic signals that can be efficiently targeted by small molecule drugs, we propose that the in-depth analysis of the phosphoproteome of CSC in combination with functional assays will allow the identification of kinases that determine the malignant phenotype of CSC. We consider this knowledge as essential prerequisite for the design of efficient combination treatments to eradicate CSC. If embedded in multimodal treatment regimens including immunotherapy, anti-CSC strategies are likely to significantly improve the overall survival of cancer patients by reducing malignant growth, metastatic spread, therapy resistance, and relapse rates.
The detailed and comprehensive analysis of rare CSC by -omics methods is a challenging endeavor, since CSC represent only a rare subpopulation of the tumor mass, posing severe constraints on the number of cells available for downstream investigations. The instrumental setup for the analysis of minute samples, therefore, has to be of sufficient sensitivity, particularly if it comes to technologies such as phosphoproteomics, where only a fraction of the respective protein molecules display posttranslational phosphorylation marks. Aside from the technological challenges, the lack of universal and unambiguous CSC markers suitable for CSC isolation needs to be taken into account for the design of the isolation procedure.
Enrichment of rare CSC by their characteristic immunophenotype distinguishing CSC from non-CSC cells of the tumor bulk has been widely used and successfully applied. However, the choice and combination of surface epitopes is often specific only to a particular malignant entity and can result in the partial isolation of characteristic subpopulations of CSC [32,33].
As an alternative, the increased activity of aldehyde dehydrogenase (ALDH) and certain efflux pumps in CSC allows to distinguish CSC from non-CSC. Increased ALDH activity can be translated biochemically into the generation of fluorescent signals. ALDH-positive cells can then readily be quantified and isolated by flow-cytometry and fluorescence-activated cell sorting, respectively. One of the first studies applying this strategy identified a rare ALDH-positive subpopulation of breast cancer cells with pronounced tumor-initiating potential consistent with ALDH-positive cells with CSC characteristics [34]. In addition, high level expression of ATPbinding cassette transporter proteins endows CSC with an efficient efflux detoxification machinery. Therefore, incubation of cancer cells with a cell permeable fluorescent dye such as HOECHST 33342 results in rapid and quantitative efflux of the dye in CSC while non-CSC retain a high intracellular concentration of HOECHST 33342. When analyzed by flow cytometry, CSC appear as dim population referred to as side population. Consistent with the dim side population being rich in CSC, HOECHST 33342 dim but not bright cells display high tumor initiating capacity [35][36][37].
CSC enrichment based on the differential immunophenotype or enzymatic activity of CSC and non-CSC is frequently applied and well established for a variety of cancer entities. However, none of these methods allows the selective expansion of CSC to readily increase CSC numbers to levels sufficient for unbiased global phosphoproteomics approaches. Compared to non-CSC, CSC have a much higher intrinsic capacity for clonal growth when cultured under specific in vitro conditions. For instance, growth of pancreatic cancer cells in 3dimensional matrix cultures results in the formation of large, tumor-initiating spheres highly enriched for CSC [38,39]. The clonogenic growth properties of CSC can therefore be used for the selective expansion of tumorinitiating CSC yielding cell numbers sufficient for elaborate phosphoproteomics studies.
Phosphoproteome analysis of cancer and cancer stem cells
The role of protein phosphorylation in the control of cellular behavior has been well appreciated and intensely studied for many years. Phosphorylation serves as one of the most important post translational modifications (PTMs) of proteins to operate and reversibly control signaling [40]. Since phosphorylation is known to affect processes such as cellular growth, cell division, and metabolism, a dysfunction in protein phosphorylation can promote the development of various diseases such as cancer. Kinases catalyze the phosphorylation of serine, threonine or tyrosine residues within proteins using ATP as substrate. The requirement of precise control of kinase activity for the integrity of an entire tissue or even organism becomes evident by the fact that genetic alterations in kinase signaling pathways are frequently associated with the development and growth of cancer Cancer Stem Cells (CSC) display enhanced chemoresistance and account for metastases and disease relapse. A tumor typically consists of a minority of CSC, which give rise to more differentiated cancer cells. These differentiated tumor cells represent the majority of cells in the primary tumor, but have a limited self-renewal capacity. Untargeted therapy (e.g. chemotherapy) mainly affects highly proliferating non-CSC. Therapy resistant CSC are spared and can subsequently lead to tumor regrowth and therapy resistance in the initially responding patient (middle panel). Anti-CSC therapy prior to or together with untargeted therapy would hinder the tumors ability to regrow (right panel). Cancer cells with CSC properties can leave the primary tumor via blood or lymphatic vessels and form metastases in distant organs (lower left panel) [41][42][43][44]. Therefore, a detailed and comprehensive knowledge of the phosphoproteome landscape of CSC is an important prerequisite for the design of efficient targeted therapies selectively blocking aberrantly active kinases and the malignant traits of CSC, respectively.
Phosphoproteome analysis or phosphoproteomics is a comprehensive technique analyzing the phosphoproteome of cells in a particular cellular state and biological context. The phosphoproteome comprises all phosphoproteins within a cell population or a single cell. According to Aebersold and Goodlett, phosphoproteomics tries to reveal the "trinity of protein phosphorylation analysis", which is "identification of the site of phosphorylation, identification of the kinase responsible for the phosphorylation, and identification of the function and role of this phosphorylation" [45]. In the past, twodimensional gel electrophoresis (2-DE) has been the dominant analysis technique for analyzing the phosphoproteome. 2-DE fractionates intact and undigested proteins by separation of the proteins by charge and molecular mass in two separate dimensions [46]. In particular, Phos-tag containing gels were developed, which enhance the separation of phosphoproteins through incorporation of Mn 2+ or Zn 2+ ions into the gel, for selective separation of phosphoproteins in SDS-PAGE gels. Followed by immunoblotting, a map of phosphorylated proteins can be created enabling the profiling of kinase activity in vitro [47].
While 2-DE has represented the golden standard for comprehensive proteome analysis for many years [48],the more generic nature of high-resolution tandem mass spectrometry coupled to one-or multidimensional high-performance liquid chromatography (HPLC-MS/ MS) [49] meanwhile has superseded the 2-DE technique. In the so called "shotgun (phospho) proteomics" approach, extracted proteins of a cell population are first digested by a specific protease before being subjected to HPLC-MS/MS for separation and detection. The breakthrough in the technical development, which enabled the use of HPLC-MS/MS as a comprehensive revelation engine for proteins and peptides, was the invention of soft ionization techniques such as ESI (electrospray ionization) [50], which enable the direct mass spectrometric analysis of biological samples from liquid, often aqueous solutions. Nowadays, mass spectrometry is the primary identification and quantification tool for comprehensive phosphoproteomics [51,52]. Moreover, identification technologies based upon the gas-phase fragmentation of peptide ions [53] and the matching of the resulting set of fragment ions with protein sequence databases [54][55][56] have laid the ground for the highthroughput identification and quantification of proteins in proteomic samples, enabling the analysis of more than 10,000 proteins in a single 12-day experiment [57].
HPLC-MS/MS workflow for phosphoproteomics
A typical experimental design of a phosphoproteomics study first involves the isolation of the phosphoproteins, which is done by cell lysis in a lysis buffer assuring phosphatase and protease inhibition. After a complex sample preparation procedure of denaturation, reduction, and alkylation, the isolated proteins are digested into peptides. This is normally done by using proteases such as trypsin, chymotrypsin, or LysC, which provide peptides of a size highly suitable for mass spectrometric investigation [58]. Combinatorial approaches that complement trypsin by multiple proteases help to overcome the drawback of tryptic digestion, which often results in missing particular cleavage sites, particularly in the case of phosphorylation or other post translational modifications [59].
In contrast to the sample preparation in proteomics, the workflow for phosphoproteomics has to be expanded by procedures for phosphopeptide enrichment. Since the complexity of the cellular proteome hinders the direct analysis of phosphopeptides that are usually present in concentrations much lower than their nonphosphorylated analogues, further fractionation and phosphopeptide enrichment is needed to investigate the phosphoproteome. Different enrichment and fractionation methods have been applied, which were recently reviewed [51]. Typically, enrichment strategies rely on affinity chromatography taking advantage of the phosphate-specific binding abilities of certain metal oxides [60] (titanium dioxide, tin oxide [61]) or of immobilized metal ions such as Fe 3+ [62] or Ga 3+ [63]. The corresponding chromatographic modes have been termed metal oxide affinity chromatography (MOAC) or immobilized metal affinity chromatography (IMAC).
Since the detection of phosphorylated tyrosines is superimposed by the higher abundant serine and threonine phosphorylations in conventional shotgun phosphoproteomics approaches, immunoprecipitation based on phosphotyrosine antibodies has been implemented as an alternative enrichment strategy. Thus, the targeted enrichment of phosphorylated tyrosines prior to HPLC-MS detection improves the coverage of the phosphoproteome, especially when focusing on tyrosine phosphorylation by tyrosine kinases [51,64,65].
Furthermore, multidimensional chromatographic separations are usually applied for extensive fractionation of (phospho-) peptides [66]. Thereby the sample complexity is reduced and the instrument sensitivity is increased. Since peptides may contain both acidic and basic side chains, they can bear, depending on the pH of the solution, a positive or negative net charge, making them amenable to both cation-and anion-exchange chromatography [67]. Moreover, phosphorylation introduces a negative charge, thereby increasing the negative or decreasing the positive charge of a peptide, usually also resulting in a more hydrophilic nature of the phosphopeptides. Therefore, hydrophilic chromatographic separation techniques or combinations of charge-based/ hydrophilic interaction modes are applicable [68].
The most commonly applied methods for separation in a first dimension are strong cation exchange chromatography (SCX) [69][70][71] or reversed-phase HPLC at high pH [72] besides electrophilic repulsion chromatography (ERLIC) [73] or hydrophilic interaction chromatography (HILIC) [74]. This first dimension is usually combined with a final (ion-pair) reversed-phase (IP-RP) separation before mass spectrometric detection via high-resolution mass spectrometry (HRMS) [51]. Offering the advantage of very high resolution and mass accuracy, highresolution hybrid mass spectrometers such as quadrupole-time-of-flight (Q-TOF) [75], linear ion trap-Orbitrap (LTQ-Orbitrap), or quadrupole-Orbitrap (Q-Orbitrap) instruments [76] are the first choice in largescale phosphoproteomics approaches. These instruments provide full scan spectra of intact peptides as well as fragment spectra of selected peptide precursor ions, which are then compared with databases for peptide identification by means of suitable computational tools [55,56,77]. Advantages and disadvantages of instruments have been reviewed elsewhere [51,78]. A short summary of a typical phosphoproteomics workflow is shown in Fig. 3.
Challenges of analyzing the phosphoproteome
Phosphoproteins and -peptides are bringing about special instrumental and sample preparation challenges. The availability of relatively high amounts of sample required for untargeted phosphoproteome analysis, typically in the range of 100 μg [68] to several milligrams of protein, may be problematic, especially when trying to analyze human material from biopsies [79]. This limitation may be overcome by employing targeted analysis by means of highly sensitive, mass spectrometry-based selected-or multiple reaction monitoring (SRM or MRM) methods [80]. Furthermore, Sequential Window Acquisition of all Theoretical Fragment Ion Mass Spectra (SWATH-MS) is evolving as a highly efficient global (phospho) proteome quantification strategy [51] and might facilitate the incorporation of tissue samples into SWATH-MS proteome maps similar to biobanks [81]. Microfluidic approaches to single-cell phosphoprotein analysis in a clinical context will be discussed in a separate section below.
Due to the sub-stoichiometric nature of protein phosphorylation, special sample preparation and phosphopeptide enrichment steps are required, as mentioned above. Besides, phosphopeptides provide significant difficulties for the mass spectrometric analysis [52,82]. Phosphopeptides show lower ionization efficiencies in positive ionization mode due to ion suppression compared to non-phosphorylated peptides [83]. In addition, in the case of phosphoproteins, the labile phosphoryl group can be easily lost during fragmentation. This leads to an incident called neutral loss of 98 Da, which usually produces a dominant fragment ion and has to be considered for the identification of the peptides. Different fragmentation techniques have been applied and combined to improve the phosphopeptide identification such as collision-induced dissociation (CID) [84], higher-energy collision-induced dissociation (HCD), and electrontransfer dissociation (ETD) [85], but until now there is no universally applicable technique [86]. Furthermore, it Fig. 3 Typical phosphoproteomics workflow. Each step in a phosphoproteomic experiment can contribute to limitations in reproducibility and phosphoproteomic depth, which can ultimately restrict the biological insight obtained from an experiment. Concerted efforts in the phosphoproteomics community to improve each step in this workflow continue to advance our ability to sample the phosphoproteome with greater speed and depth, but comprehensive phosphoproteome coverage remains out of reach. Reproduced from [51] with permission of ACS Publications © 2015 is important to localize the phosphorylation to the corresponding amino acid residue. This phosphosite localization can be even more important and challenging than the peptide identification itself craving for an appropriate algorithm [87].
Moreover, tyrosine phosphorylation occurs 100-1000 times less than Ser/Thr phosphorylation, which requires phosphotyrosine-specific enrichment strategies as described above [51]. The study of tyrosine phosphorylation is important in unraveling signaling mechanisms connected to malignancies such as cancer, especially because the majority of FDA approved kinase inhibitors applied in tumor therapy target tyrosine kinases [88,89].
In addition to the requirement of sophisticated and state-of-the-art technologies, also the dynamic nature of phosphorylation requires careful avoidance of enzymatic or chemical dephosphorylation by means of phosphatase inhibitors, making the analysis a challenging task [90]. Phosphorylation events are time dependent and thus phosphoproteomics can only provide a snapshot of the particular condition.
Quantification of changes in phosphoproteome regulation
Quantification is essential to reveal changes in the phosphoproteome. It enlightens the proteins, which are significantly regulated in the particular experimental conditions in response to e.g. a defined treatment, and helps to resolve signaling networks. There are different quantification strategies applied for phosphoproteomic approaches mainly including isotope-labeling and labelfree methods. The most dominant techniques focus on labeling of peptides or proteins mostly with stable isotopes, which can be readily distinguished by mass spectrometry [91]. Stable isotope labeling by amino acids in cell culture (SILAC) is a very common metabolic in vivo labeling method before proteolytic digestion [92]. Thereby, during protein biosynthesis the cells incorporate isotope-labeled amino acids provided in the cell culture medium or in the feed for animal models.
Peptides can also be isotope-labeled during the tryptic digestion of proteins through incorporation of heavy oxygen from H 2 18 O. Moreover, reductive dimethylation labeling using regular or deuterium-labeled forms of formaldehyde and sodium cyanoborohydride is an efficient post digestion labeling method applied for full MS quantification by comparing extracted ion chromatogram peak areas corresponding to the differently isotopelabeled peptide species [93]. Isotope-labeling of proteins or peptides can also be performed upon chemical derivatization with isotope-labeled, mostly amino-or thiolreactive agents such as Isotope-coded Affinity Tags (ICAT) [94] or Isotope-Coded Protein Labels (ICPL) [95].
Finally, tags like isobaric tags for relative and absolute quantitation (iTRAQ) [96], or tandem mass tags (TMT) [97] can be used to quantify phosphopeptides by tandem mass spectrometry [98]. Here, quantification is enabled by tandem mass spectrometry (MS/MS) after fragmentation of the phosphopeptide upon generating reporter ions to obtain ratios between controls and treatments. One of the major advantages of isobaric labeling is the economization of measurement time and expense by merging of multiple (up to ten) samples. A more time consuming but attractive method for quantitative phosphoproteomics is label-free quantification of peptide signals in independent HPLC-MS/MS analyses [99,100]. This is especially interesting for phosphoproteomics, since it does not require any further labeling and thus saves costs and prevents interferences with the phosphate group of the peptides. Nevertheless, there is a strong requirement for careful experimental design and/ or normalization strategies in order to obtain comparable signal intensities [101].
Phosphoproteomic applications for the analysis of cancer cells
HPLC-MS/MS based phosphoproteomics represents a discovery driven approach, which can help to track new drug targets and illuminate up-and downstream signaling molecules. Furthermore, phosphoproteomics can help to give new insights into phosphorylation networks and kinase-substrate interactions.
The human epidermal growth factor (HER) family of receptor tyrosine kinases was one of the first targets, which was addressed by phosphoproteomic analysis. The first studies analyzed changes in phosphorylation focusing on the analysis of phosphoproteins after enrichment by phosphotyrosine antibodies to examine the effect of epidermal growth factor (EGF) stimulation [102]. Although these studies provided insight into activation profiles of key proteins involved in epidermal growth factor receptor (EGFR) signaling and other unknown downstream proteins, they lack a global view to the phosphoproteome.
One of the first large-scale analyses of tyrosine kinase activity in lung cancer was performed in 2007 by Rikova et al., who identified 50 tyrosine kinases and over 2500 downstream substrates [103]. They confirmed wellknown tyrosine kinases involved in oncogenic signaling such as EGFR and hepatocyte growth factor receptor (HGFR or c-Met). Furthermore, it was shown that activated forms of anaplastic lymphoma kinase (ALK) and receptor tyrosine kinase (ROS) can be identified in lung cancer cells, in particular in non-small cell lung cancer cell lines (NSCLC). A first deep and extensive view of tyrosine kinase activity and downstream signaling networks was described.
Revealing phosphoproteomic dynamics becomes more and more important especially in the area of cancer research. The first study elucidating temporal dynamics of phosphorylation upon growth factor stimulation was performed by Olsen and Mann in 2006. According to their discoveries, EGF-signaling is regulated by phosphorylation of a variety of transcriptional regulators, amongst others signal transducer and activator of transcription 5 (STAT5), transcription factor MYC, and transcription factor JUND, within a short time frame of 20 min. By following regulatory changes over a particular time frame, signaling outcomes could be connected to responsible upstream or downstream events [104].
Quantitative phosphoproteomic profiling was already used to portrait different tumorigenic signaling pathways, to compare different tumor entities and to analyze the heterogeneity of tumors. Only recently Schweppe et al. applied a Super-SILAC approach for decoding global phospho-signaling networks in NSCLC patient samples. They were able to differentiate between different types of non-small cell lung cancer populations due to changes in particular oncogenic drivers such as epidermal growth factor receptor 2 (ErbB2) and RAF/MEK/ ERK signaling [105]. The RAF/MEK/ERK signaling is important for cellular growth, malignant transformation and drug resistance [106]. The regulation of stromal cells by oncogenic KRAS (Kirsten rat sarcoma viral oncogene homolog) in pancreatic ductal adenocarcinoma (PDA) cells was demonstrated by Tape et al. [107]. They performed an innovative sample preparation method called automated phosphopeptide enrichment (APE), where magnetic TiO 2 and Ti-IMAC microspheres are used to enrich phosphopeptides by employing a magnetic particle handling robot [108]. They investigated the cell-autonomous and non-cell autonomous signaling effects of oncogenic KRAS on the phosphoproteome of PDA. Thereby, a cell-autonomous activation of ERK 1/2 was determined resulting in an induction of Map kinase and cyclin dependent kinase motifs. Likewise, oncogenic KRAS was demonstrated in a quantitative proteomic analysis to control PDA cells by influencing the Sonic Hedgehog (SHH)-Smoothened (SMO)-GLI axis of stromal cells. The stromal-driven tumor cell phosphoproteome moreover differed from the oncogenic KRAS regulated cell-autonomous phosphoproteome revealing reciprocal signaling of the stromal cells. This evidence emphasizes the importance of focusing on tumor heterogeneity in cancer studies and therapy.
Phosphoproteomics and proteogenomics can help to understand mechanisms of resistance to cancer therapeutics and predict efficacy or adverse reactions relevant for personalized medicine. As a comprehensive technique, phosphoproteomics offers the opportunity to study changes in the phosphorylation of targeted proteins after treatment and thus can be used as an investigation tool for pre-clinical and clinical investigations. Thereby it can be used to improve and expand current drug treatment systems [105] by tailoring medication for therapy to individual responsiveness and tendency for side effects. By applying phosphoproteomics on metastatic castration-resistant prostate cancer (CRPC) material, Drake and colleagues could identify phosphorylation of key mediators in six major signaling pathways, including the cell-cycle pathway, DNA repair pathway, AKT/ mTOR/MAPK pathway, and the nuclear receptor pathway, which revealed potentially useful information for patient stratification and targeted therapy [109].
Proteogenomics parses the relation of genetic alterations to functional protein expression by comparison and integration of RNA and DNA sequencing data and (phospho) proteomics to infer their particular influence on the resulting phenotype [110,111]. In breast cancer, the analysis of the phosphoproteome identified several phosphorylated kinases and a G Protein-coupled receptor cluster that could not be detected at the mRNA level [110]. Previous proteogenomic characterization of highgrade serous carcinoma (HGSC), which comprises the majority of ovarian cancer cases, included phosphopeptide analysis and demonstrated the added value of protein phosphorylation data when correlating pathway activity with patient survival [112]. Another proteogenomic study characterized rectal cancer patients and used proteomics data to prioritize candidate driver genes [111].
In the past decade, the focus has shifted towards the functional and temporal analysis of changes within particular oncogenic pathways upon treatment with tyrosine kinase inhibitors as potent cytostatic drugs for the treatment of various cancers. Zhang et al. examined the global phosphoproteome after erlotinib treatment, a tyrosine kinase inhibitor for the treatment of lung cancer. They utilized lung adenocarcinoma cell lines harboring mutations in the kinase domain of EGFR, making them either sensitive or resistant to erlotinib treatment. They compared phosphorylation events and canonical pathways enriched in the sensitive or resistant cells [113]. Particular differences in EGFR connected pathways and changes in the phosphorylation patterns of regulatory proteins such as phosphorylated AKT (pAKT) and pERK (phospho-extracellular-signal regulated kinase) depending on erlotinib treatment of the resistant or sensitive cells were observed (Fig. 4). Their study gives novel impressions of phosphorylation events affected by erlotinib treatment and provides insights into possible mechanisms of drug resistance.
Only recently Wu et al. identified Focal Adhesion Kinase 2 as a modulator of tamoxifen resistance in breast cancer. They treated MCF7 breast cancer cells for 6 months with the selective estrogen-receptor modulator tamoxifen or ethanol as a vehicle control in vitro. SILAC was used to perform quantitative phosphoproteomic profiling based on HPLC-HRMS. By systematically analyzing the 2189 identified phosphorylated proteins, the focal adhesion pathway was identified as one of the most enriched signaling pathways. Protein phosphorylation was significantly elevated in the tamoxifen resistant cells. The 27 hyperphosphorylated proteins included the focal adhesion kinases FAK1 and FAK2 in the tamoxifen resistant breast cancer cells. In ongoing investigations by using real-time PCR, Western blot analyses, and immunofluorescence staining the overexpression of FAK2 in tamoxifen resistant cells was confirmed. Finally, siRNA knockdown of FAK2 significantly reduced the proliferation of the MCF7tamoxifen resistant cells and thus confirmed the pivotal role of FAK2 for tamoxifen resistance in these cells [114].
CSCan intricate challenge for proteomic and phosphoproteomic profiling CSC are of main interest both for biomedical research and clinical therapy. As it has been introduced above, CSC account for metastasis, relapse, and resistance to cancer therapeutics in different cancer entities. Analyzing CSC remains a challenge due to their low abundance and the task to specifically isolate these cells (see above).
Since phosphorylation patterns and dynamics are crucial for the regulation of normal and malignant cellular behavior, future studies are to focus on phosphoproteomics to investigate cancer stem cell signaling. Proteomic profiling has already been applied to different cancer stem cell entities. In 2010, one of the first quantitative profiling studies of pancreatic CSC was published by Dai et al. They solved the problem of the limited number of CSC gained from xenograft mouse models of primary human pancreatic adenocarcinomas by applying a twodimensional approach [115] that combined capillary isoelectric focusing and fraction collection in combination with nano reversed-phase HPLC-MS/MS followed by label-free quantification [115]. With this approach, they identified mitochondrial dysfunction as the top regulated pathway in the CSC population compared with the bulk tumor group. Moreover, other pathways known to be involved in cellular growth and proliferation such as VEGF signaling were shown to be enriched in CSC. Also, Interleukin signaling, Ras homologue gene family member A (RhoA), and integrin signaling pertaining to inflammatory and immunological pathways were found to be associated with CSC communication. Their results underline the connection between inflammation and carcinogenesis.
Recently, the proteome of sonic hedgehog driven human medulloblastoma stem-like cells was analyzed before and after retinoic acid differentiation [116]. The stem-like cells isolated from human infant medulloblastoma samples were further cultured as neurospheres in selective medium. HRMS following HPLC separation determined heat shock protein 70 as overexpressed in stem-like cells. Furthermore, the nuclear factor kappalight-chain-enhancer of activated B-cells (NF-κB) complex and tumor suppressor protein p53 were illuminated as pivotal players for cancer and stemness networks. Ongoing investigations showed that the phosphorylated p65 subunit of the NF-κB complex was highly expressed in these cancer stem cells, thereby identifying new key biological players involved in cancer stem cell biology of medulloblastoma.
To better understand dynamic signaling processes in CSC, Nilsson et al. initiated the first quantitative phosphoproteomic analysis of glioblastoma stem cells in 2010. They scrutinized glioblastoma stem cells (GSC) derived from human tumors and cultured them as neurospheres. These cells were treated with the novel JAK2/ STAT3 phosphorylation inhibitor WP1193 and/or the JAK/STAT3 activator IL-6 under normoxic and hypoxic conditions [117]. Six different conditions were compared by using TMT labeling prior to HILIC fractionation and TiO 2 enrichment. The separation was performed by RP-HPLC and detection by HRMS resulting in a total of 3414 proteins detected. Subsequent data evaluation linked 21 highly regulated proteins to STAT3, HIF1α (hypoxia inducible factor 1 alpha) and IL-6 signaling.
Several phosphoproteins linked to metabolic changes were observed under hypoxic conditions besides 11 proteins connected to HIF1α. Mitogen-activated protein kinase 1 (MAPK1)-expression in particular was increased reflecting HIF1α activation. Comparing normoxic and hypoxic conditions, they showed that hypoxic GSC were less responsive and thus more resistant to treatment with WP1193. Under treatment with WP1193 in combination with IL-6 they observed increased Insulin-like growth factor I (IGF1) signaling in both normoxic and hypoxic cells which confirmed the modulatory role of IGF1 in glioblastoma proliferation and migration [118]. Even though the effect of hypoxia on glioblastoma growth was well described based on their data, this study did not focus on the analysis of phosphorylation sites and kinase substrate interactions. Thus, they could not enlighten the deeper effect of different treatment conditions to the phosphorylation dynamics in glioblastoma stem cells.
Kozuka-Hata et al. addressed glioblastoma initiating cells two years later by investigating the effect of EGF stimulation on initiating cells from glioblastoma patients [119]. They used SILAC for quantification and TiO 2 columns to enrich the phosphopeptides prior to HPLC-MS/MS analysis. By searching against a human RNA database, they identified a novel peptide encoded by supervilin-like (LOC645954), which showed altered phosphorylation patterns upon EGF stimulation in a cell-type dependent manner. They started to look deeper into phosphorylation sites and their influence on communication and regulation of GSC. Out of 6073 phosphopeptides encoding 2282 phosphoproteins, 635 proteins belonging to the ErbB and mTOR signaling were shown to be upregulated in these CSC.
Still, our understanding of CSC regulation via phosphorylation remains largely incomplete. Only recently, the downstream signaling of stromal cell-derived factor 1 (SDF-1)/G protein-coupled receptor chemokine receptor 4 (CXCR4) in breast CSC has been examined [120]. The critical role of CXCR4 for tumor progression has already been known from O'Hayre et al., who examined the CXCL12/CXCR4 signaling network in chronic lymphatic leukemia (CLL) in 2010 but due to technical limitations, this work lacked comprehensive phosphosite analysis [121]. Yi et al. isolated CD44 high /CD24 low CSC from human mammary epithelial cancer cells (HMLER) and cultured them as tumor spheres. Phosphorylation events induced by 10 min treatment with SDF-1 with or without transient CXCR4 knockdown were compared. Phosphorylation changes were observed in several proteins with cell regulatory functions such as GTPase activating proteins and histone modification enzymes. Furthermore, they more deeply analyzed phosphorylationaffected kinases and phosphatases, among them ERK1 and serine/threonine-protein kinase 4 (PAK4), which were already known to be involved in the SDF-1/ CXCR4 signaling cascade. PAK4 was already described as being important for the development of breast cancer [122]. Besides, 44 kinases out of 50 at least 2-fold elevated kinases detected have been not known to be related to this signaling machinery before. Furthermore, 70 phosphosites of the 87 phosphosites detected in these kinases were still undiscovered. By examining kinase-substrates and phosphatase-substrates of 266 phosphoproteins with increased phosphorylation, multiple upstream kinases were found to be mediated by SDF-1/CXCR4 signaling. These were upstream kinases such as Pyruvate dehydrogenase kinase 1 (PDK-1), ERK1, GSK3 β for 5 phosphoproteins such as PKA (protein kinase A) and NF-κB. Moreover, a MAPK network downstream of SDF-1/CXCR4 signaling could be created providing novel insights into the resulting system-wide phosphorylation dynamics [120].
In spite of a remarkable progress over the decades made in the field of CSC research, analyzing the global phosphoproteome and phosphorylation dynamics of this subpopulation of cells is still not routinely practicable. CSC expansion by cell cultivation is mostly needed to obtain enough material for the analysis, which, however, can distort a realistic situation and reduce the clinical relevance.
Single cell proteomics for CSC investigation
One of the major disadvantages of current phosphoproteomics approaches is the need for relatively large amounts of cells samples, i.e. in the range of several million cells. This inevitably results in the study of heterogeneous cell populations, where the protein amount of each single cell and the respective phosphorylation pattern may vary considerably. Variability in phosphorylation-dependent signaling can influence the phenotype and quality of tumors, indeed it can be a reason for the formation of CSC [123]. CSC and bulk cancer cells are known to show inter-and intratumoral heterogeneity with marked differences in their malignant capacities. This versatility of a cancer (stem) cell population can be influenced by the microenvironment and/or intratumoral communication processes that induce different cell specific gene expression states [124] (for reviews see [3,125]).
Until now, there are still technical limitations to perform phosphoproteomics at the single-cell-level, with sensitivity being the primary constraint [90]. For a comprehensive state-of-the-art phosphoproteomics approach the protein amount of a single cell is too low. Thus, the innovative approaches are based on the implementation of microfluidic systems in combination with very sensitive detection schemes of phosphoproteomics. In particular, lab-on-a-chip technologies should enable and simplify single-cell phosphoproteomic analyses [126]. Wei et. al only recently reported the first single-cell phosphoproteomics approach to study signaling dynamics in glioblastoma with a focus on development of drug resistance. They used the single-cell barcode chip technology (SCBC) to investigate more than a dozen of proteins and phosphoproteins [127,128]. In this setup, one-cell microchambers were used to isolate single cells as illustrated in Fig. 5 [129]. These microchambers were (3) the cell is lysed due to shear stress from the expanding cavitation bubble; and (4) cellular constituents are released into the chamber. c Single cell protein levels are measured using an antibody spot. Chamber volume is 4.6 nL and results in favorable kinetics. By employing TIRF, only 10 fluorophores within 200 nm of the surface are imaged, which are assumed to be antibody/antigen bound. Reproduced from [129] with permission from the Royal Society of Chemistry © 2011 connected by programmable valves to lysis buffercontaining storage cavities, such that on-chip cell lysis could be performed. Each microchamber could be covered with a chip that featured an antibody-barcoded stripe which was used to capture the released (phospho) proteins. Detection of the (phospho)proteins was subsequently done by fluorescently labeled secondary antibodies [130,131]. Thereby activation of ERK-and proto-oncogene tyrosine-protein kinase Src signaling was detectable and linked to the cause of resistance to CC214-2an mTOR kinase inhibitor [132].
However, this approach is far from being comprehensive and unbiased. Antibody arrays are used to capture and quantify the proteins and phosphoproteins of interest. The barcode protein assay exhibited comparable dynamic ranges to commercially available ELISAs for around 12 proteins [127]. Meanwhile the number of detectable proteins was extended up to around 40 proteins per cell. Nevertheless, this targeted concept using prior knowledge about the tumor can hardly be compared with the discovery-driven process of unbiased HPLC-MS/MS based phosphoproteomics. Nevertheless, it can be feasible for implementation into the clinics, since only small amounts of material are needed and assays can be customized easily. There have been many attempts of combining this microfluidic principle with mass spectrometry, then called Chip-MS (for a review see [133]). These techniques are still in progress to be automated and improved but they combine both advantages of the downscaling feature of the microchip and the sensitive and discriminative detection capabilities of the MS instrument.
Conclusions
The highly malignant nature of rare CSC such as their exquisite capacity to initiate and fuel tumor growth, to seed metastases and their pronounced intrinsic resistance to chemo-and radiation therapya frequent cause for patients´relapse -calls for efforts to decipher the malignant code of the phosphoproteome. Understanding the complex phospho-signaling landscape of CSC will support the development of innovative multi-modal treatments including small-molecule targeting of key CSC kinases in combination with for instance, immunotherapy to significantly improve the overall long-term survival of patients.
Experimental workflows offering sufficient sensitivity and extensiveness for unbiased phosphoproteome analysis represent a real challenge in the investigation of signaling in heterogeneous populations of tumor cells. Nevertheless, in the past two decades, significant improvements in the detection techniques in terms of detection limits and structural information have enabled phosphoproteomic studies with very low amounts of sample down to the single-cell level. Moreover, the dynamic nature of phosphorylation itself provides challenges from the biological system, requiring very rapid quenching and sample preparation pipelines. Examining the phosphorylation events at single cell level is a desirable approach, but currently is restricted to the preselection of candidate phosphoproteins.
Comprehensive HPLC-MS/MS phosphoproteomics based on the analysis of single CSC represents an innovative and illuminative approach to investigate tumor initiating cells in great detail. In the future, with customized, enhanced and improved instrumentation this technique will likely become a routine part of modern clinical diagnosis and analysis as well as an essential method in the area of precision oncology.
|
2023-01-17T15:17:19.343Z
|
2017-03-29T00:00:00.000
|
{
"year": 2017,
"sha1": "8efd0cb8478b0f438ef0c81f4a92dd144b12cf39",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1186/s12964-017-0166-1",
"oa_status": "GOLD",
"pdf_src": "SpringerNature",
"pdf_hash": "8efd0cb8478b0f438ef0c81f4a92dd144b12cf39",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": []
}
|
203567477
|
pes2o/s2orc
|
v3-fos-license
|
The role of lipid metabolism in aging, lifespan regulation, and age‐related disease
Abstract An emerging body of data suggests that lipid metabolism has an important role to play in the aging process. Indeed, a plethora of dietary, pharmacological, genetic, and surgical lipid‐related interventions extend lifespan in nematodes, fruit flies, mice, and rats. For example, the impairment of genes involved in ceramide and sphingolipid synthesis extends lifespan in both worms and flies. The overexpression of fatty acid amide hydrolase or lysosomal lipase prolongs life in Caenorhabditis elegans, while the overexpression of diacylglycerol lipase enhances longevity in both C. elegans and Drosophila melanogaster. The surgical removal of adipose tissue extends lifespan in rats, and increased expression of apolipoprotein D enhances survival in both flies and mice. Mouse lifespan can be additionally extended by the genetic deletion of diacylglycerol acyltransferase 1, treatment with the steroid 17‐α‐estradiol, or a ketogenic diet. Moreover, deletion of the phospholipase A2 receptor improves various healthspan parameters in a progeria mouse model. Genome‐wide association studies have found several lipid‐related variants to be associated with human aging. For example, the epsilon 2 and epsilon 4 alleles of apolipoprotein E are associated with extreme longevity and late‐onset neurodegenerative disease, respectively. In humans, blood triglyceride levels tend to increase, while blood lysophosphatidylcholine levels tend to decrease with age. Specific sphingolipid and phospholipid blood profiles have also been shown to change with age and are associated with exceptional human longevity. These data suggest that lipid‐related interventions may improve human healthspan and that blood lipids likely represent a rich source of human aging biomarkers.
insulin-like growth factor (IGF), and adenosine monophosphate-activated protein kinase (AMPK) signaling pathways-play integral roles in regulating aging . The identification of aging biomarkers that change over time has concomitantly helped us to understand what mechanisms underlie aging. For example, nicotinamide adenine dinucleotide (NAD + ) concentrations decrease during aging and high-fat diets as well as increase in response to caloric restriction, exercise, and fasting (Verdin, 2015). Moreover, NAD + supplementation extends lifespan in mice (Zhang et al., 2016) as well as in yeast and worms (Verdin, 2015). This biomarker data (i.e., that NAD + levels decrease with age) preceded the lifespan data and paved the way for studies exploring the effects of NAD + repletion on aging.
Due to the sheer amount of time and cost required to validate a study in humans, the bulk of our aging and lifespan data come from shorter-lived yeast, worms, flies, and rodents. With the exception of research showing that caloric restriction improves health and survival in rhesus monkeys (Mattison et al., 2017), little aging work has been done in longer-lived organisms. The bulk of our understanding regarding aging comes from genetic experiments in model organisms , and we do not yet know how similar or dissimilar human aging is. For example, genome-wide association studies searching for longevity-related variants have found a lack of association with many genes known to extend lifespan in simpler animals (de Magalhaes, 2014). This is likely due to major biological differences between these organisms and humans as well as the limited genetic diversity of laboratory animal strains. As such, it is probable that a large portion of aging interventions proven in the laboratory will not yield significant clinical effects in humans (de Magalhaes, 2014). Therapies that are evolutionarily conserved between different model organisms are, however, more likely to have a therapeutic effect in Homo sapiens. Caloric restriction, for example, extends lifespan or improves health in every organism tested-including radically disparate animals such as mosquitoes (Joy, Arik, Corby-Harris, Johnson, & Riehle, 2010) and humans (Kraus et al., 2019;Most, Tosti, Redman, & Fontana, 2017).
Rather than screen every lifespan-extending intervention in humans to better understand how human aging works, another approach would be to utilize aging biomarkers. Biomarkers that strongly correlate with aging, lifespan, and healthspan can teach us about which processes are involved in human aging. They can also help us understand, independent of an individual's chronological age, how old a patient is biologically. Clinically, this could be used as an important health assessor. For example, Fleischer et al recently generated and analyzed a large dataset of genome-wide RNA-seq profiles of human dermal fibroblasts (Fleischer et al., 2018). These fibroblasts were derived from 133 people aged one to 94 years old.
By developing an ensemble machine learning method, they were able to estimate an individual's age to a median error of four years.
Testing in ten progeria patients revealed that this transcriptomic approach was capable of predicting accelerated aging (Fleischer et al., 2018). These data are impactful as they suggest that, with sufficient biomarker knowledge, patient senescence could be accurately measured by looking at objective, computer-analyzed parameters. In the clinic, this would enable precision medicine by giving doctors the ability to make patient-specific decisions based on their aging state.
Put differently, a patient's true biological age could be accurately ascertained instead of making assumptions based on their chronological age. Currently, generalized recommendations are provided given average outcomes associated with different age groups. Robust biomarkers would also allow us to rapidly test the efficacy of rejuvenative interventions in humans (Mahmoudi, Xu, & Brunet, 2019).
Myriad types of aging biomarkers exist. They can take the form of physiological and clinical data such as white blood cell count, absolute monocyte count, blood pressure, body mass index, resting heart rate, forced expiratory volume, gait speed, and grip strength (Burkle et al., 2015;Sebastiani et al., 2017;Xia, Chen, McDermott, & Han, 2017). As an example of how useful one of these biomarker parameters can be, grip strength is highly predictive of mortality, morbidity, and future disability (Leong et al., 2015). Biomarkers can also manifest as analyzed genomic, epigenetic, transcriptomic, and proteomic data. The epigenetic marker DNAm PhenoAge, which is comprised of DNA methylation information from 513 different CpGs, was shown to strongly correlate with age in every tissue tested and to be predictive of all-cause mortality as well as the age-related diseases cancer and Alzheimer's disease (Levine et al., 2018). By performing transcriptomic analyses, this marker was also associated with an increased activation of pro-inflammatory pathways as well as a decreased activation of DNA damage response genes (Levine et al., 2018). Efforts are currently underway to initiate a clinical trial that will utilize DNA methylation information to assess the efficacy of various antiaging interventions (Mitteldorf, 2019). Biomarkers can additionally manifest as molecules such as carbohydrates, apolipoproteins, glycoproteins, hormones, cytokines, and lipids (Burkle et al., 2015;Sebastiani et al., 2017;Xia et al., 2017). Interleukin-6, for instance, is a pro-inflammatory cytokine and glycoprotein that increases in concentration with age (Maggio, Guralnik, Longo, & Ferrucci, 2006). This age-related increase in interleukin-6 fits into our current understanding that the immune system gets progressively dysregulated with age and that unhealthy inflammation contributes to senescence. The upregulation of the interferon response pathway, for example, occurs during aging in multiple tissues from mice as well as in other vertebrate species such as rats, African turquoise killifish, and humans (Benayoun et al., 2019).
Ideally, a robust and practical biomarker would be one that incurs a low monetary cost and can be measured safely, repeatedly, and easily. Blood draws are especially appealing because they are inexpensive, simple, low risk, and can be taken as needed throughout a patient's lifetime. While several biomarker studies have focused on protein-based markers, the advancement of metabolomic techniques has made it feasible to look closely into a large array of metabolites. Metabolomic lipids and lipid-related proteins represent a large, rich source of potential biomarkers that are easily measured in the blood. Compounds in lipid metabolism can take many forms, such as phospholipids, triglycerides, waxes, steroids, and fatty acids.
They also play diverse physiological roles, such as forming cell membranes and lipid rafts (Pike, 2003) as well as exerting powerful cell signaling effects (Sunshine & Iruela-Arispe, 2017). Lipids are perhaps the most well-known for the paramount roles they play in both the storage and mobilization of energy.
Although lipids have been traditionally treated as detrimental and as simply associated with age-related diseases, numerous studies have shown that lipid metabolism potently regulates aging and lifespan. Jové et al, for example, assessed the plasma lipidomic profiles of 11 different mammalian species with longevities varying from 3.5 to 120 years (Jove et al., 2013). They found that a lipidomic profile could accurately predict an animal's lifespan and that, in particular, plasma long-chain free fatty acids, peroxidizability index, and lipid peroxidation-derived product content are inversely correlated with longevity (Jove et al., 2013). Similarly, Jobson et al scanned the genomes of 25 different species and reported that genes involved in lipid composition had undergone increased selective pressure in longer-lived animals (Jobson, Nabholz, & Galtier, 2010). Evidence from animals with extreme longevity also links lipid metabolism to aging. The ocean quahog clam Arctica islandica, an exceptionally long-lived animal that can survive for more than 500 years, exhibits a unique resistance to lipid peroxidation in mitochondrial membranes (Munro & Blier, 2012). The bowhead whale, another complex animal with extreme longevity that can live longer than 200 years, has lens membranes that are especially enriched with phospholipids. This unique enrichment is thought to at least partially underlie its uncanny resistance to the age-related lens disease of cataracts (Borchman, Stimmelmayr, & George, 2017). Naked mole rats, which enjoy remarkably long lifespans and healthspans for rodents, have a unique membrane phospholipid composition that has been theorized to contribute to their exceptional longevity (Mitchell, Buffenstein, & Hulbert, 2007). The importance of lipids in lifespan is further confirmed by the ability of lipid-related interventions to enhance longevity in model organisms (Huang, Withers, & Dickson, 2014).
The goal of this review was to assess the potential of lipids or lipid-related proteins to function as biomarkers of aging and to affect aging. To do this, we highlight how alterations in lipid metabolism can impact lifespan and age-related disease. We discuss how these lipidrelated interventions are distinct from those made by altering canonical aging pathways and also highlight lipid-associated signatures that correlate with extreme human longevity. Based on the existing data, we believe that lipids are a promising source of human aging biomarkers and that, clinically, they may be able to effectively determine a patient's biological age. We also believe that lipid-related interventions represent a promising clinical strategy for improving human healthspan and ameliorating age-related disease. Lastly, we propose aspects of lipid metabolism that could be clinically targeted to elongate the period of healthy life in humans. The ability of lipidspecific interventions to elongate both lifespan and healthspan in animal models demonstrates that, rather than being simply associated with age-related disease, lipid metabolism is a direct and potent regulator of aging.
| LIFE S PAN E X TEN S I ON VIA LIPID -REL ATED INTERVENTI ON S
One of the most effective ways to understand aging is to assess what interventions can modify lifespan. Different dietary, genetic, pharmacological, and surgical lipid-related interventions have been shown capable of extending lifespan in model organisms such as worms (Tables 1 and 2), flies (Table 3), and rodents (Table 4). Although several lipid-related therapies can boost longevity in yeast (Huang et al., 2014), we focus on aging studies in multicellular animals in this review. In this section, we limit our discussion to specific interventions that prolong longevity. In the subsequent section, we delve into associated mechanisms as well as observed trends between these life extension studies. Mentioned lipid synthesis pathways that are relevant to aging are visually summarized in Figure 1. The specifically highlighted pathways are triglyceride synthesis (Ahmadian, Duncan, Jaworski, Sarkadi-Nagy, & Sul, 2007;Shi & Cheng, 2009), sphingolipid synthesis (Gault, Obeid, & Hannun, 2010), fatty acid synthesis (Jump, 2011;Wakil, Stoops, & Joshi, 1983), and phospholipid synthesis (Vance, 2015; Figure 1). We have additionally created a table that briefly summarizes the primary function of each genetically targeted lipid-related protein implicated in lifespan regulation (Table 5).
| Nematodes
Different lipid-related nongenetic (Table 1) and genetic (Table 2) interventions are capable of extending lifespan in Caenorhabditis elegans, including the supplementation with fatty acids. In response to fasting, O'Rourke, Kuballa, Xavier, and Ruvkun (2013) found that expression of lysosomal lipase lipl-4 was induced, which in turn led to an enrichment of ω-6 polyunsaturated fatty acids (PUFAs). Direct supplementation of the ω-6 PUFAs arachidonic acid and di-homo-γ-linoleic acid in culture media promoted starvation resistance and extended animal lifespan. Inactivation of autophagy reversed this increase in lifespan, suggesting that autophagy underlies ω-6 PUFA-induced life extension (O'Rourke et al., 2013). Similarly, PUFA treatment with α-linolenic acid can increase lifespan in a dose-dependent manner. Oxidized α-linolenic acid generates oxylipins, and the oxylipin 9S-hydroperoxy-10E,12Z,15Z-octadecatrienoic acid further increases longevity in α-linolenic acid-treated worms (Qi et al., 2017). Feeding with 10-hydroxy-2-decenoic acid, a fatty acid component of honeybee royal jelly, similarly increases lifespan (Honda et al., 2011). This fatty acid additionally confers oxidative and thermal stress tolerance (Honda et al., 2015). Dietary supplementation with the monounsaturated fatty acids (MUFAs) oleic, palmitoleic, or cis-vaccenic acid is also sufficient to increase lifespan. In addition, the accumulation of MUFAs underlies the life extension observed in worms with a deficiency in H3K4me3 methyltransferase, an important epigenetic enzyme . The administration of fish oil containing the PUFAs eicosapentaenoic acid and docosahexaenoic acid has also been reported to enhance longevity, though too much fish oil had the effect of shortening lifespan (Sugawara, Honma, Ito, Kijima, & Tsuduki, 2013).
Other studies have reported nematode life extension in response to treatments with specific, nonfatty acid substances (Table 1). The plant lignan matairesinol, which acts as an adiponectin receptor agonist, was documented to extend mean lifespan by 25% in C. elegans (Su & Wink, 2015). Sesamin, which can inhibit delta 5 desaturase (Shimizu et al., 1991), is yet another lignan that has been reported to boost longevity (Nakatani et al., 2018). The compound α-lipoic acid has been reported to regulate lipid metabolism via the deacetylase sirtuin 1 (Chen, Kang, Wang, & Lee, 2012). α-lipoic acid, which is derived from the fatty acid octanoic acid (Solmonson & DeBerardinis, 2018), increases thermal stress resistance and elongates life in C. elegans (Benedetti et al., 2008). It additionally attenuates H 2 O 2 levels and improves chemotaxis (Brown, Evans, & Luo, 2006). The histone deacetylase inhibitor and ketone body β-hydroxybutyrate extend mean lifespan by 20% and increase thermotolerance in C. elegans. Relevant to the age-related neurodegenerative diseases Alzheimer's disease and Parkinson's disease, treatment with this ketone body additionally decreases α-synuclein aggregation and delays amyloid-β toxicity (Edwards et al., 2014). Nematodes treated with phosphatidylcholine showed an analogous protection against amyloid-β toxicity. They additionally displayed longer lifespans under conditions of oxidative stress (Kim, Kim, Park, & Park, 2019).
Nematode lifespan can also be elongated via RNAi knockdown ( Table 2). The yolk lipoprotein VIT/vitellogenin is capable TA B L E 1 Lipid-related nongenetic interventions that extend lifespan in Caenorhabditis elegans
Intervention % Lifespan increase Relevant observations Reference
Administration of α-lipoic acid 24.2 Attenuated hydrogen peroxide levels Enhanced chemotaxis in older worms Brown et al. (2006) Administration of α-lipoic acid 21 Conferred thermal stress resistance Benedetti et al. (2008) Feeding with the royal jelly fatty acid 10hydroxy-2-decenoic acid 12 Life extension was independent of the insulin signaling transcription factor DAF-16 Honda et al. (2011) Dietary supplementation with ω-6 polyunsaturated fatty acids (arachidonic acid or di-homo-γ-linoleic acid) (Seah et al., 2016). These data corroborate a prior report from the Cynthia Kenyon Laboratory, which found that RNAi knockdown against either vit-2 or vit-5 enhances longevity (Murphy et al., 2003). Using RNAi to inhibit elongation of fatty acid protein 1 (elo-1), elongation of fatty acid protein 2 (elo-2), or TA B L E 2 Lipid-related genetic interventions that extend lifespan in Caenorhabditis elegans
14-31
Neither deletion nor overexpression of hyl-1 resulted in life extension Tedesco et al. (2008) Constitutive expression of the lysosomal lipase LIPL-4 24 Long-lived worms are lean Lipid hydrolysis is induced via decreased insulin signaling Wang et al. (2008) RNAi knockdown of elongation of fatty acid protein 1 (elo-1), elongation of fatty acid protein 2 (elo-2), or the fatty acid desaturase fat-4 11 for elo-1 8 for elo-2 15 for elo-1 and elo-2 25 for fat-4 Knockdown of both elongases yielded a greater lifespan effect than either elongase alone Depletion of fat-4 produced the most significant life extension Gene knockdown was accompanied by increased resistance to oxidative stress Stefana et al. (2017) worms also lived longer. Conversely, a sphingolipid-rich egg yolk diet accelerated development and shortened lifespan. Fatty acid chain desaturation and elongation in many sphingolipid species also increased during development and aging (Cutler, Thompson, Camandola, Mack, & Mattson, 2014). Further cementing a role for ceramides and worm aging, RNAi knockdown against acid sphingomyelinase-3 (ASM-3), an enzyme that produces ceramide by hydrolyzing sphingomyelin, extends animal lifespan. Genetically inactivating asm-3 by introducing a mutation also extended life, though it did so to a reduced degree compared with RNAi-treated worms (Kim & Sun, 2012).
Several studies have found that genetically modifying lipid genes enhance longevity (Table 2). In response to dietary restriction, there is a reduction in the number of N-acylethanolamines, which are lipid-derived signaling molecules. Inducing an N-acylethanolamine deficiency via transgenic overexpression of fatty (Lucanic et al., 2011). Nematode longevity can also be increased via constitutive expression of the lysosomal lipase LIPL-4, which is an enzyme that hydrolyzes fats such as cholesterol and triglycerides.
These transgenic worms were lean and had fewer lipid droplets as well as decreased fat storage (Wang, O'Rourke, & Ruvkun, 2008).
Overexpression of the distinct lipid hydrolase diacylglycerol lipase is another route to life extension and additionally promotes resistance to oxidative stress (Lin et al., 2014). Functional loss of the ceramide synthase genes hyl-1 and lagr-1 via genetic deletion also extends nematode lifespan. This aging effect is abrogated if the autophagy-associated gene ATG-12 is knocked down, and in worms lacking hyl-1 and lagr-1, an increased number of autophagosomes are observed. These nematodes additionally display increased heat resistance as well as reduced feeding and reproduction (Mosbech et al., 2013). This work corroborates the previously mentioned data, which reported that RNA interference against hyl-1 prolongs life (Tedesco et al., 2008).
| Fruit flies
Lipids also impact lifespan in fruit flies (Table 3) Larvae that were fed a low-yeast diet and that were subsequently fed a low-yeast diet, a high-yeast diet, or a low-yeast, high-glucose diet as adults showed median lifespan increases of 20%-30%, 70%-90%, and up to 145%, respectively. An important mechanism underlying this life extension is the suppression of toxic lipids dubbed autotoxins. These toxic lipids are shed into the environment and can shorten both male and female Drosophila lifespan (Stefana et al., 2017). In contrast to the negative effects of autotoxins, dietary supplementation with various concentrations of the lipophilic compound and antioxidant butylated hydroxytoluene can increase both median and maximum lifespans in Drosophila bipectinata. Using the thiobarbituric acid test, it was also shown that flies fed butylated hydroxytoluene exhibited a decreased rate of lipid peroxidation compared with controls (Sharma & Wadhwa, 1983). The compound α-lipoic acid, which was previously reported to prolong life in worms (Benedetti et al., 2008;Brown et al., 2006), also extends life in D. melanogaster and does so in both females and males (Bauer, Goupil, Garber, & Helfand, 2004).
Separate experiments have shown that gene inactivation via
RNAi knockdown or insertional mutagenesis can boost longevity in Drosophila (Table 3). Glial-specific double RNAi knockdown against the low-density lipoprotein (LDL), receptor-related protein 1 (LRP-1), and the LDL receptor-related protein 2 (LRP-2) extends lifespan. In addition, larval growth is slowed and pupariation is delayed. These receptors were expressed in glial cells and responsible for moving lipoprotein across the blood-brain barrier. Knockdown against these receptors approximately halved the numbers of neurons positive for lipoprotein (Brankatschk, Dunst, Nemetschke, & Eaton, 2014 to shunt more diacylglycerol into 2-arachidonoylglycerol (Lin et al., 2014). Drosophila lifespan is additionally extended by the overexpression of fatty-acid-β-oxidation-related gene fatty acid-binding protein or dodecenoyl-CoA delta-isomerase. This life extension is accompanied by an enhanced tolerance to oxidative stress and starvation.
Fatty acid-binding proteins function as intracellular lipid chaperones, while dodecenoyl-CoA delta-isomerase works to catalyze the degradation of long-chain fatty acids during beta-oxidation (Lee, Lee, Paik, & Min, 2012). Overexpression of human apolipoprotein D (ApoD), a lipid-binding protein that promotes resistance to oxidative stress, is sufficient to enhance longevity and reduce age-associated lipid peroxide accumulation in Drosophila. These transgenic flies also enjoyed heightened protection against hyperoxia, dietary paraquat, and heat stress (Muffat, Walker, & Benzer, 2008). Transgenic overexpression of the fly homolog of ApoD, GLaz, analogously extends lifespan. Flies overexpressing GLaz exhibit an increased resistance to hyperoxia as well as superior walking and climbing abilities following sublethal exposure to hyperoxia. Overexpression of GLaz did not significantly alter weight, protein content, or lipid content but it did make flies more resistant to starvation (Walker, Muffat, Rundel, & Benzer, 2006
| Mosquitoes
Although the intervention is less direct, one study has found that a lipid-related genetic intervention can prolong life in two different species of mosquitoes (Table 3). The laboratory of Michael Riehle has shown that the transgenic overexpression of the kinase Akt1 exclusively in the fat body can increase survivorship in both Aedes aegypti and Anopheles stephensi mosquitoes (Arik et al., 2015). Akt1 is a member of the canonical insulin/IGF signaling pathway, and its overexpression in A. aegypti or A. stephensi mosquito species was sufficient to increase survivorship by 14%-47% or 15%-45%, respectively. Elevated expression of this protein kinase also activated the downstream signaling molecules forkhead box O and p70 S6 kinase. Survivorship differences compared to controls were only observed when mosquitoes were fed blood and were abrogated when mosquitoes were fed only sugar. Interestingly, transgenic mosquitoes also showed increased expression of the fat body vitellogenin, which is a precursor protein to egg yolk (Arik et al., 2015). The lack of a trade-off between reproduction and longevity as well as the enhancement of a reproductive protein tied to prolonged life is noteworthy.
Although the invertebrate fat body combines many of the functions of adipose tissue and liver in vertebrates (Law & Wells, 1989), it is important to note that there are important differences between this organ and adipose fat in more complex animals. While vertebrate fat is known to promote unhealthy inflammation and produce pro-
| Rodents
While not as numerous as the lifespan data in simple organisms (Tables 1-3), different interventions linked to lipid metabolism have been shown to augment longevity in rodents (Table 4). In rats, the laboratory of Nir Barzilai has shown that surgical removal of visceral fat at 5 months of age significantly increases both mean and maximum lifespan. It also reduces the incidence of severe renal disease (Muzumdar et al., 2008). Prior work from the same laboratory had shown that the removal of visceral fat improves insulin action and delays the onset of diabetes. Interestingly, the extraction of visceral fat did not alter levels of plasma free fatty acids. It did, however, decrease the expression of leptin and tumor necrosis factor-alpha in subcutaneous adipose tissue . A third, earlier study from the same research group found that increased insulin sensitivity in response to visceral fat removal was accompanied by a marked decrease in the plasma levels of insulin-like growth factorbinding protein-1 ( Barzilai et al., 1999).
We previously mentioned that, in Drosophila, overexpression of either human ApoD or GLaz, the fly homolog of human ApoD, is sufficient to extend lifespan (Muffat et al., 2008;Walker et al., 2006). Suggestive of an evolutionarily conserved anti-stress mechanism, overexpression of human ApoD is also capable of increasing survival under conditions of oxidative stress in mice (Ganfornina et al., 2008). This overexpression also prevented the rise of brain lipid peroxides postoxidant treatment. In contrast, loss of function of this gene increased the level of brain lipid peroxidation, reduced protection against oxidative stress, and impaired both learning and locomotor abilities (Ganfornina et al., 2008). Separate work by Thomas and Yao showed that, compared to wild-type mice, ApoD KO mice showed a significant increase in saturated fatty acids (16:0 and 18:0), a monoene (16:1), dienes (linoleic acid and eicosadienoic acid), and a hexane (docosatetraenoic acid) (Thomas & Yao, 2007). Prior work in these Dgat1 −/− mice had shown that, although these transgenic animals can still synthesize triglycerides, they are lean and exhibit a resistance to diet-induced obesity. Compared to controls, they displayed an increase in both energy expenditure and activity. Lactation was, however, defective in females (Smith et al., 2000). Despite being leaner, Dgat1 −/− mice consumed more food at baseline and they had higher surface temperatures. When placed in a cold environment, their hyperphagia became more pronounced (Chen, Ladha, Smith, & Farese, 2003). Although a DGAT1 deficiency recapitulates many aspects of caloric restriction-such as reduced adiposity, decreased tissue inflammation, and a lower fe- Another genetic intervention worth mentioning is the knockdown of the phospholipase A2 receptor (PLA2R1) in a mouse model of progeria (Griveau et al., 2018). Pla2r1 encodes for a transmembrane protein receptor that binds to secreted phospholipase A2 proteins and regulates various cell signaling processes (Bernard & Vindrieux, 2014;Sukocheva et al., 2019). showed a preferential use of fat as fuel, a higher metabolic rate, less triglyceride content, a markedly reduced adiposity, and decreased weight gain. Enhanced insulin sensitivity was observed in metabolic tissues, and both circulating glucose and insulin levels were reduced.
In addition to elevated uncoupling, they also exhibited increased AMPK activity and β-oxidation (Canaan et al., 2014). Renal tubular epithelial cells from FAT10 −/− mice display abrogated activation of TNF-α-induced NF-κB and a concomitantly reduced induction of NF-κB-regulated genes .
| LIPID -REL ATED AG ING MECHANIS MS
These collated lifespan and healthspan data (Tables 1-4) make a lucid argument for a significant role of lipid metabolism in aging and lifespan regulation. When analyzing the published data in worms, flies, and rodents, some interesting trends begin to emerge. In this section, we discuss the aging mechanisms revealed to us from these lifespan studies and expand upon them.
| Fatty acids
Four different studies have shown that feeding with PUFAs or MFAs can enhance longevity in nematodes (Table 1). The compound αlipoic acid, which is derived from the saturated fatty acid octanoic acid, can similarly extend lifespan in C. elegans and D. melanogaster (Tables 1 and 3), while fish oil containing eicosapentaenoic acid and docosahexaenoic acid can prolong life in C. elegans (Table 1) , which may be one mechanism by which fatty acids impact aging.
While direct feeding with fatty acids is sufficient to enhance longevity, overexpression of fatty acid proteins (e.g., fatty acid amide hydrolase or the fatty acid desaturase FAT-7 in worms and fatty acid-binding protein or dodecenoyl-CoA delta-isomerase in flies) can also yield desirable aging effects (Tables 2 and 3). Conversely, the disruption of fatty acid genes can induce physiological harm.
| Lipases, lipoproteins, and cholesterol
Studies in nematodes and fruit flies have shown that the duration of life can be extended by overexpressing a lipase enzyme (Tables 2 and 3). Moreover, life extension via increased lysosomal lipase activity has been linked to the antiaging, repair-associated process of autophagy (Lapierre et al., 2012).
Lipases catalyze the hydrolysis of fats and work to process lipids such as triglycerides and cholesterol. The enzymatic activity and mRNA levels of pancreatic lipase are decreased in older mice, and concomitantly, elderly mice exhibit decreased lipid absorption (Yamamoto et al., 2014). Lipoprotein lipase activity has analogously been reported to decrease with age in rat postural skeletal muscle (Bey, Areiqat, Sano, & Hamilton, 2001), and during physical inactivity, activity of this same lipase is suppressed (Bey & Hamilton, 2003). Inactivity also caused a local reduction in the uptake of plasma triglyceride into muscle as well as a decrease in the concentration of high-density lipoprotein (HDL) cholesterol. Treadmill walking raised lipoprotein lipase activity ~eightfold (Bey & Hamilton, 2003). Lipoprotein lipase hydrolyzes the triacylglycerol component of lipoproteins, which transport fat molecules throughout the body. Aberrant lipoprotein lipase function is associated with obesity, Alzheimer's disease, infection, insulin resistance, dyslipidemia associated with diabetes, chylomicronemia, and atherosclerosis (Mead, Irvine, & Ramji, 2002).
Monoacylglycerol lipase is highly expressed in primary tumors and promotes cancer pathogenesis via regulation of a fatty acid network (Nomura et al., 2010). These data suggest that, broadly, altering the activity of specific fat lipases may delay aging and symptoms of age-related disease. It would be invaluable to know whether or not overexpressing a lipase could extend lifespan in mice or other vertebrate models.
Lipoproteins also have an important role to play in aging. Not only does inhibition of the yolk lipoprotein VIT/vitellogenin prolong life in worms (Table 2), but the overexpression of the lipoprotein ApoD enhances survival and promotes stress resistance in flies and mice (Tables 3 and 4). In dogs, the expression of the apo-B, E lipoprotein receptors declines linearly with increasing age. These receptors are capable of binding both the apo-B-containing lowdensity lipoproteins (LDLs) as well as the apo-E-containing cholesterol-induced HDLs (Mahley, Hui, Innerarity, & Weisgraber, 1981).
In rats, plasma cholesterol levels increase with age. This increase can be attenuated by treatment with growth hormone, and this attenuation was presumed to occur via effects on lipoprotein metabolism (Parini, Angelin, & Rudling, 1999). Old mice show an impaired lipid mobilization response to fasting that includes milder fastinginduced changes in apolipoprotein gene expression compared with young mice (Araki, Okazaki, & Goto, 2004). Lipoproteins have also been correlated with various age-related ailments. For example, human serum concentrations of lipoprotein(a) are significantly associated with an increased risk of Alzheimer's disease (Solfrizzi et al., 2002) and HDL cholesterol tends to be inversely associated with cancer risk (Pirro et al., 2018). HDL cholesterol and triglycerides have been positively and negatively associated with an increased risk of age-related macular degeneration, respectively (Colijn et al., 2019). HDL cholesterol is also a predictor of major cardiovascular events in patients treated with statins (Barter et al., 2007).
| Triglycerides
One of the lipid-related, lifespan-increasing interventions in mice targeted triglyceride synthesis by creating a deficiency in the triglyceride synthesis enzyme acyl-CoA:diacylglycerol acyltransferase 1 (Table 4). Relatedly, long-lived mice lacking the ubiquitin-like FAT10 gene displayed decreased triglyceride content (Canaan et al., 2014) and an adenovirus-mediated increase in ApoD, a longevity-relevant lipoprotein (Tables 3 and 4), reduces plasma triglyceride levels in mice (Perdomo et al., 2010). These data indicate that triglycerides are closely tied to the aging process. By compared metabolic param- have shorter lifespans as well as higher levels of total hepatic triglyceride, total hepatic phospholipids, and cholesterol (Ishigami et al., 2004). Adipocytes, which store triglycerides, have also been shown to promote metastatic initiation by sensitizing melanoma cells to the cytokine TGF-β (Golan et al., 2019). These data make a strong case for triglycerides being highly relevant to aging and age-related disease. More specifically, a common theme appears to be that elevated levels of triglycerides are associated with physiological dysfunction.
| Ceramides and sphingolipids
Sphingolipids, including ceramides, have their own role to play in regulating aging (Tables 1-3). In nematodes, four different studies were able to extend the lifespan of C. elegans by inhibiting sphingolipid machinery (Table 2). These inhibited molecular targets include ceramide synthase genes, a sphingomyelinase, serine palmitoyltransferase, glucosylceramide synthase, dihydroceramide desaturase, and neutral/acidic ceramidase (Table 2). In D. melanogaster, inactivation of the ceramidase enzyme Drosophila alkaline ceramidase is sufficient to extend lifespan (Table 3). It is interesting that the impairment of sphingolipid/ceramide metabolism can prolong life in two different animal models. Data from rats show that sphingolipid catabolic enzyme activity increases during aging (Sacket, Chung, Okajima, & Im, 2009). Lifespan data from yeast further demonstrate that chronological lifespan can be elongated by reducing the rate of sphingolipid synthesis (Huang et al., 2014). Thus, disrupting the production of specific sphingolipids appears to exert pro-longevity effects.
There are situations, however, where interfering with the sphingolipid pathway can have detrimental health effects. Ceramide transfer protein is responsible for transferring ceramide from the endoplasmic reticulum to the Golgi complex. D. melanogaster flies functionally lacking this protein exhibit an increase in membrane fluidity, reduced protection against oxidative damage, decreased thermal tolerance, and shortened lifespans (Rao et al., 2007). C. elegans worms that lack sphingosine kinase have decreased lifespans, smaller brood sizes, and reduced body sizes. They additionally show worse locomotor behavior and neuromuscular function in old age (Chan et al., 2017). It has also been suggested that defects in sphingolipid metabolism contribute to the pathogenesis of different brain disorders, including the age-related neurodegenerative diseases Alzheimer's disease and Parkinson's disease (Di Pardo & Maglione, 2018). More data are required to understand why some sphingolipid-targeted interventions are pro-aging and why others are antiaging.
The ability of sphingolipids to influence aging matches up with their essential biological roles. The release of ceramide by acid sphingomyelinase, for example, is a prerequisite for CD95 signaling and apoptosis induction (Grassme et al., 2001). This is clinically significant as CD95 promotes tumor growth, and conversely, the loss of CD95 reduces both the incidence and size of tumors (Chen et al., 2010). More broadly, ceramides increase in concentration with age in mammals and have been linked to various age-related ailments, including cancer, type 2 diabetes, neurodegeneration, immune dysfunction, and cardiovascular disease. Ceramide accumulation is also correlated with increased insulin resistance and oxidative stress (Huang et al., 2014). Very recent work by Chaurasia et al. (2019) have shown that the deletion of dihydroceramide desaturase 1 improves insulin resistance and hepatic steatosis in mice. Mechanistically, ceramide was revealed to promote the uptake and storage of lipids and to impair glucose utilization (Chaurasia et al., 2019). Thus, clinical therapies that reduce ceramide concentrations may delay or ameliorate symptoms of aging in humans.
| Phospholipids
It is intriguing that, in a mouse model of progeria, deleting the phospholipase receptor PLA2R1 improved specific healthspan parameters (Table 4). In the same study, knockdown of this receptor was shown to prevent senescence in human fibroblasts (Griveau et al., 2018). PLA2R1 is associated with both cancer suppression (Bernard & Vindrieux, 2014) and idiopathic membranous nephropathy (Coenen et al., 2013). More broadly, PLA2R1 is thought to be a regulator of various biological processes, including pro-inflammatory signaling, apoptosis, senescence, and autoimmunity (Sukocheva et al., 2019). Bowhead whales and naked mole rats, two animals characterized by exceptional longevity, both exhibit unique phospholipid profiles (Borchman et al., 2017;Mitchell et al., 2007). The disruption of lipid hydrolases that regulate phospholipid metabolism has also been shown to decrease lifespan in worms (Park et al., 2018), flies (Kinghorn et al., 2015;Kunduri et al., 2014), and mice (Shinzawa et al., 2008). Conversely, treatment with phosphatidylcholine prolongs life in C. elegans under conditions of oxidative stress (Kim et al., 2019). While it has yet to be shown that modulating phospholipid machinery can lead to a statistically significant increase in vertebrate lifespan, these data all suggest that phospholipids are highly relevant to aging.
| Ketogenic diet
In the prior section, we discussed that treatment with the ketone body ß-hydroxybutyrate extends lifespan and increases thermotolerance in C. elegans (Edwards et al., 2014), while a ketogenic diet prolongs lifespan and healthspan in mice Roberts et al., 2017). In worms, ß-hydroxybutyrate has been proposed to extend life via two different antiaging pathways, the first of which would inhibit histone deacetylases and lead to increased DAF-16/ FOXO activity. The second pathway involves the mitochondrial metabolism of ß-hydroxybutyrate, which would increase the production of reactive oxygen species via increased citric acid cycle metabolism and electron transport chain activity. This would activate the SKN-1/ Nrf2 antioxidant response pathway and promote longevity (Edwards, Copes, & Bradshaw, 2015). In the original lifespan study, markers of neurodegenerative disease were attenuated in response to treatment with ß-hydroxybutyrate (Edwards et al., 2014). A more recent paper by Manzo et al. (2018) have shown that, in a Drosophila model of amyotrophic lateral sclerosis, a significant decrease and increase were observed in the levels of ß-hydroxybutyrate and carnitine conjugated long-chain fatty acids, respectively. Feeding flies either ß-hydroxybutyrate or medium-chain fatty acids improved locomotor function (Manzo et al., 2018). These data indicate that this ketone body can exert neuroprotective effects in two different animal models.
The ketogenic diet has been described as a biochemical model of fasting and works by producing ketone bodies (e.g., β-hydroxybutyrate, acetoacetate, and acetone) from fats when glycogen stores have been depleted in the liver. Ketone bodies are thought to affect neurons by inducing changes in metabolism, epigenetics, and signaling (Fedorovich, Voronina, & Waseem, 2018). Interestingly, two separate studies have shown that a ketogenic diet preserves memory performance during aging in male mice Roberts et al., 2017). Given these data and that a ketogenic diet has been used to treat human epilepsy for almost a century (Boison, 2017), it seems reasonable to hypothesize that ketone bodies exert neuroprotective effects. Indeed, a recent case study found that a ketogenic diet rescued cognition in a 71-year-old female patient with a dual diagnosis of mild Alzheimer's disease and metabolic syndrome.
The patient was heterozygous for the epsilon 4 allele of ApoE (Morrill & Gibas, 2019). It would be interesting to see whether, rather than a ketogenic diet, treatment with specific ketone bodies is sufficient to improve healthspan parameters and exert neuroprotective effects in vertebrate animal models. in C. elegans, suggesting that this ketone body is a dietary restriction mimetic (Edwards et al., 2015). Similarly, the royal jelly component 10-hydroxy-2-decenoic acid does not elongate life in eat-2 C. elegans mutants, suggesting the mechanism of action for this fatty acid overlaps with dietary restriction (Honda et al., 2015). to surgical removal of visceral fat in ad libitum-fed rats was significantly less than the life extension observed in rats that were calorically restricted. This suggests that caloric restriction works, at least in the part, by pathways that are distinct from those that are affected by the removal of visceral fat (Muzumdar et al., 2008).
| Canonical aging biology pathways
Although some overlap exists, the significantly different gene expression profiles in calorically restricted mice versus Dgat −/− mice indicate that this is yet another longevity mechanism that is not identical to caloric restriction (Streeper et al., 2012). an asm-3 deficiency and an age-1/PI 3-kinase deficiency were even longer-lived (Kim & Sun, 2012). Reduced daf-2 activity upregulates the lysosomal lipase lipl-4, and RNAi against lipl-4 partially suppresses the longevity of daf-2 mutants . Life extension generated by the overexpression of fatty-acid-β-oxidation-related genes activates dFOXO signal , and RNAi knockdown of LRP-1 and LRP-2 decreases how much AKT is phosphorylated (Brankatschk et al., 2014). Similarly, phosphatidylcholine treatment promotes the nuclear accumulation of DAF-16 (Kim et al., 2019). Conversely, the HYL-2 ceramide synthase acts independently of DAF-2 (Menuz et al., 2009) and 10-hydroxy-2decenoic acid-induced life extension in C. elegans is not dependent on DAF-16 or DAF-2 (Honda et al., 2015(Honda et al., , 2011. The TOR and AMPK signaling pathways have also been implicated in lipid lifespan studies. For example, the royal jelly component 10-hydroxy-2-decenoic acid failed to extend lifespan in already long-lived daf-15 mutants. Since daf-15 is a target of TOR, the authors conclude that this fatty acid overlaps with the TOR signaling pathway (Honda et al., 2015). DAF-15/raptor has, however, been reported to be a point of convergence for both the insulin/IGF and TOR signaling pathways (Jia, Chen, & Riddle, 2004). Overexpressing diacylglycerol lipase decreases the phosphorylation levels of S6 kinase, a target of TOR, while creating a transgenic mutant of this protein elevates the phosphorylation levels of S6 kinase (Lin et al., 2014). In a long-lived TOR pathway mutant with a defect in the worm ortholog of S6 kinase, levels of the lipid eicosapentaenoyl ethanolamide were decreased. Direct treatment of these worms with eicosapentaenoyl ethanolamide suppressed their longevity phenotype, thereby implicating N-acylethanolamines in their extended lifespan (Lucanic et al., 2011). Inhibition of TOR in C. elegans was also found to induce the expression of the lysosomal lipase lipl-4 (Lapierre et al., 2011). In mice, mTORC1 signaling was regulated in a tissue-dependent manner by a ketogenic diet (Roberts et al., 2017). Like TOR, AMPK is a metabolic sensor that regulates the synthesis, oxidation, and lipolysis of lipids (Wang, Liu, Zhai, Zhang, & Tian, 2018). In the mouse study which showed that FAT10 knockout mice live longer, the authors noted that these mice had increased AMPK activity in skeletal muscle (Canaan et al., 2014).
| Genetic lipid signatures correlated with extreme human longevity
Genome-wide association studies have identified multiple genetic factors that are correlated with exceptional longevity. Given the ability of lipid-related interventions to modulate lifespan and healthspan in model organisms (Tables 1-4), lipid-related genes would be expected to be associated with longer lifespans in humans.
This is indeed the case. Work by Atzmon et al. (2006) have shown that, among centenarians, homozygosity for the −641C allele in the APOC3 promoter (rs2542052) is significantly higher compared with controls. This genotype was associated with lower serum levels of APOC3, a favorable pattern of lipoprotein levels and sizes, a lower prevalence of hypertension, and greater insulin sensitivity (Atzmon et al., 2006). A separate study involving 338 centenarians implicated both ApoE and angiotensin-converting enzyme in human aging. The epsilon 4 allele of ApoE was significantly less common in centenarians compared with controls.
Conversely, the epsilon 2 allele of ApoE was significantly more common in centenarians (Schachter et al., 1994). Ryu, Atzmon, Barzilai, Raghavachari, and Suh (2016) have shown that the ε3/ε4 ApoE genotype is markedly depleted in centenarians, while the ε2/ ε3 genotype is substantially enriched. The epsilon 4 allele of ApoE is also a risk variant for late-onset neurodegenerative diseases and is thought to contribute to the pathogenesis of Alzheimer's disease via multiple different pathways (Yamazaki, Painter, Bu, & Kanekiyo, 2016). Given that overexpression of human ApoD or GLaz, the fly homolog of ApoD, was shown to confer antiaging benefits in flies and mice (Tables 3 and 4), it is interesting that apolipoproteins are also associated with greater human longevity. This suggests that the protective effects of apolipoproteins are highly evolutionarily conserved and that clinical interventions targeting apolipoproteins could improve human healthspan.
Rs7844965, located in an intron of the lipid hydrolase EPHX2, was linked with increased human lifespan in a group of UK Biobank participants (Pilling et al., 2017). Genome-wide association of 1 million parental lifespans has shown that gene pathways involving lipid proteins and homeostasis, synaptic function, and vesicle-mediated transport are enriched for lifespan variation (Timmers et al., 2019).
In addition, a group of Ashkenazi Jews with exceptional longevity (mean age of 98.2 years) and their offspring were reported to have an increased frequency of homozygosity for the codon 405 valine allele of the gene CETP, which encodes for cholesteryl ester transfer protein (Barzilai et al., 2003).
| Lipidomic analyses of extreme human longevity
A few studies have done broad, lipidomic analysis to assess the rela- PUFAs (Gonzalez-Covarrubias et al., 2013). Interestingly, many longlived organisms or mutants have a decreased ratio of PUFA to MUFA (Papsdorf & Brunet, 2019). Separate work utilizing NMR metabonomics and shot-gun lipidomics found that centenarians display unique changes in biosynthesis compared with elderly controls. In particular, phospholipids and sphingolipids were identified as putative markers and modulators of healthy aging (Montoliu et al., 2014 (Parthasarathy et al., 2017). Triglycerides have been reported to increase progressively with age and have been actively proposed as a biomarker of aging (Xia et al., 2017). Triglyceride levels increase in older patients and are thought to be a significant risk factor for coronary artery disease, particularly in women (LaRosa, 1997). These data suggest that triglycerides have the potential to be a useful lipid biomarker. However, it is important to note that genetically predicted triglyceride levels have been reported to be unassociated with either frailty and longevity in elderly populations (Liu et al., 2017). Ergo, while most studies indicate that triglycerides increase with age (Papsdorf & Brunet, 2019), it remains to be determined whether or not triglyceride levels can accurately predict parameters of aging.
| Lipoproteins and cholesterol as blood biomarkers of aging
Given everything mentioned heretofore, it should be unsurprising that ApoE plasma levels have been proposed as a human aging biomarker and have been reported to strongly associate with cardiovascular mortality (Mooijaart et al., 2006). The related LDL apolipoprotein B and LDL cholesterol reportedly both increase with age, and these increases are linked to a progressively reduced fractional catabolic rate of LDL apolipoprotein B. Stimulation of hepatic LDL receptor expression via the cholesterol-lowering medication cholestyramine in six old males was sufficient to increase the catabolic rate to levels identified in younger subjects. These data indicate the LDL increase with age occurs due to a reduced capacity for its removal (Ericsson et al., 1991). (Wirth et al., 2017). Offspring of individuals with exceptional longevity have significantly larger LDL and HDL particle sizes as well as a lower prevalence of cardiovascular disease, hypertension, and metabolic syndrome (Barzilai et al., 2003). Ashkenazi Jewish offspring of centenarians showed fewer and larger LDL particles compared with their same-aged partners. No differences in HDL particle phenotypes were reported (Heijmans et al., 2006). Larger LDL particle sizes as well as lower triglyceride levels were reported in offspring of nonagenarian siblings compared with controls, which were partners of the offspring. LDL particle size was associated with male longevity, while triglyceride levels were associated with female longevity (Vaarhorst et al., 2011). HDL from older subjects was reported to have an altered composition that impaired its antioxidant properties and overall function. HDL from elderly patients contained less cholesterol and had more sphingomyelin (Holzer et al., 2013).
Relatedly, a progressive decline in plasma HDL concentrations has been associated with cognitive dysfunction in centenarians .
Data also suggest that plasma cholesterol may be a viable human aging blood biomarker, though the data are disparate. Both female and male offspring of centenarians reportedly have higher plasma levels of HDL cholesterol compared with controls. Men also exhibited significantly lower LDL cholesterol levels (Barzilai, Gabriely, Gabriely, Iankowitz, & Sorkin, 2001). Work by Weijenberg, Feskens, and Kromhout (1996) found that total cholesterol decreased with age, but HDL cholesterol did not change significantly with age in Dutch men. Kreisberg and Kasim previously concluded that total cholesterol, HDL cholesterol, and LDL cholesterol change uniquely over time for both men and women (Kreisberg & Kasim, 1987). A biomarker signature comprised of multiple different biomarkers, including total cholesterol, was found to be associated with lower morbidity and mortality as well as better physical function (Sebastiani et al., 2017). More comprehensive, systematic analyses are required to better understand the relationship between cholesterol and lipoproteins with aging as well as how this relationship differs between men and women. white matter hyperintensity and lower total brain volumes. The authors concluded that this PUFA was therefore a candidate marker of brain aging (Tan et al., 2012). By assessing different fatty acids with age, it was reported that plasma saturated, polyunsaturated, and monounsaturated fatty acids increase with age. Concomitant with this, circulating concentrations of IL-6 and TNF-α increased, while IL-10 and TGF-β1 decreased over time. Certain saturated fatty acids were reported to be associated with changing levels of TGF-β1 and TNF-α (Pararasa et al., 2016).
| Fatty acids and lipid peroxidation as blood biomarkers of aging
In normal elderly people, a decrease in antioxidants and an increase in lipid peroxidation were reported compared with younger controls. The lipid peroxidation malondialdehyde was highly elevated in older patients with diabetes and hypertension (Akila, Harishchandra, D'Souza, & D'Souza, 2007). Yavuzer et al. (2016) have shown that both hypertension and aging are associated with higher lipid peroxidation in humans. In particular, they identified lipid hydroperoxide and thiobarbituric acid-reactive substances as sensitive markers for both hypertension and aging in elderly patients (Yavuzer et al., 2016). Aging is additionally associated with an increase in lipid peroxidation in cardiac muscle obtained from 59 patient donors (age range of 8-86 years) with a mean age of 56 ± 12 years (Miro et al., 2000).
| Sphingolipids and phospholipids as blood biomarkers of aging
Very little work has explicitly assessed whether or not sphingolipids, including ceramides, are candidate human aging blood biomarkers.
Plasma sphingolipids have been proposed as biomarkers for Alzheimer's disease (Mielke & Haughey, 2012) and have also been linked to the agerelated diseases of diabetes, obesity, nonalcoholic fatty liver disease, insulin resistance, and cardiovascular disease (Iqbal, Walsh, Hammad, & Hussain, 2017). The plasma ceramide C16:0 has been associated with a slower gait, an important aging marker of physical function (Wennberg et al., 2018). These data as well as the lipidomic data previously mentioned (Gonzalez-Covarrubias et al., 2013;Montoliu et al., 2014;Wong et al., 2019) nicely justify further exploring the relationship between blood sphingolipids and human aging. Indeed, a recent study in nondiabetic patients found that higher levels of plasma insulin and an increased HOMA of insulin resistance score were associated with an elevation in plasma ceramides (Lemaitre et al., 2018).
With regard to phospholipids, low plasma levels of lysophosphatidylcholines were found to be associated with impaired mitochondrial oxidative capacity in adults . Lower levels of blood phospholipids, including the lysophosphatidylcholine 18:2, were separately shown to be highly predictive of memory impairment in older adults (Mapstone et al., 2014). Low plasma levels of lysophosphatidylcholine 18:2 also predict a greater decline of gait speed in the elderly . Patients with cancer, a classical age-related disease, analogously show a decrease in the concentration of plasma lysophosphatidylcholine (Taylor, Arends, Hodina, Unger, & Massing, 2007). Given these data and that both phosphatidylcholine and phosphatidylethanolamine have been reported to decline with age in model organisms (Papsdorf & Brunet, 2019), phospholipids show substantial potential as blood aging biomarkers in humans. Interestingly, a study by Trabado et al. (2017) have reported that elderly healthy subjects have higher plasma levels of sphingomyelins and phosphatidylcholines compared with young subjects. Although this reinforces the theory that these lipids are connected to aging, it suggests that specific lipids within these families may uniquely increase or decrease with age. It also suggests that other parameters, like patient health or genetic variability, may influence the relationship between a given sphingolipid or phospholipid with age.
| CON CLUD ING REMARK S AND FUTURE DIREC TIONS
Although many questions remain to be elucidated, it is clear that lipid metabolism has an imperative role to play in regulating the aging process. Lipid-related interventions are capable of modulating lifespan in various model organisms. Moreover, specific lipids and lipid-related molecules have been shown to increase or decrease in an age-dependent manner. Lastly, lipid-related genetic markers can strongly correlate with exceptional longevity in humans. These qualities exceed the requirement for a hallmark of aging (Lopez-Otin et al., 2013) and demonstrate unequivocally that lipid metabolism is intimately connected to the aging process. They additionally highlight several different potential pathways that could be targeted to increase human healthspan (Figure 2). Since many of our proposed target pathways ( Figure 2) overlap with each other (e.g., lipase activity and fatty acid metabolism), it would be intriguing to learn what aging mechanisms are shared between each of these targets when they impact longevity.
Future work should aim to better understand the mechanisms that underlie lifespan changes in response to specific lipid-related interventions in model organisms. Additional research in vertebrate models, such as African turquoise killifish, mice, rats, and Rhesus monkeys, is especially needed. Identifying unique lipid characteristics shared among animals with extreme longevity (e.g., Greenland shark, bowhead whale, giant tortoise, and ocean quahog clam) or theoretical immortality (e.g., planarian flatworms and hydra) would also help illuminate pro-longevity lipid pathways. Another approach would be to do comprehensive analyses of healthspan parameters and the incidence of age-related disease in patients being treated with lipid-relevant pharmacological interventions or patients with lipid-related genetic mutations. This would help to identify targets and treatments that could be explicitly utilized to elongate human healthspan.
We are also hopeful that lipid signatures could be developed as reliable biomarkers to accurately predict human biological age.
Although they are usefully predictive, large human cohort studies represent a current bottleneck and it might be good to think about alternative study designs. If data sharing becomes more common, reanalyzing data may help to glean new insights from existing datasets. Another experimental approach could make use of the many apps that exist to trace daily food intake, composition, and activity. Recruiting people to document daily intake of specific food compounds could be coupled with various measurements to study lipids and aging or aging-related health outcomes in large F I G U R E 2 Proposed lipid-related pathways that could be targeted to extend human healthspan. The overexpression of lysosomal lipase enhances longevity in worms, while the overexpression of diacylglycerol lipase extends lifespan in both worms and flies. Gene inactivation or inhibition of genes encoding the sphingolipid-relevant sphingomyelinase-3, glucosylceramide synthase, serine palmitoyltransferase, dihydroceramide desaturase, neutral/acidic ceramidase, or ceramide synthase proteins extends life in Caenorhabditis elegans, while the inactivation of alkaline ceramidase increases lifespan in Drosophila melanogaster. Longevity can also be increased by feeding specific monounsaturated or polyunsaturated fatty acids to worms, by overexpressing fatty acid amide hydrolase in worms, or by overexpressing fatty acid-binding protein or dodecenoyl-CoA delta-isomerase in flies. The overexpression of apolipoprotein D enhances survival in flies and mice, and the overexpression of the fly homolog of this gene extends lifespan in flies. In worms, RNAi knockdown against the yolk lipoprotein VIT/vitellogenin prolongs life. Survival time can also be elongated by RNAi knockdown against low-density lipoprotein-receptorrelated protein 1 and low-density lipoprotein-receptor-related protein 2 in Drosophila. Creating a deficiency in the triglyceride synthesis enzyme acyl-CoA:diacylglycerol acyltransferase 1 boosts longevity in mice and knockdown of the phospholipase A2 receptor improves healthspan parameters in a mouse model of progeria. Relevant to the latter finding, treating worms with phosphatidylcholine boosts longevity. Treating C. elegans with the ketone body ß-hydroxybutyrate or feeding mice with a ketogenic diet additionally extends lifespan. There are likely additional lipid-related healthspan targets that remain to be elucidated human datasets. Given the current data, we are optimistic that certain lipid-related interventions are capable of extending human healthspan.
ACK N OWLED G M ENTS
The authors would like to acknowledge Dr. Jessica Lane (Cincinnati Children's Hospital, Cincinnati, Ohio, United States of America) for helpful discussions regarding lipid metabolism. They are also grateful to the two anonymous reviewers for their thoughtful suggestions and insights.
AUTH O R CO NTR I B UTI O N S
AAJ and AS wrote the manuscript. AAJ compiled the tables and created the figures. Both AAJ and AS contributed to the design of the manuscript.
|
2019-09-28T13:02:43.265Z
|
2019-09-27T00:00:00.000
|
{
"year": 2019,
"sha1": "49b8063295ac3ea29c3417659838ca837a55be24",
"oa_license": "CCBY",
"oa_url": "https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/acel.13048",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8160b28988c207947c571957a04c5658c127da8b",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Medicine",
"Biology"
]
}
|
53415987
|
pes2o/s2orc
|
v3-fos-license
|
Oxidative Stress Involved Autophagy and Apoptosis in Helicobacter pylori Related Gastritis
Gastritis, inflammation of gastric mucosa, is a very common condition in the world wide. There is no universally accepted classification of gastritis. Early classification was based mainly on the morphology, but recently pathogenic mechanisms have also been incorporated. The Sydney system, a classification of gastritis introduced in 1990, and updated in 1995, has included both an endoscopic and histologic divisions and is designed for an unambiguous uniform reporting system. (Dixon, Genta et al. 1996) The histologic changes of acute gastritis include hyperemia, edema, and infiltration of polymorphonuclear cells, together with variable loss of epithelium. Endoscopically, these changes can be observed as edema, petechial or submucosal hemorrhage, erosions or ulcers. A lot of factors including nonsteroidal anti-inflammatory drugs (NSAIDs) and various noxious substances may result in these acute abnormalities. Excess production of reactive oxygen species evokes oxidative stress, which can induce apoptosis and autophagy in the damaged tissue or cells. Oxidative stress induced by the NSAIDs and various noxious substances may contribute to the pathophysiologic and histopathologic alterations including autophagy and apoptosis, leading to gastritis. The discovery of Helicobacter pylori has markedly improved our understanding about the nature of chronic gastritis and other important gastroduodenal diseases. (Marshall 1983; Warren 1983) H. pylori have been accepted as the most common cause of chronic gastritis. (Suerbaum and Michetti 2002) Colonization of gastric mucosa by H. pylori is always associated with persistent inflammation. Several virulence factor derived from H. pylori may promote these inflammatory mucosal changes. H. pylori-associated chronic gastritis usually accompanies with polymorphonuclear infiltration and architectural change of the gastric mucosa. There are marked mucosal cellular and systemic humoral immunologic responses. The mucosal damage seen in patients with H. pylori may result from both the effect of immunologic response and the bacterial toxin. Apoptosis and autophagy may contribute to cell homeostasis in gastric epithelial cells subjected to H. pylori infection. The combination of antioxidant and anti-adhesion materials can be attenuated the severity of gastritis.
Epidemiology
More than 50% of population in the world is infected with this bacterium. The prevalence of infection is increased with age and thought to be a cohort effect. Epidemiologic studies show this infection is generally acquired during childhood and the majority of H. pylori transmission is through close person-to-person contact. The oral-oral, gastro-oral, or fecal-oral exposure seems the most probable explanation for infection. (Brown 2000;Amieva, El-Omar et al. 2008)
Bacterial factor and colonization
H. pylori infection is closely associated with chronic active type B gastritis, peptic ulcers, gastric cancer and gastric MALT lymphoma. The pre-neoplastic lesions, such as glandular atrophy and intestinal metaplasia, may consequently result from persistent chronic gastritis in some patients. The outcome of H. pylori infection depends on the combination of bacterial and host factors in addition to less well-defined environmental conditions. H. pylori are one of few organisms capable of colonizing the harsh environment of the human stomach. The initial step in H. pylori infection is the penetration and adherence of the bacterium to mucin and epithelial cells. H. pylori generate large amounts of cytosolic and cell surface-associated urease. The urease helps the organism to avoid the bactericidal activity of gastric acid. H. pylori can use its polar flagella to migrate rapidly to a more favorable environment below the mucus layer very close to the surface of the epithelium where the pH is near neutral. After H. pylori migrate to the gastric epithelium, the bacteria adhere to the surface of host cells and may damage them in order to obtain nutrients and establish persistent colonization. Several different adhesion molecules have been identified and classified as adhesins. (Boren, Falk et al. 1993) The best studied H. pylori adhesins are outer membrane proteins that bind carbohydrate modifications in the glycoproteins of epithelial cells. The specific bacterial gene product, BabA, act as the ligand for the fucosylated blood group antigen Lewis b receptor. (Ilver, Arnqvist et al. 1998) The SabA protein adheres to sialated glycoproteins, specifically to sialyl-Lewis-X. (Mahdavi, Sonden et al. 2002) A segment of bacterial DNA, known as the cag pathogenicity iland (cag PAI), direct the key interaction between H. pylori and the host cells. Many of the genes adjacent to cagA encode proteins that provide a type IV secretion system (TFSS) that allows the transfer of bacterial products from pathogenic bacteria into the host cell. (Censini, Lange et al. 1996;Christie, Atmakuri et al. 2005) cag PAI plays an important role in the pathogenesis of gastritis. Patients infected with cagA positive strain of H. pylori are generally associated with increase interleukin-8 (IL-8) expression and inflammation in their gastric mucosa. (Blaser and Atherton 2004) CagA protein translocates into the cytoplasm of epithelial cell where it is tyrosine phosphorylated by host Src kinases and subsequently results in the change of cell morphology and cell function. The response of epithelial cell to H. pylori infection is complex and a summary of interaction with several influencing variables, such as bacterial virulence factors, the signaling linked to specific receptors, reaction of immune and hormones. The vacA gene is present in all strains of this organism. However, only more than half of H. pylori strains are able to express the vacuolating cytotoxin (VacA). The association of the structure and function of VacA with the severity of disease has been demonstrated. (Cover 1996;Van Doorn, Figueiredo et al. 1999;Blaser and Atherton 2004). Specific vacA alleles (s1 and m1) can result in more severe disease and epithelial cell apoptosis. (Atherton, Cao et al. 1995;Cover, Krishna et al. 2003) H. pylori strains that express outer inflammatory protein A (OipA) are also associated with increased expression of IL-8, neutrophil infiltration, and more severe clinical outcomes. (Yamaoka, Kikuchi et al. 2002)
Host response
The host response to H. pylori infection plays a very important role in the pathogenesis of this organism related gastrointestinal disease. The IL-1 is known as a strong inhibitor of gastric acid secretion. The genetic polymophisms of IL-1 have been demonstrated to be associated with an increased incidence of hypochlorhydria, atrophic gastritis and gastric cancer. (El-Omar, Carrington et al. 2000;Furuta, El-Omar et al. 2002) The development of gastric cancer can be related to increase IL-1 expression, more severe gastritis and greater colonization of H. pylori. H. pylori infection can result in changes in epithelial cell morphology, disruption of the tight junction, production of cytokines, increased epithelial cell proliferation, and increased rates of epithelial cell death via apoptosis. (Amieva, Vogelmann et al. 2003;Naumann and Crabtree 2004;Ernst, Peura et al. 2006) The induction and expression of genes in epithelial cells responding to H. pylori stimulation is regulated by several transcription factors which are controlled by a series of signaling mechanisms. The nuclear factor-kappa B (NFk B) seems to be the most studied molecule in these transcription factors. NFk B activity in H. pylori infected epithelium is markedly enhanced, correlating with increased secretion of IL-8 protein and infiltration of inflammatory cell. (Naumann and Crabtree 2004) The changes of cell functions, including cell profiferation, inflammation, and survival in response to H. pylori infection are mostly regulated by mitogen-activated protein (MAP) kinase cascades. (Keates, Keates et al. 1999) The acid secretion is a major function of gastric epithelial cells. The net effect of H. pylori infection on acid secretion is related to the duration and distribution of infection and the presence of mucosal atrophy. The epithelial barrier function is altered as a consequence of both direct effects of H. pylori and its accompanying inflammatory response. Humans infected by H. pylori develop a unique inflammatory response in which infection persists despite the recruitment and activation of lymphocytes, phagocytic cells, and other immune cell populations. (Ernst, Peura et al. 2006) It is known that H. pylori can induce an infiltration with T lymphocytes, plasma cells, mononuclear phagocytes and neutrophils. Furthermore, expression of cytokines such as tumor necrosis factor (TNF), IL-1, IL-6 and IL-8 is also enhanced by the infection. However, it is not well known how the immune response and the mechanisms behind it related to disease outcome. The immunoregulatory and proinflammatory cytokines induced by H. pylori may influence the nature of the local T cell response. It is thought that helper T (Th) cells can be divided into two subsets, Th1, and Th2. The Th1 subset promotes cell-mediated immunity by producing mainly IL-2, TNF-, and interferon-gamma (IFN-γ), and the Th2 subset, which is important for antibody response produces IL-4, IL-5 and IL-10. Evidences in recent research have strongly suggested that T cell mediated immune response may play an important role in induction of disease in H. pylori infection. (Crowe, Alvarez et al. 1995;D'Elios, Manghetti et al. 1997;Mohammadi, Nedrud et al. 1997) It is still not clear whether www.intechopen.com different outcome of the disease is modulated by the different type of T cell immune response, although some studies suggested that Th1 type dominant cellular response may be involved in the Helicobacter disease. (Haeberle, Kubin et al. 1997) Little is known about whether Th2 response can be protective or whether modulation of these responses can change the outcome of infection.
Apoptosis and autophagy
Some bacterial components may reach the lamina propria where it can activate underlying phagocytosis through the damaged epithelial barrier. One of these bacterial factors is the H. pylori neutrophil-activating protein (Hp-NAP). This protein can promote the adhesion of neutrophil to the endothelial cells and stimulate chemotaxis of monocytes and neutrophils and production of reactive oxygen intermediates. (Satin, Del Giudice et al. 2000) Recruitment and activation of neutrophils and macrophages result in the release of various inflammatory mediators. H. pylori urease has been shown to bind to class II major histocompatibility complex (MHC) molecules on the surfaces of gastric epithelial cells and induce apoptosis. (Fan, Gunasena et al. 2000) H. pylori VacA can be inserted into mitochondrial membranes where it induces cytochrome c release and activates the caspase-3-dependent cell-death signaling cascade. (Galmiche, Rassow et al. 2000) Also, Th1 cytokines induced by H. pylori can stimulate apoptosis through a Fas-mediate pathway by inducing expression of the cell-surface receptor Fas and Fas ligand. (Wagner, Beil et al. 1997;Jones, Day et al. 1999;Smythies, Waites et al. 2000) The expression of inducible nitric oxide synthase (iNOS) is increased in H. pylori infected gastric mucosa. Nitric oxide (NO) and Superoxide (O 2 -) may be produced by infiltrating neutrophils. These reactive oxygen species (ROS) can react to form peroxynitrite which is a potent oxidant and reducing agent. Apurinic-apyrimidinic endonuclease-1 (redox factor-1) plays an important role in the regulation of redox-sensitive signaling and is expressed in epithelial cell during infection with H. pylori. (Ding, O'Hara et al. 2004;O'Hara, Bhattacharyya et al. 2006) The increased oxidative DNA damage by ROS is thought to play a causal role in malignant transformation. The cells which undergo apoptosis are removed by phagocytes. This engulfment of H. pylori infected epithelial cells by phagocytes plays an important role in the cytokine induction and the activation of host adaptive response. An increase of chronic inflammatory cells in the gastric mucosa indicates the presence of a chronic gastritis that may result from the increased oxidative stress. Apoptosis and autophagy are two types of programmed cell death that play a critical role in tissue homeostasis and disease development. Exacerbated production of ROS in the inflamed tissue results in substantial type I programmed cell death, apoptosis, including increases in Bax/Bcl-2 ratio, caspase-3 activity, DNA fragmentation and apoptotic cell formation in the damaged tissue. (Baik, Youn et al. 1996;Smoot, Elliott et al. 2000;Chien, Lee et al. 2001;Yu, Chien et al. 2004;Yu, Lin et al. 2005) Autophagy is type II programmed cell death and is a major lysosomal catabolic pathway for cytoplasmic macromolecules and organelles. Autophagy could be mediated by Beclin-1, a novel Bcl-2-interacting protein, to promote autophagocytosis and a cell-survival response. (Blommaart, Luiken et al. 1997;Liang, Jackson et al. 1999) Previous studies have indicated that H. pylori-induced gastric epithelial cell damage by increased Bax/Bcl-2-related proapoptotic cell death and decreased autophagy survival and/or repair. (Catrenich and Chestnut 1992;Lee, Yeo et al. 2004)
Clinical and therapeutic application
How to prevent and cure H. pylori infection associated with gastritis is an important issue. Nowadays the first-choice of therapy for H. pylori infection is one-week combination of a proton pump inhibitor and antibiotics. Following failure of the first-line treatment, secondline therapies, including alternative triple and quadruple regimens, have frequently been applied to the patients. (Chey, Wong et al. 2007;Malfertheiner, Megraud et al. 2007) Although, the current antibiotic-based therapies are generally effective, it may still fail because of the rising trend of antibiotic resistance or the low compliance. (Megraud 2004;Vakil, Megraud et al. 2007) To find out an alternative agent or mixture with preventive and therapeutic potential on H. pylori infection is therefore urgently required. Some strains of Lactobacillus and Bifidobacterium can inhibit H. pylori growth. However, probiotics do not eradicate H. pylori but maintain a lower level of this pathogen in the stomach. (Gotteland, Brunser et al. 2006) A vaccine can be used either propbylactically or therapeutically for H. pylori infection. In the mice, vaccination can result in significantly reduced H. pylori colonization but it cannot achieve satisfactory eradication or prevention of H. pylori infection. (Del Giudice, Covacci et al. 2001) A successful H. pylori infection requires the penetration and adherence of the bacterium to mucin and gastric epithelial cells. H. pylori that adheres to gastric mucosa subsequently causes gastric epithelial cell damage and atrophy via oxidative stress and the type I apoptotic or type II autophagic programmed cell death-related pathway. Sialylated glycoconjugates are responsible for the adherence of H. pylori to gastric epithelium. Cumulated studies have shown that anti-adhesive therapy using 3'-sialyllactose can prevent the binding of H. pylori to human gastrointestinal epithelial cells and decrease H. pylori colonization in rhesus monkeys without side-effects. (Simon, Goode et al. 1997;Mysore, Wigginton et al. 1999) Catechins, known as one kind of antioxidants, have been shown to possess anti-oxidative, anti-inflammatory, anti-apoptotic and cancer prevention activity. (Katiyar and Mukhtar 1996;Lin and Lin 1997;Yu, Lin et al. 2005) Besides, catechins including their major active component, epigallocatechin-3-gallate (EGCG), have antibacterial effect against H. pylori by inhibiting the activity of urease and VacA of this organism. (Mabe, Yamada et al. 1999;Matsubara, Shibata et al. 2003;Ruggiero, Tombola et al. 2006) Although catechins or 3'-sialyllactose have an inhibitory effect on H. pylori infection in vitro, these two compounds fail to effectively control infection in animal model in vivo when each is used alone. (Mabe, Yamada et al. 1999;Mysore, Wigginton et al. 1999;Matsubara, Shibata et al. 2003) However, effective prevention and treatment of H. pylori infection using a combination of catechins and sialic acid in AGS cells and BALB/c mice have been shown in a recent study. (Yang, Shun et al. 2008) The combination of catechins and sialic acid showed synergistic or additive anti-H. pylori activity and significantly reduced iNOS expression and Bax/BCl-2-mediated apoptosis but enhanced Becline-1 mediated autophage. Pretreatment with catechins/sialic acid completely prevented H. pylori infection and resulted in normal histology. Post-treatment with catechins/sialic acid decreased the bacterial load and gastritis score and eradicated up to 60% of H. pylori infectious in a dose-dependent manner. The rationale of this treatment model which can effectively control H. pylori infection includes several points. (Yang and Chien 2009) First, by reviewing the literature, we can find that monotherapy using a single drug such as PPI, bismuth salt, or antibiotics always fail to eradicate H. pylori in humans, although each of these drugs can work in vitro. (Bamba, www.intechopen.com Kondo et al. 1997) Thus, dual therapy, then triple therapy, and even quadruple therapy have been recommended step by step. Second, catechins have antioxidant and anti-microbial effects (Lin and Lin 1997;Mabe, Yamada et al. 1999), whereas sialic acid has anti-adhesive and antioxidant effects. (Simon, Goode et al. 1997;Teneberg, Jurstrand et al. 2000) Together, they may have additive or synergistic effect against H. pylori. Third, both catechins and sialic acid can exist in the foodstuff are wildly accepted to be very safe to humans.
Conclusions
It is now clear that both bacterial factors and host response play a role in the pathogenesis of H. pylori related gastroduodenal disease. H. pylori infection can cause gastric mucosal injury, including oxidative stress, inflammation, and apoptosis formation but inhibit the autophagic survival response in gastric epithelial cell as indicated in Figure 1. A new strategy for control of H. pylori infection is to interfere with the interaction between the bacteria and target cells and to eradicate bacteria but not target cells at the same time. The combination of compounds with anti-adhesive, antioxidant, and anti-microbial activities may protect gastric mucosa from infection by H. pylori and treat its related gastritis via downregulation of apoptosis and upregulation of autophagy. www.intechopen.com
|
2018-10-28T19:06:14.762Z
|
2011-09-15T00:00:00.000
|
{
"year": 2011,
"sha1": "053f089bc4e0f41fa5d6b723baaa67f923fb2cb6",
"oa_license": "CCBYNCSA",
"oa_url": "https://www.intechopen.com/citation-pdf-url/19869",
"oa_status": "HYBRID",
"pdf_src": "Adhoc",
"pdf_hash": "983b66d3a110821ff3559262d8400ac6bf5d9794",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
226351243
|
pes2o/s2orc
|
v3-fos-license
|
A preliminary risk prediction model for cannabis use disorder
Highlights • Substance use disorders are currently a major public health crisis in the US.• The prevalence of cannabis use disorder is rising due to legalization of cannabis.• This study built models to predict the risk of cannabis use disorder for a user.• Risk factors include personality traits, impulsivity and initial smoking enjoyment.
Introduction
Substance use disorders are currently a major public health crisis in the US (SAMHSA, 2016). Cannabis is the most commonly used illicit substance in the world (NIDA, 2019). With more than 200 million users of cannabis worldwide, its harmful health effects have become a serious global problem (Colizzi et al., 2020;Gunn et al., 2016;Meier et al., 2012). During the past two decades, the laws and policies related to cannabis use have also changed drastically throughout the world. For example, countries such as Canada, Spain, and Germany have legalized cannabis for medical use while some have even legalized its non-medical use, e.g., Uruguay in 2015 and Canada in 2018 (Cox, 2018). Not surprisingly, the legalization trend continues in the US, with 33 states and the District of Columbia legalizing medical marijuana use, and 11 states and the District of Columbia legalizing adult non-medical marijuana use (ProCon.org, 2020).
Regardless of the developing accord about the usefulness of medical marijuana for several serious illnesses, there is a widespread concern that this may cause adverse effects (Brown and Hasin, 2019;Hammond et al., 2020;Wall et al., 2019). According to a study on the effects of medical marijuana laws, the likelihood of current as well as regular use of cannabis among people aged 21 or older has increased after the laws came into effect (Wen et al., 2015). This also appears to have contributed to an increased prevalence of illicit cannabis use and cannabis use disorder (Hasin et al., 2017). In particular, among adult males, arrests due to illegal marijuana possession in major cities have increased by 15-20% and the treatment provided in rehabilitation facilities for such arrests have increased by 10-20% (Chu, 2014).
This article focuses on cannabis use disorder (CUD). Earlier, there was a consensus that CUD is rare, which is no longer true. It is estimated that about 34% of cannabis users develop CUD during their lifetime based on the 4th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (Marel et al., 2019). Furthermore, a recent study based on DSM-V criteria found that about 27% of cannabis users develop CUD during their lifetime (Feingold et al., 2020). Another research shows that after legalizing marijuana for recreational use, the prevalence of CUD among past year cannabis users between the ages of 12 and 17 rose from 22.8% to 27.2% (Cerdá et al., 2019). Thus, given that the prevalence of CUD is expected to increase further, it is imperative to predict the risk of developing CUD for cannabis users, especially for adolescents and emerging adults, based on their personal risk factors. Identifying individuals at high risk of CUD will allow the possibility of applying early intervention, which may potentially help stem the increasing prevalence of the disorder.
Several risk factors have been reported for substance use disorders in general and specifically for CUD. These include male sex, early exposure to traumatic events, early use initiation, family history of substance use, childhood depression, and conduct disorder symptoms (Gray and Squeglia, 2018;Meier et al., 2016;Tomko et al., 2019). High impulsivity and certain personality traits are also associated with the disorders (Beaton et al., 2014;Ketcherside et al., 2016). In particular, work by coauthor Filbey's lab showed that openness distinguishes cannabis-only users from nicotine-only users, co-morbid marijuana and nicotine users, and non-users (Ketcherside et al., 2016). The results from this study also indicate that conscientiousness is lower among cannabis users.
Some brief screening tools such as BSTAD (Brief Screener for Tobacco, Alcohol, and Other Drugs) (Levy et al., 2014) and S2BI (Screening to Brief Intervention) have been developed for adolescents (Levy et al., 2014). For example, the cutoff for CUD based on BSTAD is at least two days of marijuana use in the past one year. A relatively lengthy tool, Transmissible Liability Index, assesses the inherited risk for disorders based on a 45-item questionnaire (Tarter et al., 2015;Vanyukov et al., 2009). Also, a recent study has developed a simple cumulative risk index for substance dependence in adulthood using risk factors in childhood and adolescence (Meier et al., 2016). It can be used to screen adolescents who are likely to develop persistent disorder in adulthood. A similar study has developed a risk score by counting the number of early life risk factors present in an individual and associating it with cannabis use and CUD in early adulthood (Hayatbakhsh et al., 2009). However, a key limitation of the existing tools is that none of them provides a quantitative risk of developing the disorder based on personal risk factors, which restricts their practical utility. Models for predicting such risks have been developed for several diseases, including breast cancer (Gail et al., 1989;NCI, 2020), contralateral breast cancer (Chowdhury et al., 2018(Chowdhury et al., , 2017, heart disease (D'Agostino et al., 2008), depression (Cattelani et al., 2019;King et al., 2008), and psychiatric disorders (Bernardini et al., 2017), and they are in wide clinical use. However, currently there is no such quantitative risk prediction tool for CUD.
In this study, we build upon the findings of Ketcherside et al. (2016) in a cannabis-using adult population and perform a secondary analysis of the data. More specifically, we build a preliminary quantitative risk prediction model to estimate the chance that a cannabis user will develop CUD based on various demographic, behavioral, psychiatric, and cognitive risk factors.
Participants
The study participants are cannabis users who were recruited from the general population in the Albuquerque metro area during 2007-2010 (Filbey and Yezhuvath, 2013). The study was approved by the University of New Mexico and University of Texas at Dallas institutional review boards. All participants signed an informed consent form. The intent of the original study was to determine the neurobiological antecedents of substance use disorders (Filbey and Yezhuvath, 2013;Ketcherside and Filbey, 2015). For the secondary analysis reported in this article, the inclusion criterion was regular use of cannabis, i.e., at least 4 times a week for at least 6 months.
Data preparation
The initial data set obtained after applying the inclusion criterion consisted of 118 cannabis users. We used CUD as the outcome (response) variable, which was derived based on the DSM-IV criteria for dependence. DSM-IV is a multi-dimensional measure for diagnosing CUD and is well established in the literature (Hayatbakhsh et al., 2009;Marel et al., 2019;Meier et al., 2016). The variable selection process to identify potential risk factors was the following. First, the variables with more than 50% missing values were discarded. Then, among the remaining variables, only those that remain relatively stable over time were chosen. Given the cross-sectional nature of the data, focusing attention on such type of variables protects against using risk factors that may actually be an effect of CUD. This resulted in 30 variables. These included measures of impulsivity and personality traits. The former were obtained using two questionnaires, namely, Impulsive Sensation-Seeking Scale (ImpSS), a 19-item self-reported questionnaire from the Zuckerman-Kuhlman Personality Questionnaire (Zuckerman et al., 1993) and Barratt Impulsivity Scale (BIS), a 30-item self-reported questionnaire where the items can be grouped into six first-order factors that measure different aspects of impulsivity (Beaton et al., 2014;Stanford et al., 2009). Both ImpSS and BIS were considered because there are some characteristics of impulsivity that are captured by ImpSS but not by BIS and vice versa, and the two have been used together in several studies (Beaton et al., 2014;Martínez-Loredo et al., 2015, 2018. The personality traits were obtained using Neuroticism, Extraversion, and Openness inventory (NEO), a five-factor inventory for measuring five different dimensions of personality (Ketcherside et al., 2016). The actual measures derived from these questionnaires were total score on the ImpSS questionnaire, scores on the six factors from the BIS questionnaire, and scores on the five factors from the NEO questionnaire.
Only 46 of the 118 subjects had complete data on all 30 variables. To guard against loss of subjects due to missing data on potentially unimportant variables, univariate logistic regression models were fitted with each of these variables as a predictor. Thereafter, the predictors with univariate model p-value less than or equal to 0.3 were selected into the final set of potential predictors for a multivariate model (Hosmer et al., 2013). The resulting data set had 12 potential risk factors and 94 subjects with complete observations on them. This final data set was used for the rest of the model building exercise. Table 1 presents the 12 risk factors. They fall in three groups. Group 1 consists of three variables, namely, age at the time of data collection, age at first use of cannabis, and the level of enjoyment from initial cigarette smoking (measured on an ordinal scale from 0 to 10 through the question "How much did you enjoy smoking at first?," with higher value indicating more enjoyment). Group 2 consists of six measures of impulsivity, namely, total score on the ImpSS questionnaire (ImpSS-T) (Beaton et al., 2014;Zuckerman et al., 1993) and scores on five of the six first-order factors from the BIS questionnaire (Beaton et al., 2014;Stanford et al., 2009), namely, attention (BIS-A), cognitive instability (BIS-I), motor impulsiveness (BIS-M), perseverance (BIS-P), and cognitive complexity (BIS-C). For all variables in this group, higher values imply greater impulsivity (Patton et al., 1995). Group 3 consists of scores on three of the five personality dimensions measured using the NEO inventory (Stanford et al., 2009), namely, neuroticism (general tendency to experience negative feelings; NEO-N), openness (open to new experiences and imaginative; NEO-O), and conscientiousness (forward planning, organization and ability to carry out tasks; NEO-C). Higher values for these variables imply greater neuroticism, openness, and conscientiousness, respectively.
Data analysis
The data analysis was performed using five common statistical and machine learning models for classification (James et al., 2013), namely, logistic regression with LASSO penalty, K-Nearest Neighbor (KNN), support vector machine (SVM) with radial kernel, random forest, and gradient boosting. We chose these techniques because they work even when the sample size is small relative to the number of predictors, as is the case here (James et al., 2013). The tuning parameters involved in these models were selected using leave-one-out-cross-validation (LOOCV) (James et al., 2013). Moreover, in keeping with the common practice, the performance of these models was evaluated by examining their prediction accuracy as measured using overall accuracy (i.e., the proportion of correct classifications), sensitivity (i.e., the proportion of correct classifications among the CUD subjects), and specificity (i.e., the proportion of correct classifications among the non-CUD subjects). Further, due to the lack of independent test data, the performance measures were computed using LOOCV. By protecting against overfitting, the LOOCV-based measures provide a more accurate assessment of model performance on future unseen data than those computed directly from the training data. By default, the models use 0.5 as the cutoff for probability, that is, a study subject is classified as having CUD if their probability of CUD exceeds 0.5. If the cutoff is increased, the sensitivity will decrease and specificity will increase.
Sample characteristics
The characteristics of cases of CUD and controls (i.e., the non-CUD subjects) with respect to the 12 risk factors under consideration are shown in Table 1. Out of 94 subjects, 67 (71%) were males and 58 (61.7%) were cases. Compared to controls, the cases on average were younger (23.16 vs 25.56 years), reported lower level of enjoyment from initial smoking (2.62 vs 4.11), were more impulsive sensation seeking (10.69 vs 8.47 on ImpSS-T), and had higher scores on BIS impulsivity traits (e.g., 7.21 vs 5.94 on cognitive impulsivity, BIS-I). The cases were also more likely to experience negative feelings and were more open to new experiences than controls as the cases reported higher averages for neuroticism (21.78 vs 16.36 on NEO-N) and openness (33.52 vs 32.03 on NEO-O), respectively. In addition, the cases were less conscientious than controls as reflected by lower average values for NEO-C for cases (29.93 vs 33.56). Among the twelve risk factors, only the following six exhibit statistically significant association with CUD in a univariate logistic regression model at 10% level of significance: level of enjoyment from initial smoking, ImpSS-T, BIS-A, BIS-I, NEO-N, and NEO-C. Fig. 1 presents the variables with non-zero regression coefficients from LASSO logistic regression model and the top seven variables based on variable importance measures for the other models. The seven variables selected by LASSO, namely, age, level of enjoyment from initial smoking, ImpSS-T, BIS-I, NEO-N, NEO-O, and NEO-C, were also found to be important by the other models. In particular, except ImpSS-T and NEO-O, the remaining five were selected as important by all other models. Moreover, ImpSS-T was chosen as an important predictor by KNN, random forest, and SVM while NEO-O was indicated to be important by random forest and gradient boosting. Table 2 presents the accuracy, sensitivity, and specificity of the models based on 0.5 cutoff as computed using LOOCV as well as the AUC of the models. The associated ROC curves and the plots of accuracy versus cutoff are provided in Supplementary Materials. Although the various models performed similarly, which is reassuring, overall we may conclude that LASSO and gradient boosting outperformed the others. For example, the two are tied for the highest AUC. Nevertheless, an advantage of LASSO is that it provides estimates of regression coefficients and hence odds ratios. This allows easy interpretation of the effects of the risk factors. This important and desirable feature is not available in other models. Therefore, we choose the LASSO logistic regression model as our final model.
Results from multivariate models
It may be of interest to quantify the advantage of this model over a random guess classifier that predicts CUD with probability 0.617, the proportion of CUD cases in the data. The accuracy, sensitivity, and specificity of this classifier can be calculated to be 0.527, 0.617, and 0.383, respectively. These are much lower than the corresponding values reported in Table 2 for the LASSO model.
The final LASSO model
The final LASSO model predicted the CUD status with 66% accuracy. Its sensitivity and specificity were 0.81 and 0.42, respectively. Thus, it does a much better job of correctly identifying the CUD cases than the non-CUD controls at the probability cutoff of 0.5. This cutoff may not be appropriate in all clinical settings. The appropriate cutoff can be chosen by examining its ROC curve, presented in Supplementary Materials, for the tradeoff between sensitivity and specificity. Its AUC is 0.65. The seven variables selected by this model together with their estimated coefficients and the associated odds ratios (OR) are shown in Table 3. The higher probability of CUD was associated with younger age (OR = 0.97), lower level of enjoyment from initial smoking (OR = 0.89), higher Table 1 Characteristics of subjects in terms of mean (SD) on various risk factors whose pvalues from univariate logistic regression models are less than 0.3. To illustrate the model, we considered two subjects from the data who had the largest and the smallest predicted probability of CUD. Their true status is CUD and non-CUD, respectively. The first subject was young (age = 18); received little enjoyment from initial smoking (score = 1); had high scores on impulsivity (ImpSS-T = 12), cognitive instability (BIS-I = 12), and neuroticism (NEO-N = 32); was quite open to new experiences (NEO-O = 38); and had low conscientiousness (NEO-C = 19). The predicted probability of CUD for this subject was 0.93. The second subject was 49 years old; received much enjoyment from initial smoking (score = 10); had low scores on impulsivity (ImpSS-T = 9), cognitive instability (BIS-I = 4), and neuroticism (NEO-N = 11); was also quite open to new experiences (NEO-O = 36); and had high conscientiousness (NEO-C = 39). The predicted probability of CUD for this subject was 0.15.
Discussion
Substance use disorders are a growing public health problem and cannabis is the most commonly used illicit substance in the world (NIDA, 2019; WHO, 2020). The legalization of cannabis for medical and recreational purposes worldwide has increased cannabis use and CUD. Therefore, there is a growing need for a CUD risk prediction tool. In this study, we built a preliminary model by identifying risk factors with the help of several statistical and machine learning algorithms.
We eventually chose the LASSO logistic regression model as the final model for two reasons. First, there was no major difference among the top performing models. Second, LASSO allows the ability to interpret the effects of risk factors quantitatively, a feature unavailable in the other methods. The LASSO model gave seven risk factors with non-zero (important) coefficients. We had also explored the possibility of adding interaction terms to this model but did not eventually add any because the model with interactions had lower predictive accuracy than this model. The risk factors identified by our model are consistent with the literature (Dougherty et al., 2013;Kong et al., 2013;Lee-Winn et al., 2018;Pampati et al., 2018;Winters and Lee, 2008). In particular, previous findings indicate that younger people are more likely to develop CUD (Winters and Lee, 2008). Using ImpSS and BIS scales, numerous studies have shown that high impulsivity is prevalent among users of nicotine (Chase and Hogarth, 2011), cocaine (Ball, 1995), and alcohol (Curran et al., 2010). We also found that higher ImpSS-T increases the likelihood of dependence on cannabis. The positive association between cognitive instability and CUD status that we found is also known (Mitchell and Potenza, 2014).
Similarly, the relationship of CUD with personality trait risk factors based on NEO is consistent with the previous findings (Ball, 2005;Fridberg et al., 2011;Kotov et al., 2010). For example, cannabis users have higher openness and lower conscientiousness compared to nonusers (Fridberg et al., 2011;Ketcherside et al., 2016). Generally, high neuroticism is reported in nicotine-only users (Tate et al., 1994;Terracciano et al., 2008) and average neuroticism is reported in cannabisonly users (Terracciano et al., 2008). We found that higher neuroticism is associated with higher likelihood of CUD, which is not surprising because our sample consists of co-morbid marijuana and nicotine users.
We also found that less enjoyment from initial smoking is associated with increased likelihood of becoming cannabis dependent. This is in line with the findings from a nationally representative longitudinal study, which was conducted to identify the risk factors associated with different stages of cannabis use (Pampati et al., 2018). This study found that greater quantity of cigarette use decreased the likelihood of reinitiation of cannabis use among participants who were cannabis users prior to reaching adolescence (Pampati et al., 2018). Even though our overall findings are consistent with the literature, we did not find several risk factors for CUD that have been previously reported in the literature. Some of the risk factors such as childhood depression and conduct disorder symptoms were not available in these data. While some other factors such as early exposure to traumatic events had substantial missing data because of which they were excluded. Yet others may not have been identified due to limitations of the study as described in the following.
Our study's first limitation is the cross-sectional and observational nature of the study because of which it is difficult to establish a causal relationship between a risk factor and CUD, especially for the factors that can vary over time. To mitigate the latter issue, we only used risk factors that remain relatively stable over time. However, even then we need to be cautious about drawing any conclusion about causation as this is an observational study. The second is that there are not a large number of subjects and the participating subjects came from a specific metro area in the US, which may not be representative of the entire population of all cannabis users. The third is due to missing values on the variables. When the risk factors are jointly analyzed in a multivariate model, this leads to a loss of some subjects as those with missing values in any of the multiple variables are discarded. We tried to balance the loss of sample size with the inclusion of risk factors. Moreover, to mitigate the issue of small sample size, we chose the statistical and machine learning methods that work even when the sample size is small relative to the number of predictors. Nonetheless, availability of complete data on more subjects would have provided higher power for identifying association.
We also acknowledge that the data used for this study were acquired in 2007-2010 and may be limited in its generalizability to current cannabis use impacts. Nonetheless, New Mexico's cannabis policies may be more historically representative of current national policies (compared to other states) given that medically-indicated cannabis was legalized in New Mexico in 2007 coinciding with the study's data collection. Thus, our findings may provide insights into future trends related to continued changes in cannabis legislation in the US. Also importantly, there has been no change in rate of current marijuana use in New Mexico in recent years, although the rate has remained significantly higher than the US rate (YRBS, 2017;YRRS, 2017). Thus, cannabis use in New Mexico has been stable and should not limit the impact of the current findings. Lastly, the mechanisms that underlie the risk for CUD likely remained relatively unchanged in the last 10 years.
Despite its limitations, this study represents a novel attempt to build a CUD risk prediction tool. To address the limitations, we are working towards building a risk prediction model using longitudinal data from a large number of subjects spread throughout the US. In addition, some people may be dependent on more than one substance (Moss et al., 2014;Richmond-Rakerd et al., 2016 and in fact, there may be common risk factors for several substance disorders (Oshri et al., 2018;Richmond-Rakerd et al., 2016). Therefore, it would be of interest to model jointly the relationship between multiple substance disorders and potential risk factors. Finally, inclusion of genetic and/or imaging factors can also provide a more personalized model.
Conclusion
This study developed a preliminary relative risk model for predicting the risk of CUD based on several risk factors. Higher risk of CUD was associated with younger age, lower level of enjoyment from initial smoking, higher score on impulsivity, greater cognitive instability, higher neuroticism, i.e., more prone to experience negative feelings, greater openness to new experiences, and lower conscientiousness.
Role of funding source
This work was funded by the University of Texas at Dallas SPIRe seed grant and NIH grants K01 DA021632-01A1 and R01 DA042490. The funders had no role in study design, data analysis, interpretation of findings, and preparation of this manuscript.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
|
2020-10-28T19:20:39.568Z
|
2020-10-20T00:00:00.000
|
{
"year": 2020,
"sha1": "391bc7cc3cbaab7d1a3f42c77918dad21d5481e2",
"oa_license": "CCBYNCND",
"oa_url": "https://doi.org/10.1016/j.pmedr.2020.101228",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "fc49e3bfee615b2d3a3e4519a31da57b772114ac",
"s2fieldsofstudy": [
"Psychology",
"Medicine"
],
"extfieldsofstudy": [
"Psychology",
"Medicine"
]
}
|
204542561
|
pes2o/s2orc
|
v3-fos-license
|
Oxidative stress-mediated TXNIP loss causes RPE dysfunction
The disruption of the retinal pigment epithelium (RPE), for example, through oxidative damage, is a common factor underlying age-related macular degeneration (AMD). Aberrant autophagy also contributes to AMD pathology, as autophagy maintains RPE homeostasis to ensure blood–retinal barrier (BRB) integrity and protect photoreceptors. Thioredoxin-interacting protein (TXNIP) promotes cellular oxidative stress by inhibiting thioredoxin reducing capacity and is in turn inversely regulated by reactive oxygen species levels; however, its role in oxidative stress-induced RPE cell dysfunction and the mechanistic link between TXNIP and autophagy are largely unknown. Here, we observed that TXNIP expression was rapidly downregulated in RPE cells under oxidative stress and that RPE cell proliferation was decreased. TXNIP knockdown demonstrated that the suppression of proliferation resulted from TXNIP depletion-induced autophagic flux, causing increased p53 activation via nuclear localization, which in turn enhanced AMPK phosphorylation and activation. Moreover, TXNIP downregulation further negatively impacted BRB integrity by disrupting RPE cell tight junctions and enhancing cell motility by phosphorylating, and thereby activating, Src kinase. Finally, we also revealed that TXNIP knockdown upregulated HIF-1α, leading to the enhanced secretion of VEGF from RPE cells and the stimulation of angiogenesis in cocultured human retinal microvascular endothelial cells. This suggests that the exposure of RPE cells to sustained oxidative stress may promote choroidal neovascularization, another AMD pathology. Together, these findings reveal three distinct mechanisms by which TXNIP downregulation disrupts RPE cell function and thereby exacerbates AMD pathogenesis. Accordingly, reinforcing or restoring BRB integrity by targeting TXNIP may serve as an effective therapeutic strategy for preventing or attenuating photoreceptor damage in AMD.
Introduction
Age-related macular degeneration (AMD) constitutes a progressive, chronic disease that represents a common irreversible cause of severe loss of vision 1 . Vision loss in AMD occurs through photoreceptor damage in the macula. In particular, cellular dysfunction of the retinal pigment epithelium (RPE), a major component of the blood-retinal barrier (BRB), might lead to the disruption of photoreceptor homeostasis 2,3 . In the later stage of AMD, patients suffer from macular damage, which can occur as a consequence of geographic atrophy or choroidal neovascularization (CNV). In geographic atrophy, the RPE degenerates, leading to a progressive loss of photoreceptors. The wet form of AMD is characterized by the abnormal growth of new blood vessels that break through the BRB and grow into the retina under the macula 4 . RPE damage and a loss of BRB function comprise common features of both dry and wet AMD, in which vascular endothelial growth factor (VEGF) plays a major pathogenic role 5 .
The RPE is particularly metabolically active, highly oxygenated, and vulnerable to oxidative damage because it is exposed to photosensitizers such as antioxidants and lipofuscin. This sensitivity leads to a variety of age-related changes that reduce RPE function and increase cell death 6 . Among the risk factors for AMD, oxidative stress serves as a key component of AMD pathogenesis. Oxidative stress refers to the progressive cell damage caused by reactive oxygen species (ROS), which contribute to protein misfolding and evoke dysfunction during RPE cell senescence 7 . Numerous studies have reported that the cumulative amount of damage increases with age due to impairments in the DNA repair system along with intensified oxidative stress and decreased antioxidant defense. Moreover, the much less effective recovery systems for mitochondrial DNA damage can cause oxidative stress and the accumulation of the resulting aberrant products 8,9 . However, little is known regarding the specific factors that mediate oxidative stress in RPE cells. One potential candidate, autophagy, is especially crucial for the maintenance of RPE homeostasis, as RPE cells are exposed to sustained oxidative stress 10 . In the pathogenesis of AMD, insufficient digestion resulting from impaired autophagy in the RPE leads to the accumulation of damaged organelles, extracellular drusen deposits, and lipofuscin. However, once the RPE is damaged beyond a critical point, such as in the later stage of AMD, functional autophagy might instead cause cell death and exacerbate the disease 11 .
Thioredoxin-interacting protein (TXNIP), an α-arrestin family member, acts as a multifunctional adaptor protein for different signaling pathways 12 . The primary role of TXNIP is the negative regulation of thioredoxin (TRX) function by inhibiting its reducing capacity and promoting cellular oxidative stress 13 . The inhibition of TRX by TXNIP carries lethal consequences for cells and accelerates destructive inflammation 14 . In human aortic endothelial cells with TXNIP knockdown cultured under high-glucose conditions to promote oxidative stress, a decrease in the amount of ROS generated was observed compared with that observed in control cells. This suggests that high levels of TXNIP inhibit the redox activity of cytoplasmic TRX1 associated with an increase in ROS levels 15 . Conversely, ROS appear to negatively regulate TXNIP expression in vascular smooth muscle cells, as the pretreatment of these cells with antioxidative agents inhibits the TXNIP downregulation that occurs following H 2 O 2 stimulation 16 . These results demonstrate the importance of ROS for the expression of TXNIP 16 .
p53 is a well-known tumor suppressor with the ability to cause cell senescence and apoptosis 17 . Although p53 can indirectly influence cell growth and proliferation by activating cell cycle inhibitors or the transcriptional activation of proapoptotic proteins 18,19 , p53 also regulates autophagy [20][21][22] . For example, autophagy can be induced in a p53-dependent manner in response to genotoxins. p53-induced autophagy occurs through the activation of AMP-activated kinase (AMPK), which results in the rapid, acute inhibition of the mammalian target of rapamycin complex 1 (mTORC1) through the activation of tuberous sclerosis (TSC) 1 and 2 kinases. In addition, p53 activation can also contribute to the long-term inhibition of mTOR by inducing the upregulation of phosphatase and tensin homologue and TSC2 at the transcriptional level 23 . Another mechanism of p53-induced autophagy involves the transcriptional activation of damage-regulated autophagy modulator 20 . Autophagy induced by p53 may facilitate p53 cell cycle arrest activity and synergize with accelerated cell death in response to p53 activation. However, the role of TXNIP in oxidative stress-induced RPE cell dysfunction and the mechanistic link between TXNIP and autophagy remain to be clarified.
In this study, we observed that TXNIP expression was rapidly downregulated in RPE cells under oxidative stress. Notably, we showed that the knockdown of TXNIP induces autophagic flux through increased p53 activation. Furthermore, we demonstrated that TXNIP is involved in neovascularization through the regulation of hypoxiainducible factor-1α (HIF-1α) activation.
Cell culture and transfection
The human retinal pigment epithelium cell line ARPE-19 was purchased from ATCC. The ARPE-19 cells were cultured in Dulbecco's modified Eagle's medium/nutrient mixture F-12 (DMEM/F-12, Gibco) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin in a humidified incubator at 37°C. shTXNIP (NM_006472.3-1666s21c1) and non-target shRNA control (SHC002) cloned in a PLKO.1 lentiviral vector were purchased from Sigma (St. Louis, USA). The overexpression of TXNIP in ARPE-19 cells was achieved by lentivirus-mediated transduction of full-length human TXNIP subcloned into a pLKO.1 and a pLVX-EF1α-IRES-Puro lentiviral vector (Clontech, USA). To generate stable transfectants, the lentiviral vector was cotransfected into Lenti-X-293T (Clontech) cells with virus packaging mix (Sigma) using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions. The virus, along with 5 µg/ml polybrene (Santa Cruz Biotechnology), was added to ARPE-19 cells. After 20 h, the medium was removed and replaced with fresh media containing 3 μg/ ml puromycin (Santa Cruz Biotechnology). Puromycinresistant clones were selected by culturing for 2 weeks in the presence of puromycin. TXNIP expression levels were analyzed by western blotting. For rescue experiments, RNAi-resistant human eGFP-TXNIP was transfected into TXNIP KD cells. The cells were transfected with GFP-LC3 and mRFP-GFP-LC3 with Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions and cultured for 12 h. All experiments were performed 32 h after transfection. siRNA against human LC3 and p53 and nonspecific control siRNA were purchased from Santa Cruz Biotechnology. For siRNA experiments, 1 × 10 6 cells were transfected with 100 pmol of control siRNA, siLC3 and sip53 using the Neon transfection system (Invitrogen) (conditions: 1600 V, 10 ms, 2 pulses) and then cultured for 48 h.
Cell viability assay
The cytotoxicity of H 2 O 2 was assessed by an MTT (M5655, Sigma-Aldrich, USA) assay. Cells (1 × 10 4 cells/ well) were seeded into 96-well plates. After overnight incubation, the culture medium was removed, the cells were rinsed with phosphate buffered saline (PBS), and the cells were treated with the indicated concentration of H 2 O 2 in culture medium containing 1% FBS. After 24 h of H 2 O 2 treatment, 0.5 mg/ml MTT was added to each well and incubated for 4 h to allow mitochondrial dehydrogenase to convert MTT to insoluble formazan crystals. At the end of treatment, MTT was added to the culture medium for 4 h. The medium was then aspirated, and the formazan was solubilized by the addition of 100 μl of DMSO. The absorbance at 570 nm was measured using an enzymelinked immunosorbent assay (ELISA) microplate reader. Each experiment was performed in triplicate and repeated three times to assess the reproducibility of the results.
Cell proliferation assay
The proliferation of wild-type (WT), shCtrl, and shTXNIP ARPE-19 cells was determined using a Wst-1 assay. Cells (1 × 10 4 cells/well) were seeded into 96-well plates. After overnight incubation, the culture medium was removed, and the cells were rinsed with phosphatebuffered saline (PBS). The cells were treated with or without H 2 O 2 in culture medium containing 5% FBS. After a certain period of time (24, 48, or 72 h), 10 μl of Wst-1 reagent (ab155902, Abcam, USA) was added to each well. After incubation with Wst-1 reagent for 2 h, the medium was collected, and the absorbance of the treated and untreated samples was measured using an ELISA microplate reader at 440 nm. Each experiment was performed in triplicate and repeated three times to assess the reproducibility of the results.
Cell cycle analysis
The shCtrl and shTXNIP ARPE-19 cells, were cultured in normal growth medium for 48 h. After 48 h, the cells (4 × 10 5 cells/well) were detached and resuspended in PBS. The cells were stained with 5 μM Draq5 TM (#424101, BioLegend, USA) for 15 min at room temperature. After incubation with Draq5 TM , the cells were spun down and washed with PBS. After the cells were resuspended, DNA content was directly analyzed by FACS.
Cell random motility
The random motility of the cells was determined using specialized bottom-glass confocal dishes (Ibidi, Munich, Germany). Cell suspensions (100 μl) at a density of 1 × 10 4 cells were seeded in confocal dishes for live-cell motility analyses. After the cells were incubated for 6 h, the dishes were transferred to a live-cell incubating chamber (Live Cell Instrument, Seoul, South Korea) at 37°C under 5% CO 2 on the stage of an inverted fluorescence microscope (Olympus, IX81-ZDC) with a UPLSAPO 20X objective lens. Random cell motility was monitored over a 4-h period by capturing images every 10 min; data analysis was performed using MetaMorph version 7.1 (Universal Imaging, Media, PA).
Coimmunoprecipitation (Co-IP) experiments
Cells were lysed in RIPA buffer supplemented with proteinase and phosphatase inhibitor cocktails. The cell lysates were centrifuged for 10 min at 13,000 rpm at 4°C to remove cellular debris. To preclear the cell lysates, 50 μl of normal serum was added to 500 μg of each cell lysate and mixed with 10 μl of a protein A-conjugated agarose bead slurry. The cell lysates were incubated at 4°C under rotary agitation for 1 h. After preclearing, the cell lysates were centrifuged at 4000 rpm at 4°C for 10 min, and the supernatant was kept for the IP experiments. The supernatants were incubated for 4 h with 2.5 μg of an anti-MDM2 or anti-TXNIP antibody at 4°C under rotary agitation. Next, 80 µl of protein A-conjugated agarose beads was added, and the resulting mixtures were incubated for 8 h at 4°C under rotary agitation. The immunoprecipitated samples were washed three times with RIPA buffer, eluted with 2× loading buffer at 95°C for 10 min, and analyzed by western blotting.
Immunofluorescence assays
To visualize tight junctions, focal adhesions and F-actin formation, cells were attached to glass coverslips coated with 10 μg/ml fibronectin. The cells were washed once in PBS, fixed for 10 min in 3.7% formaldehyde, permeabilized for 20 min at room temperature with 0.2% Triton X-100, washed in PBS, and blocked for 30 min at room temperature in 1% bovine serum albumin in PBS. After incubation for 1 h with a ZO-1 (#13663, 1:1000, Cell Signaling) or paxillin (PM1071, 1:500, ECM Biosciences, USA) antibody, the cells were washed in PBS and incubated for 1 h with FITC-conjugated anti-mouse antibodies and TRITC-labeled phalloidin (P1951, 1:2000, Sigma-Aldrich) in PBS. the cells were washed with PBS and stained with DAPI to visualize the nuclei, and coverslips were mounted on the slides. Images were captured with an Olympus DP30BW digital camera and processed using MetaMorph version 7.1 (Universal Imaging, Media, PA).
Enzyme-linked immunosorbent assay (ELISA) for VEGF
The VEGF concentration was assayed in the culture medium using a Human VEGF ELISA Kit (Abcam, UK) according to the manufacturer's instructions. Cell culture medium was collected and supplemented with 2% fetal bovine serum before being stored at −20°C to maintain the stability of VEGF. All reagents and standards were freshly prepared and added during the assay as instructed by the manufacturer. The concentration of VEGF was measured by the color intensity of the solution using a microplate reader (Spectra Max i3X, Molecular Devices, USA) at 450 nm and 570 nm. The readings at 570 nm were subtracted from the readings at 450 nm to allow the correction of optical imperfections. The VEGF concentration was determined by comparing the corresponding readings with those of the standard curve using known concentrations of VEGF. Each experiment was performed in triplicate and repeated three times to assess the reproducibility of the results.
Tubule network formation assay
To examine the paracrine effect of VEGF on RPE cells, the transwell assay system was used. In detail, 1.2 × 10 5 HRMECs were seeded in M199 medium supplemented with 1% FBS in the lower chamber of the transwell compartment coated with growth factor-reduced Matrigel (Corning, NY, USA) in 24-well plates. Then, 1 × 10 6 RPE cells were seeded in the upper chamber of the transwell compartment (0.4 μm pore size, Corning, NY, USA) coated with 0.1 mg/ml bovine fibronectin. After 18 h, tube networks were observed under bright field microscopy, and the relative tube area was analyzed using ImageJ software.
Quantification and statistical analysis
Statistical analysis was carried out using Windows Microsoft Excel 2013. Statistical differences between two experimental groups were calculated using the unpaired two-tailed Student's t-test. A significance level of p < 0.05 was used throughout the study.
Results
The loss of TXNIP constitutes a major cause of the oxidative stress-induced suppression of RPE cell proliferation To elucidate the role of TXNIP in RPE cells under chronic oxidative stress, the cytotoxic effect of H 2 O 2 was first determined. Human ARPE-19 cells were treated with various concentrations of H 2 O 2 for 48 h. As shown in Supplementary Fig. 1a, no apoptotic (or cytotoxic) effect was observed with a concentration of H 2 O 2 up to 0.25 mM. We next addressed whether TXNIP expression in RPE cells is regulated under oxidative stress. Treatment with H 2 O 2 (0.25 mM), a noncytotoxic concentration, rapidly and markedly abrogated the level of TXNIP (Fig. 1a, b) and resulted in a significant decrease in the proliferation in ARPE-19 cells (Fig. 1c). Consistent with decreased proliferation, the expression of cell cycle regulators such as cyclin A and cyclin D1 was reduced in H 2 O 2 -treated ARPE-19 cells (Fig. 1d). To determine whether the loss of TXNIP by H 2 O 2 is involved in the reduction in RPE cell proliferation, we generated TXNIP knockdown RPE cell lines by the lentivirus-mediated transduction of a TXNIP-specific shRNA (shTXNIP) (Supplementary Fig. 1b). We found that the knockdown of TXNIP significantly decreased the proliferation of ARPE-19 cells (Fig. 1e). Consistent with these results, a marked reduction in cyclin A and cyclin D1 was also observed in TXNIP-depleted ARPE-19 cells, whereas the expression of a cyclin-dependent kinase inhibitor 1 (p21 Cip1/Waf1 ) was increased (Fig. 1f). These results indicated that the loss of TXNIP by oxidative stress reduces RPE cell proliferation.
Inhibition of autophagy attenuates the TXNIP loss-induced suppression of RPE cell proliferation
Increasing evidence has shown that autophagy and proliferation can act cooperatively, antagonistically, or synergistically to regulate cellular homeostasis 10 . To explore whether the TXNIP loss-induced suppression of proliferation in RPE cells results from an interplay between autophagy and proliferation, we first evaluated the effects of oxidative stress on the induction of autophagy in ARPE-19 cells. The conversion of microtubuleassociated protein 1 A/1B-light chain 3 (LC3)-I to LC3-II was markedly increased in a dose-dependent manner in ARPE-19 cells treated with H 2 O 2 (Fig. 2a). Moreover, the total cellular expression levels of p62 were timedependently decreased in 0.25 mM H 2 O 2 -treated RPE cells, which is indicative of an increase in autophagic activity, as p62 is selectively incorporated into the autophagosome through direct binding to LC3 (Fig. 2b). Autophagy induction was further supported by the increased number of cytoplasmic punctae expressing green fluorescent protein (GFP)-LC3 fusion proteins in RPE cells (Supplementary Fig. 2a). In contrast, a ubiquitous, diffuse pattern of cytosolic green fluorescence was observed in untreated RPE cells.
Next, to determine whether the loss of TXNIP is involved in oxidative stress-induced autophagy, we examined the conversion and formation of LC3 in (Fig. 2c). In contrast, the overexpression of TXNIP in RPE cells markedly blocked H 2 O 2 -induced LC3 conversion ( Supplementary Fig. 2b, c). Although the accumulation of LC3-II or an increased number of cytoplasmic LC3 punctae indicates the induction of autophagy, this phenomenon may result from the interruption of autophagolysomal maturation or the completion of autophagy 24 . Thus, we performed turnover assays for LC3 to determine whether the overall autophagic flux was induced. As shown in Fig. 2d, TXNIP-depleted ARPE-19 cells exhibited an increase in LC3-II accumulation in the presence of bafilomycin A1, a lysosomal inhibitor, indicating that an increased amount of LC3 in autophagosomes was delivered to lysosomes for degradation. In addition, based on the difference in acidic stability between GFP and red fluorescence protein, TXNIP-depleted ARPE-19 cells displayed more yellow and red puncta compared with that in shCtrl cells, clearly indicating that autophagic flux was increased in TXNIP-depleted ARPE-19 cells (Fig. 2e). To determine the role of TXNIP loss-induced autophagy, TXNIP-depleted ARPE-19 cells were transfected with LC3-specific small interfering RNA (siRNA) or control siRNA (siCtrl), and their respective cell proliferation was compared. LC3-specific siRNA treatment significantly decreased the TXNIP lossinduced suppression of proliferation in TXNIP-depleted ARPE-19 cells (Fig. 2f). These results demonstrate that TXNIP loss-induced autophagy is involved in the suppression of RPE cell proliferation under oxidative stress.
TXNIP loss-mediated p53 activation regulates autophagy
Next, we scrutinized the signaling pathways that are critically involved in TXNIP loss-induced autophagy in RPE cells. Among the various signaling regulators, p53 is considered a crucial factor in autophagy regulation 20,23 . To determine whether p53 is involved in TXNIP loss-induced autophagy, we examined the level of p53 in TXNIPdepleted ARPE-19 cells. As shown in Fig. 3a, a marked increase in p53 levels was detected in TXNIP-depleted ARPE-19 cells compared with shCtrl cells. Notably, p53 was (Fig. 3b). Similarly, these phenomena were also observed in H 2 O 2 -treated RPE cells ( Supplementary Fig. 3a). To determine whether the increase in p53 is directly regulated by TXNIP, TXNIP-depleted ARPE-19 cells were transfected with eGFP-TXNIP. The reexpression of TXNIP in TXNIP-depleted ARPE-19 cells decreased the TXNIP loss-enhanced p53 level (Fig. 3c), suggesting that TXNIP regulates p53 stability or degradation in RPE cells. Considering that decreased cytoplasmic p53 triggers AMPK activation, thereby inhibiting mTOR/Unc-51-like autophagy activating kinase (ULK1) signaling and inducing autophagy 25 , we next demonstrated that the phosphorylation of AMPK at Thr 172 and ULK1 at Ser 555 was increased Fig. 3 TXNIP loss-mediated p53 activation regulates autophagy. a LC3 and p53 expression levels were significantly increased in shTXNIP cells. Cells were lysed and subjected to western blotting using the indicated antibodies. b RPE cells were plated on coverslips coated with 10 μg/ml fibronectin. Immunostaining was performed on fixed cells with an anti-p53 antibody. Scale bars, 50 μm. c shTXNIP cells were transfected with eGFP or eGFP-TXNIP constructs. The cells were lysed and blotted with the indicated antibodies. d The phosphorylation of AMPK at Thr 172 and ULK1 at Ser 555 was increased in shTXNIP cells, but the phosphorylation of mTOR at Ser 2448 and ULK1 at Ser 757 was decreased. e RPE cells were transfected with 10 μM p53 siRNA for 48 h. Cells were lysed and blotted with the indicated antibodies. f RPE cells were transfected with siCtrl or siLC3B and cultured for the indicated time periods. Cell proliferation was analyzed by the Wst-1 assay. g Cell lysates were immunoprecipitated with an anti-MDM2 antibody, blotted, and probed with anti-p53 and anti-TXNIP antibodies. h, i The activation of p53 at Ser 15 was notably increased in shTXNIP cells. Cells were lysed and subjected to western blotting using the indicated antibodies (h). Immunostaining was performed on fixed cells with anti-p53 and phospho-p53 Ser 15 antibodies. Scale bars, 30 μm (i). # p < 0.01. The error bars indicate the SEM in TXNIP-depleted ARPE-19 cells, whereas that of mTOR at Ser 2448 and ULK1 at Ser 757 was markedly decreased (Fig. 3d). To further clarify the involvement of nuclear p53 in TXNIP loss-induced autophagy, TXNIP-depleted ARPE-19 cells were transfected with p53-specific siRNA. p53 knockdown reduced LC3-II accumulation and resulted in a significant increase in the proliferation of TXNIP-depleted ARPE-19 cells (Fig. 3e, f), indicating that the TXNIP lossmediated accumulation of nuclear p53 induced autophagy and led to the suppression of RPE cell proliferation.
It has been shown that the binding of p53 to MDM2 promotes p53 proteasomal degradation, thus providing a negative autoregulatory mechanism of p53 stability and activity 26 . Therefore, we examined whether TXNIP is involved in the interaction between MDM2 and p53. Coimmunoprecipitation experiments revealed that MDM2 strongly interacts with p53 but not with TXNIP in RPE cells. However, TXNIP strongly binds p53 (Supplementary Fig. 3b). Notably, the interaction between MDM2 and p53 was markedly suppressed in TXNIP-depleted ARPE-19 cells (Fig. 3g). Moreover, as the phosphorylation of p53 constitutes an important event in the control of p53 stability by blocking its binding with MDM2 27 , we further demonstrated that TXNIP loss resulted in the increased phosphorylation of p53 at Ser 15 in RPE cells (Fig. 3h, i). Collectively, these results suggested that TXNIP might regulate the interaction between p53 and MDM2 by directly binding to p53 and controlling p53 phosphorylation.
TXNIP is involved in BRB integrity through the regulation of Src kinase activation
A previous study reported that oxidative stress induces BRB permeability dysfunction through the disruption of tight junctions accompanied by an increase in actin stress fibers in RPE cells 28 . To investigate the role of TXNIP in RPE cell junctional integrity and motility, we performed immunofluorescence analysis. This clearly revealed that zona occludens-1 (ZO-1) was expressed at the cell-cell borders in shCtrl cells, whereas discontinuous or disrupted ZO-1 was observed in TXNIP-depleted ARPE-19 cells (Fig. 4a). Moreover, actin stress fibers were clearly increased in TXNIP-depleted ARPE-19 cells (Fig. 4b), indicating that the disruption of tight junctions resulted from stress fiber-induced tension acting on cell-cell junctions. Next, we examined whether the disruption of tight junctions by TXNIP loss has an effect on cell motility. Single cell random motility analysis revealed enhanced cell motility in TXNIP-depleted ARPE-19 cells (Fig. 4c).
To elucidate the mechanism by which TXNIP regulates RPE cell motility, we examined the activation of focal adhesion kinase (FAK) and Src, as Src kinase activity plays an important role in cell motility and RPE cell integrity 29,30 . TXNIP-depleted cells displayed significantly increased phosphorylation of Src, FAK, and paxillin (Fig. 4d). Moreover, cell locomotion is a dynamic process consisting of repeated cycles of the protrusion of the leading edge, the formation of new matrix adhesions, and the retraction of the trailing edge 31 . In particular, previous reports haves indicated that focal complexes (FXs) are involved in protrusion formation at the leading edge, whereas focal adhesions (FAs) are involved in the contractile actomyosin system to pull the cell body and restrain the migration process 32 . To determine the effect of TXNIP on FX and FA formation, we performed immunostaining for FAK and paxillin. This revealed a significant increase in FX, along with a notable increase in the size of FAs, in the cortex of membranes in TXNIPdepleted ARPE-19 cells (Fig. 4e). These results clearly indicate that the formation of larger, more stable FAs with tethered actin stress fibers, which facilitate cell spreading and are strengthened by Src kinase activation, lead to the disruption of BRB junctional integrity.
Loss of TXNIP increases HIF-1α and VEGF expression
Oxidative stress increases HIF-1α expression and prevents its hydroxylation, which ultimately leads to the increased expression of VEGF, the major growth factor that triggers CNV in wet AMD 33,34 . To determine whether TXNIP can regulate HIF-1α expression in RPE cells, we performed immunostaining for HIF-1α. As shown in Fig. 5a, minimal HIF-1α expression was observed in the cytoplasm of control cells. However, the loss of TXNIP in ARPE-19 cells significantly increased HIF-1α expression in both the nucleus and cytoplasm. Consistent with this, western blot analysis showed that HIF-1α expression was increased, whereas the proline hydroxylation of HIF-1α was decreased in TXNIP-depleted ARPE-19 cells (Fig. 5b). Next, to determine whether the increased HIF-1α expression in TXNIP-depleted ARPE-19 cells has an effect on VEGF expression and secretion, we first evaluated the VEGF expression level. TXNIP-depleted cells displayed significantly increased VEGF expression compared with that displayed by control cells (Fig. 5c). Moreover, VEGF secretion was also increased in TXNIPdepleted ARPE-19 cells (Fig. 5d). Similarly, treatment with H 2 O 2 in RPE cells also increased VEGF expression and secretion levels ( Supplementary Fig. 4a, b). We also investigated whether the increase in VEGF secretion induces angiogenesis via paracrine activation. Tubule network formation by human retinal microvascular endothelial cells (HRMECs) cocultured with TXNIP knockdown cells was enhanced relative to that observed in control cells (Fig. 5e). These results indicate that the reduction in TXNIP expression levels in RPE cells by chronic or sustained oxidative stress might promote CNV through the induction of VEGF.
Discussion
In this study, we showed that the downregulation of TXNIP causes RPE cell dysfunction via at least three pathways: (1) the induction of RPE cell autophagy through p53-mediated AMPK activation; (2) an increase in tight junction disruption and cell motility through Src kinase activation; and (3) the induction of CNV by HIF-1αmediated VEGF secretion. Together, these pathways, which are regulated by TXNIP scaffold properties, lead to accelerated AMD pathogenesis. The results of the current study therefore indicate that TXNIP might serve as a therapeutic target for AMD by means of the maintenance of RPE functions.
Oxidative stress has been implicated in AMD pathogenesis, to which autophagy can contribute through the deregulation of cellular defense against such stress. However, aberrant autophagy has been frequently found to be associated with AMD; moreover, autophagy itself is considered a type of programmed cell death, supporting its involvement in cell death in AMD 35,36 . The tumor suppressor protein p53 acts as a key trigger for the induction of autophagy through AMPK activation 25 . Once cellular energy becomes depleted, AMPK is activated and then inhibits mTORC1 through the phosphorylation of TSC2 and raptor, thereby reducing ULK1 phosphorylation at Ser 757 . However, the activation of ULK1 at Ser 555 and Thr 575 by AMPK promotes autophagy 37 . Consistent with these phenomena, we observed that the activation of AMPK at Thr 172 and ULK1 at Ser 555 was increased in TXNIP-depleted ARPE-19 cells, whereas the phosphorylation of mTOR at Ser 2448 and ULK1 at Ser 757 was markedly decreased (Fig. 3d), suggesting that the depletion of TXNIP in ARPE-19 cells induced autophagy through p53-mediated AMPK activation. However, the method by which TXNIP regulates p53 activation remains unclear.
The activation of p53 occurs through multiple mechanisms, including increased protein concentration resulting from decreased p53 degradation, nuclear translocation, and posttranslational modifications such as phosphorylation and acetylation [38][39][40] . In particular, the phosphorylation of p53 at Ser 15 blocks p53 binding with MDM2 27 . Accumulating evidence supports that p53 is inhibited during normal cell growth by MDM2 through Fig. 5 Loss of TXNIP increases HIF-1α and VEGF expression. a RPE cells were plated on coverslips coated with 10 μg/ml fibronectin. Immunostaining was performed on fixed cells with an anti-HIF-1α antibody. b RPE cells were lysed and blotted with the indicated antibodies. c The expression level of VEGF in RPE cells was analyzed by reverse transcription PCR (RT-PCR, left panel) and qPCR (right panel). d The secretion level of VEGF in RPE cells was determined by ELISA. Each experiment was performed in triplicate and repeated three times to assess the reproducibility of the results. e Representative images of capillary-like tube formation in HRMECs induced by coculture with RPE cells after 18 h. The relative wound and tube areas were calculated using ImageJ software. *p < 0.01 versus shCtrl. The error bars indicate the SEM. Scale bars, 30 μm (a) and 100 μm (e) either ubiquitin-dependent p53 degradation in the cytoplasm 26 or the repression of p53 transcriptional activity in the nucleus 41 . In the present study, we also observed that the loss of TXNIP notably increased the phosphorylation of p53 at Ser 15 and inhibited p53 binding with MDM2 (Fig. 3g, h). Under normal conditions, p53 is synthesized in the cytoplasm and transported between the cytoplasm and the nucleus in a cell cycle-dependent manner. It accumulates in the cytoplasm during G1 phase, localizes to the nucleus during the transition from G1 to S phase and is shuttled back to the cytoplasm shortly after the start of S phase 39 . However, our study showed that p53 was strongly expressed in the nucleus in TXNIP-depleted ARPE-19 cells. Moreover, p53-mediated cell growth suppression occurs via the transcriptional activity of the p53 target gene CDKN1A, which encodes p21 42 . Our study clearly showed that TXNIP loss significantly increased p21 expression levels, indicating that the sustained activation of p53 in the nucleus inhibits cell growth through p21 induction.
TXNIP constitutes a scaffold protein that serves as a multifunctional adaptor protein. Mutation in the chromosome region maintenance 1 (CRM1) binding domain of TXNIP suppresses HIF-1α nuclear export in cancer cells 43 . Poly (ADP-ribose) polymerase 1-mediated poly (ADP-ribosyl)ated p53 is unable to undergo nuclear export, as its interaction with CRM1 is impeded, which promotes the accumulation of p53 in the nucleus, where p53 exerts its transcriptional function 44 . Moreover, TXNIP plays important roles in the maintenance of hematopoietic cells. TXNIP interacts directly with p53 via cysteine residues and regulates p53 stability 45 . Thus, we suggest that the activation of p53 might be derived from the blockage of binding with MDM2 or the loss of the CRM1 binding domain of TXNIP, which causes p53 localization to the nucleus.
Tight junctions in RPE cells contribute to a restricted diffusion barrier between the retinal and choroidal perfusion. Oxidative stress has been reported to affect the disruption of RPE junctional proteins. The distribution of ZO-1 and occludin in tight junctions and cadherin in adherens junctions are disrupted by oxidative stress 2,46 . In the current experimental model, the depletion of TXNIP was associated with a defect in cell tight junctions and increased cell motility. This may be related to the RPE detachment observed during early AMD 47 . In our study, the downregulation of TXNIP by oxidative stress resulted in the disruption of ZO-1 along with the reorganization of perijunctional actin rings and an increase in actin stress fiber formation. However, the molecular mechanism by which oxidative stress mediates tight junction disruption in RPE cells remains unknown. In colorectal cancer cells, the oxidative stress-induced disruption of tight junctions is mediated by the activation of cSrc 48 . The TXNIP domain contains Src homology 3 binding domains and PPxY motifs. PPxY motifs in TXNIP interact with Src homology phosphatase-2 (SHP2), which directly regulates C-terminal Src kinase activation. In turn, the activation of SHP2 increases the phosphorylation of Src at Tyr 527 , which inhibits cSrc activity 49 . Herein, we propose that TXNIP-mediated Src kinase activation regulates RPE cell integrity. This is consistent with a report by Spindel et al. that showed that the downregulation of TXNIP in endothelial cells increases the phosphorylation of Src at Tyr 416 through the regulation of SHP-2 plasma membrane localization, which results in increased actin stress fibers and the disruption of cell-to-cell junctions 50 .
The retina is the most metabolically active tissue in the human body and is highly sensitive to a reduction in oxygen tension. Therefore, any disturbances in oxygen delivery into the retina or local occlusive vascular diseases associated with inflammation in the eye ultimately leads to hypoxic conditions in the retina that may elicit the development of AMD [51][52][53] . Moreover, ROS and HIF-1α are directly involved in stimulating angiogenesis, both in tumors and in the retina 54,55 . ROS increase the gene expression of the nuclear transcription factor HIF-1α and prevent the hydroxylation of the HIF-1α protein, which is necessary for the elicited transcriptional activity of factors such as VEGF, a major stimulator of CNV. Consistent with this, VEGF is strongly expressed in surgically excised CNV membranes from human AMD eyes 56 . However, the mechanism by which oxidative stress mediates HIF-1α expression in RPE cells has not been clearly resolved. A previous study showed that the CRM1 region of TXNIP binds with HIF-1α and the ubiquitin ligase von Hippel-Lindau tumor suppressor (pVHL) and that the TXNIP-pVHL-HIF-1α complex is able to promote HIF-1α degradation 43 . Furthermore, the suppression of TXNIP by microRNA-224 leads to the nuclear translocation of HIF-1α 57 . In accordance with these results, our study showed that TXNIP is a negative regulator of HIF-1α expression, independent of its interaction with TRX and ROS. We therefore suggest that TXNIP can form a complex with HIF-1α, which leads to HIF-1α degradation, along with the inhibition of its nuclear translocation and ability to upregulate VEGF.
In conclusion, our results demonstrate that TXNIP plays a significant role in the regulation of p53-mediated autophagy induction and Src kinase-mediated tight junction disruption. Moreover, TXNIP regulates the activation of the transcriptional properties of HIF-1α, leading to increased VEGF secretion, which might promote CNV. Therefore, the dysfunction of RPE cells by a redox-dependent reduction in TXNIP is likely due to a reduction in the scaffold properties of TXNIP. Together, these findings support the potential role of TXNIP as a therapeutic target for AMD.
|
2019-10-15T14:40:00.056Z
|
2019-10-01T00:00:00.000
|
{
"year": 2019,
"sha1": "e0728f3dc8e13bf0248cfb51173ac276509a380e",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s12276-019-0327-y.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "e0728f3dc8e13bf0248cfb51173ac276509a380e",
"s2fieldsofstudy": [
"Medicine",
"Environmental Science",
"Biology"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
42580356
|
pes2o/s2orc
|
v3-fos-license
|
Correlation between relative growth rate and specific leaf area requires associations of specific leaf area with nitrogen absorption rate of roots
Summary • Close correlations between specific leaf area (SLA) and relative growth rate (RGR) have been reported in many studies. However, theoretically, SLA by itself has small net positive effect on RGR because any increase in SLA inevitably causes a decrease in area‐based leaf nitrogen concentration (LNCa), another RGR component. It was hypothesized that, for a correlation between SLA and RGR, SLA needs to be associated with specific nitrogen absorption rate of roots (SAR), which counteracts the negative effect of SLA on LNCa.• Five trees and six herbs were grown under optimal conditions and relationships between SAR and RGR components were analyzed using a model based on balanced growth hypothesis.• SLA varied 1.9‐fold between species. Simulations predicted that, if SAR is not associated with SLA, this variation in SLA would cause a 47% decrease in LNCa along the SLA gradient, leading to a marginal net positive effect on RGR. In reality, SAR was positively related to SLA, showing a 3.9‐fold variation, which largely compensated for the negative effect of SLA on LNCa. Consequently, LNCa values were almost constant across species and a positive SLA–RGR relationship was achieved.• These results highlight the importance of leaf–root interactions in understanding interspecific differences in RGR.
Introduction
Plant species vary widely in their relative growth rate (RGR), even when grown under uniform, close-to-optimal conditions. In recent decades, intensive attempts have been made to determine the physiological and morphological traits that cause inherent differences in RGR. The general approach is to break RGR down into three components, as follows: where SLA is the specific leaf area (leaf area per unit leaf mass), NAR is the net assimilation rate (rate of dry mass production per unit leaf area and time), and LMR is the leaf mass ratio (ratio of leaf mass to total dry mass). NAR may be further broken down into two components: leaf nitrogen productivity (LNP) and area-based leaf nitrogen concentration (LNC a ), as follows: 1990; Lambers & Poorter, 1992;Cornelissen et al., 1996;Atkin et al., 1998;Poorter & Van der Werf, 1998;Reich et al., 1998a;Wright & Westoby, 2000; but see also Shipley, 2006). However, SLA might have a less positive effect on RGR than one would conclude from these analyses because, according to the balanced growth hypothesis (Brower, 1962;Davidson, 1969;Hilbert, 1990;Garnier, 1991), any increase in SLA should inevitably cause a decrease in other components of RGR. The balanced growth hypothesis describes plant nitrogen concentration (PNC) as the ratio of the total nitrogen absorbed by a plant per unit time to the total biomass produced per unit time: where SAR is the specific absorption rate (amount of nitrogen absorbed per unit root mass and time) and RMR is the root mass ratio (ratio of root biomass to total plant biomass). Given that NAR is a product of LNP and LNC a and that mass-based leaf nitrogen concentration (LNC m ) is proportional to PNC (LNC m = aPNC, where 'a' is a constant), Eqn 3 becomes: Since LNC a is a product of LNC m and SLA −1 , this equation is further changed to: Thus, Eqn 4 Equation 4 predicts that any increase in SLA should cause a decrease in LNC a if other variables remain constant. The decrease in LNC a may be compensated if plants decrease their LMR/RMR (leaf : root ratio) with increasing SLA. In either case, however, an increase in SLA should be paid by a decrease in other RGR components, LNC a or LMR, leading to a marginal net positive effect of SLA on RGR.
The reason that many previous studies nonetheless report a strong association between SLA and RGR could be that SLA correlates with other factors that have positive effects on LNC a . If so, the decreases in LNC a caused by SLA are compensated for, and an 'apparent' correlation between RGR and SLA may be established. From Eqn 4, the negative effect of SLA can be compensated if LNP or 'a' negatively, or SAR positively, correlates with SLA. However, LNP may generally positively correlate with SLA (Wright & Westoby, 2000), and nitrogen allocation between organs may not be systematically different between species (Osone & Tateno, 2005b). Thus, the most likely candidate appears to be SAR. In fact, Osone & Tateno (2005b) demonstrated, by means of a comparative experiment and a physiologically controlled experiment, that SAR positively affects leaf : root ratio, leaf nitrogen, and maximum photosynthesis rate. Furthermore, limited studies showed that SLA correlates positively with SAR across species, and the close associations between SAR and RGR reported in these studies also suggest a strong involvement of SAR in determining RGR Garnier, 1991;Poorter et al., 1991;Reich et al., 1998b;Wright & Westoby, 2000). However, how SAR interacts with each of these leaf parameters and how much SAR contributes to RGR remain poorly understood.
The purpose of the present study was to examine the interaction between SAR and RGR components in determining RGR. We hypothesized that SAR is positively associated with SLA across species, and that SAR offsets the negative effects of SLA on LNC a , and thereby allows a positive correlation between SLA and RGR. If these hypotheses are true, however, it is difficult to quantify the gross positive effects of SAR and gross negative effects of SLA on LNC a from the observed pattern of LNC a , because the observed patterns only represent the outcome after these effects of SAR and SLA have been offset. To estimate these opposing effects on LNC a , we performed simulations based on the balanced growth hypothesis. We also investigated whether these hypotheses could apply to other datasets in the literature.
Plant materials
Five deciduous woody species and six herbs typical of open sunny habitats such as secondary forest, abandoned cropland, and roadsides were used in this study. The tree species comprised Prunus lannesiana var. speciosa Makino, Zelkova serrata (Thunb.), Morus bombycis Koidz., Salix integra Thunb. ex Murray, and S. gilgiata Seemen. The herbs comprised two perennial species, Polygonum cuspidatum Sieb. et Zucc. and Boehmeria nipononivea Koidz., and four annuals, Chenopodium album L., Bidens frondosa L., Achyranthes fauriei Lev. et Van, and Ipomoea nil (L.) Roth. Seeds were collected from individual populations in the Tochigi and Ibaraki prefectures, Japan. In the case of Z. serrata, seedlings with a cotyledon and two or three leaves were collected from a field at the Forest Products Research Institute in Ibaraki prefecture in late April 2005.
Growth conditions
The experiments were conducted in a naturally illuminated growth chamber with a temperature and humidity control unit at the Forestry and Forest Products Research Institute in Tsukuba, Japan (36°01′N, 140°25′E), between May and PNC SAR RMR early July 2005. The tree species, which tend to grow slowly, were germinated in advance of the herbs in order to reduce the variation in initial seedling mass between the two groups. Seeds of all tree species except Salix spp. and Z. serrata were germinated on moist vermiculite or moist filter paper, and the seedlings with two leaves were transplanted individually into 1.6 l polyethylene pots filled with washed river sand. The collected Z. serrata seedlings were first transplanted onto moist vermiculite for 1 wk to allow the roots to recover from any damage or injury sustained at the time of collection and then transplanted individually into pots. Seeds of the herb species and Salix spp. were germinated directly in polyethylene pots filled with washed river sand then the seedlings were thinned to only one per pot. Nutrient solution was given to allow the species to absorb nitrogen at their potential rates through the entire experimental period. Since the maximum nitrogen absorption rate reported in an earlier study was 0.1 g N g −1 d −1 for a herb species (Poorter et al., 1991), the nitrogen content of the pots was maintained well above a concentration that allowed absorption at this maximum. In the early stages of the experiments, each plant was given 500 ml of nutrient solution once a day, then later, when the plants were larger, twice a day to avoid nutrient depletion. Every 2 d, the pots were flushed with tap water to avoid salt accumulation. The air temperature in the chamber was maintained between 22 and 26°C over each 24 h period to follow the daily time course of the photosynthetic photon flux density (PPFD). Relative humidity was maintained at 75% throughout. The seedlings were exposed to full sunlight at all times (the chamber was made of glass) and were randomly rearranged within the chamber every 2 d.
Harvesting
Because growth parameters vary with plant size, they should be compared among individuals of a similar size. To ensure plants of a similar size, woody species were harvested at 14 d intervals and herb species at 8 d intervals three times. The first harvest was performed on 8 June 2005 for woody species and 13 June 2005 for herbs. At each harvest, four to six individuals per species were sampled. The plants were divided into leaves, stems, and roots and leaf area was determined immediately using a flatbed scanner and image analysis software (Image J, http://rsb.info.nih.gov.IJ/). The dry mass of each plant part was determined after oven-drying at 70°C for 3 d. The nitrogen content of each plant part was then determined using an automatic N/C analyzer (NC-80; Shimadzu, Kyoto, Japan).
Data analysis
Growth components were calculated following the classical growth analysis method, in which changes in leaf area and biomass between two harvests are assumed to be exponential (Hunt, 1982). LNP was calculated as (log e (X 2 ) − log e (X 1 ))/ (X 2 -X 1 ) × (Y 1 -Y 2 )/(t 2 -t 1 ) where X n is total leaf nitrogen, Y n is plant biomass and t n is time at the nth sampling. SAR was calculated analogously as X n is root biomass and Y n is total plant nitrogen at the nth sampling. For these calculations, the data obtained at the first and second harvests, at which the plant mass substantially overlapped among species (0.37 ± 0.05 g at the first harvest and 2.21 ± 0.28 g at the second), were used. Regression analyses were performed to examine the relationships between growth parameters using the statistical package SPSS (version 12.0 J; SPSS Japan Inc., Tokyo, Japan).
Comparison with other datasets
In order to make comparisons with earlier studies, data were collected from a series of experiments conducted by Poorter and co-workers Poorter et al., , 1991 and Reich et al. (1998a,b) using DIGITIZEIT 1.5.7 (www.digitizeit.de; ShareIT! Inc., Cologne, Germany). LNC a , leaf : root ratio and LNP, which were not presented in these studies, were calculated as the leaf nitrogen content per leaf mass divided by SLA, LMR divided by RMR, and NAR divided by LNC a , respectively.
Using the pooled data collected from the present and previously mentioned studies, direct and indirect relationships between variables were analyzed through Path analyses using the statistical package Amos (version 6.0; SPSS Japan Inc., Tokyo, Japan). Before Path analysis, all data were lntransformed. Based on the balanced growth hypothesis, our path model assumed that SLA, LNP, leaf : root ratio and SAR affect LNC a , and that the leaf : root ratio adjusts the balance between carbon inflow (SLA and LNP) and nitrogen inflow (SAR). The relationships between LNP, SLA, and SAR were set as covariations, that is, with no direct effects between these variables. However, it is conceivable that SLA, in part, directly affects LNP, since it is mostly determined by lamina thickness and the density of the leaf tissues, which affect the net photosynthetic rate per unit leaf nitrogen by influencing the degree of attenuation of photons passing through and the rate of CO 2 diffusion in the leaves (Hikosaka, et al., 1998;Poorter & Evans, 1998). However, anatomical factors are not the only factors affecting photosynthetic nitrogen-use efficiency. Hikosaka et al. (1998) showed that differ-ences in allocation of nitrogen to ribulose-1,5-bisphosphate carboxylase, a key enzyme of photosynthesis, and specific activity of ribulose-1,5-bisphosphate carboxylase also affect photosynthetic nitrogen-use efficiency. Thus, we assumed that SLA and LNP would covary with each other so as to follow the 'plant trait syndrome', that is, multiple plant traits having evolved in a coordinated manner showing a specific combination between species (Reich et al., 1992;Grime, 2001).
Research 421
LNC a (Fig. 2a, broken line). In contrast to the simulation, the measured LNC a showed only a small decrease with SLA ( Fig. 2a, closed circles). This inconsistency suggests that there are one or more factors compensating for the negative effect of SLA on LNC a preventing LNC a of larger SLA species from decreasing. Figure 3 shows the relationships between SLA and the other parameters in Eqn 4. The leaf : root ratio decreased slightly with SLA (Fig. 3a). Although the trend was not statistically significant, this may to some extent compensate for the negative effect of SLA. To estimate how much of an effect this might have, we calculated species LNC a values using Eqn 4, assigning measured species values to SLA and the leaf : root ratio, while fixing other parameters with the values obtained from Z. serrata between species (simulation 2). LNC a values calculated in this simulation (Fig. 2a, open circles) were generally larger than those in simulation 1, but still smaller than the measured LNC a values. This suggests that the leaf : root ratio partly compensated for the negative effect of SLA on LNC a , but there should be factors also affecting LNC a positively. LNP increased with SLA (Fig. 3b). Since an increase in LNP negatively affects LNC a (Eqn 4), this should not compensate for the negative effect of SLA. In a similar manner to simulation 2, the coupled effects of SLA and LNP on LNC a values were calculated assigning measured species values to SLA and LNP in Eqn 4 while fixing other parameters (simulation 3). The calculated LNC a values showed only a slight decrease compared with simulation 1 (Fig. 2b), suggesting that the negative effect of LNP was small, even though LNP showed the same degree of variation as SLA (Fig. 1a,c). The coefficient of the relationship between mass-based leaf nitrogen concentration and plant nitrogen concentration, a, was almost constant across species (Fig. 3c), and therefore had little effect on LNC a (Fig. 2c, simulation 4). Finally, SAR showed a large increase with SLA (Fig. 3d). LNC a values calculated using Eqn 4 by assigning observed values to SLA and SAR with all other parameters kept constant across species were much larger than those in simulation 1 and close to the measured LNC a values (Fig. 2d, simulation 5). This suggests that the negative effect of SLA on LNC a was largely compensated for by SAR.
If SAR does not correlate with SLA, what would be the relationship between RGR and components of RGR? We calculated species LNC a values using Eqn 4, assuming SAR to be constant across species and taking other parameters as measured for each species, and then calculated RGR with Eqn 2, using the obtained LNC a . Because of the smaller LNC a of species with a larger SLA, the calculated RGR increased only marginally with increases in SLA (Fig. 4a). Although LNP by itself had a less negative effect on LNC a (Fig. 2b), the slope of RGR on LNP was also reduced because of the correlation with SLA (Fig. 4b). These results indicate that for there to be a positive correlation between RGR and SLA there must also be a positive relationship between SLA and SAR.
The interactions between SLA and SAR in determining LNC a can be also shown on a graph where species are grouped by SLA (group 1, species with SLA < 30 m 2 kg −1 ; group 2, species with SLA ≥ 30 m 2 kg −1 and SLA ≤ 36 m 2 kg −1 ; and group 3, species with SLA > 36 m 2 kg −1 ; Fig. 5a). Across all species, LNC a showed no significant relationship with SAR but instead showed convergence. However, within the five species with similar SLA (SLA between 30 and 36 m 2 kg −1 , closed circles), LNC a showed a strong positive relationship with SAR (r 2 = 0.89, P = 0.001). As the variation in SLA was small among these species, this increase in LNC a represents the positive effects of SAR on LNC a , excluding the effects of SLA. On the other hand, a between-group comparison showed the effects of SLA on LNC a excluding the effects of SAR. At the same SAR, species with smaller SLA (open squares) generally showed a larger LNC a than the five species described (located above the regression line of the five species), while those with larger SLA (open circles) showed a smaller LNC a (located below the regression line of the five species), indicating that SLA negatively affected LNC a . The apparent convergence of LNC a when all species were included was thus a result of these opposing effects of SLA and SAR counteracting each other. Similar patterns were observed for the leaf : root ratio when grouping species by SLA (Fig. 5b); at a similar SAR, the leaf : root ratio was smaller in larger SLA species, while at a similar SLA, it was larger in larger-SAR species. This suggests that the leaf : root ratio was also affected by SLA and SAR in a coordinated manner. Figure 6 shows the data collected from studies by Poorter and co-workers Poorter et al., , 1991 and by Reich et al. (1998a,b) together with the present data. Note that the parameters were plotted against SAR so as to contrast parameters that affect LNC a negatively on the y-axis with SAR, which affects LNC a positively, on the x-axis. As in the present study, SLA and LNP were strongly correlated with SAR in these earlier studies (Fig. 6a,b), although LNP showed distinct between-study differences. The leaf : root ratio increased with SAR slightly in Poorter's dataset, but showed no consistent relationship in Reich's dataset (Fig. 6c). In Poorter's study, LNC a decreased with SAR only slightly, Fig. 4 Simulated relationships between relative growth rate and specific leaf area (a), and relative growth rate and leaf nitrogen productivity (b). Closed circles, measured values; open circles, simulated values. For simulated relative growth rate (RGR) values, area-based leaf nitrogen concentration (LNC a ) was first calculated using Eqn 4, assuming that parameters other than specific nitrogen absorption rate of roots (SAR) were constant across species (assuming that there was no positive effect of SAR); RGRs were then calculated with Eqn 2, assigning these LNC a values. Regression relationships: (a) y = 0.005x − 0.028 (r 2 = 0.55, P = 0.009) for measured values, y = 0.002x + 0.069 (r 2 = 0.70, P = 0.001) for calculated values; (b) y = 0.028x − 0.050 (r 2 = 0.43, P = 0.017) for measured values, y = 0.013x + 0.060 (r 2 = 0.60, P = 0.003) for calculated values.
Research 423
suggesting that the negative effect on LNC a was largely compensated for by SAR, as in the present study (Fig. 6d). On the other hand, in Reich's study, LNC a decreased with SAR more greatly. At first glance, this was different from the present result, but in fact did not contradict the balanced growth hypothesis. In Reich's dataset, slopes of SLA and LNP on SAR were larger than those in the present study and Poorter's study, indicating that the negative effects of SLA and LNP on LNC a were more dominant over the positive effects of SAR on LNC a than in these other studies. Therefore, if the balanced growth hypothesis is true, LNC a should decrease more steeply than the other datasets. Equation 4 showed that in Reich's dataset, where 3.6-and 3.4-fold variations in SLA and LNP were associated with a 7.6-fold variation in SAR, compensation of the negative effects of SLA and LNP by SAR should be 40% (LNC a should decrease by 60% with increasing SAR). This is almost in accordance with the observed 64% decrease in LNC a in the dataset, suggesting that the behavior of LNC a followed the balanced growth hypothesis. As a result of the combined behaviors of LNC a and LNP on SAR, NAR increased with SAR in the present and Reich's study and showed no consistent pattern with SAR in Poorter's study (Fig. 6e).
To further confirm whether these correlations really reflect the hypothetical causalities, we performed path analyses (Fig. 7). The three datasets were pooled, excluding those for grass species in Poorter's study as they have a specifically smaller leaf : root ratio than the other species when ln-transformed for the analyses; this gave a total of 33 species. The path model that was derived was not significantly rejected (χ 2 = 2.5, 1 df, P = 0.11). Further, each of the path coefficients predicted by the model was significantly different from zero (P < 0.05) and all signs of the path coefficients were in the predicted direction except for that from LNP to the leaf : root ratio. The lack of a significant relationship between LNP and the leaf : root ratio was caused by the relatively large betweenstudy differences in the two variables, which might reflect the differences in experimental conditions between studies. When the path was removed, the fit to data was improved (χ 2 = 4.0, 2 df, P = 0.14).
The present study suggests, however, that these correlations would not hold without an association between SLA and SAR, which compensates for the negative effect of SLA on LNC a . Simulations predicted that, if SAR was not correlated with SLA, the 1.9-fold variation in SLA observed in the present study would cause a 47% decrease in LNC a along the SLA gradient (Fig. 2a), and thus, the SLA showed only a small increase with RGR (Fig. 4a). According to the balanced growth hypothesis (Eqn 4), to fully compensate for the negative effect of SLA (which varied 1.9-fold between species) on LNC a , SAR or leaf : root ratio should vary at least 3.6 (=1.9 2 )-fold, being positively correlated with SLA, or a or LNP should vary 3.6-fold, being negatively correlated with SLA. In the present (r 2 = 0.09, P = 0.78) for all species included, y = 9.67x + 1.18 (r 2 = 0.89, P = 0.010), for species with SLA between 30 and 36; (b) y = 15.4x + 2.5 (r 2 = 0.19, P = 0.16) for all species included, y = 44.6x + 1.5 (r 2 = 0.67, P = 0.056) for five species with SLA between 30 and 36. Bars represent ± SE of the mean. study, 3.9-fold variation in SAR was associated with SLA, which was just enough to compensate for the negative effect of SLA resulting in a minimal decrease in LNC a with increasing SLA (Fig. 2). Negative relationships between SLA and LNC a (or NAR) were previously indicated in earlier studies Biere, 1996;McKenna & Shipley, 1999;Wright & Westoby, 2000;Shipley, 2002). However, to our knowledge, this is the first paper to quantify the negative effect of SLA on LNC a and show that a positive correlation between RGR and SLA requires a positive relationship between SLA and SAR.
confronting effects (Fig. 5b). This might be why the leaf : root ratio or LMR generally does not show a consistent pattern between species in comparative growth analysis (reviewed by Aerts & Chapin, 2000;Poorter & Nagel, 2000).
The simulations indicated that the negative effects of SLA on LNC a were so large that they would cause only marginal net positive effects on RGR (Fig. 4a). This may give the impression that having larger SLA is of little advantage in increasing RGR. However, in fact, our calculations may have overestimated the negative effects of SLA and LNP on LNC a and LMR because we did not consider a functional relationship between LNC a and LNP. Since NAR is generally a curvilinear function of LNC a , with the x-intercept larger than zero (Hirose, 1986;Van der Werf et al., 1993;Osone & Tateno, 2005a,b), LNP is maximized at LNC a when the tangent from the origin touches the NAR-LNC a curve, and is decreased with LNC a moving away from the point. If this relationship were included, the decreases in LNC a of larger SLA species would be less than what was actually calculated, because the concomitant decreases in LNP should alleviate further decreases in LNC a (see Eqn 4).
Theoretically, species differences in the nitrogen absorption rate per root mass can be caused by both differences in morphological (e.g. root surface area or root length per root mass) and physiological properties (e.g. numbers of channels on the root surface) of roots. The correlation between SAR and specific root length (root length per root mass) was reported in Reich et al. (1998b) and Osone & Tateno (2005b). On the other hand, Reich et al. (1998b) showed there were also substantial differences in area-based nitrogen absorption rate (nitrogen absorbed per root surface area per unit time), suggesting an involvement of root physiology to species differences in SAR. In the present study, SRL varied approx. fivefold between species and positively related with SAR (r 2 = 0.41, unpublished data), but the area-based nitrogen absorption rate differed only 1.5-fold between species. Thus, species differences in SAR may be largely caused by species differences in root morphology in the present study.
Generality of the relationships
The datasets from earlier studies Poorter et al., , 1991Reich et al., 1998a,b) also showed positive SLA-SAR and LNP-SAR correlations (Fig. 6a,b). Consistent with the present study, the negative effects of SLA and LNP on LNC a were largely compensated for by SAR in Poorter's dataset (Fig. 6d). However, in Reich's study, the negative effects were only compensated for by 36% (a 64% decrease in LNC a , Fig. 6d). This lower degree of compensation might be caused by differences in the experimental conditions. In Reich's study, plants were given nutrient solution with a nitrogen concentration almost half that of the present study once a day, compared with twice a day in the present study. Therefore, species with higher carbon gain (i.e. higher SLA and LNP) might experience nitrogen depletion as they grow and absorb nitrogen at a lower rate than their potential, while other species might absorb nitrogen near their potential. This should result in a relatively small SAR to SLA (a steep SLA-SAR slope), and thus a decreased LNC a in larger SLA species. From these results, we tentatively conclude that the covariation of SAR with SLA and LNP is a general trend, but that the degree of compensation by SAR may differ between datasets or experimental conditions. To generalize what experimental conditions causes partial or full compensation requires further studies.
There were distinct differences in LNP and the leaf : root ratio between the present and Poorter's study Poorter et al., , 1991; LNP was smaller and the leaf : root ratio was larger in the present study (Fig. 6b,c). This might be because of the difference in light conditions. In the present study, where plants were grown in a naturally illuminated growth chamber, they gradually closed their stomata from early afternoon when the light intensity was maximized (1700 µmol m −2 s −1 ) and accordingly the photosynthetic rate decreased. On the other hand, in Poorter's study, where plants were grown under a constant weaker photon flux density of 315 µmol m −2 s −1 under 14-h-day period, plants were possibly able to photosynthesize at a constant rate during the daytime. This might have caused a larger daily assimilation (NAR), and thus LNP, in Poorter's study than in the present study (Fig. 6b,e), although the average light intensity during the daytime was lower in the former. When carbon inflow is limiting, plants generally increase leaf : root ratio to compensate for the small carbon inflow (Hirose, 1987;Hilbert et al., Fig. 7 A hypothetical path model for leaf and root properties. Singleheaded arrows represent direct relationship or causality. Doubleheaded arrows represent indirect relationship or covariation. RGR, relative growth rate; SLA, specific leaf area; LNP, leaf nitrogen productivity; LNC a , leaf nitrogen per area; L : R ratio, leaf : root ratio; SAR, specific absorption rate. e represents an error term. Numbers beside each arrow represent standardized path coefficients. 1991). Thus, the larger leaf : root ratio in the present study might be a response to the smaller daily assimilation (Fig. 6c). It is notable, however, that the LNC a -SAR relationships were still conservative between the two studies (Fig. 6d), despite the large difference in LNP. These results suggest that the leaf and root physiological properties and biomass allocation are strictly coordinated between species.
Implications for the determinant of RGR
Despite broad agreement on the relationship between RGR and SLA, no consensus has been reached with regard to the relationships between RGR and NAR, and RGR and LMR, in studies of comparative growth analysis (reviewed by Poorter & Van der Werf, 1998). A possible explanation for the inconsistency is the differences in ambient light adopted in different studies (Poorter, 1999;Ryser & Wahl, 2001;Shipley, 2002Shipley, , 2006; see also Medek et al., 2007). In an experiment where plants experience a weak ambient light at which photosynthesis is not light-saturated, the species difference in NAR is small, and thus SLA remains the main determinant of RGR. Conversely, in an experiment where higher ambient light is adopted, species differences in NAR are highlighted, and thus NAR will become the main determinant of RGR. Shipley (2006) provided strong support for this contention by meta-analysis, but also pointed out that substantial inconsistencies between studies still remain that cannot be accounted for by differences in light conditions. The present study may provide a supplementary explanation for this issue. From our hypotheses, the pattern of species LNC a in a dataset depends on the net effects of parameters that affect LNC a both positively (SAR) and negatively (SLA, LNP). Therefore, theoretically, LNC a could either decrease or increase independently with LNP between species. Since NAR is a product of LNC a and LNP, these inconsistent behaviors of LNC a and LNP may interfere with a consistent pattern of NAR between studies. In a dataset where LNC a and LNP are negatively related, a positive relationship between LNP and RGR is always offset to a certain extent by the negative relationship between LNC a and RGR. In this case, the extent of the relationship between RGR and NAR will be determined by the relative strength of the relationships between LNP and RGR, and between LNC a and RGR in the dataset. The datasets of Poorter and Reich in Fig. 6 provide good examples of this. In the two datasets, the relationships between SAR and other parameters presented in Fig. 6 can be equated by the relationships between RGR and the parameters, since SAR was strongly related to RGR (P < 0.001). In Reich's study, LNP increased and LNC a decreased with SAR (and thus RGR), and LNP showed a much steeper increase with SAR than the decrease in LNC a with SAR (Figs 6b,d). Consequently, NAR, the product of LNP and LNC a , showed a positive relationship with SAR (and thus RGR) (Fig. 6e). In Poorter's dataset, LNP and LNC a also showed inverse trends with SAR, but the increase in LNP balanced the decrease in LNC a (Fig. 6b,d). As a result, NAR was almost constant with SAR (and thus RGR) (Fig. 6e). In a dataset in which LNC a does not show a distinct converse trend with LNP, without such countervailing effects between LNC a and LNP, a positive relationship between NAR and RGR might exist. An example of this is provided by the present study. In the present study, LNC a was almost constant to RGR (Fig. 1d), while LNP increased with RGR (Fig. 1c), which results in NAR being positively related to RGR (Fig. 1b). Similarly, Wright & Westoby (2000) suggested that the negative correlation between LNP and LNC a and the different degrees of correlation between studies could cause the lack of a general pattern in the RGR-NAR relationship between studies.
|
2018-04-03T05:22:02.908Z
|
2008-06-28T00:00:00.000
|
{
"year": 2008,
"sha1": "b992ba698433f89fae2a33344f284b6671809708",
"oa_license": "implied-oa",
"oa_url": "https://europepmc.org/articles/pmc7192335?pdf=render",
"oa_status": "GREEN",
"pdf_src": "PubMedCentral",
"pdf_hash": "c2417cfb7843bf28f5334acfc2057082d08abbe8",
"s2fieldsofstudy": [],
"extfieldsofstudy": [
"Medicine",
"Biology"
]
}
|
247548124
|
pes2o/s2orc
|
v3-fos-license
|
Adie’s Pupil: A Diagnostic Challenge for the Physician
Adie’s pupil, also called tonic pupil, is mainly seen in young women. Most patients have unilateral eye involvement. The pupil of the affected side is significantly larger than that on the healthy side. The direct and indirect light reflection from the pupil on the affected side disappears. The pupil on the affected side is sensitive to low concentrations of pilocarpine. The pathogeneses of Adie’s pupil are complex, some of which are insidious and lack corresponding specific diseases. Through a literature review, we found that Adie’s pupil is mainly associated with infectious diseases, most commonly syphilis, followed by immune diseases and paraneoplastic syndromes. The ophthalmological symptoms and pupil abnormalities can disappear after active treatment of the primary disease. Pilocarpine can be used to treat ophthalmologic symptoms, such as blurred vision, for which patients might visit an ophthalmologist or neurologist. It is essential for clinicians to improve their understanding of the disease to avoid misdiagnosis. Differential diagnosis between Adie’s pupil, oculomotor nerve palsy, anticholinergic drug overdose, Argyll-Robertson pupil, and congenital mydriasis need to be identified by the physician. Here, the clinical manifestations, pathogenesis, relationship between Adie’s pupil and diseases, and differential diagnosis of Adie’s pupil are reviewed.
Background
The size of the pupil is adjusted by the pupillary sphincter and dilator. The contraction of the pupillary sphincter muscle when innervated by the parasympathetic nerve fibers shrinks the pupil. However, the contraction of the pupillary dilator muscle when innervated by the sympathetic nerve fibers dilates the pupil [1]. There are 2 types of pupil reflection. The first is the light reflex, which causes bilateral pupils to shrink when 1 pupil is illuminated; the second is the accommodative reflex, also known as the near reflex, which is manifested by the bilateral pupil constriction when staring at a nearby object [2].
Adie's pupil, also called tonic pupil, was first reported by the British neurologist William John Adie in 1931 [3]. It manifests as unilateral or bilateral pupil dilation, direct and indirect loss or weakening of light reflection, abnormal adjustment reflex, and pupil contraction disorder. Adie's pupil is predominantly seen in young women, with an age of onset of 20 to 40 years. Patients usually observe that 1 pupil is larger than the other when looking in the mirror. Unilateral involvement is observed in 80% of patients [4]. Adie's pupil usually exists in isolation, but it should be noted that it can be associated with other symptoms or can be part of other symptoms. When it is accompanied by weakening or disappearance of deep reflexes, it is known as Adie syndrome or Holmes-Adie syndrome [5]. When Adie's pupil is accompanied by weakened or absent deep reflexes and segmental anhidrosis, it is known as Ross syndrome [6]. The complexity of the pathogenesis often leads to difficulty in identifying Adie's pupil and possible related diseases by ophthalmologists and neurologists.
Clinical Manifestations of Adie's Pupil
The characteristics of Adie's pupil can be summarized as follows: the affected pupil is significantly larger than the normal pupil; the direct and indirect reflection of light from the affected pupil disappears; and the affected pupil is sensitive to low concentrations of pilocarpine (Figure 1).
Pathogenesis of Adie's Pupil
Adie's pupil is usually secondary to eye diseases, including infections, tumors, autoimmune diseases, and trauma; however, it can also be associated with systemic diseases with autonomic dysfunction. It is generally considered a peripheral neuropathy caused by damage to the ciliary ganglion and its parasympathetic postganglionic fibers [3]. The ciliary ganglion contains fibers that innervate the ciliary muscle (responsible for adjusting the lens) and the pupillary sphincter. The number of fibers that innervate the ciliary muscle far exceed those that innervate the pupillary sphincter. Thus, when the ciliary ganglion is damaged, the fibers innervating the ciliary muscles have a greater chance of survival. This causes the fibers that originally innervated the ciliary muscles to innervate the pupillary sphincter [40]. Compared with normal pupil fiber regeneration, this abnormal regeneration leads to pupillary contraction and abnormal accommodation.
Relationship Between Adie's Pupil and Diseases
Through the literature review, we found that Adie's pupils are mainly associated with infectious diseases, most commonly syphilis, followed by immune diseases and paraneoplastic syndromes ( Table 1). On encountering Adie's pupils, the first disease that needs to be ruled out is syphilis, followed by other diseases, such as Vogt-Koyanagi-Harada syndrome and Sjogren syndrome, and other infections, autoimmune diseases, and paraneoplastic syndromes (Figure 2). The ophthalmologic symptoms and pupil abnormalities can disappear after treatment of the primary disease. A diagnosis of idiopathic Adie's pupil can be considered after exclusion of all possible primary diseases. Pilocarpine can be used to treat ophthalmologic symptoms, such as blurred vision [40].
Differential Diagnosis of Adie's Pupil
Adie's pupil needs to be distinguished from pupil abnormalities caused by the following factors: Oculomotor nerve palsy: In addition to the dilated pupils and loss of light reflection, there are other clinical manifestations, such as ptosis and restricted eye movements. Brain magnetic resonance imaging can reveal damage to the oculomotor nerve nucleus and fibers [43], and the pilocarpine test is usually negative [44].
Drug overdose: Atropine is an anticholinergic drug that can lead to a series of anticholinergic symptoms after excessive intake, including dilated pupils, hallucinations, agitation, tachycardia, e934657-2 delirium, and fever [45]. In addition, Datura, a traditional Chinese medicine, can also cause the above symptoms when ingested in large quantities [46].
Argyll-Robertson pupil:
It is generally thought to be related to neurosyphilis. Patients usually present with bilateral miosis, loss of direct and indirect light reflection, and presence of accommodation and convergence reflexes. The affected pupil dilates after atropine instillation [47].
Congenital mydriasis: Mydriasis and associated loss of light reflex are congenital, and both pupils can be affected. It is more common in women, and the pathogenesis is unclear. A combination of medical history and negative pilocarpine test can assist in identification of congenital mydriasis [48].
Conclusions
Adie's pupil demonstrates sex predilection, having a greater incidence in women than men. Blurred vision is the main clinical manifestation of this disease. However, some patients have no ophthalmologic symptoms. The characteristics of Adie's pupil can be summarized as follows: the affected pupil is significantly larger than the normal pupil; the direct and indirect reflection of light from the affected pupil disappears; and the affected pupil is sensitive to low concentrations of pilocarpine.
Adie's pupil is mainly associated with infectious diseases, most commonly syphilis, followed by immune diseases and paraneoplastic syndromes. After active treatment of the primary disease, the ophthalmologic symptoms and pupil abnormalities can disappear. Pilocarpine can be used to treat ophthalmic symptoms, such as blurred vision. Differential diagnosis between Adie's pupil, oculomotor nerve palsy, anticholinergic drug overdose, Argyll-Robertson pupil, and congenital mydriasis need to be identified by the physician. Therefore, when evaluating patients with manifestations similar to Adie's pupil, physicians should first identify the diseases mentioned above. Low concentrations of pilocarpine, detailed history, and physical examination can be helpful. The primary disease spectrum behind Adie's pupil should be actively screened.
Acknowledgements
We would like to thank Editage (www.editage.com) for English language editing.
Declaration of Figures' Authenticity
All figures submitted have been created by the authors, who confirm that the images are original with no duplication and have not been previously published in whole or in part. e934657-6
|
2021-12-12T17:47:26.332Z
|
2021-12-06T00:00:00.000
|
{
"year": 2021,
"sha1": "582ee1a21804e7b80a058ad020ad4eb836035998",
"oa_license": "CCBYNCND",
"oa_url": null,
"oa_status": "CLOSED",
"pdf_src": "PubMedCentral",
"pdf_hash": "fb7c731db4cbd68dec18ff3c14029aa116ab4ecd",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
257665383
|
pes2o/s2orc
|
v3-fos-license
|
Short version of the smartphone addiction scale: Measurement invariance across gender
The Smartphone Addiction Scale Short Version (SAS-SV) has been widely used in research, but little is known about the measurement invariance across gender. The current study measured SAS-SV invariance between male and female college students in a sample of 1112 participants. Single- and multiple-group confirmatory factor analyses (CFAs) of smartphone addiction symptom ratings were conducted using R program with RStudio. SAS-SV was psychometrically robust in measuring the severity of smartphone addiction among college students, as well as the gender-based invariance. The differences in SAS-SV between male and female participants were likely to represent true gender differences, and meaningful comparisons could be made.
Introduction
Smartphones have led to dramatic changes in daily activities and behaviors. Applications allow communication with others, e-mail access, enjoyment of music/videos/films, game playing and schedule management. Smartphones may be beneficial, expanding horizons, promoting safety, alleviating stress, maintaining relationships and finding useful information [1][2][3] which has contributed to their indispensability [4]. However, improper smartphone use leads users towards unintentional time-wasting, and immoderate use carries the risk of smartphone addiction with an impact on physical and mental health.
Smartphone addiction describes a lack of control of smartphone use with negative consequences [5], and is considered a technological or behavioral addiction. Smartphone addicts may display the six core performances of addiction: salience, mood modification, tolerance, withdrawal, conflict and relapse [1,6]. Smartphone addiction has been positively correlated with mental distress, such as depression, anxiety, loneliness, stress and boredom in empirical studies [7][8][9][10] and linked to poor sleep quality, impaired learning and acquisition and premature cognitive decline [11]. Adverse physical effects have also been reported, such as dry eye [12], musculoskeletal pain [13], hypertension [14], body dysfunction and weakened immunity [15]. These conditions are associated with a decrease in psychological well-being [16] and reduced life satisfaction [17]. Therefore, measures to prevent or treat this addictive behavior are needed, with the establishment of an accurate diagnostic criterion being the first step. . An interesting outcome of SAS-SV-based research is the gender difference in smartphone addiction severity.
Gender differences were tested during SAS-SV development, but with controversial results. The mean scores of females were significantly higher than those of males [23], and females from Hong Kong were more likely to be addicted to smartphones than males [26]. Similar results were found among Japanese college students [32]. However, other studies have found higher scores for males than females and greater likelihood of addiction among the male population [30, 33,34], and no gender difference has also been an outcome [35][36][37][38]. The last outcome indicated homogeneous symptoms among males and females and equivalent levels of smartphone addiction, although content viewed and motivation/justification might be varied.
The above controversy has arisen due to the assumption that "the measurement and the structure of the underlying construct are equivalent across groups" [39][40][41] and the failure to perform SAS-SV invariance tests. It remains unclear whether the functions of SAS-SV are identical across male and female participants, and thus, verification of the lack of cross-gender measurement bias is required. Measurement invariance of the SAS-SV across gender groups should be examined.
Measurement invariance refers to whether the same metric produces identical measurements under different conditions during observation and research [42]. For a given factorially defined construct, mathematical equality of corresponding measurement parameters, such as loadings and intercepts, across two or more groups would indicate measurement invariance [43]. Indeed, measurement invariance has been considered a prerequisite for the comparison of gender, age, culture and other dimensions [41]. The demonstration of measurement invariance across the cis genders would allow for the meaningful comparison of their mean scores [44]. Measurement invariance testing is often performed through an iterative process. A series of increasingly restrictive confirmatory factor analytic (CFA) models determine the extent to which measurement parameters (loadings and intercepts) are equivalent across different samples or time points [40,45,46]. Associations between manifest indicators and latent constructs, such as regression intercepts, factor loadings and residual (error/uniqueness) variances, are involved. In practice, measurement invariance includes four hierarchically nested parts: configural, metric, scalar and error variance invariance. Accordingly, the four aspects of SAS-SV should be tested to provide evidence for the gender-based measurement invariance of this scale.
Cross-culture and cross-time SAS-SV invariance have been investigated previously, and the measurement equivalence is verified [31,47]. Cross-gender equivalence has been established for the revised Chinese Smartphone Addiction Scale (SAS-RC) [48]. However, this scale has been adapted from the Smartphone Addiction Scale long version [24] and lacks several items of the SAS-SV. University students engage in high levels of smartphone use [49], and addiction is prevalent in China [50,51], as elsewhere. Therefore, the aim of the current study was to examine the cross-gender equivalence of SAS-SV in a population of Chinese university students.
Participants and procedure
Data about participants' levels of smartphone addiction was collected by an anonymous online survey. A total of 1112 college students, 529 of whom were male and 583 female, with a mean age of 20.28 ± 1.43 years (range: 18-25), participated in the study.
Ethical approval was granted by the research ethics committee of the Inner Mongolia Normal University (College of Psychology). Verbal informed consent was given by all participants and their head teachers for the use of anonymized data for research and a link to the questionnaire sent via WeChat or QQ groups during participants' spare time.
Measurements
SAS-SV is a self-reported measure for assessment of smartphone addiction severity. The scale consists of 10 items [23]. All participants were told that "please indicate to what extent you agree that this is true for you", and were asked to answer the following question: "Is missing planned work due to smartphone use?". Responses were recorded by a 6-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Cronbach's alpha for the scale across the whole group was 0.883, with 0.895 and 0.872 for male for female participants, respectively. The construct validity of SAS-SV has been verified in various cultures and countries [23, 25-31], and it is correlated with numerous measures [52,53], thus providing convergent and concurrent construct evidence for the validity of this scale.
Statistical analysis
SPSS 25.0 was used for descriptive statistics, normality evaluation and reliability assessment. "lavaan" [54] and "semTools" [55] in R programming language with RStudio [56] were used for confirmatory factor analysis (CFA) and evaluation of configural, metric, scalar and error variance invariance across genders. Data normality was assessed by skewness and kurtosis. Item scores were regarded as normally distributed if the absolute values of the two indices did not exceed 2 [57]. Weighted least squares means and variance adjusted (WLSMV) estimator was used for single-and multiple-group CFAs [58]. Multiple indices were adopted to examine the goodness of fit of CFA models, including the significance of χ 2 , ratio between χ 2 and degrees of freedom (df), root mean square error of approximation (RMSEA), comparative fit index (CFI) and standardized root mean square residual (SRMR). A good fit of the CFA model to the data was indicated by a non-significant χ 2 , χ 2 /df < 5, RMSEA < 0.08, CFI > 0.95, TLI > 0.95, SRMR < 0.08 [59][60][61][62].
Two single-group CFAs were conducted to assess the fit of the model in the two gender groups separately. Multiple-group CFA was then used to analyze SAV-SV measurement invariance across the two groups. Four hierarchically nested models were constructed and compared in an iterative manner. The raw score means were compared between groups to test whether males and females had an equivalent levels of smartphone addiction.
SAS-SV measurement invariance was assessed via the overall fit for each model and comparisons between nested models. Measurement invariance was supported if fit indices were acceptable and nested model comparisons fulfilled the following criteria: (1) RMSEA change (ΔRMSEA) < 0.050; (2) CFI change (ΔCFI) < 0.004; (3) SRMR change (ΔSRMR) < 0.010 [63,64]. Chi-square change was not a criterion due to its sensitivity to sample size and limited capacity to distinguish invariant from non-invariant models [41,44,45,65] and results are reported merely for completeness.
Descriptive statistics
Descriptive statistics for all male and female participants are shown in Table 1
Single-group confirmatory factor analysis
Single-group CFAs for the whole group and for male and female participants are presented in Tables 2 and 3. SAS-SV yielded adequate model fit statistics of a significant χ 2 , χ 2 /df = 5.227, RMSEA = 0.062 CFI = 0.985, TLI = 0.980, SRMR = 0.029 with standardized factor loadings ranging from 0.581 to 0.782 for the whole sample. Although χ 2 was significant. the value of χ 2 / df was slightly higher than the recommended cutoff value [60]. As the chi-square statistic was sensitive to sample size [41,65] and all other model fit indices were within acceptable ranges, the model was considered to be appropriate.
In order to test whether the unidimensional model is the best fitting model, according to the suggestions of Ronald F. Levant and his colleagues [66], variance composition of the SAS-SV was assessed based on the common factors, hierarchical, bifactor model. However, none of these models could be identified. Therefore, the unidimensional model of SAS-SV was considered acceptable and regarded as the best fitting model for this scale.
Fit indices indicated good data fitting to the model for the male sample with a significant χ 2 , χ 2 /df = 3.634, RMSEA = 0.071, CFI = 0.983, TLI = 0.978, SRMR = 0.032 and standardized factor loadings ranging from 0.586 to 0.802 and for the female sample with a significant χ 2 , χ 2 / df = 3.263, RMSEA = 0.062, CFI = 0.982, TLI = 0.977, SRMR = 0.034 and standardized factor loadings ranging from 0.575 to 0.792. According to the criteria mentioned above, the unidimensional model of SAS-SV was considered acceptable for both male and female groups.
Measurement invariance across gender
Multiple-group CFAs are presented in Table 4. The configural invariance model (M1) was estimated without applying any equality constraint, and the model fit indices suggested good fitting of the data to the model (χ 2 /df = 3.426, RMSEA = 0.066, CFI = 0.983, SRMR = 0.033). Thus, configural invariance was supported.
The metric invariance model (M2) was tested by restricting factor loadings to be equivalent between the two gender groups. The model fit the data well (χ 2 /df = 2.534, RMSEA = 0.053, CFI = 0.983, SRMR = 0.033). ΔRMSEA was 0.013, ΔCFI 0.000 and ΔSRMR 0.000 between M1 and M2 which were all within the recommended guidelines. Thus, metric invariance of SAS-SV across gender was supported.
The scalar invariance model (M3) was assessed by constraining factor loadings and item intercepts to be identical across the two gender groups and good fitting of the data to the model was shown (χ 2 /df = 2.288, RMSEA = 0.048, CFI = 0.985, SRMR = 0.033). ΔRMSEA was 0.005, ΔCFI 0.002 and ΔSRMR 0.000 between M2 and M3 which were all within the recommended criteria. Therefore, SAS-SV scalar invariance across gender was confirmed.
Strict invariance of the scale (M4) was assessed by holding factor loadings, intercepts and residual item variance equal across male and female groups and good fitting of the data to the model was shown (χ 2 /df = 2.425, RMSEA = 0.051, CFI = 0.981, SRMR = 0.036). ΔRMSEA was 0.003, ΔCFI 0.004 and ΔSRMR 0.003 compared with M3, which were all within the recommended criteria. Therefore, strict SAS-SV invariance across gender was demonstrated.
Lastly, the raw score means were compared between the two groups. The results of the independent t test were: t (1110) = -1.01, p = 0.31, indicating that males and females had an equivalent level of smartphone addiction.
Discussion
The SAS-SV has been frequently used to assess the severity of smartphone addiction but without full characterization of cross-gender equivalence. The current study tested SAS-SV measurement invariance among male and female participants. Single-group CFA indicated acceptable data fitting for the unidimensional model in the whole group and in both gender sub-groups, consistent with previous Chinese language and other studies [25][26][27][28][29][30][31]. However, the previous Chinese study was restricted to Hong Kong residents [26] and may be subject to some cultural differences from mainland China. Therefore, testing of SAS-SV factor structure in the mainland Chinese population was required. The present study suggests that this scale is a robust and useful tool for both male and female residents of mainland China. Scalar invariance is regarded as an imperative procedure for comparing latent means, which conforms to the identical operational definition, including equal intervals and zero points across the indicators between the two gender groups [67]. Thus, intercepts of items of the scale were equivalent across gender. In addition, multiple-group CFA demonstrated SAS-SV strict invariance, indicating homogeneous residual variances of manifest indicators (items) across male and female university students [68]. Thus, an observed variable (item) may be predicted from the underlying construct (smartphone addiction) with an identical degree of measurement error across gender [44]. Moreover, raw score means of the two gender groups were compared, and the results indicated that males and females had an equivalent level of smartphone addiction.
SAS-SV's unidimensional structure was found to be equivalent for male and female university students. The configural, metric, scalar and strict invariance measurement invariance tests also demonstrated the gender invariance of the psychometric features. Identical SAS-SV scores from the two gender groups represented equal levels of smartphone addiction in male and female students. Any gender differences might reflect the true difference in the severity of smartphone addiction, but not measurement bias. Therefore, the SAS-SV may be used with confidence to measure and compare the degree of smartphone addiction across gender in Chinese university student samples. Previous research has identified different content preferences between male and female smartphone users, with males preferring games apps and females preferring social media apps [69], However, the current study suggests no significant gender differences in addictive symptoms. SAS-SV items were interpreted in a comparable way by male and female university students and the scale was robust. The current results support the usage of SAS-SV, free from measurement bias, in research and clinical contexts.
We acknowledge several limitations to the current study. First, only measurement invariance across two self-identified genders (male and female) was evaluated, and non-binary and transgender individuals were not taken into account, giving an incomplete view according to Ronald F. Levant and his colleagues [70]. Therefore, future studies are needed to investigate SAS-SV measurement invariance across different gender identities. Second, only college student samples were included in this study, and smartphone addiction is also prevalent in other populations, such as children, adolescents and adults. Thus, gender invariance of the scale should also be investigated in different sample populations. Third, the current study was conducted on a convenience sample, which could reduce the generalizability of the results. Future studies with different sampling methods are warranted.
Conclusions
The structural validity and measurement equivalence of the SAS-SV was assessed by single-and multiple-group CFAs. Empirical evidence is presented to support the unidimensional structure and full measurement equivalence of the SAS-SV across gender. Differences in SAS-SV scores between male and female college students indicated obvious gender differences in smartphone addiction, which were not due to measurement deviation of the scale items. The current study contributes to meaningful interpretations of gender comparisons of smartphone addiction. There are practical implications for researchers, clinicians and the general public.
|
2023-03-23T06:17:31.556Z
|
2023-03-22T00:00:00.000
|
{
"year": 2023,
"sha1": "455128396dff560441eb86139790dfb6d47802df",
"oa_license": "CCBY",
"oa_url": "https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0283256&type=printable",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "843b91aee3a46c0eb2be78c3280edfbfd6146339",
"s2fieldsofstudy": [
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
235712478
|
pes2o/s2orc
|
v3-fos-license
|
Changes in Left Ventricular Ejection Fraction after Mitral Valve Repair for Primary Mitral Regurgitation
This study sought to identify the short- and long-term changes in left ventricular ejection fraction (LVEF) after mitral valve repair (MVr) in patients with chronic primary mitral regurgitation according to preoperative LVEF (pre-LVEF) and preoperative left ventricular end-systolic diameter (pre-LVESD). This study evaluated 461 patients. Restricted cubic spline regression models were constructed to demonstrate the long-term changes in postoperative LVEF (post-LVEF). The patients were divided into four groups according to pre-LVEF (<50%, 50–60%, 60–70%, and ≥70%). The higher the pre-LVEF was, the greater was the decrease in LVEF immediately after MVr. In the same pre-LVEF range, immediate post-LVEF was lower in patients with pre-LVESD ≥ 40 mm than in those with pre-LVESD < 40 mm. The patterns of long-term changes in post-LVEF differed according to pre-LVEF (p for interaction < 0.001). The long-term post-LVEF reached a plateau of approximately 60% when the pre-LVEF was ≥50%, but it seemed to show a downward trend after reaching a peak at approximately 3–4 years after MVr when the pre-LVEF was ≥70%. The patterns of short- and long-term changes in post-LVEF differed according to pre-LVEF and pre-LVESD values in patients with chronic primary mitral regurgitation after MVr.
Introduction
Mitral valve surgery is recommended for chronic primary mitral regurgitation (MR) depending on the patient's symptoms and left ventricular (LV) systolic function according to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) guidelines [1]. LV systolic function is an important factor in determining the timing of surgery, and early intervention can be performed before the onset of LV systolic dysfunction [1]. LV ejection fraction (LVEF) and LV end-systolic diameter (LVESD) are used to assess the LV systolic dysfunction in patients with chronic MR [1].
Some studies have investigated the changes in LVEF after mitral valve surgery in patients with chronic severe MR [2][3][4][5][6][7]. In a previous study that included patients with severe degenerative MR with normal preoperative LVEF (pre-LVEF), LV dysfunction was observed in 18% of the patients immediately after mitral valve repair (MVr) and the LV dysfunction recovered in only two thirds of the patients during long-term follow-up [7]. Another study investigated the changes in postoperative LVEF (post-LVEF) at several time points, and observed an early decrease followed by a gradual increase for up to 5 years after surgery [4]. However, it may be difficult to conclude how LVEF serially changes 2 of 12 postoperatively based only on these studies. Because the measurement time points were arbitrarily divided into distinct groups, the detailed changes within the same time interval could have been neglected in previous studies [4,7].
No study to date has examined the serial long-term changes in LVEF after MVr in patients with chronic primary MR with a wide range of pre-LVEF values. Identifying such changes may be helpful in predicting long-term outcomes after MVr in these patients based on pre-LVEF values. In addition, the extent of the decrease in LVEF immediately after MVr has not been extensively evaluated in patients with chronic MR, although a decrease in LVEF has been known to occur immediately after MVr in patients with chronic MR [4,7]. In the current study, we aimed to investigate the short-and long-term changes in LVEF after MVr in patients with chronic primary MR with a wide range of pre-LVEF values.
Study Population
This retrospective observational study included patients who underwent MVr for chronic MR with grade ≥ moderate caused by excessive motion of mitral valve leaflets, at Asan Medical Center (Seoul, Korea) from January 2005 to July 2015. Patients were excluded if remnant MR with grade ≥ moderate (moderate was defined as proximal isovelocity surface area radius ≥4 mm and <8 mm at the aliasing velocity of 40 cm/s) was detected at the immediate postoperative echocardiography; if they had coronary artery disease, concomitant aortic valve disease of grade ≥ mild, infective endocarditis, rheumatic MR, and acute MR; if they underwent redo surgery, concomitant coronary artery bypass graft surgery, or repeated mitral valve surgery after the index hospitalization; and if they did not have at least one postoperative echocardiographic examination. The research protocol was approved by our Institutional Review Board (AMC IRB 2017-0052), which waived the requirement for written informed consent. Data were acquired from a retrospective review of electronic medical records.
Echocardiographic Data
All patients underwent transthoracic echocardiographic examination before and during the follow-up after MVr. The details of the echocardiographic examination, including the assessment of MR severity, are described in the supplementary methods. Our institution followed the standards and techniques recommended by the American Society of Echocardiography for measuring MR severity [8,9].
Preoperative echocardiographic data obtained closest to the day of surgery were used in the analysis. At our institution, all patients undergo routine echocardiographic evaluation before discharge. The details of the analyzed immediate postoperative echocardiographic data are described in the supplementary methods.
The measurements for LVEF, LV end-systolic and LV end-diastolic volume index (LVESVI, LVEDVI), left atrial (LA) diameter, ratio of peak early diastolic velocity of mitral inflow to mitral annulus early diastolic velocity (E/e'), and the pressure gradient calculated from peak tricuspid regurgitation (PG TR ) were collected.
In addition, we calculated the midwall fractional shortening (mFS) to assess LV contractility. The details of the calculation of actual and predicted mFS, circumferential end-systolic wall stress (cESS), and stress corrected mFS (sc-mFS) are described in the supplementary methods.
During the follow-up, re-developed MR was defined as the recurrence of MR with grade ≥ moderate, regardless of the measurement time point. All-cause mortality was obtained during the follow-up.
The P for trend test using linear regression analysis or Spearman's correlation analysis was performed to investigate whether the immediate postoperative echocardiographic parameters had a linear trend across the four groups. To evaluate the effect of time on post-LVEF, linear mixed models were constructed, with group, time, and the interaction between group and time as fixed effects, and patient effect as random effects. Kaplan-Meier analysis was performed to compare long-term mortality using log-rank sum test, and Bonferroni correction was used as the post hoc test (adjusted α = 0.05/6 = 0.0083). Multivariable Cox proportional hazards regression analysis was performed to evaluate the impact of pre-LVEF groups on long-term mortality. p < 0.05 was considered statistically significant. The other details are described in the supplementary methods.
Results
During the study period, 1030 patients underwent mitral valve surgery for MR. Among them, 569 did not satisfy the inclusion criteria ( Figure 1); thus, the remaining 461 patients were evaluated. No patients underwent heart transplantation or coronary artery intervention during the follow-up. There were 49 patients who were diagnosed with heart failure.
suggested in the ACC/AHA guidelines for patients with chronic primary MR [1].
The P for trend test using linear regression analysis or Spearman's correlation analysis was performed to investigate whether the immediate postoperative echocardiographic parameters had a linear trend across the four groups. To evaluate the effect of time on post-LVEF, linear mixed models were constructed, with group, time, and the interaction between group and time as fixed effects, and patient effect as random effects. Kaplan-Meier analysis was performed to compare long-term mortality using log-rank sum test, and Bonferroni correction was used as the post hoc test (adjusted α = 0.05/6 = 0.0083). Multivariable Cox proportional hazards regression analysis was performed to evaluate the impact of pre-LVEF groups on long-term mortality. p < 0.05 was considered statistically significant. The other details are described in the supplementary methods.
Results
During the study period, 1030 patients underwent mitral valve surgery for MR. Among them, 569 did not satisfy the inclusion criteria ( Figure 1); thus, the remaining 461 patients were evaluated. No patients underwent heart transplantation or coronary artery intervention during the follow-up. There were 49 patients who were diagnosed with heart failure.
A total of 2654 echocardiographic examinations were performed during 3.6 (1.8-7.1) years of follow-up. The immediate postoperative echocardiography, performed at 4.0 (3.0-5.0) days after the surgery, was analyzed in 455 patients ( Figure 1).
The immediate post-LVEF values were 39.0 ± 9.4%, 50.9 ± 9.9%, 53.8 ± 8.6%, and 56.5 ± 7.2% in Gr<50, Gr50-60, Gr60-70, and Gr≥70, respectively (Figure 2A). The higher the pre-LVEF was, the greater was the decrease in immediate post-LVEF (P for trend <0.001) ( Figure 2B). When considering pre-LVESD ≥ 40 and < 40 mm separately in the same pre-LVEF range, the immediate post-LVEF decreased more in patients with pre-LVESD ≥ 40 mm than in those with pre-LVESD < 40 mm (each p < 0.001) (Figure 3). We could not perform this comparison in Gr<50, because there was no patient with pre-LVESD < 40 mm in this group. The result about immediate post-LVEF in patients excluding those with preoperative atrial fibrillation (n = 81) are described in the Supplementary Table S1. The preoperative sc-mFS (pre-sc-mFS) values were >100%, except in Gr<50 ( Figure 4A). Conversely, the immediate postoperative sc-mFS (post-sc-mFS) values were < 100% in all groups, and were similar in the comparison of pairs among Gr50-60, Gr60-70, and Gr≥70 ( Figure 4B). The preoperative sc-mFS (pre-sc-mFS) values were >100%, except in Gr <50 ( Figure 4A). Conversely, the immediate postoperative sc-mFS (post-sc-mFS) values were <100% in all groups, and were similar in the comparison of pairs among Gr 50-60 , Gr 60-70 , and Gr ≥70 ( Figure 4B). The preoperative sc-mFS (pre-sc-mFS) values were >100%, except in Gr<50 ( Figure 4A). Conversely, the immediate postoperative sc-mFS (post-sc-mFS) values were < 100% in all groups, and were similar in the comparison of pairs among Gr50-60, Gr60-70, and Gr≥70 ( Figure 4B). With respect to long-term follow-up, the patterns of changes in post-LVEF differed according to pre-LVEF ( Figure 5A). In Gr <50, post-LVEF increased until about 1 year after MVr, and thereafter showed a plateau of approximately 50% . In Gr 50-60 and Gr 60-70 , post-LVEF increased until about 3-4 years after MVr, and thereafter formed a plateau of approximately 60% in the long-term. Conversely, in Gr ≥70 , post-LVEF decreased and thereafter subsequently increased until approximately 3 years, after which it seemed to decrease to a lower level than that in Gr 50-60 over time. After excluding patients with re-developed MR (n = 59), similar patterns were observed in the long-term changes in post-LVEF (Supplementary Figure S2). Furthermore, we constructed separate models according to pre-LVESD (≥40 versus <40 mm) in the same pre-LVEF range ( Figure 5B). Across all groups, for long-term changes in post-LVEF, higher post-LVEF was observed in patients with pre-LVESD < 40 mm than in those with pre-LVESD ≥ 40 mm. after subsequently increased until approximately 3 years, after which it seemed to de crease to a lower level than that in Gr50-60 over time. After excluding patients with re-de veloped MR (n = 59), similar patterns were observed in the long-term changes in post LVEF (Supplementary Figure S2). Furthermore, we constructed separate models accord ing to pre-LVESD (≥40 versus <40 mm) in the same pre-LVEF range ( Figure 5B). Acros all groups, for long-term changes in post-LVEF, higher post-LVEF was observed in pa tients with pre-LVESD < 40 mm than in those with pre-LVESD ≥ 40 mm. A total of 33 patients died during the follow-up. Kaplan-Meier analysis showed a difference in mortality rate among the groups (p = 0.0019), with the lowest mortality rate in Gr 60-70 ( Figure 6). After adjustment for age, sex, and Charlson comorbidity index, the morality rate was lower in Gr 60-70 than in Gr 50-60 . The hazard ratio of mortality in Gr 50-60 was 3.82 (95% confidence interval, 1.77-8.27; p = 0.001), when Gr 60-70 was considered as a reference group.
A total of 33 patients died during the follow-up. Kaplan-Meier analysis showed a difference in mortality rate among the groups (p = 0.0019), with the lowest mortality rate in Gr60-70 ( Figure 6). After adjustment for age, sex, and Charlson comorbidity index, the morality rate was lower in Gr60-70 than in Gr50-60. The hazard ratio of mortality in Gr50-60 was 3.82 (95% confidence interval, 1.77-8.27; p = 0.001), when Gr60-70 was considered as a reference group.
Discussion
The current study found that the short-and long-term changes in LVEF after MVr differed according to pre-LVEF and pre-LVESD values in patients with chronic primary MR. The principal findings were as follows: (1) the higher the pre-LVEF was, the greater was the decrease in LVEF immediately after MVr; (2) the long-term post-LVEF reached a plateau of approximately 60% when the pre-LVEF was ≥50%, but seemed to show a downward trend after reaching a peak at approximately 3-4 years after MVr when the pre-LVEF was ≥70%; (3) among patients with the same pre-LVEF, the post-LVEF was lower in those with pre-LVESD ≥ 40 mm than in those with pre-LVESD < 40 mm during both shortand long-term follow-ups; (4) the long-term mortality rate was lowest in patients with 60% ≤ pre-LVEF < 70%.
Theoretically, the decrease in LVEF immediately after MVr can be attributed to a decrease in preload, an increase in afterload, or a decrease in contractility of left ventricle. When exploring Gr60-70 and Gr≥70, or the patient groups with a LVEF considered normal for chronic primary MR according to the 2017 ACC/AHA guidelines [1], our results showed that post-LVEDVI decreased more in Gr≥70 than in Gr60-70. This might have led to a greater decrease in post-LVEF in Gr≥70 than in Gr60-70. We found that immediate postoperative cESS decreased more in Gr≥70 than in Gr60-70 (Table 2); therefore, it theoretically makes sense for immediate post-LVESVI to decrease more in Gr≥70 than in Gr60-70, because LVESVI is mainly dependent on afterload. However, our results showed that immediate post-LVESVI did not change in Gr≥70 but that it decreased in Gr60-70. From these findings, we could not exclude the effect of intrinsic LV contractility, which could be revealed in the absence of volume overload, on the immediate post-LVEF.
In this study, we assessed sc-mFS which provided an afterload-independent estimate of LV systolic function [10]. Because motion at the endocardial surface is greater than that predicted by sarcomere shortening alone as a result of cross-fiber shortening, LVEF does not necessarily reflect myocardial contractility [11]. To overcome this drawback, mFS was compared in our study. It is notable that the immediate post-sc-mFS values were <100%
Discussion
The current study found that the short-and long-term changes in LVEF after MVr differed according to pre-LVEF and pre-LVESD values in patients with chronic primary MR. The principal findings were as follows: (1) the higher the pre-LVEF was, the greater was the decrease in LVEF immediately after MVr; (2) the long-term post-LVEF reached a plateau of approximately 60% when the pre-LVEF was ≥50%, but seemed to show a downward trend after reaching a peak at approximately 3-4 years after MVr when the pre-LVEF was ≥70%; (3) among patients with the same pre-LVEF, the post-LVEF was lower in those with pre-LVESD ≥ 40 mm than in those with pre-LVESD < 40 mm during both short-and long-term follow-ups; (4) the long-term mortality rate was lowest in patients with 60% ≤ pre-LVEF < 70%.
Theoretically, the decrease in LVEF immediately after MVr can be attributed to a decrease in preload, an increase in afterload, or a decrease in contractility of left ventricle. When exploring Gr 60-70 and Gr ≥70, or the patient groups with a LVEF considered normal for chronic primary MR according to the 2017 ACC/AHA guidelines [1], our results showed that post-LVEDVI decreased more in Gr ≥70 than in Gr 60-70 . This might have led to a greater decrease in post-LVEF in Gr ≥70 than in Gr 60-70 . We found that immediate postoperative cESS decreased more in Gr ≥70 than in Gr 60-70 (Table 2); therefore, it theoretically makes sense for immediate post-LVESVI to decrease more in Gr ≥70 than in Gr 60-70 , because LVESVI is mainly dependent on afterload. However, our results showed that immediate post-LVESVI did not change in Gr ≥70 but that it decreased in Gr 60-70 . From these findings, we could not exclude the effect of intrinsic LV contractility, which could be revealed in the absence of volume overload, on the immediate post-LVEF.
In this study, we assessed sc-mFS which provided an afterload-independent estimate of LV systolic function [10]. Because motion at the endocardial surface is greater than that predicted by sarcomere shortening alone as a result of cross-fiber shortening, LVEF does not necessarily reflect myocardial contractility [11]. To overcome this drawback, mFS was compared in our study. It is notable that the immediate post-sc-mFS values were <100% in all groups, indicating that LV systolic function might have been impaired during the immediate postoperative period. Myocardial stunning, which occurs after cardiopulmonary bypass, may be a reason for this observation. However, myocardial stunning typically resolves over 48-72 h after ischemia [12]. Considering that immediate postoperative echocardiographic examination was performed at 4.0 (3.0-5.0) days after surgery, we inferred that immediate post-LVEF and immediate post-sc-mFS may represent preoperative intrinsic LV contractility that was masked by a compensation mechanism for LV volume overload, rather than myocardial stunning.
Circumferential fiber contraction has been reported to aid in maintaining global ventricular systolic function in patients with severely impaired longitudinal fiber function, so that LVEF can be preserved [13][14][15]. We showed that the pre-sc-mFS increased with an increase in pre-LVEF and that most of the pre-sc-mFS values were >100%. These results may imply that the pre-LVEF increased with greater activation of the compensation mechanism for LV volume overload. Patients with supra-normal LVEF may potentially have more severe MR and hence substantially reduced afterload, resulting in a higher LVEF. Indeed, our results showed that cESS in Gr ≥70 was lower than that in Gr 60-70 , although there was no statistically significant difference. Furthermore, preoperative PG TR was higher in Gr ≥70 than in Gr 60-70 , with similar E/e' and LA diameter, showing that all values exceeded the normal ranges (Supplementary Figure S1). These findings implied a similarly increased LA pressure but a much higher systolic pulmonary arterial pressure in Gr ≥70 than in Gr 60-70 , suggesting the potential for the development of reactive pulmonary hypertension in Gr ≥70 . We speculated that a preoperative supra-normal LVEF may reflect the condition in which the left ventricle maximally compensates for volume overload. Supra-normal LVEF may be a surrogate for a greater MR burden, rather than being the generally known concept of LVEF.
With respect to long-term changes, post-LVEF reached a plateau of approximately 60% in patients with preoperative LVEF ≥ 50%. Notably, the mortality rate was lower in Gr 60-70 than in Gr 50-60 . The 2017 ACC/AHA guidelines offer LVEF 60% as a cut-off for LV dysfunction, and suggest that mitral valve surgery is reasonable before LVEF reaches 60% in asymptomatic patients with chronic primary MR, with a level of evidence of B or C-LD [1]. A previous study investigated the mortality rate according to pre-LVEF (<50%, 50-60%, and ≥60%) in patients with primary MR after surgical correction and found hazard ratios 2.8 and 1.8 in patients with pre-LVE < 50% and 50-60%, respectively, compared to those of patients with pre-LVEF ≥ 60% [16]. Most previous studies referred in the guidelines adopted LVEF 60% for LV dysfunction and then compared outcomes between surgically and medically treated patients with normal pre-LVEF [17,18], or compared the recovery of post-LVEF with the normal level irrespective of the measurement time points [4,5,7,19]. Few studies have demonstrated the optimal cut-off of LVEF for predicting long-term outcomes. Meanwhile, our findings may suggest the cut-off of LVEF for defining LV dysfunction in terms of long-term outcomes after MVr, supporting the current guidelines. However, adjustment for many possible covariates is needed to determine this association.
In Gr ≥70 , the long-term post-LVEF seemed to decrease after reaching a peak. A previous study reported that pre-LVEF > 60% did not guarantee LVEF recovery during 10 years of follow-up after MVr in patients with primary chronic MR, in that LVEF returned to the preoperative level only in two thirds of patients with postoperative LV dysfunction although all patients showed pre-LVEF > 60% [7]. Moreover, a supra-normal LVEF (defined as LVEF ≥ 65%) was recently reported to show higher mortality rates than LVEF 60-65% in a large, heterogeneous clinical cohort [20]. We think that our results in Gr ≥70 may be in line with this recent report, in that both studies suggest that supra-normal LVEF should not be considered the same as normal LVEF. Further studies are definitely needed to elucidate this issue. Various modalities for LV assessment, including cardiac magnetic resonance imaging to evaluate myocardial fibrosis [21], B-type natriuretic peptide as a biomarker for LV dysfunction, LV global longitudinal strain to assess LV function, left atrial strain as a marker of reversible cardiac dysfunction [22], and ventricular-arterial coupling as a recognized parameter of global cardiovascular performance [23], may improve the understanding of the biomechanics of LV change.
In Gr <50 , further studies including larger numbers of such patients are necessary to identify the long-term changes in post-LVEF, because there were only 15 patients in this group.
LVESD is known to be indicative of reduced LV systolic function in patients with chronic MR [24][25][26]. A previous study demonstrated that postoperative LV dysfunction after correction of MR could be predicted with a reduced pre-LVEF and larger pre-LVESD [2].
Another study showed the additive value of pre-LVESD to pre-LVEF for predicting post-LVEF < 50% after MVr [5]. Likewise, the clinical significance of LVESD was supported by our observation that the post-LVEF was higher in patients with pre-LVESD < 40 mm than in those with pre-LVESD ≥ 40 mm with the same pre-LVEF range.
The present study had some limitations. First, this was a retrospective observational study. The measurement time points for follow-up echocardiography were determined at the physician's discretion. Therefore, the intervals between the measurement time points differed according to individual patients and were variable even in the same patient. Some patients underwent echocardiographic examination less frequently over time after surgery, and the interpretation of these estimates of long-term results may be limited. This should be considered when interpreting the results. In addition, systolic blood pressure measurement was available in a few patients; therefore, the same was true for sc-mFS. Second, since the present study had a small number of deaths, the number of covariates included in the multivariable model was limited to avoid a potential problem of overfitting. Further studies including a large number of patients are needed to confirm independent association between pre-LVEF and long-term mortality. In addition, our result of multivariable analysis showed a wide range of 95% confidence interval that may be attributed to a small number of deaths, which should be taken into account when interpreting the result. Third, we did not have information about MR-related symptoms, symptom duration, or the time interval between the onset of symptoms and MVr, which may be related to the extent of LV remodeling. Therefore, these factors may be indicative of the LV function even in the same LVEF range. Further studies are needed to elucidate these issues.
Conclusions
In terms of short-term change, the higher the pre-LVEF was, the greater was the decrease in the immediate post-LVEF, and the immediate post-LVEF decreased more with larger pre-LVESD in the same pre-LVEF range. In terms of long-term changes, post-LVEF showed a plateau in patients with pre-LVEF > 50%, and lowest mortality was observed in patients with 60% ≤ pre-LVEF < 70%. In addition, there may also be a possibility that post-LVEF showed a decreasing trend in the long-term in patient with pre-LVEF ≥ 70%. Further studies are needed to confirm our findings.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/jcm10132830/s1, Figure S1: Preoperative PG TR , E/e , and LAD in patient with chronic primary mitral regurgitation, Figure S2: Restricted cubic spline curves of LVEF changes in patients in whom MR with grade ≥ moderate did not develop again during the follow-up after MVr for chronic primary MR, Table S1: Immediate postoperative LVEF after excluding patients with preoperative atrial fibrillation. Informed Consent Statement: Patient consent was waived due to all data being acquired from a retrospective review of electronic medical records.
Data Availability Statement:
The data presented in this study are available on request from the corresponding author with a reasonable reason.
Conflicts of Interest:
The authors declare no conflict of interest.
|
2021-07-03T06:17:01.671Z
|
2021-06-26T00:00:00.000
|
{
"year": 2021,
"sha1": "ebb4981c67c53fe57972bfdacd9208d479c15dba",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2077-0383/10/13/2830/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "e00c18eae189e4e500fa6283749f01cd2e2c641b",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
1661650
|
pes2o/s2orc
|
v3-fos-license
|
CLIoS: Cross-lingual Induction of Speech Recognition Grammars
We present an approach for the cross-lingual induction of speech recognition grammars that separates the task of translation from the task of grammar generation. The source speech recognition grammar is used to generate phrases, which are translated by a common translation service. The target recognition grammar is induced by using the production rules of the source language, manually translated sentences and a statistical word alignment tool. We induce grammars for the target languages Spanish and Japanese. The coverage of the resulting grammars is evaluated on two corpora and compared quantitatively and qualitatively to a grammar induced with unsupervised monolingual grammar induction.
Introduction
The localization of spoken dialogue systems is currently gaining interest because of the commercial demand to apply those systems to many different languages. We report project work that starts out from the EU-project TALK 1 , which focused on the development of new technologies for adaptive multimodal and multilingual human-computer dialogue systems and produced (amongst others) the SAM-MIE system (Becker et al., 2006), a flexible spoken language MP3-player interface for in-car application. Given a well designed system architecture, the relevant language-dependent modules of a spoken dialogue system are the speech recognition and the language generation component. We focus on the localization of the speech recognition and interpretation component, at the example of the rule-based grammar of the SAMMIE system. Even though this work was developed focusing a specific application, we expect that it is relevant for many possible dialogue system applications. To construct a speech recognition grammar for a specific language and domain, it is necessary to collect a lot of data to estimate which expressions the user is likely to use, typically in a Wizard-of-Oz experiment. Then, the evaluated data has to be incorporated into the speech recognition grammar. This is time consuming and expensive, and it would be beneficial if the result of data gathering and grammar building in one language could be transferred to other languages (semi-)automatically. In this paper, we focus on the construction of a rule-based grammar for a new target language by porting a grammar from a source language. There are three options to consider: 1. Data collection and construction of target language grammar for every language separately.
2. No data collection, but direct translation of the grammar by a human expert.
3. No data collection and semi-automatical translation of the source grammar. 1 www.talk-project.org Option 1 is very labour-intensive, Option 2 has high error potential and requires human expertise in many fields. Therefore we chose Option 3. The key feature is to separate the task of translation from the task of grammar writing. The translation is done by a human translator whereas the grammar construction is done by an automatic induction algorithm.
In contrast to traditional grammar induction algorithms, which try to find bracketings for a corpus of sentences by judging similarities and differences (van Zaanen, 2000;Kuhn, 2004;Kuhn, 2005), we exploit the syntactic rules and semantic information of the source grammar in addition to the sentence corpus. Grammar induction may not result in a perfect target language grammar. Nevertheless, it saves work. A small test corpus of speech data can be gathered to evaluate the grammar's coverage and add expressions that did not emerge through the translation approach, to improve the quality of the target language grammar. Furthermore, this approach provides us with a parallel corpus for spoken dialogue which may be relevant for other applications as well and enables us to profit from findings in the fields of machine translation and cross-lingual knowledge induction, cf. (Kuhn, 2004;Kuhn, 2005). This paper is organized as follows: the next section explains the grammar induction in detail: sentences and syntax trees are generated, the sentences are translated and the resulting bi-corpus is word-aligned. Using information from the word-alignment, the terminals in the source syntax trees are substituted by the target language terminals, the linear order in the resulting syntax tree is adapted to form a valid target language syntax tree, which is split into production rules. The production rules are merged to obtain the target grammar. Preliminary evaluation results are included at the corresponding positions within Section 2. In Section 3, the final grammar evaluation results are discussed. Section 4 summarizes the results and gives an outlook to further work.
Related Work
Making grammars re-usable for new languages is a goal also followed by (Ranta, 2004) via a "Grammatical Frame-work" (GF), a type-theoretic grammar formalism that addresses four aspects of grammars: multilinguality, semantics, modularity and grammar engineering, and re-use of grammars in different formats and as software components.
In (Johannson, 2006) GF was used to globalize and localize a Swedish grammar for dialogue system utterances to obtain a set of grammars for Swedish, Spanish and English.
In (Perera and Ranta, 2007) GF was used for spoken dialogue system grammar localization, porting the English SAMMIE speech recognition grammar 2 to GF and then introducing multilinguality, thus creating GF grammars for English, Finnish, French, German, Spanish and Swedish. This produced a second German SAMMIE grammar that was compared to the original one, however, the German GF SAMMIE grammar did not match the coverage of the original German SAMMIE grammar. MedSLT (Buillon et al., 2007) is a grammar-based medical speech translation system. The system supports simple medical examination dialogues about throat pain between an English-speaking physician and a Spanish-speaking patient. General feature grammar resources from the REGU-LUS toolkit (Rayner et al., 2006) are compiled into flatter, domain specific grammars, translation is realized via an interlingua. Alignment Based Learning (ABL) by (van Zaanen, 2000) is an unsupervised grammar induction system based on the idea of substitutability. It can be applied to an untagged corpus of natural language sentences and produces a bracketed version of that corpus. By clustering and selecting the bracketing hypotheses, a grammar is induced which covers the original corpus of sentences plus more similar sentences. Our approach is different from both GF and MedSLT in the respect that we do not use resource grammars. Even though resource grammars and the idea of re-using grammars is attractive, we wanted to implement a simple solution to the localization problem that does not rely on the introduction of a framework that requires ample resources in turn. Compared to ABL, our approach requires more resources -the generated sentences, their translations, and the source grammar production rules compared to a monolingual text corpus of the target language only. On the other hand, the grammar which we induce is used for natural language interpretation, whereas ABL can so far only create a grammar that determines if a given sentence is covered by the grammar or not.
Cross-lingual Grammar Induction
The CLIoS (Cross-Lingual Induction of Speech Recognition Grammars) system consists of four steps to successfully induce a target grammar: first, the source grammar must be converted to a corpus of sentences which can be translated by human experts. In our approach, the source language speech recognition grammar is used as a generation grammar which generates the source sentences and their derivations., i.e. the syntax tree 3 containing information on the production rules that were used to generate 2 a manual translation from the German SAMMIE grammar which we discuss in this paper 3 technically a "generation" or "derivation tree".
it. Second, this corpus of sentences has to be translated. Third, a statistical word alignment for the parallel corpus is estimated and manually corrected where required. Fourth, using the bi-corpus of source and target phrases and the original syntactic and semantic rules of the source grammar (in the form of single syntax tree derivations of the source grammar), a new target grammar is induced. This is done by combining the rules with the translated words via alignment projection and adapting the grammar rules to the target language. This is an overview of the tasks necessary for the grammar induction: 1. Generate corpus of source sentences and set of source syntax trees.
2. Translate sentence corpus to target language.
3. Obtain word alignment for the parallel corpus.
4. Induce target grammar by: (a) Substituting source terminals in source syntax trees by target terminals through alignment projection.
(b) Adapting syntax trees to mirror the correct target sentences.
(c) Splitting syntax trees into production rules.
(d) Merging production rules to form the target grammar.
Tasks 3 and 4 comprise the grammar induction phases which are shown in more detail in Figure 1. Five phases are distinguished: in Phase (I), a statistical word alignment is provided by the GIZA++ tool (Och, 2003) and manually corrected where necessary. In Phase (II), the alignments are used to substitute source language terminals in the syntax tree by target language terminals, for details see Section 2.4. The resulting unordered target syntax tree is adapted in Phase (III) to comply with the linear order given in the correct translated target sentence. This is done by the swivel operation (cf. Section 2.4.). In Phase (IV), the re-ordered syntax tree is split into subtrees of depth 1, which can be interpreted as the production rules that define a trivial local grammar, see Section 2.5. This is done for all the sentences in the bi-corpus and results in a large set of production rules. Those rules are merged in Phase (V) to form the target grammar.
Source Grammar Format and Phrase Generation
Our designated source grammar is a context free 4 grammar in Nuance GSL format 5 (Nuance Communications, 2003) that does not contain recursive rules. Grammars with comparably flat syntax and semantics can be modeled without recursion and common speech recognizers (e.g. Nuance Communications, Microsoft) do not allow direct or indirect left recursion in particular; right and middle recursion are used infrequently. grey background indicates the absence of source terminals Figure 1: Induction Phases. Phase (I) shows the statistical word alignment that is used in Phase (II) to substitute source terminals by target terminals in the syntax tree. The syntax tree is re-ordered in Phase (III) and split into local structures which define production rules in Phase (IV). In Phase (V), those rules are merged to form the target grammar.
In our source grammar, NPs are modeled "semantically distinguishable": there is no generic NP production rule, but the NPs also carry their semantic value in their name, e.g. NP ALBU M , NP SON G etc. On the right-hand side of its rules three operators are used, namely concatenation, alternatives, and optionality.
Semantic interpretation is realized via slot filling, whereas slots or interpretation tags are not limited to contain semantic information only; meta data and dialogue state information is also transported via slots. Sentences are produced by generating a syntax tree (with interpretation tags and production derivations) and then projecting the syntax tree to a flat sentence.
Sample Selection
As exhaustive generation with the SAMMIE grammar is impossible due to practical restrictions (the grammar is able to generate about 10 17 different sentences), an optimal generation strategy was chosen that provides much fewer sentences (3856) by leaving out redundant rule expansions (retaining the first expansion data and re-using it at the following expansions), without losing information. The resulting 3856 sentences are close to a minimal expansion of the grammar, which produces a corpus of 1840 sentences, but at the cost of ignoring optional and alternative constructs and omitting terminal symbols.
Phrase Translation
We use the speech recognition grammar to generate phrases, which can be translated easily by a common translation service. This general approach has the advantage that no language pairing is preferred, and that the approach is feasible as long as there is a human translator available for the language pair in question. It also makes it possible to specify the grammar and behaviour of the spoken dialogue system in one language and expect that the specification will be implemented consistently for the different, localized versions of the spoken dialogue system. The benefits of using human translators instead of statistical machine translation include accurateness and the incorporation of context knowledge: if there is a whole sentence to be translated, many of the ambiguities that occur with the translation of single words cease to exist. Also, human translators will produce grammatically correct sentences, which is important to our approach and cannot be guaranteed by machine translation systems. Through the translation of the 3856 sentences, we have obtained a bilingual corpus that is sentence-aligned. In addition to the sentence alignment, the semantic interpretation corresponding to the sentences is known from the interpretation tags. A central prerequisite for the success of our strategy is the cross-lingual validity of our semantic tag categories. The existence of the parallel corpus gives us the opportunity to confirm this prerequisite. Knowledge from the source language can be re-used in a straightforward manner if these semantic tag categories can be re-used (cf. next section). We tested the hypothesis that the interpretation tags are consistent or at least transferable over languages: For the language pairs German-Spanish and German-Japanese, analysis results show that there is either a direct translation for the interpretation feature in the target languages or a corresponding mapping that a human translator will use automatically when translating the source sentence to the target language. The semantic transfer happens automatically via the implicit information that the translator uses. Note that some interpretation categories may not be relevant for certain languages, e.g. the addressing style for English (cf. examples below). The cross-lingual grammar induction approach described here requires that all interpretation categories relevant for the designated target languages are contained in the source grammar even though they are not directly needed there. In our case, porting grammars from German to Spanish and Japanese, this was not necessary, which is one of the reasons why we chose to start out from a German grammar. Examples: 1. phrase mood: phrase moods like "indicative", "imperative", "interrogative" exist in most European languages (Bodmer, 1997). Japanese has similar features and with the translation of phrases, the interpretation tag on phrase moods is meaningful and correct for the target languages considered here.
2. addressing style: "duzen" and "siezen", a German way of addressing a person familiarly or formally by using different grammatical persons, non-existent in English: In Spanish, this distinction is realized similarly as in German, only that the third person singular instead of the third person plural is used to be formal. In Japanese, there are two styles that are mainly used; the simple form that is used when speaking with family and friends, and a more formal style (called desu-masu-style), that is used for polite conversation with colleagues. These two styles were automatically chosen by the human translators to express the corresponding German sentences. If we considered English as a target language for the grammar induction, we would generate a grammar with production rules that distinguish between the formal and the informal addressing style, although the generated and accepted sentences would be identical for both styles. However, the dialogue manager could still make us of this information, for instance to decide whether to call the user by his or her given name or last name. If that distinction is not wanted, there is no harm in having too many production rules, but if the grammar engineer would like to remove the unnecessary productions, he or she could easily attain that goal due to the semantic and syntactic information coded into the production rule name (e.g., by searching the grammar for the keyword "FORMAL").
3. album noun: one example of meta information that the grammar interprets is which noun of a group of synonyms the user utilizes to express a semantic entity, e.g. an album. The system echoes the synonym that the user speaks, and uses the word "album" if the user asks for "albums", but "record" if the user demands for "records". Since it is improbable that those synonym groups contain the exact same number of entries in all languages, and even more improbable that there is a direct one-to-one-mapping between the elements of those synonym groups, this meta information may not be transferred completely automatically, but as all the words in one group mean the same thing, being synonyms, this is not a problem.
Word Alignment and Terminal Substitution
The bilingual corpus resulting from the manual translation is aligned on sentence level. For the terminal substitution, we need an alignment on word level to transfer the syntactic and semantic information from the source to the target language. This word alignment is obtained by the GIZA++ tool (Och, 2003) plus manual correction where necessary. The necessity is determined by human inspection first, but can also be done automatically by finding language-pair and phrase mood typical characteristics. Also, wrong alignment links can be found by inspecting those sentence pairs from which reordering problems emerge, see Section 2.4. According to the alignments found, the source language terminal symbols in the syntax trees are substituted by the target terminals. Source terminals without matching target terminals are deleted, target terminals without matching source terminals are inserted at the appropriate position in the syntax tree automatically, but only after the reordering, see Section 2.4.1.
Alignment Definitions
Terminal substitution occurs in blocks, as we adhere to a general definition of alignments that allows for many-tomany alignments (Och, 2003): an alignment is defined on a source string s J 1 = s 1 , ..., s j , ..., s J that is aligned to a target string t I 1 = t 1 , ..., t i , ..., t I . We define an alignment between the two word strings as a subset of the Cartesian product of the word positions; that is, an alignment A is defined as A ⊆ {(i, j) : j = 1..., J; i = 1..., I}. Such an arbitrary relation between source and target language positions allows for a source word to be aligned to none, one, or many target words and vice versa, where the many source or target words need not form a sequence. An alignment block is a subset of the previously defined alignment relation with the restriction that the one or more source words s y x = s x , ..., s y and the one or more target words t r q = t q , ..., t r that form a connected component are also sequences, i.e. there is no source word s k with x < k < y and no target word t l with q < l < r that is not part of the alignment block. The terminal substitution algorithm is designed to substitute words blockwise, but often, the alignment blocks are trivial one-to-one mappings. Connected components that are not alignment blocks can also be substituted by the algorithm, but pose a problem to the swivel algorithm in Phase III, which is why we implemented a fallback solution (cf. Section 2.4.) for uncontinuous many-to-many alignment mappings.
Alignment Quality
Current cross-lingual approaches work with large corpora, for instance the EUROPARL (Koehn, 2002) corpus which contains 34K sentences and dwarfs our small bilingual corpus by a factor of 10. However, the sentences in our domain are less complex in structure and meaning than the political debates in the EUROPARL corpus, and consist of a smaller vocabulary. In a pilot study, 30 sentences of the generated 3856 were translated manually and word-aligned by GIZA++, using a manually crafted dictionary and resulting in an F-Score 6 of 0,76. In order to calculate precision and recall, a gold standard was established. The comparatively high median frequency of words in our corpus, due to the limited vocabulary, suggested that further improvements in alignment quality could be expected when the complete corpus of 3856 bi-sentences is considered. As Table 1 shows, this is true, the F-Score for the full corpus with dictionary (0,94) proved to be significantly higher than for the mini corpus considered before. To determine the F-Score, we established a gold standard for the full corpus. This took one annotator about 5 hours, as the alignments were already of quite high quality due to the use of a weighted dictionary. The alignments that had to be corrected were mainly function words. The induction process was conducted with the manually corrected alignment where incorrect word alignments were indicated by a problem to the "swivel" operation described in the next section.
Parse Tree Adaptation
To establish order in the modified syntax trees, we compare them to the original target language sentences and change the linear order of the syntax tree's terminals to match the well-formed, translated target sentence. The reordering is done by the "swivel" operation, which changes the order of children in a node, but cannot alter node dominance. Yamada and Knight also used that view on syntax tree structure (Yamada and Knight, 2001), which Gildea compared to an "Alexander-Calder-Mobile" (Gildea, 2003). We found the "swivel" operation to be sufficient to produce valid adaptions of a syntax tree in 97% of the cases, since the trees are of fairly flat structure. However, 3% of the Spanish corpus contained disconnected alignments, i.e. one-tomany alignments where the many terminals did not form an alignment block as defined in Section 2.3.1. Three courses of action were possible: change the alignment paradigm to align only adjacent terminals, eliminate the problematic sentences from our corpus and not consider them for the merging of the grammar, or implement a fallback solution for this special case. We chose to implement an alternate solution for the 135 sentences that could not be reordered completely by the swivel operation, which is to cut the branch off at the appropriate level and reassign it to its rightful parent node. This is done only for the problematic alignment link in question, the remaining reordering of the tree is done via the "swivel" operation.
Insertion of Terminals
So far, the syntax tree contains only the terminals that could be projected across the languages. Therefore, the target terminals that are still missing in the tree have to be inserted. There are three options for the insertion strategy: post-order (attach to the parent of the terminal that precedes the inserted terminal in the correct sentence), pre-order (attach to the parent of the terminal that succeeds the inserted terminal), or in-order (both post-and pre-order, i.e. attach the terminal in question under the first common parent of the preceding and the succeeding terminal). Figure 2 shows an example where an English sentence (Figure 2(a)) is projected to Spanish sentence (Figure 2(b)) by literal translation of the single words. The correct Spanish translation of the sentence, however, reads "Pedro quiere a María" 7 . Figure 2(c) shows the terminal "a" that needs to be inserted -to the resulting string, the insertion strategy is of no relevance, but it affects the structure of the syntax tree, as can 7 In Spanish, prepositions are used for case identification whereas in English, case identification is shown via word order. be seen in Figure 2(d-f), where the syntax tree structures resulting from the different insertion strategies are displayed. Note that the syntactic category that "a" belongs to is not known to us, therefore, the terminal "a" is inserted directly under on of the discussed nodes. Analyzing the correct target sentence with a Part-Of-Speech tagger could probably give us the syntactic category for "a", which could be used to structure the induced grammar more nicely, however, we chose to insert the words directly to avoid overgeneration, as the syntactic category found by a POS tagger would be too general for our approach. We chose the in-order insertion strategy as we expect it to produce the most reliable results because it uses two information sources instead of one (the successor and the predecessor). After the insertion of missing terminals at the appropriate places, the corrected syntax trees reflect a valid derivation for the corresponding target sentences, and implicitly contain the production rules that were needed for their derivation.
Production Rule Merging
Since a grammar is defined by a set production rules (which implicitly contain the information needed for a formal definition, namely set of terminals, set of nonterminals, and start production), we obtain a grammar by combining and merging all the production rules that the grammar should contain. Every syntax tree is split in its participating production rules, all rules are collected, "uniqued" (only one of a set of identical production rules is used) and merged. The sum of merged production rules form the induced target grammar. Merging must take account of the three grammar rule operators introduced in Section 2.1. above. They are treated by the following rules: 1. Alternatives with matching identifiers are merged to alternative lists.
2. Concatenation lists are merged "modulo" optional operators if the non-optional elements are identical.
3. If two concatenation lists cannot be merged, they are "ORed" by adding a new alternative containing the two concatenation lists.
To guarantee the correctness of our implementation, the split and merge operations were tested monolingually by splitting the source sentence syntax trees without substituting terminals, and then merging the resulting productions back together. This resulted in a source grammar equivalent to the initial source grammar. Equivalence was shown by generating 200,000 random sentences with one grammar and successfully parsing them with the other grammar, and vice versa, several times.
Evaluation
We evaluated the complete induced target grammar according to three criteria: 1. Does the grammar correctly parse and interpret the training corpus of target sentences?
2. Does the grammar perform in a sufficient way on a test corpus of user utterances in the target language?
3. How does the grammar compare to a grammar which is obtained by monolingual grammar induction based on the training corpus only (van Zaanen, 2000), with respect to coverage on the training corpus and the test corpus?
Results for Coverage and Interpretation
Coverage and correct interpretation was tested on two corpora for each language: the "training corpus" of the 3856 generated and translated Spanish/Japanese sentences that were used to induce the two CLIoS grammars, and the "test corpus" of Spanish/Japanese user utterances collected via a Wizard-of-Oz experiment.
Training Corpus
We found that of the 3856 sentences from our Spanish training corpus, i.e. the generated and manually translated Spanish sentences, 100% are parsed and interpreted correctly by the induced grammar. As expected, the same can be said about the Japanese training corpus, the induced Japanese grammar covers 100% of the training corpus.
User Utterance Corpus
Designing a grammar is an iterative process: the more potential speakers are consulted on how they would express a given concept, the more possible user utterances emerge. These utterances tend to converge, of course, so that the probability that a given utterance was already made and added to the grammar before converges asymptotically to 100%, but unknown utterances can always emerge and a (non-recursive) grammar can only cover a fix amount of possible phrases. This situation can be compared to Nielsen's view (Nielsen, 1993) on heuristic evaluation, where "individual experimenters can perform a heuristic evaluation of a user interface on their own, but the experience from several projects indicates that any single evaluator will miss most of the usability problems in an interface. However, since different evaluators tend to find different problems, it is possible to achieve substantially better performance by aggregating the evaluations from several evaluators." We can see the subjects in our evaluation as heuristic evaluators of the grammar, finding problems (out-of-grammar utterances). Projecting Nielsen's concept of heuristic evaluation to our case, complete grammar coverage over all possible utterances is improbable to be achieved for small numbers of evaluators. The user utterance corpora were collected via a Wizard-of-Oz-style experiment with native speakers, 10 subjects for Spanish and 5 for Japanese. The subjects were shown the German SAMMIE system and received a short introduction relying to them the possible actions within the MP3 domain which they could carry out by speaking to the system (e.g. listening to an album or modifying playlists). Then, the investigator explained the tasks to them in German to prevent delivering certain words to the subjects that should be chosen freely, and the subject formulated what he/she would say in this situation in the car, in his/her native language (Spanish/Japanese). If the subject could think of several possible utterances, all the utterances that were made entered the evaluation corpora. The whole session was recorded and the user utterances were transcribed manually. For Spanish, this resulted in a corpus of 281 utterances for 27 different tasks, for Japanese we collected 135 utterances for 27 tasks. Figure 3 shows a picture of the experimental setup. Figure 3: Experimental Setup. The subject in the mock-up system's driver seat, investigator recording utterances and explaining tasks from passenger seat.
Of the 281 Spanish utterances, 17 (6,04%) could not be interpreted by the induced grammar. Analysis showed that of the 17 problematic utterances, 15 (88%) addressed command tasks such as stopping playback or scrolling the screen. The problem source was the use of words that do not exist in the grammar so far, to some extent because the words that the subjects used were colloquial or uncommon 8 .
To summarize: 93,96% of the Spanish user utterance corpus was parsed and interpreted correctly. The remaining 6.04% could be inserted easily into only 5 different production rules of the induced grammar. Of the 135 Japanese utterances, 121 (89,63%) were accepted and interpreted correctly by the induced grammar.
Similarly to the Spanish utterances, one problem source was the use of words that do not exist in the grammar so far, but could be added easily. Another problem was that some users chose to say "I-want-to" sentences that would formally begin with "watashi-wa" ("I"), but left out the "watashi-wa", which was not an option in our induced grammar. By making this part of the production rule optional after the induction, which was possible through simple search and replace commands, the utterances lacking the "watashi-wa" could be parsed and interpreted correctly.
Results for Comparison of Monolingual and Cross-Lingual Grammars
We used ABL (van Zaanen, 2000) to monolingually induce a grammar (called ABL-ES) from the 3856 Spanish sentences, and likewise a Japanese ABL grammar (ABL-JP) from the 3856 Japanese sentences. We compare the two ABL grammars to the two created by cross-lingual induction (called CLIoS-ES and CLIoS-JP) with respect to coverage on the training corpus and the test corpus. However, the ABL framework does not permit the interpretation tags that are projected to the CLIoS grammar from source to target language via the word alignment. Thus the ABL grammar can be used for syntactic analysis rather than semantic interpretation. Qualitatively, it can be said that the ABL grammars are more difficult to comprehend by humans than the CLIoS grammars, first due to the lack of production rule names which convey an idea of what to expect from the right side of the production rule to the grammar engineer, and second due to the amount of recursive rules which complicate human comprehension.
Left recursion in the ABL grammars had to be removed as common speech recognition grammar compilers, in our case Nuance, cannot handle left recursion (Moore, 2000;Nuance Communications, 2003). The left recursion was removed by using the algorithm from (Moore, 2000) as the standard algorithm could not be applied due to memory demands resulting from the complexity of the induced ABL grammars. Table 3 shows a comparison of size in KB and the number of nonterminals for the six respective grammars (ABL induced with left recursion for Spanish and Japanese, ABL induced without left recursion for Spanish and Japanese, and CLIoS induced for Spanish and Japanese Table 3) and the ABL-ES grammar parses only 16,35%. Of course, the ABL-ES grammar can not parse more than the 93,96% that the CLIoS-ES grammar parses, as we already established that the problem with these utterances was the use of words unknown to the grammar, i.e. words that did not occur in the target sentence corpus. For Japanese, the CLIoS-JP grammar correctly parses 89,63% of the user utterance corpus and the ABL-JP grammar parses 11,57%.
Conclusion and Outlook
We have presented a strategy for the localization of speech recognition grammars that separates the task of translation from the task of grammar generation (CLIoS). The approach taken is a pragmatic combination of automated NLP methods and manual translation. A speech recognition grammar is induced cross-lingually for the target language by using the production rules of the source language, manual translation and a statistical word alignment tool. Two grammars were induced and evaluated from a generated corpus of 3856 German sentences and their Spanish/Japanese translations. Evaluation showed that the two induced grammars correctly parse and interpret 100% of the training corpora, both Spanish and Japanese respectively. Of the test corpus of collected user utterance corpora for both languages, the Spanish grammar successfully interpreted 93,96% and the Japanese grammar 89,63%. We were able to show that the CLIoS approach results in grammars that are easier to read by humans and therefore easier to improve afterwards than a current state of the art monolingual unsupervised induction approach (ABL). In Table 3: Comparison of the grammars induced by ABL and CLIoS. ES stands for Spanish, JP for Japanese, LR stands for the left recursion contained in the original ABL grammars that could not be processed by the Nuance compiler (hence no coverage data). Note that the size and the nonterminal numbers for the CLIoS grammars are taken from their BNF forms instead of the EBNF forms, to allow for a meaningful comparison.
addition to that, the CLIoS grammars performed much better on a test corpus of user utterances than the ABL induced grammars.
We evaluated the method proposed in this paper on the interpretation grammar of one specific dialogue system. However, the approach will clearly apply with similar results to other grammars, as long as they have no recursive rules and a comparably shallow syntax. Due to the shallow syntax, realizing semantic interpretation via slot-filling works well, therefore the semantic interpretation slots are integrated into our approach and mapped across languages. It will be a matter of future research to investigate how the induction approach generalises to grammars with a deeply structured syntax and more complex semantic interpretation rules.
Acknowledgements
We would like to thank Menno van Zaanen for letting us use his ABL grammar induction and for helpful discussions, and Robert C. Moore for removing the left recursion from the ABL-induced grammars. We are grateful for Stephan Thoma's invaluable programming help and for the fruitful discussions with Sebastian Padó and Hans Leiß. The SAM-MIE grammar was written for the TALK research project 9 by Jochen Steigner, Peter Poller, and Rosemary Stegmann.
|
2014-07-01T00:00:00.000Z
|
2008-05-01T00:00:00.000
|
{
"year": 2008,
"sha1": "624999af830e8c1b38b57dd3165e48ee9579c91e",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "ACL",
"pdf_hash": "624999af830e8c1b38b57dd3165e48ee9579c91e",
"s2fieldsofstudy": [
"Computer Science",
"Linguistics"
],
"extfieldsofstudy": [
"Computer Science"
]
}
|
267212355
|
pes2o/s2orc
|
v3-fos-license
|
Opportunity Spaces for Children ́s Informal Learning in Public Environments
This paper introduces my PhD research on how a tangible and embodied technological layer can enhance and enlarge an existing public space to promote children’s informal learning (IFL). IFL is increasingly recognized as the basis for education, as children learn through social, creative, and playful interactions. IFL is rooted in daily lives and expands the children´s sphere of action beyond their confined social spaces in public and educational institutes. Using the urban environment makes education more accessible to a broad group. Therefore, I will create conditions and structures that make informal educational processes possible and available. I seek to achieve opportunity spaces by fostering alternative experimentational approaches. Through implementing design interventions in the public, I will playfully appropriate the existing and create opportunity spaces for IFL. The design intervention should open up children’s creativity, encourage social interaction, and enhance mental health, which is often neglected in the education system.
INTRODUCTION
Informal learning (IFL) is a spontaneous and unplanned method of acquiring knowledge and skills through everyday experiences [2].Technology can create additional layers to existing surroundings, reimagining and reappropriating built environments [26].Interaction technologies have been used in educational and developmental settings to support and improve learning [7,14].The proposed PhD aims to investigate creating children's opportunity spaces for IFL by adding a technological layer onto existing spaces through environmental appropriation.Interventions will be designed to playfully appropriate public areas, fostering children's creativity, encouraging social interaction, and enhancing mental health.The research will combine concepts from multiple disciplines, including ethnomethodological observations, critical artistic research, and design speculations.The research will investigate the impact of a critical, playful, and somatic perspective on the environment children inhabit.The outcomes will include interaction strategies, design interventions, and recommendations on designing everyday appropriation to enlarge children´s opportunity space through technology.
MOTIVATIONS
I am a dyslexic and creative individual who struggled with learning in traditional school settings.Why that?Because I am a playful experiencer who makes use of the body to perceive and sense the world around me.I do have a background in speculative design, social design, and exhibition design, where I create objects and spaces but also invite people to participate in workshops to discuss and design together on how we want to inhabit our world.In my PhD, I will create spaces for and with children who want to experiment through alternative lenses, supporting their creativity and allowing them to explore the colorful world of learning in public spaces.The thesis aims to integrate IFL into a technologyenhanced environment, promoting a playful, free, and creative Figure 2: Through the appropriation of space, an additional tech layer will be topped onto the base layer of the public space, whereas an opportunity space can evolve for IFL learning method [1,8,10,11].Therefore, I will create interventions for children's appropriation, allowing them to open up their creativity, encourage social interaction, and enhance mental health, as these categories are often left out in formal institutions [1,3,7,22].I will also draw on a tangible and embodied approach to create alternatives to conventional education.The design language of the envisioned opportunity spaces should be expanded to include multiple sensory levels, such as sound, touch, or smell.The use of technology in public spaces allows for expanded, child-centred planning and can take into account the sensory and qualities of the physical space [23].I want to ensure that my research includes the impact of co-existing bodies, objects and buildings on children's environments [21].The city is a living space for many identities, people, genders, groups, and minorities, and immersive technologies offer opportunities to overlay digital experiences onto the physical world or create expanded places for more than one reality [6,17,19,25].The vision of the project is to create a playful, tangible, embodied, sonic, or sensual space that supports informal learning in public space and transforms traditional pedagogic approaches for a greater variety of personalities and identities [14].
RELATED WORK 3.1 Informal Learning and the Use of Playfulness
Children spend 20% of their days in school, while 80% of their days are spent outside of institutional contexts, where they experiment, explore, and interact socially [9].Education can be divided into formal, non-formal, and informal learning (IFL), which includes experimental, incidental, and random learning [22].IFL is nonstructured, non-plannable, and cannot be evaluated [22,24].It can be found at home, on playgrounds, in schools, on public spaces, with family, friends, objects, digital media, and through playful interaction [4].Children create spaces through their playful activities in existing surroundings that they (re)appropriate and temporarily inhabit.They learn in those spaces and create the foundation of their knowledge [4].I will look at literature and design work that promotes the use of technology with a playful, social, and mindful learning concept.Research has shown that the combination of creativity with technology can have significant educational potential [10].The optimal conditions for creative and playful learning are presented through self-directed, object-based play with a selfdetermined objective.Parten's and Corsara focused on social behavior, interaction strategies, and sharing routines [2,18] and Wyeth also included in her study the children's interactions with each other, with the environment, the use of resources during creative and constructive play [29].Building up on that, Wyeth identified implications for designing playful technology, emphasizing that technology should be open for different uses and interaction opportunities [29].Furthermore, she also identified three types of free play activities through her observations: calm, play, and artistic.Jarusriboonchail et al. defined several points when designing with technology, including flexibility, simplicity, open-endedness, anytime visitability, and larger design creations in physical dimension [10].Tabel et al. suggested that children's free play and creative learning are influenced by social interaction and collaboration, allowing multiple users to participate in the creation process and that technology should be easily imitated, allowing children to learn from their peers' creations [10].
The Appropriation of Urban Space
Children sometimes playfully create a place for themselves, using structures and objects to transform existing spaces and appropriate the place [4].They invent places to linger, play, interact, and be creative.Studies also confirm that there are potential and benefits to creating space for children´s education and playfulness in the public environment [4,9].According to Hassinger et.al., revitalizing public space can be important for unprivileged children.Open space can be designed as an embodied learning landscape and an expansion of the educational surroundings.As some children do not have the same educational possibilities as others, access to a public learning space can also be a valuable, equitable opportunity for them to learn outside educational establishments [9].
The unique qualities of public places are a result of how people utilize, appropriate, reinterpret, and define them.The use of the public space is a citizen's right, as Rodriguez supports [12].He further defines the usage as the temporary act of using public places for individual or group activities other than those for which they were initially intended.The term "appropriation" is rooted in environmental psychology to refer to an ephemeral phenomenon that suggests a dynamic procedure of interaction between the person and their environment [12].
It is worth noting here that cities have a predefined layout, public spaces are clearly defined, and the in-between spaces with no connotation are limited [19][20][21]28].The planning of the cities is often not oriented towards the demands and needs of the young generation, who is also hardly involved in the design process of future urban space [9].
Multiple layers of Technology
Technology can be used to (re)appropriate and revitalize (public) space [17].The trend of integration into the built environment of existing cities is more and more going towards ubiquitous computing, embedded systems, and the Internet of things [27].Also, the research field of Urban Interaction design, Architectonic Interactions and Media Architecture focuses on technology with multiple layers.As Wiberg states, the new opportunities that appear in the physical environment open space for reimagining and reappropriating the built environment [26].The main objective of media architecture is the technological enhancement of public space to improve how people engage with and claim ownership of the environment.Nguyen et al. describe that media architecture focuses on humancentered urban technology, which also often uses playful designs, asks humans to liner, helps to explore new locations, invites people for mental rest and fosters a feeling of community through social contact [17].Implementation options are using artefacts, visual or acoustic elements that implicitly or apparently involve the user in interacting with their physical environment [6,17,23].
Tangible and haptic interaction interfaces in the environment are considered highly beneficial for children´s development, especially with the emergence of screen-based technology in recent years [7,8,13,15].Liang et al. conducted studies on Tangible User Interfaces for children's creative learning education and psychology in Human-Computer Interaction [15].Jarusriboonchai et al. state that technological education has the power to cultivate, but traditional digital tools and curricula do not really foster open-ended imaginative, playful, and creative behavior [10].Common educational kits mostly teach the basics of coding, whereas Tangible User Interfaces designs sometimes do not allow abstractness, openness, richness, and complexity [15].Within digital learning in schools, the technology lacks the possibility of free play and constructive play ´in the wild´and often follows the restrictions of the predefined learning purpose.Studies investigated how technologies can be included more naturally into the ways children are already playing and self-directed learning [10].
Build on Theoretical Background
My work aligns closely with the previously mentioned research studies of the related work.The motivation of my thesis is to integrate IFL which includes experimental, incidental, and random learning into a surrounding enhanced with technology [22].Working on the topic of education I will promote a playful learning approach as free, constructive, open-ended and creative learning [7,8,11].I will be using some guidelines and lenses of the related work for the observations, artistic research and design interventions.I will find out how to naturally implement a tech layer into the spaces children inhabit to provide an opportunity space for IFL.The use of media architecture, tangible and embodied interaction technology will support my research intentions to open children's creativity, encourage social interaction, and enhance mental health in public spaces.Through my research, I will contribute to literature and design works that promote the use of technology with a playful, social, and mindful learning concept.
APPROACH
I plan to develop a theoretical framework for my PhD research on informal learning and space appropriation.I will review the literature on these topics and draw on approaches from playful interaction and urban planning to understand how informal learning can be created in public spaces.The research will involve observation of children in public spaces and their interactions with the environment and each other, using a qualitative approach.The basis for this is a qualitative approach, whereby ethnomethodology is suitable for observing the peculiarities of people's everyday lives at a micro level and capturing the multi-layered dimensions of the used environment [16].In a design workshop, children will design their opportunity space for informal learning.I will make use of a speculative design approach, challenging societal, cultural, and technological norms.I will engage in reflective and reflexive practices, using speculative and imaginative techniques to envision alternative futures.Narrative and storytelling techniques will be used with the children to convey findings while questioning societal norms and assumptions.Critical design research is a powerful tool for sparking conversations, raising awareness, and encouraging critical thinking about the impact of design on our lives and society [5].The artistic research case study will critically investigate the multi-dimensional spaces of existing places and their opening and closing with different materials through embodied and tangible technology.The aim is to provoke thought, raise questions, and stimulate discussions about appropriation and mental health.The results of the case studies will be used to provide guidelines for designing opportunity spaces for informal learning.
RESEARCH QUESTIONS
RQ: How can we create children´s opportunity spaces for IFL by adding a technological layer onto an existing space through environmental appropriation?
• SRQ1: How can existing spaces be (critically) designed to create an opportunity for children´s informal learning?(Opportunity for informal learning) • SRQ2: What materiality and technology enable temporal opportunity spaces and the appropriation of the city to facilitate informal learning (with multiple used layers)?• SRQ3: How can media architecture, playful, embodied, and tangible interaction design help children critically appropriate space for informal learning?(Playful and tangible appropriation) 6 RESEARCH ACTIVITIES 6.1 How can existing spaces be (critically) designed to create an opportunity for children´s informal learning?(Opportunity for informal learning) (SRQ1) This observation focuses on children's interactions in public spaces and their ability to transform and appropriate existing spaces.It uses critical lenses to examine individual circumstances, such as places, materials, nature, technology, learning, embodied interactions, time, interest, opening and closing.The study investigates the emergence of temporary, invisible spaces where education can occur and disappear afterwards.The study will observe everyday places, such as school routes or bus stops, to understand the daily habits of individuals and groups.The collected data will be used to design a participatory workshop to create new spaces for children.
6.2 What materiality and technology enable temporal opportunity spaces and the appropriation of the city to facilitate informal learning (with multiple used layers)?(SRQ2) In this Artistic research, I plan to expand existing spaces through material and materiality studies.In the material exploration, I will explore the quality and properties of materials in relation to the body, time, and ephemerality within temporal spaces.The research aims to make the invisibility of appropriated places visible and emphasize social and mental engagement in creating space.The study will explore various media and materials, such as sound, vibration, light, shadow, water, textile, movement, and nature.The connection between technology and materialities will be tested, evoking emotional relations and expanding mental space in public spaces.
6.3 How can playful -, embodied-, and tangible interaction design help children critically appropriate space for informal learning?(Playful and tangible appropriation) (SRQ3) In this Design Research, I plan to speculate on future opportunity spaces in everyday situations.Therefore I plan to co-design temporal opportunity spaces for informal education through a speculative design workshop with children, exploring existing spaces, investigating different groups, and envisioning technological extensions in urban spaces.The inclusion of children in the creative process will help plan future developments and provide alternative perspectives.I will include the data collected in the observation study, exploration findings of the material, artistic research study, and the outcome of the speculative design workshop to design Interventions for everyday interaction in public spaces for informal education, aiming to encourage children's creativity, social interaction, and improve well-being.The next step will be to evaluate opportunity spaces for informal learning through operationalizations, such as open questioners, participatory data physicalization, and observations, using noticeable criteria and theoretical constructs.
TIMETABLE / PLANED PUBLICATIONS
I am starting my second year of my PhD with conducting observations in public spaces.I plan to investigate how children playfully appropriate space in public environments for informal learning.In the second and third years, I will explore artistically temporal opportunity spaces with multiple layers of technology through place-based experiences with different materials, sound, vibration, haptic, light, and AR.In my second year, I will conduct a participatory speculative design workshop, inviting children to design playful, embodied, and tangible interaction technologies in the built environment.In the third year, I will explore materials and technology that enrich the sensory, embodied, and tangible experience in temporal opportunity spaces.I will design interventions for public spaces, building tangible, embodied objects and tools, and I will be installing them in the city environment.I plan to evaluate these interventions in the third year to determine if they support children's informal learning in public environments.I am currently doing physical experiments and material studies, which I will bring into my design workshops with the children and implement into the design interventions.For this, I use haptic materials and observe how their change and movement through the body affect the situations in the space.I hope that by participating at the TEI DC, I can get feedback and knowledge about the extensibility of tangibility.I want to generate more constructive knowledge to make sense of the design of tangible spaces and the use of embodied materials with technology.
CONTRIBUTION
Through the design of Interventions, I want to probe and prove how design and interaction technology can enhance the opportunity spaces for IFL.The Interventions will be examples of opportunity spaces for the public space, which will be installed and can be tried out by the public.I want to provide insights into how we can design opportunity spaces for IFL.These insights can be a guide for future urban planning with tangible technology, where we orientate towards the children´s needs.A catalogue of inspiration will be created for researchers, sociologists, urban designers, architects, and technicians.This catalogue will serve as a tool for adults to expand their imagination for technological implementation and make the world of children more visible.
Figure 1 :
Figure 1: Visualisation of the research plan, including all layers of research activities
Figure 3 :
Figure 3: Design interventions will be placed in the public space
ACKNOWLEDGMENTSI
want to thank my supervisors, Christopher Frauenberger and Verena Fuchsberger, for their guidance.This research project is part of the Paris-Lodron University of Salzburg and Salzburg University of Applied Sciences doctoral college doc.hci -Designing Meaningful
|
2024-01-26T14:10:22.371Z
|
2024-02-11T00:00:00.000
|
{
"year": 2024,
"sha1": "8c2a674515eb5db3199f7a44cbb1d5cfeef5f565",
"oa_license": "CCBY",
"oa_url": "https://dl.acm.org/doi/pdf/10.1145/3623509.3634897",
"oa_status": "HYBRID",
"pdf_src": "ACM",
"pdf_hash": "313b9857819a938d67fea589306467d35caae942",
"s2fieldsofstudy": [
"Education"
],
"extfieldsofstudy": [
"Computer Science"
]
}
|
3965070
|
pes2o/s2orc
|
v3-fos-license
|
Cancer incidence in Stockholm firefighters 1958–2012: an updated cohort study
Objectives Previous studies on firefighters indicate an increased risk of cancer although findings regarding which cancer sites are in excess have been inconsistent. The aim of this study was to investigate the cancer incidence among Swedish firefighters. Methods This updated cohort study included 1080 men who worked at least 1 year as a firefighter in the city of Stockholm, Sweden during 1931–1983. First-time diagnoses of cancer were identified through the Swedish Cancer Registry from 1958 until 2012. Employment as a firefighter was determined from the annual fire station enrolment records. Standardized incidence ratios were calculated using the Stockholm population as reference. Results Firefighters in Stockholm had a low overall risk of cancer (SIR = 0.81 95% CI 0.71–0.91). However, firefighters were at an increased risk of stomach cancer (SIR = 1.89 95% CI 1.25–2.75). Firefighters had significantly low risks for prostate cancer (SIR = 0.68 95% CI 0.52–0.87) and malignant melanoma of the skin (SIR = 0.30 95% CI 0.06–0.88). There was a statistically significant trend of increasing overall risk of cancer with increasing employment duration, although there was still no excess of cancer overall in any of the categories of employment duration. Conclusion Stockholm firefighters had an increased risk of stomach cancer but a low overall risk of cancer. The trend of increasing overall risk of cancer with increasing employment duration could potentially be related to the carcinogenic exposures at work.
Introduction
Firefighters are during their work potentially exposed to a wide range of carcinogenic substances. Through the smoke, they may be exposed to carcinogens such as benzene, 1,3-butadiene, formaldehyde and polycyclic aromatic hydrocarbons (PAHs) (IARC 2010). Firefighters could also be exposed to asbestos, crystalline silica and polychlorinated biphenyls (PCBs) depending on the characteristic of the fire site (IARC 2010) as well as to diesel exhaust from the firefighting vehicles (Froines et al. 1987). Additionally, most firefighters work shift which could disrupt the circadian rhythm and potentially increase the risk for some types of cancer (IARC 2010). The occupational exposure may vary between countries, depending on the regulations and use of protective gears, extinguishing methods and other practices used in the occupation. Based on limited evidence in humans and inadequate evidence in experimental animals "occupational exposure as a firefighter" was classified as possibly carcinogenic to humans (Group 2B) by the International Agency for Research on Cancer (IARC) in 2007 (IARC 2010).
Several studies have investigated firefighters' risk of cancer and some have found an increased risk for various cancer sites. In 2006, LeMasters et al. conducted a review of 32 studies and meta-analysis of 26 studies and found a probably increased risk for multiple myeloma, non-Hodgkin lymphoma, prostate cancer and testicular cancer as well as a possibly increased risk for cancer of the skin, brain, rectum, buccal cavity and pharynx, stomach, colon, leukemia and malignant melanoma (LeMasters et al. 2006). Another comprehensive review and meta-analysis conducted by IARC in 2010 included in addition some studies that were reported after the review by LeMasters et al. and found strongest evidence for increased risk of non-Hodgkin lymphoma, prostate cancer and testicular cancer (IARC 2010). A recent study from the Nordic Occupational Cancer (NOCCA) project including over 16,000 firefighters in the five Nordic countries showed an increased incidence for adenocarcinoma of the lung, prostate cancer, malignant melanoma and non-melanoma skin cancer, as well as an excess risk for all cancer sites combined (Pukkala et al. 2014). Decreased incidence of testicular cancer among firefighters was also observed (Pukkala et al. 2014). Even though there have been many studies on firefighters and cancer conducted, many are limited to information from death certificates which limit the possibility to study cancer sites with high survival rates.
There are a limited number of studies on Swedish firefighters' risk of cancer. Tornling et al. conducted in 1994 a cohort study on all firefighters who worked in Stockholm, Sweden for at least 1 year during 1931-1983(Tornling et al. 1994. Cancer incidence was studied from 1958 to 1986. The study showed an increased incidence of stomach cancer and a tendency for increasing incidence of brain-and stomach cancer with increasing number of fires (Tornling et al. 1994). Since this Swedish study was conducted, several international studies have been published indicating an increased cancer risk among firefighters (Daniels et al. 2014;Glass et al. 2016;IARC 2010;LeMasters et al. 2006;Pukkala et al. 2014) which have resulted in increased concern for occupational cancer among the firefighters themselves. Therefore, we decided to update the previous Stockholm cohort study (Tornling et al. 1994) with information on work duration and cancer incidence until 2012, adding 26 years of follow-up. The aim of this study was to investigate the cancer incidence among Swedish firefighters.
Methods
The study is based on an update and extended follow-up of a previous cohort study described in detail elsewhere (Tornling et al. 1994). The cohort includes men who worked at least 1 year as a firefighter in Stockholm, Sweden during 1931Sweden during -1983. The cohort was identified through annual enrolment records from 15 fire stations in Stockholm and comprised 1153 men. Of these, 63 had died or emigrated before 1958 when the follow-up began and 10 men had worked less than 1 year, resulting in 1080 men participating in the study.
All men were followed up for cancer incidence from 1 January 1958, when the National Cancer Register was established, to 31 December 2012. This resulted in a follow-up time of 29-54 years, depending on time of employment. All first-time cancers for each specific cancer site were included. The cancer sites were identified using the ICD7-codes (International Classification of Diseases, 7th Revision). By law in Sweden, all malignant tumors are to be reported, and the Cancer Registry has a high coverage of 96% (Barlow et al. 2009). Date of death and migration status were retrieved from population records held by Statistics Sweden.
Information on employment duration was collected through the annual enrolment record listings that were held at each fire station. Information on starting-and ending year of employment for each firefighter was used to calculate employment duration and no data were missing for any of the participants. The employment duration also included years worked before 1931 and years worked until 2012, if they fulfilled the criterion of working at least 1 year between 1931 and 1983. In the analyses, employment duration was classified in 10-year groups, resulting in 1-9, 10-19, 20-29, and ≥ 30 years of employment. The starting year of employment was classified as; 1903-1939, 1940-1959 and 1960-1983. We updated the information on employment duration up until 2012, but unfortunately, updating of data on number of fires fought was not possible to obtain for the extended follow-up.
Ethical approval for this study was obtained by the Stockholm ethical review board (Dnr: 2013/2126-31/3 and 2015/787-32). The study was funded by AFA Insurance in Sweden (Grant no. 130104).
Statistical analyses
Standardized incidence ratio (SIR) is the observed number of cases in the study population divided by the number of cases expected to occur given the age specific incidence in the reference population. SIRs were calculated for all cancer sites combined and for specific cancer sites one by one. Expected number of cancer cases were obtained from the general male population in Stockholm County by the person year method. The total number of men in the population was about 600,000 in 1958 and just over 1 million in 2012. Standardization was carried out with the reference population of Stockholm County grouped by birth year and calendar year in 5-year intervals. Trends for risk with age, employment duration and starting year of employment were estimated and tested with a log-linear poisson model adjusted for age. Confidence intervals were constructed by the exact method by Garwood (Garwood 1936). We also analysed cancer incidence stratified on age (< 50, 50-64, and ≥ 65 years), employment duration (1-9, 10-19, 20-29, ≥ 30 years) and starting year of employment (1903-1939, 1940-1959, 1960-1983). The analyses were done for the full follow-up (1958-2012) as well as for the former follow-up only and for the extended follow-up only . In the analyses of the former follow-up only, three new cancer cases were identified (two cases of stomach cancer and one case of brain cancer) that had not been identified in the previous publication. SAS software was used for all statistical analyses.
Results
The cohort consisted of 1080 Swedish men who were followed up in the cancer registry for up to 54 years, resulting in 265 cancer cases. The firefighters had been working for 1-44 years, with a mean employment duration of 26 years (Table 1). Many firefighters started working early in their lives and the mean age at employment as a firefighter was 25 years. Table 2 presents the observed and expected incidence for all cancer sites with at least two observed cancer cases, sub-divided into three time intervals; for the entire follow-up , for the former follow-up only and for the extended follow-up only . The total cancer incidence in firefighters was lower than expected (SIR = 0.81 95% CI 0.71-0.91). This was also seen when looking at the extended follow-up only (SIR = 0.67 95% CI 0.56-0.79) but not when looking at the former follow-up only. The risk of stomach cancer was increased in firefighters (SIR = 1.89 95% CI 1.25-2.75) with 27 observed cases among firefighters compared to 14.5 expected cases, something shown also in the former follow-up (SIR = 2.21 95% CI 1.35-3.41). For the extended follow-up only, the power was lower but the tendency towards an increased risk of stomach cancer was present also for this period with 7 observed cases compared to 5.2 expected (SIR = 1.35 95% CI 0.54-2.78). Prostate cancer, on the other hand, showed a significantly decreased risk both in the full follow-up (SIR = 0.68 95% CI 0.52-0.87) and in the extended follow-up only (SIR = 0.48 95% CI 0.33-0.69) but not in the former follow-up. Firefighters also had a low risk for malignant melanoma of the skin in the full follow-up (SIR = 0.30 95% CI 0.06-0.88) as well as in the extended follow-up (SIR = 0.27 95% CI 0.03-0.97). There were no observed cases of testicular cancer among firefighters in the entire follow-up, compared to 1.5 expected cases. Table 3 presents the cancer incidence for total cancer, stomach cancer and prostate cancer among firefighters in the entire follow-up, stratified by age, employment duration and starting year of employment. We only included the cancer sites that showed significantly increased or decreased risks in the entire follow-up, although malignant melanoma of the skin was omitted since there were too few cases for stratification. A trend was found for employment duration where firefighters who had been employed 1-9 years had a significantly lower cancer incidence overall (SIR = 0.47 95% CI 0.30-0.75) compared to the Stockholm reference population. Firefighters who had been employed for 20-29 or ≥ 30 years had a SIR of 0.98 (95% CI 0.77-1.26) and 0.84 (95% CI 0.72-0.98). The trend of increasing cancer incidence overall with increasing employment duration was tested statistically significant with a p-value of 0.03. There was an overall low risk of total cancer for all age groups, with the lowest risk in the youngest age group and then steadily increasing risk with increasing age. Firefighters younger than 50 years had a SIR for overall cancer of 0.40 (95% CI 0.15-0.86) while firefighters 65 years or older had almost the same risk as the reference population (SIR = 0.92 95% CI 0.80-1.05). There was a statistically significant trend (p < 0.01) of higher risk for cancer for firefighters employed early compared to firefighters employed later (1940)(1941)(1942)(1943)(1944)(1945)(1946)(1947)(1948)(1949)(1950)(1951)(1952)(1953)(1954)(1955)(1956)(1957)(1958)(1959) but the risk was not increased in any of the categories of starting year of employment.
Firefighters showed an increased risk of stomach cancer in all ages, for employment duration over 10 years and for all starting years of employment. There was no trend of increasing stomach cancer incidence with later starting year of employment (p-value = 0.69) and no trend of increasing stomach cancer incidence with increasing employment duration (p-value = 0.19). Younger firefighters had a tendency of higher risk of stomach cancer than older firefighters.
There was no trend of increasing prostate cancer incidence with increasing employment duration (p-value = 0.75). Firefighters 50-64 years of age had a significantly lower risk of prostate cancer (SIR = 0.50 95% CI 0.24-0.92) as did firefighters 65 years or older (SIR = 0.72 95% CI 0.53-0.95) compared to the reference population. Firefighters employed 1960-1983 had a lower risk of prostate cancer (SIR = 0.20 95% CI 0.08-0.47) compared to firefighters employed earlier.
Discussion
This study showed that firefighters in Stockholm had an overall low risk of cancer but an increased risk of stomach cancer compared to the reference population. Prostate cancer and malignant melanoma of the skin showed significantly low risks. There was an increasing overall risk of cancer with increasing employment duration but there was no excess of cancer overall in any of the categories of employment duration. The extended follow-up (1987-2012) provided essentially the same conclusions as the full followup . The overall low risk of cancer in Stockholm firefighters found in this study is in contrast with many previous studies on firefighters where a slightly increased risk have been found (Daniels et al. 2014;Glass et al. 2016;IARC 2010;Pukkala et al. 2014). The low cancer risk among firefighters found in this study could potentially be explained by a selection of healthier individuals to the occupation, the so called healthy worker bias. All men working as firefighters in Sweden must pass a mandatory physical test at start of employment as well as regularly throughout their employment, which most likely makes them in better physical condition and health than the general population in Sweden (Arbetsmiljöverket 2005). However, this is also mandatory in other countries where increased risks among firefighters have been shown. The overall low cancer risk for this cohort of firefighters could also possibly be explained by the left truncation bias. The enrollment to the cohort started in 1931 but the outcome could only be studied from 1958, and therefore all firefighters who died or emigrated between 1931 and 1958 were excluded from the cohort, leaving a slightly healthier population of firefighters at start of follow-up. A sub-analysis of only firefighters employed after 1958 showed a lower overall cancer risk (SIR = 0.34 95% CI 0.23-0.49) compared to the entire cohort (SIR = 0.81 95% CI 0.71-0.91). The interpretation of this result is a bit difficult. A lower SIR indicates that there is not a left truncation bias in the overall results; however, the later population is much younger and would be expected to have a lower SIR than the older population according to our analyses stratified on age.
Even though the overall cancer risk was low, the risk increased with increasing employment duration. Analysis showed a significant trend (p-value = 0.03) for increasing risk of overall cancer with increasing employment duration, indicating that carcinogenic exposure at work could be of importance for the cancer risk among firefighters. The increasing overall cancer risk with increasing age and with earlier starting year of employment could also possibly be related to carcinogenic exposure since age, starting year of employment and employment duration are highly correlated in this study. However, there was no excess of cancer overall in any of the categories of age, starting year of employment or employment duration, compared to the general population.
This study showed an increased risk of stomach cancer (SIR = 1.89 95% CI 1.25-2.75). The increased risk is in line with Tornlings results from 1994 based on the same material (Tornling et al. 1994). Looking only at the extended follow-up, there was also an increased risk of stomach cancer, although not statistically significant (SIR = 1.35 95% CI 0.54-2.78). This indicates that the increased risk of stomach cancer is not just a chance finding. Tornling et al. also found in the former follow-up that the risk of stomach cancer increased with increasing number of fires the firefighters fought, which might indicate that the occupational exposure is involved in the etiology (Tornling et al. 1994). Unfortunately, in this extended follow-up with updated employment information until 2012, it was not possible to update the information on number of fires fought. LeMasters review states that firefighters are at a "possible" risk of stomach cancer (LeMasters et al. 2006). Known risk factors for stomach cancer are helicobacter pylori infection, occupational exposure in rubber production industry, tobacco smoking and ionizing radiation (Cogliano et al. 2011). There is also some evidence indicating that asbestos, lead compounds, nitrate, pickled vegetables and salted fish are risk factors for stomach cancer (Cogliano et al. 2011). A recent review and meta-analysis by Lee et al. showed a significant association between occupational crystalline silica exposure or silicarelated working conditions and stomach cancer (Lee et al. 2016). It is possible that firefighters are exposed to rubber compounds if the burning material consists of these compounds or asbestos and crystalline silica dust if the fire site holds these materials. An increased risk of lung cancer and pleura cancer/mesothelioma would be expected if firefighters were exposed to asbestos. This was not shown for lung cancer in our material (SIR = 0.79 95% CI 0.52-1.15) but for pleura cancer we observed 2 cancer cases compared to 0.8 expected cases. This might indicate an increased risk for pleura cancer although the numbers are too few to draw any conclusions. A review from Raj et al. concluded that "dusty occupations" could be a cause of stomach cancer (Raj et al. 2003). Firefighters often get in contact with dust when tearing down burning material, and during the last stage of the firefighting (overhaul). The decreased risk of prostate cancer found in firefighters in this study is somewhat surprising since several previous studies and meta-analyses have found an increased risk of prostate cancer in firefighters and since shift work possibly could increase the risk of prostate cancer (IARC 2010). It is possible that Swedish firefighters are not disturbed as much during shift and night work compared to firefighters in other countries, with less impact on the circadian rhythm, although we do not have such information.
Firefighters in this study had no observed cases of testicular cancer, with 1.5 cases expected. Both LeMasters et al. and IARC found an increased risk of testicular cancer (IARC 2010;LeMasters et al. 2006). However, the Nordic study NOCCA found in 2014 a significantly decreased risk of 0.51 (95% CI 0.23-0.98) for testicular cancer (Pukkala et al. 2014). Also a recent study on 30,000 US firefighters followed from 1950 to 2009 found a decreased testicular cancer incidence (Daniels et al. 2014). These two studies make it less likely that our finding is due to chance. The risk factors for testicular cancer are mainly unknown, and so is the reason for a reduced risk among firefighters.
This study has several strengths and weaknesses. This is a cohort study with a long follow-up time in the cancer registry of up to 54 years. This long follow-up allows for cancer incidence to be studied. Cancer incidence was collected from the Swedish Cancer Registry with a high coverage of 96%. Unfortunately, there were no individual data for lifestyle factors like tobacco smoking available and therefore analyses could not be adjusted for potential confounders. Firefighters' employment duration was used as a proxy for their exposure which is a weakness and no information on the use of respiratory protection and other protective equipment was available. It should also be considered that this is an occupational cohort and that the reference population is the Stockholm population. This can increase the risk of bias due to a selection of healthy individuals to the occupation.
The generalizability of this study is thought to be high in Sweden, but could be lower internationally due to differences in the firefighting occupation between countries, such as type of work activities, type of materials that may burn, methods of fire extinguishing and use of protective equipment. The exposure to carcinogens may also differ between work in urban and rural locations and between different periods of time (Fritschi and Glass 2014). Smoking habits may differ between countries but a recent study on lung cancer among firefighters in Europe, Canada, New Zeeland and China showed that smoking was similarly common in firefighters as in non-firefighters (Bigert et al. 2016). The diagnostic criteria between countries are not likely to differ; however, the reporting of cancer cases could differ as Sweden has a mandatory cancer register with a very high coverage of 96% (Barlow et al. 2009).
Conclusion
This study showed that Stockholm firefighters had an overall low risk of cancer. However, firefighters had an increased risk of stomach cancer compared to the reference population. There was an increasing overall risk of cancer with increasing employment duration, which potentially could be related to the carcinogenic exposures at work, although there was still no excess of cancer overall in any of the categories of employment duration.
|
2018-04-03T02:46:35.652Z
|
2017-11-21T00:00:00.000
|
{
"year": 2017,
"sha1": "4de65cc5d7da7bd8d55d39e934245440dbbb0c37",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s00420-017-1276-1.pdf",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "4de65cc5d7da7bd8d55d39e934245440dbbb0c37",
"s2fieldsofstudy": [
"Environmental Science",
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
67789820
|
pes2o/s2orc
|
v3-fos-license
|
Multicultural Education and HuManisM tHEory as an Effort to iMprovE tHE social sEnsibility of priMary scHool studEnts
Data from the Indonesian Child Protection Commission noted that 34% Children deal with the law, one of the cases is bully. The bully case is one proof of the social sensitivity of the elementary students. The program for basic education that can be pursued to increase social sensitivity is through multicultural education and the theory of humanism. This study uses a literature study method that collects several reference journals and textbooks. The results of the research are collaborative methods of multicultural-based education and humanism theory when classroom learning is more effective for improving social sensitivity for elementary school students.
introduction
The embryo of individual character determination begins early, since the school at elementary level that is elementary school children is introduced how to speak and behave politely.The way these children acquired through formal education and informal education.Although the role of parents in educating children has a huge role, the role of formal schools is also very important in fostering the behavior of children.If in formal education then known the term students to call children.The formal school appointed in this discussion is the Elementary School, the elementary school students will mingle and socialize with their peers.
In the socialization process is required social sensitivity attitude among students.social sensitivity can be interpreted as a form of concern among fellow students, have a sense of belonging, and can tolerate in communicating.The fact that should be aligned in the socializing activities of children seemed to fade for children domiciled in cities in Indonesia, the nature of individualism is more dominant.Indeed, there is a point if humans are individual beings as well as social beings, only with increasingly sophisticated technology so that the rapidly growing digital era in Indonesia does not make students to be able to socialize in accordance with the development of its age directly.Social sensitivity can be felt if children interact directly with their social environment.
One example of a large city that the author observes is Jakarta.In the city of Jakarta elementary students are quite diverse from some cultural background of parents who come from several regions in all regions of the country of Indonesia.This form of multiculturalism has an effect on different behavior when responding to a problem.
Data from the Indonesian Child Protection Commission noted that 34%Children deal with the law, one of the cases is bully.The context of this problem, in essence as a human being, is to be tolerant of mutual tolerance despite the colorful background of everyone's culture.
Humanism theory, which means humanizing human being is one theory that can underlies the formation of social sensitivity.This will be another challenge for teachers who teach in urban areas, especially Jakarta.So that students can socialize with peers without having to leave one of the Indonesian cultural identity is "caring", based on research from various sources of journals and book sources that mentions that multicultural education and humanistic theory can be one solution of the problem of social sensitivity of students SD then the discussion of this paper is "Multidimensional Education and Humanism Theory: Efforts to Improve Social Sensibility of Elementary School Students."
Method
The type of this research is qualitative research with literature study method.According to Sukardi (2017) literature study conducted by each researcher with the main objective is to find the foundation of foundation or foundation to acquire and build the theoretical base, the frame of thinking, and determine the suspicion of temporary or often referred to as research hypothesis, so that researchers can understand, locate, organize , and then use the library variation in the field.By conducting literature studies, researchers have a deeper and deeper understanding of the issues to be studied.Various sources of literature that can be used by researchers are journals, research reports, scientific magazines, newspapers, relevant books, seminar results, unpublished scientific articles, sources, letters -letters, and others.
result and discussion 1. implementation of Multicultural Education in primary schools
Indonesia is a nation of diverse tribes, cultures, religions, and races.The diversity makes Indonesia rich with culture.The religion can also be called multiculturalism.Differences that want to be unity, does not mean that with the diversity makes Indonesia split.However, with the diversity it will happen the process of learning other things that can be the same is owned by certain groups, from the learning process occurs the process of cultural diffusion that increasingly makes Indonesia rich through the potential possessed by each individual.
Based on John (Banks, 2010) multicultural education is education that values diversity and includes the perspectives of a variety of cultural groups on a regular basis.Its proponents believe that children of color should be empowered and that multicultural education benefits all students.Based on John (Bennett, 2011) an important goal of multicultural education is equal educational opportunity for all students, with result of closing the gap in academic achievement between mainstream students and students from underrepresented groups.
The occurrence of multicultural education is a transmigration process.The process of transmigration has brought changes in many aspects, in the educational aspect of the impact is quite influential on the behavior of students in interacting in the social environment.In the opinion of Sofan (2013), "education has a purpose to make children happy to get education and can socialize well in society."Therefore, the role of multicultural education plays a role in diversity that helps students to ISSN: 0852-0976 be accepted, acknowledged, and comfortable around.
As for how to implement multicultural education in elementary school listed in the research (Maulani, 2012) that is through the role of some of the following aspects: 1.The Role of Teachers and Schools in Developing Religious Schools The role of teachers in this case includes; first, a teacher / lecturer should be able to be democratic.2. Development of Multicultural Based Religious Education Material In order to build inclusive religious school there are some religious education materials that can be developed with multicultural feel.
In the journal on multicultural education in Finland mentioned that the discourse on multicultural education and the concept of intercultural competence in the European and Nordic countries in Finland.(Fred, Dkk: 2012) said to address the discourse on multicultural education is then focused on the decision makers, researchers, and also student teachers.In this way, the article is an attempt to evaluate how intercultural competencies can and should be conceptualized in today's global scholarship.
The national core curriculum for basic education (Finnish National Education Council / FNBE, 2004) there are several different goals for the majority (ethnic) and (ethnic) minorities.The focus of the research is to embrace cultural acculturation or social processes that integrate the elements of minority culture into the majority to create a new culture without eliminating the original elements.Based on the research can be used as a reference for elementary school teachers in Indonesia in teaching students who have diversity in the classroom, the process of cultural integration can be one of the solutions for students in communicating so that social sensitivities arise among students.
the role of Humanism theory in Elementary learning
According Nur'aini (A. Malik, 2005: 181) said that "education is a process of humanization or in other words the process of humanizing human beings.Education, then, is a process, an endeavor, an effort to provide students with a strong foundation for humanizing human beings." In the theory of humanistic psychology, Maslow assumes that human beings are actually good creatures, so that humans have the right to realize their true self in order to achieve self-actualization.In the process of realization of identity is needed recognition from the community, but often humans are hampered by the circumstances of society that is not in line even the rejection of the talent and interest so that there is a social imbalance.Minderop (Krech, 2011: 48) In the journal written by (NurHikma, 2015: 1) obtained the results of research on the analysis of novel shoes dahlan which shows that there are attitudes that can be a role model that the children can grow into a more mature, strong, independent, when face problems.The problem with Dahlan's figure is about the economic limitations, that everyone who can enjoy the process in the limitations and have a good work ethic attitude then the process of self-actualization can be achieved.
Based on the description of the journal's example of the novel analysis, it can be attributed to elementary students' learning but the type of material given will be slightly different.For example, for stages of elementary school age children teachers can introduce to novel students KKPK or Small-Small Punya Karya.In the novel review which contains several sub chapters and different stories each series and novel is the work of elementary school students who have the talent to write.It can be an example of loving attitudes of friends, doing good to both parents, as well as many other positive stories that can be an example for students.so that, in accordance with the theory of humanistic learning that can make students can develop self-actualization.
Research by (Halloluwa, 2014) Sri Lanka is a country with a high standard of education.Literacy rate reached 91.2% with a ratio of 1:20 ratio which means 1 teacher to teach students as many as 20 people.Based on the ratio, there is an imbalance that happens is the unevenness of the number of teachers teaching in rural areas, when the observation dikedasaan and urban impact on decreased teacher qualifications and decreased quality of education in downtown Colomba.Based on these research journals it is necessary for humanist social sensitivity to periodically exchange roles of teachers coming from the city to communicate with teachers in rural Colomba.Proving that the practice of the theory of humanism is needed on aspects of education that are not only limited to students but the role of teachers so that student actualization can be realized.
Research results show that: (1) model development begins with a preliminary study, development study, and implementation which then produces a model of humanist learning in character education in SD, (2) humanistic model religious character in elementary school education demonstrates a good level of execution, meets criteria very effectively, very practically and validly based on a rational thought base with strong and relevant supporting theories, (3) a model of religious humanist learning in character education that is effectively developed for use in character education in primary schools, and (4) learners have a very positive response to the model of religious humanist learning in character education in elementary schools." In accordance with the philosophical views associated with religion, the human soul will not be dry if in life has the guidelines that regulate his relationship with God.It has been proved that the humanist theory that juxtaposed with the emphasis that has been developed in the elementary school learning resulted in a positive response in elementary students
Elementary school students
Literally sensitivity comes from a sensitive word that means easy to feel or easily aroused or a condition that easily reacts.Social sensitivity will elicit students' reactions in seeing the social phenomena around their own students, in addition it is expected to appear active or action as a follow-up to a reaction that leads to a positive thing.(Kiky, 2015: 265-266) The emergence of the term moral degradation is not just a term, but a problem to be solved together.The role of teachers, parents, and the environment becomes the atmosphere for character formation of elementary students.They can see, observe, imitate the behavior of those around them.As adults, it is expected to be able to influence the positive behavior on elementary school students for social sensitivity to be formed early on.Good impression and a solid initial foundation can determine the power of good behavior by elementary students.
Many efforts have been done by the government and implemented by elementary school teachers through character education.In a learning process that involves character, students' emotions are guided to be a polite person so as to produce balanced psychomotor, affective, and cognitive aspects.The good synergy is the initial capital of students to become complete human beings who can actualize themselves through the theory of humanism.
The difference lies not in the difference, but the color that blends in the difference.So that the term multicultural education that can integrate the elements of culture, introducing cultural diversity, and how to address the diversity that fused in the learning in the classroom.Efforts to accept differences and convince students that the diversity that brought Indonesia into a country rich in ethnicity, culture and language.Emphasizing to students there is no superior group and disempowered group if the student is in a majority group.So as not to bring up negative stereotypes that have an impact on discrimination.
Here is an example of a journal analysis written by (Kelly, 2014: 1) This study investigates cases of cyberbullying among ISSN: 0852-0976 students.A total of two hundred and eighty two students have attended a survey activity on students' intentions reporting cyberbullying cases.However, the cyberbullying case reporter only focuses on the victims.While students who know their friends involved in cases of cyberbullying tend to be passive.
The case of cyberbullying means that the lack of awareness of each person to have a sense of mutual love and compassion.As a result, there is a group that feels more power, more self, and the right feeling.Justification of the perception of subjectivity that causes a social gap in the community.Target cyberbullying is often the case in adolescents who have a tendency of active users of social media.Millennial generations seem to be spoiled by the power of digital that should be more wise in using social media.
The existence of these journals is evidence that the need for preventive and supervisory measures for elementary students in particular khususus using the internet.In order for cyberbullying cases do not penetrate the joints of students' thinking elementary school teachers can perform various actions.
In the study (Annisa, 2016) School programs to prevent and deal with bullying in MuhammadiyahMlangi Elementary School have not been determined in detail.The first action taken is to advise the perpetrators of verbal bullying cases.If the student is still repeating the bullying act then the class teacher, Bk teacher, and guardian of students discuss to solve the problem.The characteristics of students who have social sensitivity in the environment are those who are able to have social sensitivity so that it can determine the right step if there is a social problem around it.Such social sensitivity is gained through the social insights gained from schools, the playing environment, and families.
The above description has shown the aspects that affect the social sensitivity of elementary students from various sources of book references and journals, hence efforts to increase social sensitivity through multicultural education and humanism theory is that teachers can integrate the color of cultural diversity based on the background of kebudaayan parents through giving the material about bhineka single ika which automatically entered on the themes of each learning conclusion and suggestion 1. conclusion Religion without science will be paralyzed whereas science without religion will blind (Albert Einstein).The phrase leads to the thought that it takes the foundation of religious thought in the process of selfactualization.When the attainment has been achieved, man no longer seeks empty space on him because the human soul has been filled with a belief that can lower the arrogance in self.The righteous and best feeling that can lead to negative stereotypes that can lead to discrimination processes will not happen.
The melting of primordialism can be through multicultural education and humanistic learning theory that can prepare the superior generation of elementary school.The two things are put together and then packed into the practice of learning in elementary school.The younger generation is the successor of the ideals and as the hope of the older generation.Therefore, the development of learning based on the character so that students can be trained and educated mentally from childhood in the hope of improving the quality of morality in Indonesia, so that people no longer hear the term moral degradation that occurs in Indonesia's young generation.If the mentality of social sensitivity is established to be solid from the beginning then the hope for the next generation that is still 7-13 years old (elementary school aged), five to ten years to come will be a dignified generation.
suggestion
To face the global challenge it is necessary to have a character that has a good mentality in communicating in the global ISSN: 0852-0976 environment.Such procedures need to be established since elementary school-aged children to have a recurring habit of social sensitivity.Parental guidance is needed to support children socialize globally to meet with friends of different tribes, religions, or races so that tolerance appears naturally in children.The role of teachers is important in formal education, both the bad character of students is also influenced by the teaching of teachers.Therefore, it is necessary to recognize the diversity of Indonesian culture and how to overcome such diversity, for example through multicultural education.
|
2018-12-29T21:44:30.647Z
|
2018-07-25T00:00:00.000
|
{
"year": 2018,
"sha1": "0f3e0129b1e9b7be5c17793c2877d55acb2047ae",
"oa_license": "CCBY",
"oa_url": "http://journals.ums.ac.id/index.php/varidika/article/download/6547/3995",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "0f3e0129b1e9b7be5c17793c2877d55acb2047ae",
"s2fieldsofstudy": [
"Education",
"Sociology"
],
"extfieldsofstudy": [
"Psychology"
]
}
|
245105723
|
pes2o/s2orc
|
v3-fos-license
|
Diagnostic value of shear wave velocity in polycystic ovarian syndrome
Abstract Aim: In polycystic ovarian syndrome, the ovaries become stiffer due to chronic anovulation. We aimed to compare tissue elasticity in terms of shear wave velocities measured using acoustic radiation force impulse imaging technique between the ovaries of polycystic ovarian syndrome women and non-polycystic ovarian syndrome women. Material and methods: The study was designed as a retrospective data analysis of women who underwent transvaginal ultrasound and acoustic radiation force impulse imaging in a university hospital between July 2014 and March 2015, for various reasons. There were 32 polycystic ovarian syndrome patients and 32 patients without a diagnosis of polycystic ovarian syndrome. Age, body mass index, fasting glucose levels, cycle day 3 follicle stimulating hormone, luteinizing hormone, thyroid stimulating hormone, prolactin, antimullerian hormone levels, and menstrual patterns with clinical hyperandrogenism were evaluated. On the menstrual cycle days 2–4, by performing a transvaginal ultrasound scan, the ovarian volumes and antral follicle counts in both ovaries were recorded for each woman. The ultrasound system was converted into the elastography mode, and acoustic radiation force impulse imaging was performed. Shear wave velocity (m/sec) was measured at least 5 times for each ovary, and the mean value was calculated for each polycystic ovarian syndrome and non-polycystic ovarian syndrome woman. Results: Age, body mass index, fasting glucose levels, cycle day 3 follicle stimulating hormone, luteinizing hormone, thyroid stimulating hormone, and prolactin levels were similar between the groups (p >0,05). Antimullerian hormone levels, antral follicle counts, and mean ovarian volumes were statistically different between the groups (p <0,05). Mean shear wave velocity values for both ovaries were 2.12 ± 0.82 (0.78–4.9) m/sec in the polycystic ovarian syndrome group, and 1.18 ± 0.41 (0.77–2.0) m/sec in the non-polycystic ovarian syndrome group, which was statistically significantly different (p = 0.016). Conclusion: In our study, we found significantly higher shear wave velocity levels in polycystic ovarian syndrome women than non-polycystic ovarian syndrome women, which indicates an impact of the condition on shear wave velocity. The increased acoustic frequencies cause a decreased response in time to transition, and motion becomes out of phase; in other words, scattered waves are faster in stiffer ovaries. Our results are thus compatible with the pathophysiology of the disease. Shear wave velocity is a beneficial tool for evaluating ovarian elasticity in polycystic ovarian syndrome patients in whom the levels are found to be significantly higher than non-polycystic ovarian syndrome women. In light of these findings, shear wave velocity is expected to be slower than polycystic ovarian syndrome levels in ovulatory women.
Introduction
Elastography is a useful tool providing information on tissues by converting stiffness values into an anatomically significant map. Sonoelastography is an ultrasound (US) imaging technique. When the tissue is harder, the sound waves move faster, and their speed provides information about tissue hardness (1) . Both qualitative and quantitative types of elastography are available. Acoustic radiation force impulse imaging (ARFI) is a recently developed noninvasive dynamic tissue imaging technique of quantitative elastography. ARFI correlates tissue flexibility with conventional gray-scale images obtained by US. Shear wave velocity (SWV) is measured by ARFI to establish the stiffness of tissues in meters per seconds (m/sec). With gradually increasing acoustic frequencies, tissues fail to respond in time to the transitions between positive and negative pressures, and the motion becomes out of phase with the acoustic wave frequency. As a result, energy is deposited in the tissues, which causes momentum transfer and an increase in tissue temperature. The resulting displacement of tissue is detected and used for obtaining additional information beyond B-mode imaging (2) .
Changes in tissue elasticity are a useful tool used for identifying fibrosis and differentiating between malignant and benign lesions in the tissues of the liver, thyroid, and breast (3) . Tissue elasticity measured by ARFI and SWV is used in gynecologic and obstetric practice, e.g. in placental tissues, endometrium, myometrium, and cervix (4)(5)(6) .
Polycystic ovaries are seen in polycystic ovarian syndrome (PCOS) and sometimes also in normally ovulating women (polycystic ovarian morphology). PCOS is diagnosed using the Rotterdam Criteria, based on the presence of at least two criteria of the total three, provided that other etiologies are ruled out (7) . One of the criteria is having ≥12 antral follicles (follicles of 2-9 mm) or ≥10 ml ovarian volume in one or both ovaries.
Stein and Leventhal were the first to report this syndrome, describing the ovaries as rubbery and naming the condition "hyperthecosis" to reflect chronic anovulation (8) .
Chronic anovulation is a vicious cycle with stromal tissue producing more androgens, leading to follicles to rest in the antral stage, not progressing to the preantral stage, or not being able to gain dominance due to hypoestrogenic inner environment.
We conducted this study to gain insights into the elasticity pattern and SWV in PCOS ovaries. The primary outcome measure was to compare SWV values between PCOS and non-PCOS women. The secondary outcome measure was to compare the menstruation patterns, clinical properties, and laboratory and other ultrasonographic parameters between PCOS and non-PCOS women.
Material and methods
The study was designed as a retrospective data analysis of women who underwent vaginal US and ARFI in a university hospital between July 2014 and March 2015, for various reasons. There were 32 patients diagnosed with PCOS based on the Rotterdam Criteria. Also, there were 32 age-matched normally ovulating women having less than 12 antral follicle count per ovaries, selected as the non-PCOS group. For the diagnosis of PCOS, the Rotterdam Criteria were used, comprising: i) menstrual abnormalities such as oligomenorrhea or amenorrhea, ii) clinical or biochemical hyperandrogenism, and iii) ovaries having ≥12 antral follicles (counted with transvaginal US probe resolution of <8 MHz) or ≥10 ml ovarian volume. Having at least 2 of these 3 criteria is considered to be indicative of PCOS (7) .
Menstruation patterns were recorded as normal menstruation (menstruation/21-35 days), polymenorrhea (menstruation more frequent than 21 days), oligomenorrhea (less frequent than 45 days, or less than 8 cycles per year) or amenorrheaeither primary (no menstruation until 17 years of age) or secondary (no menstruation for more than 3 months in a woman who previously had menstruation) (9) .
Clinical hyperandrogenism was defined as having acne or hirsutism. Hirsutism is evaluated with the modified Ferriman-Gallwey (mF-G) scores; ≥8 is considered as clinical hirsutism (10) .
We excluded women with abnormal thyroid function tests, elevated prolactin levels, very high DHEA-S levels (>700 pgr/dl), and high 17-hydroxyprogesterone levels.
Ultrasound and elastography technique
On days 2-4 of the menstruation cycle, all women were evaluated using the conventional gray-scale US, and ovarian volumes were calculated for each patient by measuring the largest dimensions in the antero-posterior, transverse and longitudinal planes of both ovaries (0.5 × length × width × thickness), and the total number of antral follicles (2 to 9 mm in diameter) was calculated. Peak systolic velocity (PSV) and resistance index (RI) of both ovaries were measured by spectral Doppler imaging. The system was converted into the elastography mode, and ARFI was performed. Each patient was asked to lie supine and hold her breath. The US probe was placed transvaginally for both ovaries, and the examination was performed in the vein-free area with a region of interest of 1 × 0.5 cm, with depth set at 6 cm. SWV (m/sec) was measured at least 5 times for each ovary, and the mean value was calculated in all women. For each woman, the mean SWV of the left ovary was calculated, and the mean SWV of the right ovary was calculated; in the were expressed as percentages. The Shapiro-Wilk test was used for assessing normality. All continuous variables were found to be normally distributed. Parametric tests were used for variables with normal distribution. Differences in mean values were analyzed with Student's t-test. The characteristics between groups were compared with chisquare test. A p value ≤0.05 was considered as statistically significant.
Results
The study comprised 32 patients in the PCOS group, and 32 women in the non-PCOS group. Age, body mass index, fasting glucose levels, and hormone levels on days 2-4 of the menstrual cycle (FSH, LH, and TSH and PRL) are shown in Tab. 1. There were no statistically significant differences among these parameters in both groups (p = nonsignificant (NS)) (Tab. 1).
The menstruation parameters were significantly different between the PCOS and non-PCOS women. There were 8 (25%) normally menstruating women in the PCOS group, but there were 29 (90.6%) normally menstruating women in the non-PCOS group (p = significant (S)). Oligomenorrhea was seen in 23 (71.87%) patients in the PCOS group and in 2 (6.25%) women in the non-PCOS group (p = S). All PCOS patients had clinical hyperandrogenism (100%); in 12 (18.75%) women, both acne and hirsutism were seen, in 2 (6.25%) patients only hirsutism was present, and in 18 (56.25%) only acne was observed. In the non-PCOS group, there were 6 (18.75%) patients with clinical hyperandrogenism. The difference was statistically significant (p = S). Biochemical hyperandrogenism was seen in 28 patients (87.5%) in the PCOS group, and in 2 patients in the non-PCOS group (6.25%). AMH levels were statistically higher among the PCOS women than in the non-PCOS group (p = S) (Tab. 2). AFC levels were statistically higher in the PCOS than in the non-PCOS group (p = S). Mean ovarian volumes were statistically different between the two groups (p = S). Peak systolic velocities and resistance indices were statistically different between the PCOS and control groups (p = next step, the mean ovarian SWV was calculated by adding the mean left and the mean right SWV values and dividing the result by two (SWV = left ovarian SWV + right ovarian SWV/2). ARFI (Acuson S3000TM Ultrasound System, Siemens Health Care, Mountain View, Ca, USA) imaging was routinely performed along with conventional US of the ovaries in all cases. The gray-scale US and ARFI studies were performed using a transvaginal 8 MHz probe.
We excluded women with any cystic or solid mass in the ovaries on days 2-4 of the menstrual cycle detected via US. The data obtained from PCOS women and non-PCOS women were evaluated and compared.
Statistical analysis
Statistical analysis was performed using the SPSS version 22 (Statistical Program for Social Sciences, IBM, Chicago, IL). Demographic continuous data were characterized by means and standard deviations (SD), and nominal variables
Discussion
Early application of elasticity measurements in soft tissues to identify tissue hardening due to fibrosis or tumor was first examined in the liver, thyroid and breast tissues (11)(12)(13) . Elasticity can be measured via ARFI, which is a non-invasive and cost-effective diagnostic tool using the advantages of B-mode ultrasonography. It is generally accepted as a safe and easy method (14) .
Tissue elasticity is investigated in various obstetric and gynecologic conditions. In their study, Bildaci et al. used ARFI to evaluate placentas in patients with gestational diabetes, revealing no difference between the patients and the control groups in shear wave velocities. This could be due to the transient pathology of gestational diabetes, since the pathology can improve quickly with dietary regulations on most occasions (15) .
Ovarian cysts were evaluated by ARFI in a few studies. In their study, Ciledag et al. found that elastography scores might be useful for differentiating between benign and malignant ovarian cystic lesions. On that basis, unnecessary interventions could be avoided for benign cysts with solid components (16) .
PCOS is a result of chronic anovulation. Chronic high levels of androgens cause more theca cells (hence the former name of hyperthecosis); the connective tissue in between follicles is surrounded by androgen-producing cells. Based on the two-cell-two-gonadotropin theory, the hyperandrogenemic environment cannot be beaten due to anovulation, so more androgens are produced, and thus a chronic vicious cycle occurs. This pathologic condition makes the ovarian tissue more stiffer than in normal estrogenic cases.
The diagnosis of PCOS is established based on the 2003 Rotterdam Criteria, after the exclusion of other pathologies, as having 2 of the 3 diagnostic criteria, one of them being oligomenorrhea. In our study, there were significantly prominent menstruation abnormalities in the PCOS group. One criterion is having hyperandrogenism, established either clinically or biochemically. In our study, the PCOS patients were significantly (both clinically and biochemically) hyperandrogeniemic. Another criterion is having polycystic ovarian morphology, i.e. more than 12 antral follicles or more than 10 ml of ovarian volume. In our study, the PCOS patients had a significantly more pronounced polycystic ovarian morphology than the control group.
In their study, Ozdemir et al. compared ovarian stromal blood flow with Doppler ultrasonography in patients with PCOS and healthy controls, revealing that PI was significantly different in the PCOS group (17) . In our study, we also found different PI and PSV values in the PCOS group.
In the literature, there are three studies on polycystic ovaries and elastography (18)(19)(20) . In one of them, the researchers studied the elasticity pattern and strain ratio. Ciraci et al. classified the elasticity patterns of the ovaries in patients with PCOS as hard, medium and soft, and examined the strain rates. They revealed that the elasticity patterns and strain ratios might demonstrate morphological stiffness changes in polycystic ovaries. They also reported that it was possible to evaluate PCOS by real-time elastography, which might help to show morphological changes. Elastography features of ovarian stroma may have a role in the diagnosis of PCOS, like gray-scale US, especially using the strain ratio (18) .
In their study, Ertekin et al., reported no significant differences in SWV between the PCOS and control groups. Their conclusion was that the role of elastography in the diagnosis of PCOS was controversial. However, in their study, the sample size of the control group was small, which could have potentially affected the accuracy of their results (19) .
In a recent study by Altunkeser et al.; 66 PCOS patients were compared with 72 controls for SWV and elastography values. The authors found that the PCOS group had a mean SWV of 3.89 ± 1.81 for the right ovary, and 2.88 ± 0.65 for the left ovary. They compared the results with the controls, and found no significant difference (20) . Our mean SWV results were 2.12 ± 0.82 (0.78-4.9) in the PCOS group, and 1.18 ± 0.41 (0.77-2.0) in the control group (p = 0.016).
There are different phenotypes in PCOS, so confirming 2 of the 3 criteria is enough for the diagnosis according to the Rotterdam Criteria; women having hyperandrogenism and oligomenorrhea can have the diagnosis without the morphological appearance of PCOS (21) . However; chronic anovulatory condition leads to the hyperandrogenic inner state of the ovaries, resulting in the development of symptoms. SWVs are significantly faster in the ovaries of PCOS patients due to stiffness.
There are women with a normal menstrual pattern, without clinical or biochemical hyperandrogenism, but having
|
2021-12-12T16:59:46.215Z
|
2021-11-29T00:00:00.000
|
{
"year": 2021,
"sha1": "f0551ec4d13a73efe7925e63bb1bb00f63ecc0d5",
"oa_license": "CCBYNCND",
"oa_url": "http://www.jultrason.pl/artykul.php?a=989",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "beaa411d50a21ce2f617b2812536b3674f769348",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
221747774
|
pes2o/s2orc
|
v3-fos-license
|
The effect of nitric oxide, endothelial nitric oxide synthetase, and asymmetric dimethylarginine in hemorrhoidal disease
AIM: The aim of this study was to examine the roles of nitric oxide (NOx), endothelial nitric oxide synthetase (eNOS), and asymmetric dimethylarginine (ADMA), which is the major endogenous inhibitor of nitric oxide synthases (NOS), in the pathophysiology of hemorrhoidal disease. METHODS: This study included 54 patients with grades 3 and 4 internal hemorrhoidal disease and 54 patients without the disease who attended the General Surgery Clinic. NOx, eNOS, and ADMA levels were measured with the Enzyme-Linked ImmunoSorbent Assay (ELISA) method. RESULTS: The patients had higher NO and eNOS levels and lower ADMA levels than the control subjects (p<0.001). A significant highly positive correlation was found between NO and eNOS (p<0.001). Nevertheless, there was a highly negative correlation between ADMA and NO-eNOS(p<0.001, p<0.001). CONCLUSION: This preliminary study reveals that higher NOx and eNOS activities and lower ADMA levels in the rectal mucosa are observed in patients with hemorrhoidal disease than in those with normal rectal tissue. The imbalance between endothelium-derived relaxing factors, such as NO and endogenous competitive inhibitor of NOS, ADMA, may cause hemorrhoidal disease. Our study proposes that hemorrhoids display apparent vascular dilatation and present with bleeding or swelling. ADMA is an effective NOS inhibitor and may be a promising therapeutic option for hemorrhoidal disease. were analyzed using the chi-square test ( χ 2) test. Normally distributed continuous variables were presented as mean ± standard devia-tion. Differences in the two groups were analyzed with the Student’s t -test. To test the relationship between the variables, the Pearson Correlation was used. Indic-ative accuracy of distant markers was contrasted by analyzing the area below the receiver operating char-acteristic (ROC) curve, which was used to compare the diagnostic accuracy of the various markers. From the be another important factor of hemorrhoidal disease1.
INTRODUCTION
Hemorrhoidal tissues are normal anatomical and structural elements of the anal canal consisting of veins and muscle fibers. Hemorrhoidal disease is caused by the extension of these tissues due to several factors, including constipation, diarrhea, straining, and pregnancy. In recent studies, increasing microvascular density in hemorrhoidal tissue has been observed, suggesting that neovascularization might histopathological results were normal were included in the study as the control group. The biopsies of the control group were endoscopically performed with biopsy forceps on the normal rectal mucosa close to the anus. Resected pieces were washed with saline solution. The study and control groups were referred to as Group 1 and Group 2, respectively. The samples were stored at −80°C for biochemical evaluation.
Preparation of tissue samples
The Group 1 specimens were homogenized in a four-fold volume of phosphate-buffered solution (PBS, pH: 7.4) using a homogenizer (Next Advance Bullet Blender Storm 24). To remove debris, the homogenate was centrifuged at 3000 g for 10 minutes. The clear upper supernatant was taken, and tissue analyses (NOx, eNOS, and ADMA) were performed. All the experimental procedures were performed at +4°C.
Measurement of NOx, eNOS, and ADMA levels in tissue Tissue NOx, eNOS, and ADMA levels are measured by Enzyme-Linked ImmunoSorbent Assay (ELISA) kit (SinoGeneClon Biotech Co., Ltd., HangZhou, China) as per the manufacturer's instructions. The coefficients of intra-and inter-assay variation (% CVs) of NO, eNOS, and ADMA were less than 10%.
Statistical analysis
The number of patients in the groups was calculated for NO, eNOS 3, and ADMA parameters using a testing power of 95%. The minimum number of patients in the groups with and without the disease (hemorrhoidal disease) was calculated as 16 for NO and eNOS and 18 for ADMA according to Power analysis. In our study, a total of 108 patients were measured, 54 patients in each group.
Statistical analysis was performed using the Statistical Package for the Social Sciences (version 21.0). All data were first checked for normality. Then, the categorical variables were analyzed using the chisquare test (χ 2) test. Normally distributed continuous variables were presented as mean ± standard deviation. Differences in the two groups were analyzed with the Student's t-test. To test the relationship between the variables, the Pearson Correlation was used. Indicative accuracy of distant markers was contrasted by analyzing the area below the receiver operating characteristic (ROC) curve, which was used to compare the diagnostic accuracy of the various markers. From the be another important factor of hemorrhoidal disease1.
Nitric oxide (NO) is a potent vasodilator that is synthesized from L-arginine by one of the following three nitric oxide synthases (NOS): inducible NOS (iNOS), neuronal NOS (nNOS), or endothelial NOS (eNOS). NOS was reported to increase significantly in hem-orrhoids2. Although the vasodilator functions of NO are amply studied in the gastrointestinal tract, the relative contributions of NOS isoforms to hemorrhoids are unclear.
Asymmetric dimethylarginine (ADMA) is the major endogenous inhibitor of all three NOS isoforms. ADMA is produced from the proteolysis of the proteins that contain methylated arginine 3 . A competitive inhibitor of endogenous NOS, ADMA results in a reduction of NO production 4 . Increasing plasma ADMA levels have been reported in disease pathology in a variety of conditions that were characterized by endothelial dysfunction, including hypertension, hypercholesterolemia, renal failure, tobacco exposure, and hyperglycemia 5,6 . ADMA has also been indicated as an independent risk factor for coronary heart disease and endothelial dysfunction 7 .
There is no adequate information in the literature regarding the relationship between human hemorrhoids and ADMA. Therefore, this study aimed to investigate the roles of NO, eNOS, and ADMA in the pathophysiology of hemorrhoidal disease.
METHODS
The protocol was approved by the local Ethics Committee of the Istanbul Education and Research Hospital (verdict number: 2019/2037) and was conducted in accordance with the Declaration of Helsinki. This study included 54 patients with grades 3 and 4 internal hemorrhoidal disease who attended the General Surgery Clinic. All subjects were of Turkish descent. They all provided informed consent for inclusion before study participation was initiated. In our clinic, the Milligan Morgan procedure was performed on patients with grades 3 and 4 internal hemorrhoids under general anesthesia. All patients with hemorrhoidal disease in the rectal mucosa underwent rectal biopsy. The area was washed with saline solution and evaluated. The control group was comprised of patients who underwent colonoscopy; those with hemorrhoidal disease, malign disease, and inflammatory bowel disease were excluded from the study. Patients who underwent rectal biopsy and whose results of the ROC curve, odds ratios were calculated for cut-off points by multivariate analysis. Differences were considered significant when p< 0.05.
RESULTS
The subject characteristics and circulating concentrations of biochemical parameters are found in Table 1. The patients (Group 1) had statistically significantly higher NOx and eNOS levels (for both p<0.001) and lower ADMA levels (p<0.001) than the control subjects (Group 2).
In addition, the AUC of NOx was 0.959 (95% CI: 0.909-1.00) with 100% specificity and 94.4% sensitivity for a cut-off point at 125.60 mmol/L ( Table 3). Multivariate analysis showed that if eNOS is greater than 140.65 pg/mL, the disease risk increases 2010-fold (p=0.0002), and if NOx is greater than 125.60 mmol/L, the disease risk increases 185-fold (p=0.0006).
DISCUSSION
This preliminary study revealed that higher NOx and eNOS activities and lower ADMA levels in the rectal mucosa were observed in patients with hemorrhoidal disease than in those with normal rectal tissue. An imbalance between a endothelium-derived relaxing factor, such as NO, and the endogenous competitive inhibitor of NOS, ADMA, may cause hemorrhoidal disease. This study found that hemorrhoids display apparent vascular dilatation and manifest with bleeding or swelling. ADMA is an effective NOS inhibitor and it may be a promising therapeutic option for hemorrhoidal disease.
Hemorrhoidal disease is more common in people over 30 years old8. In this study, the mean age of the subjects was 37.5 years in Group 1 and 37.3 years in Group 2, consistent with what is found in the literature. There was no remarkable difference in age between Groups 1 and 2. There was also no remarkable difference in the incidence rates of hemorrhoidal disease in men and women8. the sex distribution of the patients included in this study was approximately the same, consistent also with the literature. There was no remarkable difference in sex between Groups 1 and 2. NO exerts physiological functions in the nervous and immune systems, contributing to behavior regulation, defense mechanisms against infectious disease, tumors, and gastrointestinal motility9. In this study, the amount of NO in Group 1 was remarkably higher than that in Group 2. Hemorrhoidal disease occurs due to the dilatation of hemorrhoid lumps that normally exist in the rectal area10. We consider that NO has an important role in the etiology of hemorrhoids, since it is released to lower pressure against causes that increase the pressure, like coughing, straining, and pregnancy. It supports our opinion that the NO level in rectal tissue excised from the patients with hemorrhoidal disease was higher than that of the control group.
Indeed, eNOS has received even more attention than NO due to its instability and the regulatory mechanisms of eNOS on NO production11. In hemorrhoids, NOS, an enzyme that synthesizes nitric oxide from L-arginine, was reported to increase remarkably12. García-Martín et al. 13 reported that the eNOS level was higher in people with migraines and had a history of migraines in their families and that eNOS inhibitors could be used in the treatment. There are publications revealing that a low level of eNOS is related to coronary artery disease14,15, essential hypertension16, and multiple sclerosis17. In our study, the eNOS level of Group 1 was remarkably higher than that of Group 2. It supports the opinion that eNOS and NOx have an effect on the occurrence of hemorrhoidal disease.
Lohsiriwat et al.2 observed NOS protein expression in tissue extracts of hemorrhoid and rectal tissue by Western blot analysis. Furthermore, they compared the expression levels to those of human microvascular endothelial cells. They also studied the distribution of all NOS isoforms in the tissue sections using immunohistochemistry. They provided further evidence that hemorrhoids have a higher protein expression of all NOS isoforms than the rectal tissue. There are considerably higher levels of nNOS and eNOS in the rectal tissue of patients with hemorrhoidal disease than in those with normal rectal tissue, suggesting that blood vessels in hemorrhoids are exposed to higher NO concentrations than those of normal rectal tissue. It appears that the bleeding or swelling caused by vascular dilatation might play an important role in hemorrhoidal symptoms and could be a potential target for medical treatment. NOS reduction, by applying NOS inhibitors, could likely improve these symptoms. There are publications revealing that ADMA, released as an endogenic and natural inhibitor of eNOS, increases hypercholesterolemia, coronary artery disease, and diabetes mellitus18. Even in a healthy population, high levels of circulating ADMA may be associated with higher rates of all-cause death19. Nevertheless, there is no adequate information in the literature regarding the relationship between human hemorrhoids and ADMA. Ragina et al.20 reported a clear and remarkable rise in systemic ADMA levels after laparoscopic colorectal surgery, even in the absence of surgical complications. The ADMA level of Group 1 was remarkably lower than that of Group 2. ADMA was very strongly negatively correlated with NO and eNOS. It is considered that vasodilatation of the hemorrhoidal masses is caused by the increase in ADMA release against increased NO and eNOS. Thus, NO release inhibition or insufficient ADMA levels can be one of the underlying causes.
Vasodilation in hemorrhoidal veins develops due to causes such as constipation, pregnancy, and coughing, which increase intraabdominal pressure. Therefore, the usability of NOS inhibitors in hemorrhoidal disease against increasing vasodilator NO and eNOS should be further studied.
We firmly believe that if our study had been conducted with a larger patient series, the results would have provided more guidance. The fact that ADMA, an endogenous NOS inhibitor, is associated with hemorrhoidal disease and other diseases has limited our ability to compare our study with others.
This study shows more detailed evidence that the rectal tissue of patients with hemorrhoidal disease has intense NOx and eNOS activities and lower ADMA levels than the normal rectal tissue, indicating that hemorrhoids are associated with noticeable vascular dilatation, high blood perfusion, tissue swelling, and bleeding tendency. If known changes in the blood flow in hemorrhoids can be explained by changes in NOx, eNOS, and ADMA levels described herein, a decrease in eNOS could potentially improve the hemorrhoid symptoms.
ADMA is an effective NOS inhibitor that may be a promising therapeutic option for hemorrhoidal disease. Further investigations are necessary to elucidate this hypothesis.
|
2020-09-17T13:06:18.842Z
|
2020-08-01T00:00:00.000
|
{
"year": 2020,
"sha1": "5cdd8cd7a6ee11a8a08746ee29788f79f49a123f",
"oa_license": "CCBYNC",
"oa_url": "https://www.scielo.br/j/ramb/a/fpq7Ln9SB6YRvTxppCVjQBf/?format=pdf&lang=en",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "4fa1ea8566ea3a84f7f33b9a59f2597e79faadb9",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
13806698
|
pes2o/s2orc
|
v3-fos-license
|
Patient-reported outcomes and associations with pleural effusion in outpatients with heart failure: an observational cohort study
Objectives We aimed to study whether patient-reported outcomes, measured by quality of life (QoL) and functional class, are sensitive to pleural effusion (PLE) in patients with heart failure (HF), and to study changes in QoL and functional class during follow-up of PLE. Methods A cohort of 62 patients from an outpatient HF clinic was included. The amount of PLE was quantified using a pocket-sized ultrasound imaging device. Self-reports of QoL and functional class were collected using the Minnesota Living with Heart Failure Questionnaire (MLHFQ) and the New York Heart Association (NYHA) functional classification. Results At baseline, 26 (42%) patients had PLE of which 19 (31%) patients had moderate to severe amounts of PLE. Patients with no to mild PLE had a lower MLHFQ score (mean 42, SD 21) compared with patients with a moderate to severe amount of PLE (mean 55, SD 24), p=0.03. For 28 patients (45%) with follow-up data, we observed a linear improvement of the MLHFQ-score (3.2, 95% CI 1.2 to 5.1) with each centimetre reduction of PLE. Correspondingly, patient-reported NYHA-class followed the same pattern as the MLHFQ-score. Conclusions Our study indicates that patient-reported outcome measures as MLHFQ may be sensitive tools to identify patients with HF at highest risk of symptomatic PLE and that treatment targeting reduction of PLE during follow-up is essential to improvement of QoL and functional capacity of outpatients with HF. Trial registration number NCT01794715; Results
INTRODUCTION
Pleural effusion (PLE) is a common sign associated with congestion and worsening of heart failure (HF) 1 and is often followed by dyspnoea. 2 However, the degree of dyspnoea may not correlate with the amount of PLE 3 4 and screening for PLE is often restricted to patients with HF with clinical signs of PLE.
It is well known that patients with HF experience significant reductions in quality of life (QoL). 5 6 Thus, initiatives to find treatment strategies that improve QoL have become an important objective in the follow-up of patients with HF. 7 In cancer care, initiatives to reduce PLE seem to be an essential palliative step, which has proven to improve QoL and decrease symptom burden without particular side effects. 8 9 However, poor self-reported QoL or functional capacity are not conventionally used indicators for screening and treatment of PLE in patients with HF. This may be due to the fact that it is largely unknown if QoL or functional class is associated with the presence of PLE in patients with HF. Further, to the best of our knowledge, no previous studies have investigated whether QoL and functional class improves in patients with HF when PLE is successfully treated. Thus, we aimed to study whether patient-reported outcomes (QoL and functional class) are sensitive to PLE, and to study changes in QoL and functional class during follow-up of PLE.
Strengths and limitations of this study
▪ Patient-reported outcome measures of quality of life (QoL) and objective measurements of pleural effusion (PLE) by point-of-care ultrasound was followed up repeatedly in a heart failure (HF) outpatient clinic. ▪ QoL is increasingly emphasised in treatment strategies and point-of-care ultrasound is recently recommended as a tool to identify PLE in European Society of Cardiology guidelines. ▪ The study was conducted at one single HF outpatient clinic and the sample size was small. However, the presented population is previously shown to be comparable with studies from other HF clinics with regard to HF severity, age, sex, New York Heart Association class and comorbidity. This was a prospective follow-up study with inclusion of 62 patients admitted to the outpatient HF clinic at the non-university Levanger Hospital, Norway between 15 April and 21 June 2013. 10 11 The sample size was determined in order to validate the feasibility and reliability of point-of-care ultrasound examination. 10 11 The HF diagnosis was confirmed according to the European Society of Cardiology (ESC) guidelines by clinical examination, 12 medical history and echocardiographic examination by one of four experienced cardiologists. 10 11 All patients provided their written informed consent to participate. The exclusion criteria were inability or unwillingness to consent or HF worsening requiring hospital admission at entry. The study observed patientreported QoL in the follow-up of a clinical trial regarding feasibility and reliability of point-of-care ultrasound in order to determine PLE (ClinicalTrials.gov: ID: NCT01794715). Data regarding feasibility and influence of point-of-care ultrasound have been recently published. 10 11 The observational part of the study was conducted according to the Second Declaration of Helsinki.
Follow-up
The follow-up of patients at the outpatient HF clinic was managed by two nurses specialised in cardiovascular and intensive care, working in close cooperation with four cardiologists. Briefly, the visits included patient education, self-management counselling and optimisation of the treatment for HF according to the 2012 ESC HF guidelines 12 in order to reduce the amount of PLE and improve the functional capacity, QoL and prognosis of the patients. None of the patients with HF needed therapeutic drainage of PLE. The follow-up schedule of the outpatient HF clinic was individualised depending on the condition of the patients. The number of visits ranged from one to four visits during the study period. The final follow-up was defined as the visit when the condition of the patient was satisfactory and stable. Twenty-eight (45%) patients had a final follow-up visit in the study period (figure 1).
Measurement of PLE
The nurses received a short, but dedicated training programme in focused point-of-care ultrasound examinations of the pleural cavities. The high feasibility and excellent reliability for quantification of PLE by point-of-care ultrasound performed by the nurses are comprehensively described in a recent publication. 10 Ultrasound examinations were performed using the Vscan (GE Ultrasound, Horten, Norway) at each visit. With the transducer placed in the intercostal space, the liver and spleen were used as reference points to identify the diaphragm of the right and left hemithoraces, respectively. During quiet breathing, the posterior chest was scanned along the paravertebral, scapular, posterior and medial axillary lines, continuously focusing on the diaphragm as a reference point. PLE was assessed in the costodiaphragmatic angle by assessing the dimension between the diaphragm and the lung surface (measured in the middle between the transducer and the mediastinum). The amount of PLE was categorised in four groups as (1) no PLE, (2) insignificant when present in the costodiaphragmatic angle only, (3) small to moderate when the measurement as described above was <3 cm and (4) significant when the measurement was ≥ 3 cm. We also dichotomised the PLE measures as no to mild (groups (1) and (2)) and moderate to severe (groups (3) and (4)).
Patient-reported outcome measures
The patients completed the questionnaires before the clinical examination at the baseline visit and at the final follow-up visit as recommended for patient-reported outcomes. 13 The functional class was self-reported according to the New York Heart Association (NYHA) classification 14 on the validated Norwegian version of this instrument. 15 QoL was reported by the validated Minnesota Living with Heart Failure Questionnaire (MLHFQ), which quantifies how patients with HF perceive their own QoL associated with the HF symptoms and treatment the preceding month. The questionnaire consists of 21 items where the patients are asked to rate how each item prevents them from living the life they desire using a sixpoint Likert scale from 0 to 5 (0= not at all, 5= all the time). The MLHFQ items include: physical HF symptoms (dyspnoea, fatigue, peripheral oedema and sleeping difficulties); psychological HF symptoms (anxiety and depression) and social/functional impairment due Figure 1 Shows the flow chart of study recruitment and follow-up. The follow-up schedule of the outpatient HF clinic was individualised depending on the condition of the patients. The number of visits ranged from one to four visits during the study period. The final follow-up was defined as the visit when the condition of the patient was satisfactory and stable. Twenty-eight (45%) patients had a final follow-up visit in the study period. HF, heart failure.
to HF (walking, climbing stairs, work, household or labour, need to rest, going places away from home, doing things with family and friends, eating, concentration, memory, loss of self-control, being a burden to others and sexual activity). The total score range from 0 to105, where the higher the score, the poorer the QoL. A reduction of 5 units or more in the MLHFQ-score by treatment is considered as clinically meaningful. 16 We defined a MLHFQ-score >40 to be an indicator for poor QoL as this cut-off includes the patients that scores in the upper quartile of the moderate QoL category. 17 Clinical examination, blood tests and medications The specialised nurses examined the patients at all visits. The clinical examinations included ECG (sinus rhythm or not), measurement of the blood pressure (mm Hg), heart rate (beats/min) and weight (to nearest 0.5 kg). 10 11 Blood samples were collected at each visit and analysed at the hospital's International Electrotechnical Commision (IEC) 17025-accredited laboratory. Creatinine was analysed by an enzymatic method developed by Abbott and N-terminal pro-brain natriuretic peptide (NT-proBNP) by electrochemiluminescence immunoassay (ECLIA). Estimated glomerulus filtration rate (e-GFR) was calculated by the chronic kidney disease epidemiology collaboration (CKD-EPI) formula. 18 The use of β-blockers was coded as 0 (no use) or 1 (use of ) and the same coding was applied for the prescription of ACE inhibitors and angiotensin receptor blockers (ARBs). For diuretics, we recalculated the doses of furosemide, hydrochlorthiazide and bendrofumethiazide to equivalent doses of bumetanide. In the analyses, 40 mg furosemide, 12.5 mg hydrochlorthiazide and 2.5 mg bendrofumethiazide were each recalculated to 1 mg bumetanide. 11
Statistical analysis
Comparisons of the baseline characteristics between the participants with no to mild PLE and the participants with moderate to severe PLE and between those participants that self-reported a MLHFQ-score ≤40 compared with those who scored higher were performed by t-test for continuous variables and χ 2 -test for categorical data. One-way analyses of variance were used in order to assess the differences between MLHFQ-scores in the different categories of PLE and NYHA-classes. For analyses of trend, we treated the PLE categories and NYHA-class as continuous variables in a logistic regression model. Pearson correlation was used to investigate the correlation between patient-reported NYHA-class and MLHFQ-score. Sensitivity, specificity, positive predictive and negative predictive values of the MLHFQ cut-off to detect PLE were calculated at baseline.
Generalised estimating equation with robust SEs and exchangeable correlation structure and inbuilt correction for age was used to model the change of the MLHFQ-score associated with change of PLE over time. For NYHA-class, we used the log link function of generalised estimating equation. In our second model, we adjusted for sex. In the third to seventh model, we added potential confounders (β-blockers, bumetanide-equivalent diuretic doses and ACE inhibitors/ARBs, change in weight and change of systolic blood pressure). These variables were added one by one to model 1, as the sample size was limited. We used a two-way graph based on linear regression to model the change in MLHFQ-score by change in centimetre (cm) PLE. We investigated the potential effect modification by sex and defined the critical p value to be <0.10 for the interaction term. For all other analyses we considered p <0.05 as significant. Statistical analyses were performed in Stata SE/13.1 for windows (© Stata Corp LP, Collage Station, Texas, USA). Table 1 shows the baseline characteristics of the patients (n=62) by PLE amount at baseline. As expected, the group with no or mild PLE had lower NT-proBNP at baseline than the group with moderate to severe PLE. There was a higher proportion of patients with sinus rhythm (49%) in the group with no to mild PLE compared with the group with moderate to severe PLE (21%). However, the two groups were comparable in age, sex, blood pressure, heart rate and the use of HF medication (all p>0.1). Self-reported MLHFQ-score >40 at baseline was associated with higher NYHA-class compared with MLHFQ-score ≤40 ( p<0.001) (data not shown).
RESULTS
As shown in figure 2, the patient-reported MLHFQ-score at baseline depended on the PLE quantity ( p=0.02). Approximately half of the patients with no PLE and most patients with moderate-to severe PLE had a MLHFQ-score >40. The mean MLHFQ-score in the patients with no or mild PLE was 42 (SD 21) compared with 55 (SD 24) in the patients with moderate to severe PLE ( p=0.03).
As shown in figure 3, self-reported NYHA-class and MLHFQ-score correlated well at baseline (r=0.63, p<0.001). MLHFQ-score was higher with more severe symptoms (NYHA-class 1 to 4) with p=0.001 for trend.
A MLHFQ-score >40 was observed for seven of nine patients with severe PLE amounts and for 15 of 19 patients with moderate to severe PLE (sensitivity 78% and specificity 49%). The corresponding positive and negative predictive values were 40% and 84% for the detection of moderate to severe PLE.
In total, the mean reduction of PLE during follow-up was 0.8 cm (SD 2.1 cm; range from −9.7 to +0.7 cm). For obvious reasons, the reduction of PLE was greatest among those with moderate to severe amount (−2.6 cm, SD 3.5, range from −9.7 to +0.7 cm) compared with those with only mild or no PLE (−0.07 cm, SD 0.2, range from −1.0 to 0 cm). During follow-up, the patients improved their MLHFQ-score by an average of 17 points (SD 27, range from −79 to +42). A total of 17 patients (61%) had a clinically significant improvement in MLHFQ-score of >5 points reduction and 25 (86%) patients reported an improvement in NYHA-class. There was no difference in the proportion of patients who experienced clinically meaningful improvement in MLHFQ-score by dichotomised PLE groups ( p=0.2). The improvement in MLHFQ-score correlated with improvement of NYHA-class (r=0.44, p=0.02). We observed a linear improvement in the MLHFQ-score (3.2 points, 95% CI 1.2 to 5.1) with each cm reduction in PLE. This improvement in MLHFQ was not explained by any of the adjustments for age, sex, β-blockers, bumetanide-equivalent, ACEI/ARB, change in weight or change in blood pressure over time (see table 2). Figure 4 shows the linear two-way association of MLHFQ-score with the reduction of PLE over time. The odds for reduction in NYHA-class per cm reduction in PLE was 1.06 (95% CI 1.04 to 1.08, p<0.001) and none of the seven adjustments models explained the association (data not shown). ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blockers; eGFR, estimated glomerulus filtration rate; MLHFQ-score, Minnesota Living with Heart Failure Questionnaire score; NT-proBNP, N-terminal pro-brain natriuretic peptide; NYHA-class, New York Heart Association functional classification.
DISCUSSION
Our study of outpatients with HF demonstrates that patient-reported QoL, by MLHFQ, is a sensitive tool for identifying patients with HF with severe amount of PLE. Furthermore, we found an improvement in MLHFQ-score and NYHA-class associated with reduction of PLE. A reduction of 1 cm of PLE corresponded to a reduction of 3.2 points in MLHFQ-score and 1.06 higher odds of improvement in functional class. In the models used, it seems that the change in PLE over time explains the improvement in MLHFQ-score and NYHA-class. Thus, paying attention to and treating PLE in follow-up of patients with HF may be important also with respect to improved symptoms and QoL.
To the best of our knowledge, no study has assessed the sensitivity of MLHFQ to detect PLE among outpatients with HF. We find that the MLHFQ is a sensitive tool for detecting severe PLE in HF follow-up, and that a MLHFQ-score >40 should warrant further examination to search for PLE. The MLHFQ-score cut-off at 40 ensured that the impact on QoL for most of the patients with moderate-severe PLE was observed. However, half of the patients without PLE had MLHFQ-score >40 and treatment of PLE based solely on QoL is not advisable. However, as early detection of PLE is crucial for optimal follow-up of patients with HF, a moderate or worsened QoL should drive further diagnostic approach, and thereafter treatment. The MLHFQ-score is an easy and cheap method to raise the suspicion of PLE in patients with HF, and thus, should be routinely applied.
Our study is also the first to describe the associations of reduced amount of PLE with improved self-reported MLHFQ-score and NYHA-class among patients with HF. The relation between HF symptoms and patientreported outcomes is an area outlined in need of future research. 19 The improvement in MLHFQ-score and NYHA-class were intercorrelated and in line with a previous study that showed improvement of MLHFQ-score during follow-up in a multidisciplinary HF programme. 20 Point-of-care ultrasound of the pleural cavities is a sensitive and specific test for detection of PLE, 1 21 and is previously shown to be superior to chest X-ray for diagnostics of PLE. 22 The current study indicates that caring for improvement of patient-reported outcomes is an additional reason for systematic use of ultrasound to assess PLE in patients with HF, 21 as excessive volume status could be an explanation for the poor QoL. However, if the availability of ultrasound is limited, a MLHFQ-score >40 identifies ∼80% of those with moderate to severe PLE and these should be referred to ultrasound.
A determinant for the improvement in MLHFQ-score in an earlier study was found to be correlated with increased systolic blood pressure. 20 In our study, adjustment for change in systolic blood pressure did not alter the improvement in MLHFQ-score associated with PLE reduction. However, the blood pressure was considered before titrating HF medication in order to prevent hypotension.
Respiratory symptoms have been reported to reduce QoL in patients with cancer 23 and PLE is associated with respiratory symptoms in patients with HF. 2 4 The observed improvement in QoL and NYHA-class may be influenced by the expected improvement in respiratory symptoms through other mechanisms causing dyspnoea in HF and not only by the PLE reduction per se. However, in palliative care, intermittent pleural drainage is often associated with superior palliation, improvement of QoL and decreased morbidity compared with no treatment of the PLE, 9 24 and we observed a reduction of PLE to be copresent with improvement in symptoms following targeted treatment. Nevertheless, the study also has limitations. The study was conducted at a single HF outpatient clinic and the sample size was small with limited number of patients with severe HF and large PLE. However, the presented population is previously shown to be comparable with studies from other HF clinics with regard to HF severity, age, sex, NYHA-class and comorbidity and thus, it is plausible that our results are generalisable to other HF clinics. 11 Assessment of HF symptoms is difficult because of the subjective nature of the symptoms. Incorporation of patient-reported outcome measures in clinical practice may be important to improve management and care for patients with HF. 13 25 26 Poor patient-reported MLHFQ-score and NYHA-classification are previously shown to be a strong predictor for early death in patients with HF. 5 20 25 27 Improvement of MLHFQ-score during follow-up of HF is previously shown to predict event-free survival. 20 28 29 The low number of participants do not allow for survival analyses. However, self-report to detect PLE is useful as patients may prefer a better QoL over prolonged life. 30 As patient-reported QoL is increasingly emphasised in treatment strategies, further studies should be performed in order to evaluate the need for more systematic identification and earlier intervention aiming to reduce PLE in follow-up of patients with HF.
CONCLUSIONS
The MLHFQ-score seems sensitive to detect severe PLE. Patient-reported outcome measures like the MLHFQ-score and NYHA-class were associated with the amount of PLE. Further, the long-term improvement in MLHFQ-score and NYHA-class were associated with reduced amount of PLE. Thus, reducing PLE may be important to improve the QoL in patients with HF. Routinely including patient-reported outcomes in follow-up of outpatients with HF may identify those at highest risk of having symptomatic amount of PLE and may allow for further improvement in the care for these patients. Further studies are needed to draw definitive conclusions regarding the interrelationship between PLE, QoL assessment and outcomes.
|
2018-04-03T04:14:03.349Z
|
2017-03-01T00:00:00.000
|
{
"year": 2017,
"sha1": "a269c4172f53f0ad07135e7c8c3933cd24cd91bb",
"oa_license": "CCBYNC",
"oa_url": "https://bmjopen.bmj.com/content/bmjopen/7/3/e013734.full.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "a269c4172f53f0ad07135e7c8c3933cd24cd91bb",
"s2fieldsofstudy": [
"Medicine",
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
233304879
|
pes2o/s2orc
|
v3-fos-license
|
Chemoradiotherapy alone or in combination with Endostar for patients with advanced non-small cell lung cancer: A systematic review and meta-analysis
Previous studies show inconsistent effect estimates for the efficacy of Endostar in patients with advanced non-small cell lung cancer (NSCLC) undergoing chemoradiotherapy. Therefore, this meta-analysis aimed to determine the effectiveness and safety on the basis of data obtained from available randomized controlled trials (RTCs). We find relevant articles reporting the use of Endostar combined with chemoradiotherapy regimens in the treatment of advanced NSCLC. The retrieval period was from June 2008 to June 2018. A total of 11 RTCs that recruited a total of 735 patients were included. Overall, the results indicated that patients who received Endostar plus chemoradiotherapy showed a significantly increased incidence of objective response rate (ORR) (relative risk [RR] = 1.48; 95% confidence interval [CI] = 1.31–1.67; P < 0.00001) and disease control rate (DCR) (RR = 1.17; 95% CI = 1.09–1.25; P < 0.00001) compared with those who received chemoradiotherapy alone. However, no significant difference was noted between groups for 1-year survival rate (RR = 1.06; 95% CI = 0.91–1.23; P = 0.48). Furthermore, combined Endostar with chemoradiotherapy did not yield a high incidence of stable and elevated Karnofsky performance score (RR = 1.06; 95% CI = 0.91–1.23; P = 0.48). Moreover, no significant difference was noted in the incidence of total toxicity between the two groups. The findings of our study indicated that treatment with Endostar plus chemoradiotherapy yielded a high incidence of ORR or DCR, but did not trigger excess adverse events in patients with NSCLC.
INTRODUCTION
Lung cancer is the most frequently occurring malignant tumor and is a threat to human health (1,2) . Non-small cell lung cancer (NSCLC) accounts for approximately 85% of lung cancers (3,4) . Since early diagnosis is still difficult, by the time a diagnosis is made, 70%-80% of the patients have missed the opportunity for radical resection (5) . Despite the variety of treatments present, the overall survival rate for advanced lung cancer is only 4-6 months, with a five-year survival rate of approximately 4.2% (6) . Chemoradiotherapy is the primary method of treatment for advanced NSCLC (7) , and was recommended in 2008 by the National Comprehensive Oncology Network of the United States as the standard regimen of unresectable NSCLC. However, this treatment is restricted by toxicity and side effects, and its therapeutic effect often plateaus (8) .
An increased understanding of tumor molecular biology has led to biological targeting drugs that have enriched the treatment of lung cancer and have become an important weapon in the treatment of cancer.
In 1971, Folkman proposed the theory of antiangiogenesis. Tumor progression is divided into two stages, prevascular and vascular phases. During prevascularization, the diameter of the tumor is less than 3 mm, and no angiogenesis is noted. During this period, nutritional uptake and excretion of metabolites by tumor cells are accomplished by simple diffusion.
During the vascular phase, neovascularization begins in the body of the tumor and establishes the microcirculation of the tumor itself. During this period, the tumor grows rapidly, and its malignant characteristics are revealed.
The transformation from prevascularization to vascular phase is known as the "angiogenesis switch" (9) . Thus, the idea for treating tumor by targeting antiangiogenesis was proposed. Injection of recombinant human vascular endostatin (Endostar, YH-16) is a novel multitargeting antiangiogenic drug that was developed by a gene recombination technique. Its mechanism of action is inhibition of tumor neovascularization by selectively inhibiting the migration of vascular endothelial cells to block the nutrient supply to tumor cells, using antiproliferation and antimigration effects, and promoting apoptosis. Since its target action on vascular endothelial cells is less toxic to normal tissue cells, it is less likely to cause bone marrow suppression and gastrointestinal reaction. In addition, the genotype of tumor vascular endothelial cells is stable, and does not tent to lead to drug resistance. Therefore, Endostar has a broad spectrum, low toxicity, and no drug resistance.
The Chinese version of lung cancer diagnosis and treatment guidelines were recommended, combining with vinorelbine and cisplatin regimen for the treatment of stage III/IV NSCLC for initial or recurrent treatment (10,11) . Recent studies showed that Endostar combined with chemoradiotherapy can improve the efficacy and 2 quality of life (QoL) of patients with NSCLC. However, the sample size of each single study was small and the quality is different. Therefore, we used a systematic evaluation and meta-analysis to systematically and objectively evaluate the efficacy and safety of Endostar combined with chemoradiotherapy in the treatment of NSCLC to provide more evidence-based medical evidence for its future application in the treatment of advanced NSCLC.
MATERIALS AND METHODS
This review was conducted and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis Statement issued in 2009 (Checklist S1).
Collection of study variables
The data that we extracted included: (1) number of patients in each RCT, (2) publication date of literature, (3) clinical characteristics, (4) clinical intervention methods, (5) objective response rate (ORR), disease control rate (DCR), and one-year survival rate, and (6) QoL and adverse effects (AEs).
Outcome definition
Outcomes were defined for RCTs using Endostar combined with chemoradiotherapy versus chemoradiotherapy in the treatment of advanced NSCLC.
Quality assessment of included randomized controlled trials
Quality evaluation was conducted according to the Cochrane Handbook (version 5.0.1) as follows: (1) methods used to randomize groups of patients, (2) how to perform adequate setting blinding, (3) how to perform an adequate allocation and conceal the sequence, (4) withdrawal and its handling, with or without a description of the number and reasons for withdrawal: low risk of bias, unclear risk of bias, and high risk of bias (12,13) .
Statistical analysis
The meta-analysis used RevMan 5.3 and STATA15 Software. The relative risk (RR) was used to measure the treatment in the study. The effect quantity was expressed as 95% confidence interval (CI). Heterogeneity among the results of the study was tested using chi-square test. The fixed effect model combination analysis was applied if the similarity among the studies in the subgroup was sufficient (I 2 < 50, P > 4.5). Conversely, using the random-effects model, the sensitivity of ORR and DCR was analyzed by removing single study methods, and subgroup analysis was carried out according to average age, pathological type, and quality evaluation grade.
Visual inspections of funnel plots were conducted. Egger and Begg tests were also used to statistically assess the publication bias for investigated outcomes. All reported P-values were two-sided, and P < 0.05 was considered statistically significant for all included studies.
Selection of studies
After preliminary screening, 65 articles were retrieved: 31 articles were summary or nursing reports, or did not used radiotherapy, two were infratests, and one study was combined with other tumors. Furthermore, eight studies were descriptive and lacked controls, three were not randomized, five were without synchronous chemotherapy, three did not treating pulmonary lesions, and one was a repeat article, leaving a final total of 11 articles included in the analysis (14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24) (figure 1). Manual searches of the references of the retrieved studies did not yield any further studies that met the inclusion criteria.
Baseline characteristics
RCTs of Endostar combined with chemoradiotherapy versus chemoradiotherapy alone to treat NSCLC were selected in this study. The baseline characteristics of the studies and patients are summarized in [15] Unclear No use No use Sufficient 2 Ding Y [16] Unclear No use Sufficient Sufficient 4 Jiang ZG [17] Unclear No use No use Sufficient 2 Yang Y [18] Unclear No use No use Sufficient 2 Chen XJ [19] Unclear No use No use Sufficient 2 Liu HW [20] Unclear No use No use Sufficient 2 Zhang Y [21] Unclear No use No use Sufficient 2 Zang Y [22] Unclear No use No use Sufficient 2 Liu L [23] Unclear No use No use Sufficient 2 Xu H [24] Sufficient No use No use Sufficient 3 Table 2. Quality analysis of included studies.
8
One-year survival rate There were five RCTs included in this study (15,16,19,23,24) . A fixed effect model meta-analysis was chosen because I 2 = 36%. The results showed that there was no significant difference in one-year survival rate between the two groups (RR = 1.06, 95% CI = 0.91, 1.23, P = 0.48). Analysis of sensitivity by removing a single item method did not have a significant effect on the overall result (figure 8).
QoL
There were four RCTs included in this study (1,9,11,15) . A random effect model meta-analysis was chosen because I 2 = 69%. The results showed that there was no significant difference in QoL between the two groups (RR = 1.20, 95% CI 0.95 = 1.51, P = 4.56). The funnel plot graph was asymmetric, indicating a possible publication bias (figure 9).
AEs
The incidence of radiation pneumonia, radiation esophagitis, bone marrow depression, liver function, renal function, digestive tract reaction, and cardiac toxicity was analyzed. There was no significant difference in any side effects between the two groups (I 2 < 50 for all, fixed effect model analysis) (table 3). Table 3. Comparative analysis of side effects between two groups.
This mechanism is mainly due to the presence of hypoxic cells in solid tumor tumors. The radiosensitivity of hypoxic cells is only one-third of that of oxygen-enriched cells. Vascular endothelial growth factor (VEGF) plays a key role in hypoxic cell generation and radiation resistance (33,34) . After radiotherapy, tumor cell proliferation was accelerated and the tumor blood vessels were damaged, making the existing blood vessels unable to supply oxygen effectively and aggravating the hypoxia of the tumor cells. Hypoxia can increase the activity of hypoxia-inducible factor. Furthermore, the hypoxia-inducible gene, VEGF, was activated (35) , leading to its overexpression (36) . High expression of VEGF in tumor tissues may cause tumor angiogenesis. Neoplastic vessels lack the characteristics of normal mature vascular tissues, and are distorted or disordered, and may form giant capillaries, known as sinusoid vessels. Abnormal arteriovenous anastomosis leads to increasingly poor circulation and exacerbates the anoxia of the tumor. This vicious circle eventually leads to tumor resistance to radiotherapy (37) . Recombinant human vascular endothelin (Endostar) can reduce the expression of VEGF, temporarily normalize the vascular structure of the tumor, improve the function of tumor blood vessel, enhance the cooperative use of tumor oxygen, and enhance the sensitivity of tumor cells to radiotherapy (38) . Endostar not only increases the sensitivity of radiotherapy, but also normalizes the tumor blood vessels and the microenvironment, and makes the vascular structure regular and the vascular basement membrane intact, increases perivascular Sertoli cells and the nutrition ability of the vascular supply, and enhances the antierosion ability, making it easier for drugs to 10 act on tumor cells and have synergistic effects with chemotherapy (39) . Moreover, Endol itself has the effect of influencing cell cycle distribution and inducing apoptosis (40) . Our results also showed an improvement in stabilizer rate of QoL was significantly higher in Endostar combined with chemoradiotherapy group (85.86%) compared with the chemoradiotherapy group (66.67%). This is likely related to the higher ORR and DCR in the combined Endostar and chemoradiotherapy group.
In addition, we analyzed the one-year survival rate of the Endostar plus chemoradiotherapy group (73.23%) and found it was higher than that of the chemoradiotherapy group (69.29%), but this was not statistically significant (RR = 1.06, 95% CI = 0.91, 1.23, P = 0.48). Moreover, analysis of the literature on QoL showed that the increased rate of Karnofsky performance score in the Endostar plus chemoradiotherapy group was more significant than that of the chemoradiotherapy group. The potential reason for this could be the high incidences of ORR and DCR in patients treated with Endostar plus chemoradiotherapy. However, the difference for QoL between groups was not statistically significant (RR = 1.20, 95% CI = 0.95-1.51, P = 0.12). These non-significant differences for one-year survival rate and QoL may be attributed to the smaller number of included studies that reported these findings; moreover, the power might not be enough to detect potential differences. Therefore, further large-scale RCTs should be conducted to verify these findings.
Previous studies reported that Endostar may reduce microvessel density, and induce cardiomyocytes, leading to cardiotoxicity (41,42) . Moreover, this is the main adverse reaction in clinical use of Endostar and is an important factor that limits its use (25,43) . However, in this study, we did not find a significant increased risk of serious cardiac toxicity in patients treated with Endostar. However, the probability of abnormal electrocardiogram in Endostar combined with chemoradiotherapy group (9.48%) was higher than that in chemoradiotherapy alone group (4.76%), but the difference was not statistically significant (RR = 1.99, 95% CI = 1.00, 3.96, P = 0.05). These results may show a publication bias, and more clinical studies are needed to verify this finding. In addition, the incidence of radiation pneumonia, radiation esophagitis, bone marrow depression, nausea, and vomiting were not increased in the Endostar combined with chemoradiotherapy group, which was consistent with previous studies (44,45) . The non-significant differences may be attributed to the low incidence of AEs, and the power was not enough to detect the potential differences. Therefore, we suggested that the combination of Endostar with chemoradiotherapy is safe and effective for use in the treatment of advanced NSCLC. This study has several limitations. First, most of the included studies lacked adequate subgroup analysis of data such as progression free survival and one-year survival rate. Second, the quality of the 11 articles included in this study was not high and there may have been a bias that affected the accuracy and reliability of the results. Third, the sample size of some studies is too small, and most patients were from China (because Endostar was approved by the China State Food and Drug Administration and applied in treatment of lung cancer), which may lead to geographical and ethnic differences. Finally, there are a few reports on the long-term curative effects; therefore, the long-term effects of Endostar plus chemoradiotherapy remains unclear.
CONCLUSIONS
We can conclude that Endostar combined with chemoradiotherapy may improve the ORR and DCR of patients with advanced NSCLC, and improve the QoL of patients. Furthermore, it was not shown to increase side effects, and is, therefore, worth considering in clinical practice. More high-quality clinical trials are required to verify this conclusion.
Declarations of interest: none.
Research involving human participants and/ or animals：none. (This paper is a metaanalysis) Informed consent: none
|
2021-04-20T07:42:34.512Z
|
2021-01-10T00:00:00.000
|
{
"year": 2021,
"sha1": "e810acc2871fc2aa511c72dc682a463a401997fa",
"oa_license": "CCBYNC",
"oa_url": "http://ijrr.com/files/site1/user_files_fad21f/nakisarezakhani-A-10-2055-80-d8472cd.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "0b9783a952d51b8468c4144ca14ecbd467259b72",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
119636000
|
pes2o/s2orc
|
v3-fos-license
|
Pluripolar graphs are holomorphic
We prove that the graph of a continuous function $f$, defined on a domain of ${\mathbb C}^n$, is pluripolar if and only if $f$ is holomorphic.
Introduction
A function ϕ defined on a domain U ⊂ C n with values in [−∞, +∞) is called plurisubharmonic in U if ϕ is upper semicontinuous and its restriction to the components of the intersection of a complex line with U is subharmonic.
A set E ⊂ C n is called pluripolar if there is a neighbourhood U of E and a plurisubharmonic function ϕ on U such that E ⊂ {ϕ = −∞}. By a result of B. Josefson [J], the function ϕ in this definition can be chosen to be plurisubharmonic in the whole of C n (i.e. U = C n ).
In 1963 T. Nishino raised the following question in connection to his paper [N1]: Let ∆ be the unit disc in C z and let f : ∆ → C w be a continuous function such that its graph Γ(f ) is a pluripolar subset of C 2 z,w . Does it follow that f is holomorphic?
The main result of this paper gives a positive answer to Nishino's question and can be formulated as follows.
Theorem.
Let Ω be a domain in C n and let f : Ω → C be a continuous function. The graph Γ(f ) of the function f is a pluripolar subset of C n+1 if and only if f is holomorphic.
As a consequence of Theorem one can easily obtain the following more general statement.
Corollary.
Let Ω be a domain in C n z and let E be a closed subset of Ω × C w ⊂ C n+1 z,w such that the fibers E(z) = {w ∈ C w : (z, w) ∈ E} of E are finite and depend continuously on z ∈ Ω in the Hausdorff metric. Assume that the number #E(z) of points in the fiber E(z) is bounded from above in Ω. Then E is a pluripolar subset of C n+1 z,w if and only if it has the form where the functions a 1 (z), a 2 (z), ..., a m (z) are holomorphic in Ω.
Note that the proof of Theorem can not be directly applied to the set E described in Corollary. Namely, the topological argument used in the proof of Lemma 3 and based on the fact that the first homology group H 1 (Ω×C w \Γ(f ), Z) is nontrivial does not work in this case. In the last section of the paper we construct an example of a compact subset E of∆×C w ⊂ C 2 z,w (∆ = {z : |z| < 1}) with finite fibers E(z) depending continuously on z ∈∆ in the Hausdorff metric such that H 1 (∆ × C w \ E, Z) = 0. In particular, there is a neighbourhood U(E) of E which does not contain any subset of∆ × C w defined by a Weierstrass pseudopolynomial (i.e. defined by the equation (1) with a 1 (z), a 2 (z), ..., a m (z) being continuous functions in Ω).
Remark. In the special case when the function f is assumed to be C 1 -smooth and its graph Γ(f ) is assumed to be complete pluripolar (i.e. Γ(f ) = {ϕ = −∞} for some function ϕ, plurisubharmonic in a neighbourhood of Γ(f )), a positive answer to Nishino's question was given by Ohsawa [O] using L 2 estimates for ∂. In this case one can also apply Pinchuk's method adapted to C 1 -surfaces in [ČH, p. 59-62] and construct, to get a contradiction, a one-parameter family of holomorphic disks {D α } attached to a totally real piece of Γ(f ) by an arc on the boundary. Restricting the plurisubharmonic function ϕ such that Γ(f ) ⊂ {ϕ = −∞} to each of these disks, we get ϕ ≡ −∞ on D α and, hence, α D α ⊂ {ϕ = −∞} which gives the desired contradiction, since the set α D α has real dimension 3. Note that neither of the methods mentioned here can be applied to prove our Theorem.
Preliminaries
For bounded nonempty sets E 1 and E 2 in C w , the Hausdorff distance is defined as A family of compact sets E(z) in C w parametrized by z ∈ Ω ⊂ C n z is said to be continuously dependent on z in the Hausdorff metric if, for each sequence {z n } of points in Ω converging to a point z 0 ∈ Ω, one has d(E(z n ), E(z 0 )) → 0 as n → ∞. In particular, if Ω is a domain in C n z and E is a nonempty closed subset of Ω×C w with bounded fibers E(z) = {w ∈ C w : (z, w) ∈ E} depending continuously on z ∈ Ω in the Hausdorff metric, then each fiber E(z), z ∈ Ω , is nonempty.
For a compact set K in C n , the polynomial hullK of K is defined aŝ The set K is called polynomially convex ifK = K.
The first simple lemma is classical and follows, for example, from the theorem 4.3.4 in [H].
Lemma 1 A compact set K in C n is polynomially convex if and only if for any point Q ∈ C n \ K there is a function ϕ, plurisubharmonic in C n , such that sup z∈K ϕ(z) < ϕ(Q). (2) Lemma 2 Let K be a polynomially convex compact set in C n and let E be a pluripolar compact set in C n . Then the set ( K ∪ E) \ K is pluripolar.
Proof. From pluripolarity of the set E it follows that there is a function ϕ E , plurisubharmonic in C n , such that E ⊂ {ϕ E = −∞}. To prove Lemma 2, we shall prove that ( ∈ K, and since the set K is polynomially convex, it follows from Lemma 1 that there is a function ϕ K , plurisubharmonic in C n , such that sup z∈K ϕ K (z) < ϕ K (Q). Then, for ε positive and small enough, one also has that sup z∈K (ϕ K (z) + εϕ E (z)) < ϕ K (Q) + εϕ E (Q). Since ϕ E (z) = −∞ for z ∈ E, it follows that sup z∈(K∪E) (ϕ K (z) + εϕ E (z)) < ϕ K (Q) + εϕ E (Q). By Lemma 1 applied to the function ϕ K + εϕ E , we get that Q / ∈ ( K ∪ E). This gives the desired contradiction.
The next statement was first proved by H. Alexander (see Corollary 1 in [A]). For the convenience of reading we include here its proof.
Lemma 3 Let U be a bounded domain in C z × R u ⊂ C 2 z,w (w = u + iv) and let g : bU → R v be a continuous function. Then U ⊂ π( Γ(g)), where Γ(g) is the graph of g and π : C 2 z,w → C z × R u is the projection.
Proof. Consider an approximation of the domain U by an increasing sequence {U n } of domains with smooth boundary. Further, consider a sequence of smooth functions {g n }, g n : bU n → R v , which approximate the function g, i.e., Γ(g n ) → Γ(g) in the Hausdorff metric. Then it follows from the definition of polynomial hull that lim sup n→∞ Γ(g n ) ⊂ Γ(g), where convergence is understood to be in the Hausdorff metric. Hence, it is enough to prove the statement of Lemma 3 in the case where the domain U has a smooth boundary and the function g is smooth. Now we argue by contradiction and suppose that there is a point Q ∈ U \ π( Γ(g)). Without loss of generality we may assume that Q is the origin O in C z × R u . We know by Browder [B] thatȞ 2 ( Γ(g), C) = 0 (hereȞ 2 ( Γ(g), C) is the secondČech cohomology group with complex coefficients). Then, by Alexander duality (see, for example [S], p.296), we get is the first singular homology group with complex coefficients). On the other hand, since O ∈ U \ Γ(g), it follows that the curve γ R consisting of the segment {z = 0, u = 0, −R ≤ v ≤ R} and the half-circle {z = 0, w = Re iθ , − π 2 ≤ θ ≤ π 2 } do not intersect the set Γ(g) for R big enough. Moreover, the linking number of Γ(g) and γ R is not equal to zero. Therefore, H 1 (C 2 z,w \ Γ(g), C) = 0. This is a contradiction and the lemma follows.
Lemma 4 Let U be a simply connected domain in C z and let f (z) = u(z)+iv(z) : U → C w be a function such that both u(z) and v(z) are harmonic in U. If the graph Γ(f ) of the function f is a pluripolar subset of C 2 z,w , then f is holomorphic.
Proof. If f is not holomorphic, we argue by contradiction and suppose that the set Γ(f ) is pluripolar. Then there is a function ϕ, plurisubharmonic in C 2 z,w , such that Γ(f ) ⊂ {ϕ = −∞}. Letṽ be the harmonic conjugate function to u in the domain U such thatṽ(z 0 ) = v(z 0 ) for some fixed point z 0 ∈ U. Then the set {z ∈ U :ṽ(z) + ε = v(z)} is nonempty and consists of real analytic curves for all ε small enough. Therefore, each of the holomorphic curves Γ ε = {(z, w) : z ∈ U, w = u(z) + i(ṽ(z) + ε)} intersects the set Γ(f ) ⊂ {ϕ = −∞} in real analytic curves. Since a real analytic curve is not polar (see, for example [T,Th.II.26,p.50]), it follows that Γ ε ⊂ {ϕ = −∞} for all ε small enough. This implies that ϕ ≡ −∞ in C 2 z,w and gives the desired contradiction.
Proof of Theorem and Corollary
Proof of Theorem. If the function f is holomorphic, then the same argument as in the proof of Lemma 4 shows that Γ(f ) is pluripolar. Namely, the function To prove that f is holomorphic we consider two cases.
1. The special case n = 1. In this case Ω is a domain in C z and f (z) = u(z) + iv(z) : Ω → C w is a continuous function such that its graph is pluripolar. Since holomorphicity is a local property, we can restrict ourselves to the case when Ω is a disc in C z and, moreover, to simplify our notations, we can assume without loss of generality that Ω = ∆ = {z : |z| < 1} is the unit disc and that the function f is continuous on its closure∆. It follows from Lemma 4 that either the function f is holomorphic or at least one of the functions u and v is not harmonic. Since both cases can be treated the same way, we can, to get a contradiction, assume that the function u is not harmonic. Denote byũ the solution of the Dirichlet problem on ∆ with boundary data u. Since u is not harmonic, one has thatũ = u in ∆. Without loss of generality we can assume that for some z 0 ∈ ∆. Let Consider the set Lemma 5 The set K is polynomially convex.
Proof. To prove polynomial convexity of K we use Lemma 1. Consider an arbitrary point (z * , w * ) ∈ C 2 z,w \ K. If the point (z * , w * ) belongs to the set then inequality (2) will be satisfied for the point Q = (z * , w * ) and the function and consider a functionũ ε harmonic on the whole of C z such that max z∈∆ |ũ(z) −ũ ε (z)| < ε. Since for (z, w) ∈ K one has u ≥ũ(z) ≥ũ ε (z) − ε, and since u * =ũ(z * ) − 3ε <ũ ε (z * ) − 2ε, it follows that inequality (2) will be satisfied for the point Q = (z * , w * ) and the function we conclude from Lemma 1 that the set K is polynomially convex. This completes the proof of Lemma 5.
Consider now the domain
in C z ×R u and the real-valued function g(z, u) = v(z) on bU. Since sup z∈∆ |u(z)| ≤ C, one has sup z∈∆ |ũ(z)| ≤ C and henceũ(z) ≤ u(z) + 2C ≤ 3C. It then follows from the definitions of U and g that the graph Γ(g) of the function g is contained in the set Γ(f ) ∪ K. Therefore, we get Γ(g) ⊂ (Γ (f ) ∪ K). Since, by Lemma 3, π( Γ(g)) ⊃ U, we conclude that Consider the following open subset of U : Inequality (3) obviously implies that the setŨ is nonempty. Since, by the definition of the sets K andŨ , π(K) ∩Ũ = ∅, it follows from (4) that Since, by our assumption, the graph Γ(f ) of f is pluripolar, we conclude from Lemma 2 and Lemma 5 that the set ( Γ(f ) ∪ K) \ K is pluripolar, i.e., for some plurisubharmonic function ϕ.
From (3) one has that there is a neighbourhood V of the point z 0 in C z such that u(z) <ũ(z) for all z ∈ V . For each a ∈ C consider the complex line ℓ a = {(z, w) ∈ C 2 : z = a} and the set It follows from (5) and (7) that for a ∈ V the projection of E a on the real line ℓ a ∩ {v = 0} contains an open segment. Since a polar set in C has Hausdorff dimension zero (see, for example [T,Th.III.19,p.65]), it cannot be projected on an open segment in R. Therefore, the set E a is not polar. It then follows from (6) that ϕ ≡ −∞ on ℓ a . Since this argument holds true for all a ∈ V , we conclude that ϕ ≡ −∞ on C 2 z,w . This contradiction proves Theorem in the case n = 1.
Since, by our assumptions, the set Γ(f ) is pluripolar, there is a function ϕ, plurisubharmonic in C n+1 , such that Γ(f ) ⊂ {ϕ = −∞}. For all points a except for a pluripolar set in C n one obviously has that the function ϕ a k (z k , z n+1 ) = ϕ(a 1 , . . . , a k−1 , z k , a k+1 , . . . , a n , z n+1 ) is not identically equal to −∞ in C 2 z k ,z n+1 . For all such points a we can use the argument from case 1 and conclude from continuity of the function f a k : Ω a k → C z n+1 and from the inclusion Γ(f a k ) ⊂ {ϕ a k = −∞} that the function f a k is holomorphic. Since the complement of a pluripolar set is everywhere dense, it follows from continuity of f that the functions f a k are holomorphic for all a ∈ Ω. This argument holds true for any k = 1, 2, . . . , n, so we conclude from the classical Hartogs theorem on separate analyticity that the function f is holomorphic. The proof of Theorem is now completed.
Proof of Corollary. Since, by our assumption, the number #E(z) of points in the fiber of E is bounded from above in Ω, we can consider k = max z∈Ω #E(z) and then the open subset U = {z ∈ Ω : #E(z) = k} of Ω. Let z 0 be a point of U and let h i (z), i = 1, 2, ..., k, be the functions defining single-valued branches of E(z) in a neighbourhood U of z 0 . Since, by our assumption, E(z) depends continuously on z ∈ Ω in the Hausdorff metric, we conclude from Theorem that the functions h i (z) are holomorphic in U. Hence, F (z) = Π i =j (h i (z) − h j (z)) is a well defined holomorphic function in U such that for each z ′ ∈ b U ∩ Ω one has F (z) → 0 as z → z ′ , z ∈ U. Then the functioñ is continuous in Ω and holomorphic in U = Ω \ {z :F (z) = 0}. Therefore, by Radó's theorem (see, e.g. [Č], p. 302),F is holomorphic in Ω. In particular, the set {z ∈ Ω :F (z) = 0} is an analytic hypersurface. Consider now the function Π k i=1 (w − h i (z)) = w k + a 1 (z)w k−1 + ... + a k (z). Since a 1 (z), a 2 (z), .., a k (z) are symmetric functions of h 1 (z), h 2 (z), ..., h k (z), they are well defined and holomorphic in U. Moreover, since E(z) depends continuously on z ∈ Ω in the Hausdorff metric, these functions are locally bounded near the set Ω\U = {z :F (z) = 0}. It follows then from removability of analytic singularities that the functions a 1 (z), a 2 (z), ..., a k (z) are holomorphic in the whole of Ω. Since, by our construction, E = {(z, w) ∈ Ω×C w : w m +a 1 (z)w m−1 + ... +a m (z) = 0}, the corollary follows.
Remark. The statement of Corollary was first proved in [Sh] for sets represented by Weierstrass pseudopolynomials by a different (and more complicated) method. It was later observed independently by the author and by A. Edigarian [E] that the methods of chapter 4 in [N2] give a simpler proof for these sets.
Example
We first prove the following simple lemma.
Lemma 6 Let f and g be holomorphic functions, defined in a neighbourhood U of a point a ∈ C z , such that f (a) = g(a) and f ′ (a) = g ′ (a). Let r be a positive number such that∆ r (a) = {z ∈ C z : |z − a| ≤ r} ⊂ U and f (z) = g(z) for z ∈∆ r (a) \ {a}. Then for all sufficiently small ε > 0 the complex curve Σ ⊂ ∆ r (a) × C w defined by the equation is a branched covering over the disk ∆ r (a) with two branches and two branching points Proof. Equation (8) is quadratic with respect to w, hence Σ is a branched covering over ∆ r (a) with two branches. A point b is a branching point of Σ if for some w b one has 0 = G ) and then (8) Hence, in view of our choice of r, b → a as ε → 0. Then, using Taylor expansions of f and g at the point a, we conclude from (10) and the assumption f (a) = g(a) Construction of the set E. Let ρ be a smooth real-valued function defined on the segment [0, 1] such that for 2 3 ≤ t ≤ 1. Consider the set where, as above, ∆ = {z ∈ C z : |z| < 1} is the unit disk. This set has two branches over the disk ∆2 3 (0) with one branching point at z = 0. The branches are glued to each other along the circle A = {|z| = 2 3 , w = 0} and become one branch {w = 0} for 2 3 ≤ |z| ≤ 1. Consider some points A 1 = (a 1 , 0) and A 3 = (a 3 , √ a 3 ) of E 1 and a point A 2 = (a 2 , C) with a 1 , a 2 , a 3 and C real and positive such that 2 3 < a 1 < 1, 0 < a 3 < 1 3 and a 3 < a 2 < a 1 . Further, consider the complex line L ′ passing through the points A 2 and A 1 and the complex line L ′′ passing through the points A 2 and A 3 . Let a 1 , a 2 , a 3 be already chosen and consider C so big that the line L ′′ intersects E 1 in two points A 3 and A 3 ), with a ′ 3 real such that 0 < a ′ 3 < a 3 , and the line L ′ intersects E 1 only at the point A 1 . The set E will be constructed as a small deformation of the set E 1 ∪ ((L ′ ∪ L ′′ ) ∩ (∆ × C w )) near the points A k , k = 1, 2, 3, that creates, as in Lemma 6, two branching points instead of each self-intersection point.
Let r > 0 be so small that the disks∆ 1 =∆ r (a 1 ),∆ 2 =∆ r (a 2 ) and∆ 3 = ∆ r (a 3 ) neither intersect each other nor the circle {|z| = 2 3 } and, moreover, do not contain the point a ′ 3 . Denote by E 1 the set (E 1 ∪ L ′ ) ∩ (∆ 1 × C w ), by E 2 the set (L ′ ∪ L ′′ ) ∩ (∆ 2 × C w ) and by E 3 the connected component of the set Then each of the sets E k , k = 1, 2, 3, is the union of the graphs of two holomorphic functions f j k , j = 1, 2, having the same value and different derivatives, both of them real (which is easy to check by direct calculation) at the center of the respective disk ∆ k . Therefore, we can apply Lemma 6 to each of these sets and, if ε is small enough, we will get branched coverings Σ 1 , Σ 2 and Σ 3 over the disks ∆ 1 , ∆ 2 and ∆ 3 , respectively, with two branches and two branching points contained in the smaller disks ∆ ′ 1 = ∆ r 3 (a 1 ), ∆ ′ 2 = ∆ r 3 (a 2 ) and ∆ ′ 3 = ∆ r 3 (a 3 ). Moreover, since for each k = 1, 2, 3 the derivatives at the centers of the disks ∆ k of the functions f j k , j = 1, 2, are real, we conclude from (9) that one of the two branching points contained in ∆ ′ k is contained in the half-disk {z ∈ ∆ ′ k : Im z > 0}, while the other is contained in the half-disk {z ∈ ∆ ′ k : Im z < 0}. Since both branching points of each set Σ k are contained in the respective disk ∆ ′ k , the set Σ k ∩ ((∆ k \ ∆ ′ k ) × C w ) will be the union of the graphs of two holomorphic functionsf j k , j = 1, 2, defined on ∆ k \ ∆ ′ k and, moreover, if ε is small enough, then each functionf j k will be close enough to the corresponding function f j k . Define the functions for z ∈ ∆ k \ ∆ ′ k , k = 1, 2, 3, j = 1, 2. LetΣ k be the union of the graphs off 1 k and f 2 k . Now we can define the set E as Define also the set E reg as E with the circle A, the point A ′ 3 of the transversal self-intersection of E and all the branching points of E being removed. Then, by our construction, E reg is a smooth connected 2-dimensional surface transversal to the w-direction.
Note that each fiber E(z) of the set E has at most four points and that the fibers E(z) depend continuously on z ∈∆ in the Hausdorff metric.
Proof. Let a be a real positive number such that a 3 ≤ a < 1 3 . Consider the point A = (a, − √ a) ∈ E and a diskD s = {(z, w) : z = a, |w + √ a| ≤ s} so small that it intersects the set E only at the point A. We first prove that the circle C s = bD s is homological to zero in ∆ × C w \ E.
Consider the curve z(t) in C z defined as If π z : C 2 z,w → C z is the projection, then the curve z(t) admits a uniquely defined lifting by π −1 z to the piece-wise smooth curve γ in E with the initial point A. The curve γ is transversal to the w-direction and has one point of selfintersection, namely, the endpoint ( 2 3 , 0), where two smooth curves on the side {|z| < 2 3 } meet each other. The geometric description of the curve γ looks as follows. We start from the point A = (a, − √ a) and then, over the segment {a ≤ Re z < 2 3 , Im z = 0}, the curve γ is contained in the "lower" branch of the set E 1 , while over the segment { 2 3 ≤ Re z ≤ a 1 −r, Im z = 0}, γ is contained in the only branch {w = 0} of E 1 for {|z| > 2 3 }. Since both branching points of Σ 1 are contained in ∆ 1 = {|z−a 1 | < r}, and since only one of them is contained in the half-disk {z ∈ ∆ 1 : Im z > 0}, we conclude that over the segment {a 1 − r ≤ Re z ≤ a 1 + r, Im z = 0} the curve γ will "change from the branch E 1 to the branch L ′ ". Then, over the half-circle {|z − a 1 | = r, Im z > 0} and the segment {a 2 + r ≤ Re z ≤ a 1 − r, Im z = 0} , γ is contained in L ′ . After that, applying the same argument as we used for the segment {a 1 − r ≤ Re z ≤ a 1 + r, Im z = 0}, we conclude that, over the segment {a 2 −r ≤ Re z ≤ a 2 +r, Im z = 0}, the curve γ will "change from the branch L ′ to the branch L ′′ ". Then, over the segment {a 3 + r ≤ Re z ≤ a 2 − r, Im z = 0} and the half-circle {|z − a 3 | = r, Im z > 0}, γ is contained in L ′′ . After that, the same argument as above shows that, over the segment {a 3 − r ≤ Re z ≤ a 3 + r, Im z = 0}, the curve γ will "change from the branch L ′′ to the branch E 1 ". And finally, over the segment {a 3 + r ≤ Re z ≤ 2 3 , Im z = 0}, the curve γ is contained in the "upper" branch of E 1 up to the endpoint ( 2 3 , 0), where we meet the first part of the curve γ which is (for |z| < 2 3 ) contained in the "lower" branch of E 1 . For each z 0 ∈ π z (γ) and each s > 0, consider the sets and Ω s (z 0 ) = {(z 0 , w) : min Then, for s small enough, each set Ω s (z 0 ) is the union of finitely many (at most three) disks in {z 0 }×C w , which are disjoint if z 0 is far enough from the circle {|z| = 2 3 }, and is the union of two connected components, one of which is a disk and the other one is the union of two disks having nonempty intersection, if {|z 0 | < 2 3 } and z 0 is close enough to the circle {|z| = 2 3 }. As |z 0 | → 2 3 from the side {|z| < 2 3 }, the centers of the two disks constituting the second connected component of Ω s (z 0 ) become closer to each other, and for {|z 0 | ≥ 2 3 } this component becomes just one disk. Each set Ω s (z 0 ) has a natural orientation induced from C w and, hence, Γ s (z 0 ) = bΩ s (z 0 ) has also a natural orientation.
Consider the set Since the curve γ is piece-wise smooth, it follows from the definition of Γ s (z 0 ) that the set T s is a piece-wise smooth surface of dimension two in ∆ × C w with the boundary on the above chosen circle C s . Moreover, since γ is oriented, and since each set Γ s (z 0 ) is oriented, we can also orient the surface T s . Topologically, T s is a torus with a disk removed, C s being the boundary of this disk. Since the curve γ ⊂ E is transversal to the w-direction, we conclude that T s ⊂ ∆ × C w \ E for s sufficiently small. This implies that the homology class [C s ] of the circle C s in H 1 (∆ × C w \ E, Z) is trivial. Now we observe that, for each point (z, w) ∈ E reg , the circle C s (z, w) = {(z, w ′ ) : |w − w ′ | = s} is homological to zero, if s > 0 is small enough. Indeed, since the set E reg is connected, there is a smooth curveγ ⊂ E reg connecting the points A and (z, w). Then, for s > 0 small enough, the set M s = {(z, w ′ ) : |w − w ′ | = s, (z, w) ∈γ} is a smooth "cylinder" of dimension two which is contained in ∆ × C w \ E and has its boundary on C s (z, w) and C s . Therefore, the circles C s (z, w) and C s represent the same homology class in H 1 (∆ × C w \ E, Z). Since C s is already proved to be homological to zero in ∆ × C w \ E, it follows that C s (z, w) is also homological to zero in ∆ × C w \ E.
Finally, let C be any smooth closed curve in ∆ × C w \ E. Then, there is a 2-dimensional disk D smoothly embedded into ∆×C w such that C = bD. We can assume that the disk D is in general position, in particular, that D intersects E in finitely many points {(z p , w p )} k p=1 which are contained in E reg . Without loss of generality, we can also assume that D is parallel to the w-direction in a neighbourhood of each point (z p , w p ). Then the disks D s (z p , w p ) = {(z p , w ′ ) : |w p − w ′ | ≤ s} are contained in D for s > 0 small enough. Therefore, C = bD is homological to k p=1 bD s (z p , w p ) in ∆ × C w \ E, the homology being D \ k p=1 D s (z p , w p ). Since each circle C s (z, w) = bD s (z p , w p ) is already proved to be homological to zero in ∆ × C w \ E, we conclude that C is also homological to zero. The proof of the claim is now completed.
As an application of Claim 1 we show the following property of the set E.
Claim 2. There exists a neighbourhood U(E) of the set E which does not contain any subset of∆ × C w defined by a Weierstrass pseudopolynomial.
Proof. Assume, to get a contradiction, that every neighbourhood U(E) of E contains a subset defined by a Weierstrass pseudopolynomial. For R big enough consider the circle C R = {z = 0, |w| = R} ⊂ ∆ × C w \ E oriented counterclockwise in the w-variable. Then, in view of Claim 1, there is a 2-chain S such that bS = C R and supp S ⊂ ∆ × C w \ E. The last inclusion implies that there exists a neighbourhood U(E) of E such that supp S∩U(E) = ∅. By our assumption, there is a subsetẼ of U(E) which is defined by a Weierstrass pseudopolynomial, i.e. it has the form (1) with a 1 (z), a 2 (z), ..., a m (z) being continuous functions. Since supp S ∩Ẽ = ∅, the homology class [C R ] of the circle C R in H 1 (∆ × C w \Ẽ, Z) is trivial. Consider the continuous map Φ : ∆ × C w \Ẽ → S 1 defined by Φ(z, w) = w m + a 1 (z)w m−1 + ... + a m (z) |w m + a 1 (z)w m−1 + ... + a m (z)| .
Then, on one hand, [C R ] = 0 in H 1 (∆ × C w \Ẽ, Z) and, hence, Φ * ([C R ]) = 0 in H 1 (S 1 , Z). On the other hand, the term w m in the numerator of formula (11) will dominate for (z, w) ∈ C R , if R is big enough. Therefore, the degree of the restriction of Φ to C R (it is a map from S 1 to S 1 ) is equal to m. Hence, Φ * ([C R ]) = m [S 1 ] = 0 in H 1 (S 1 , Z). This gives the desired contradiction and proves the claim.
|
2019-04-12T09:11:52.756Z
|
2003-01-17T00:00:00.000
|
{
"year": 2003,
"sha1": "26984fe3c3ebf1eefa6078d8642e3e1a788516ef",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "62005e967b91a795288e38cc663a1070a601ae58",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
213420822
|
pes2o/s2orc
|
v3-fos-license
|
Multicriteria decision-support system to assess the potential of exclosure-based conservation in Ethiopia
Abstract Land degradation is a global challenge that affects lives and livelihoods in many communities. Since 1950, about 65% of Africa's cropland, on which millions of people depend, has been affected by land degradation caused by mining, poor farming practices and illegal logging. One-quarter of the land area of Ethiopia is severely degraded. As part of interventions to restore ecosystem services, exclosures have been implemented in Ethiopia since the 1980s. But the lack of tools to support prioritization and more efficient targeting of areas for large-scale exclosure-based interventions remains a challenge. Within that perspective, the overarching objectives of the current study were: (i) to develop a Geographic Information System-based multicriteria decision-support tool that would help in the identification of suitable areas for exclosure initiatives; (ii) to provide spatially explicit information, aggregated by river basin and agroecology, on potential areas for exclosure interventions and (iii) to conduct ex-ante analysis of the potential of exclosure areas for improving ecosystem services in terms of increase in above-ground biomass (AGB) production and carbon storage. The results of this study demonstrated that as much as 10% of Ethiopia's land area is suitable for establishing exclosures. This amounts to 11 million hectares (ha) of land depending on the criteria used to define suitability for exclosure. Of this total, a significant proportion (0.5–0.6 million ha) is currently under agricultural land-use systems. In terms of propriety river basins, we found that the largest amount of suitable area for exclosures falls in the Abay (2.6 million ha) and Tekeze (2.2 million ha) river basins, which are hosts to water infrastructure such as hydropower dams and are threatened by siltation. Ex-ante analysis of ecosystem services indicated that about 418 million tons of carbon can be stored in the AGB through exclosure land use. Ethiopia has voluntarily committed to the Bonn Challenge to restore 15 million ha of degraded land by 2025. The decision-support tool developed by the current study and the information so generated go toward supporting the planning, implementation and monitoring of these kinds of local and regional initiatives.
Introduction
Land and water resources degradation remains one of the major global threats. It is negatively impacting livelihoods and returns from large-scale development activities [e.g., sedimentation of water infrastructure (Obalum et al., 2012)] in most African countries. Recent estimates put the extent of degraded land in Ethiopia at more than a quarter of the area of the country. This affects nearly a third of the population (Chirwa, 2014;Gebreselassie et al., 2016). Depending on the type of land cover and the method of estimation, the average rate of soil erosion in the Ethiopian highlands ranges from 6 to 33 tons ha −1 yr −1 (Hurni et al., 2015). The total estimated loss of fertile topsoil in the highlands of Ethiopia varies between 941 million tons and 1.5 billion tons per year (Tamene and Vlek, 2008;Hurni et al., 2015). UNDP (2002) has estimated nutrient depletion as a result of soil loss at 30 kg ha −1 yr −1 of nitrogen and 15-20 kg ha −1 yr −1 of phosphorus. Such massive soil movement has on-site and off-site impacts.
The main causes of land and water degradation are complex and interactive. They include deforestation, soil erosion, agricultural land expansion and overgrazing (Tekle, 1999;Paulos, 2001;Nyssen et al., 2004;Hurni et al., 2005;Yirga and Hassan, 2010;Adimassu et al., 2017), as well as underlying drivers such as weak regulatory frameworks and institutions, demographic growth, unclear land-use rights, low empowerment of local communities and poverty (Haileslassie et al., 2005;Kirui and Mirzabaev, 2014). In Ethiopia, the direct cost of soil and essential nutrient loss due to unsustainable land management was estimated in agricultural GDP. A recent modeling study by Sonneveld (2002) estimated that the loss of agricultural value during the period 2000-2010 was about US$7 billion. A more recent study indicated that the annual cost of land degradation associated with land use and land cover change in Ethiopia is about US$4.3 billion (Gebreselassie et al., 2016). To address this problem, we need integrated actions covering different landscapes and involving various stakeholders (Pistorius et al., 2016). The government of Ethiopia, in collaboration with local and international communities, has been implementing several forms of land and water management interventions (Humphrey, 1999;Tamene and Vlek, 2008;Merrey and Gebreselassie, 2011). These interventions include construction of soil and water conservation structures, afforestation and establishment of exclosures in degraded landscapes. The Ethiopian government's voluntary commitment in the context of the Bonn Challenge to restore 15 million ha of degraded land is part of these efforts. In the context of rural Ethiopia, exclosure-based forest establishment and ecosystem service restoration provides a promising approach to reversing the widespread land degradation and supporting the livelihood expectations of millions of smallholder farmers.
Exclosures are areas fenced off from anthropogenic interventions such as wood harvesting, grazing by livestock and other agricultural activities with the objective of allowing natural regeneration of vegetation and restoration of degraded communal grazing lands and other suitable land cover and land use types (Mekuria et al., 2011a). In Ethiopia, establishment of exclosures started in the northern highlands four to five decades ago (Mekuria and Aynekulu, 2013).
This practice can also be applied to sensitive ecosystems (hillside forest stands) to limit anthropogenic interventions. It can be implemented in wetlands, around river or buffer areas as well as in degraded grazing and agricultural lands. Establishment of exclosures incurs less cost compared to other land restoration interventions such as bunds and terraces. It has its pros and cons . The most commonly cited opportunity costs of exclosures include decrease in communal grazing lands and curtailed availability of and access to fuelwood (Mekuria et al., 2011b(Mekuria et al., , 2015Mekuria and Aynekulu, 2013;Adimassu et al., 2017). Nevertheless, some of these opportunity costs can be minimized though integrating income-generating activities (e.g., apiaries) within exclosures Mekuria et al., 2017).
Scaling of exclosure-based ecosystem restoration in Ethiopia is constrained by several factors including the lack of a countrywide prioritization and targeting tool and a management plan following the establishment of an exclosure. The lack of incentives for forgoing limited short-term economic benefits is another complaint heard from local communities .
Geographic Information System (GIS) tools can be applied to develop a framework to support the targeting of appropriate areas for various interventions. Multicriteria analyses too are appropriate for land suitability assessment (Venkatesan et al., 2010;Worqlul et al., 2015;Akyol et al., 2016;Worqlul et al., 2017;Schmitter et al., 2018). However, there has been limited application of a multicriteria framework to identify and target potential exclosure areas in Ethiopia. The objectives of this study therefore were: (i) to develop a GIS-based multicriteria decision-support tool that helps us to identify suitable areas with potential for exclosure-based ecosystem service restoration; (ii) to provide spatially explicit information on potential exclosure areas disaggregated by river basin and agroecology and (iii) to conduct an ex-ante analysis of the potential of exclosure areas for improving ecosystem services in terms of an increase in above-ground biomass (AGB) production and carbon storage.
The current study has multiple benefits: first, the tool we developed from this study can be applied elsewhere to out-scale exclosure. The methodology relies on open source software and open license data sources, which allows users to improve the tool. Secondly, information generated through this study could support planning, implementation, monitoring and evaluation of land restoration through exclosures. For instance, it can support ongoing efforts to sustain the benefits of water infrastructure (e.g., hydropower dams) by providing information on priority areas for establishing exclosures and help in the protection of man-made and natural water bodies from erosion and sedimentation. This study also attempts to provide information on changes in the above-ground carbon stock following the establishment of exclosures. Therefore, programs such as REDD + (Reduction of Emission from Deforestation and Degradation) and Ethiopia's voluntary commitment to the Bonn Challenge and the project to plant 4 billion trees could benefit from this study.
Study area and selection of validation points
This study focused on Ethiopia at multiple but interactive scales. The overall analysis was done at the national level and disaggregated for the 11 river basins of the country (Fig. 1A). Further, two sample basins were selected and disaggregated by agroecology ( Fig. 1B). At the lower end of the scale were three microwatersheds nested into the two sample basins and used for validation (note the dots in Fig. 1B).
Overall, we followed a two-step procedure to validate the analyses. First, we selected two basins: the Abay River and the Rift Valley lakes basins ( Fig. 1B and C). The key criteria we used to select the two basins included: (i) severity of land degradation and sedimentation of water bodies; (ii) the presence of large-scale development interventions (e.g., irrigation and hydropower developments); (iii) availability of diverse agroecological zones and ecosystem sensitivity to disturbance and (iv) potential for future development (McCartney and Girma, 2012;Haregeweyn et al., 2016). Secondly, we selected three sample watersheds, the Koga, Gumera and Hawassa-Zeway lake watersheds ( Fig. 1B and C), in the two basins for validation. The key criterion for selecting these watersheds was the presence of heterogeneous biophysical settings where exclosures have been implemented. A total of 385 ground truthing points were set up across the three watersheds.
General procedure
We developed a spatial analysis procedure using the GIS-based Multi Criteria Decision Analysis (MCDA) method to determine potential areas for exclosure establishment. The framework was used to estimate land suitability for scaling exclosures across scale in Ethiopia. The MCDA method enables us to structure decision problems and design, evaluate and prioritize alternative scenarios or decisions (Malczewski, 2006;Worqlul et al., 2015;Worqlul et al., 2017;Schmitter et al., 2018). MCDA also enables us to combine geographical data and value judgments (the decision-maker's preferences) to obtain information for decisionmaking (Malczewski, 2006).
Renewable Agriculture and Food Systems S89
The applied procedure involves seven steps ( Fig. 2): (i) definition of the conceptual framework; (ii) assigning a score/rank to each factor (Fig. 2B); (iii) creation of a constraint layer ( Fig. 2A); (iv) integration of thematic layers and spatial models (Fig. 2B); (v) validation of suitability mapping; (vi) overlay of the identified suitable areas on the actual land-use map to determine suitability in agricultural and non-agricultural land-use areas (Fig. 2C) and (vii) aggregation of suitability within river basins and disaggregation by agroecology (Fig. 2C). These steps are elaborated upon in the following sections.
Conceptual framework
To determine the suitability factors to aid us in identifying potential land for exclosures, we explored two scenarios based on: (i) the Food and Agriculture Organization [FAO (Sheng, 1990)] exclosure suitability criteria and (ii) the land-use policy of the Federal Democratic Republic of Ethiopia [FDRE (MoA, 2016)]. In both scenarios, the variables considered in the framework included slope, soil depth, biomass and proximity to wetlands (streams and lakes).
The criteria for suitability of land for exclosures vary as per FAO (Sheng, 1990) and FDRE (MoA, 2016) guidelines. According to the FAO criteria, areas with a slope above 46%, independent of land-use type, are suitable for establishing exclosures whereas the FDRE land-use policy suggests that degraded areas with a slope >50% are suitable. According to FAO, agricultural lands with a soil depth of 0-20 cm are suitable for establishing exclosures while the FDRE guidelines set this value at 0-25 cm.
With sedimentation and pollution posing an increasing threat to lakes and rivers, the Ethiopian River Basin Development Authority has launched several initiatives to delineate buffer zones around wetlands and put them under exclosures while investing in enrichment planting. Bekele-Tesemma et al. (1993) define and suggest attributes to delineate areas with low biomass potential (Bereha).
The key attribute to define these areas could be long-term biomass productivity. In this study, we applied the time series vegetation index [enhanced vegetation index (EVI)] to detect areas with longterm low biomass productivity. The following subsections provide details of the ranking of each of the factors used in the framework.
Assigning scores/ranks to each factor
The data identified and acquired for the analysis need to be reclassified so that they have the same range of values (Saaty, 1977;Schmitter et al., 2018). Accordingly, we classified each data layer as per a scale of 1-5, with 5 being the most suitable environment for exclosure land use and 1 being the least suitable ( Table 1). Ranking of classes within each factor was done on the basis of literature review (Sheng, 1990;MoA, 2016) and expert opinion. The ranking/reclassifying of values in each factor was done independent of the other factors. Figure 2B shows the factors that were included and reclassified. These were slope, soil depth, biomass, proximity to water bodies (streams and lakes).
(A) Slope: Slope is one of the factors that govern runoff and soil erosion and directly influence the infiltration of rain water.
Renewable Agriculture and Food Systems S91
The relationship between runoff and sediment yield and steepness of the slope is well-established (Haileslassie et al., 2005;Hurni et al., 2015). Runoff volume and sediment yield are higher in sloping terrain than in flat terrain (Hurni et al., 2015). A digital elevation model (DEM) with 30 m resolution from Earth Explorer web portal (http:// earthexplorer.usgs.gov/) was used to derive a slope layer (in percentage.) A score of 5 was given to a steep slope because higher runoff contributes to higher erosion. According to the FAO (Sheng, 1990), areas with a slope >46%, regardless of the land-use type, are suitable for exclosures whereas the FDRE land-use policy suggests a slope >50% (Table 1). (B) Soil depth: The depth of soil determines the use of land for a specific purpose: it limits the root depth and influences the drainage pattern. The shallower the soil, the less its capacity to store water, rendering it more prone to degradation. Implicitly, such land has to be put under exclosure and ecosystem service restoration. In this study, the soil depth layer was created from depth to bedrock with 250 m resolution as per the International Soil Reference and Information Centre [ISRIC (Hengl et al., 2015)]. According to the FAO-based scenario (Sheng, 1990), areas with a soil depth of 0-20 cm are suitable for exclosure land use while the FDRE land-use policy recommends a soil depth of 0-25 cm. In both scenarios, a score of 5 was assigned to shallow soil and a score of 1 to deep soil (Table 1) , 2012) and Google Earth highresolution satellite imagery. Areas with low EVI average were assigned a score of 5 since these areas have low biomass production compared with high EVI average areas, which received a score of 2 (Table 1). (D) Proximity to water bodies: The DEM was used to derive the stream network. Once the streams were derived, a layer showing the distance to stream was created. The distance to stream layer was reclassified into two classes: 0-30 m and above 30 m. A score of 5 was assigned to areas close to a stream, assuming that these areas are prone to erosion and needing protection. Areas far from the river were assigned a score of 2 (Table 1).
Lakes and reservoirs are affected by the human activities occurring around them. First, the water bodies layer was created from a land cover map released by the European Space Agency (Ramoino et al., 2016). This map, available for free, was created using Sentinel-2. The second layer created was a water occurrence layer based on images available since 1984 showing areas occurring from 0 to 100 of the observation periods and derived in GEE. This layer provided spatial information on water bodies. The water bodies layer from Sentinel-2 land cover and the water occurrence layer from GEE were merged into a single layer. This was then used to create the distance to lakes layer and reclassified into four classes: 0-150, 150-180, 180-8000 and greater than 8000 m. The areas close to lakes were assumed to contribute to sedimentation if not protected; and those areas far from lakes were taken to have less influence on sedimentation of lakes. Therefore, areas close to the lakes were assigned a score of 5 (Table 1).
Constraint layer
Naturally, there are areas where exclosure cannot be practised: for example, water bodies, extreme cold areas and extreme desert areas. Therefore, constraint factors, as illustrated in Table 1, were defined. A constraint layer was created for exclosure-unsuitable lands in agroecological zones, lakes and built-up areas ( Fig. 2A).
The input data for deriving the agroecological zones in this study were the 1 arcsecond scenes of the Space Shuttle Radar Topography Mission (SRTM), the DEM covering Ethiopia and the mean WorldClim rainfall data at 900 m resolution (http:// www.worldclim.org/). A total of 95 1 arcsecond scenes DEM were merged to derive a DEM covering the whole of Ethiopia. Two agroecological zones were identified as not suitable for agricultural land use (Bekele-Tesemma et al., 1993): (a) areas below 500 m a.s.l. with precipitation less than 900 mm where rainfed agriculture is not possible (desert); and (b) areas too cold regardless of the amount of rainfall (frost above 3700 m a.s.l.). Table 2 shows the different agro-ecological zones of Ethiopia based on rainfall and elevation as indicated in Bekele-Tesemma et al. (1993). Areas falling within the two classes were excluded from suitability analysis. Also, areas with a long-term EVI average below 0.15 were excluded as productive land-use practices are not practical there due to a combination of land-and weather-related factors.
Built-up areas are supposed to be excluded from the analysis. To derive the town layer, the National Oceanic and Atmospheric Administration (NOAA) Night Light data were used. The NOAA data were acquired in GEE with 4 km resolution (Gorelick et al., 2017). This and the slope layer (<15%) were used to derive a built-up constraint layer. Water bodies by themselves are assumed to be not directly under exclosure areas. Therefore, lakes and reservoirs were considered as a constraint layer. This constraint layer was created from the Sentinel-2 land cover and GEE water occurrence layer. All the individual constraint layers were then merged to create a single layer excluding areas falling in any one of the several constraints.
Integration of thematic layers and spatial model
Integration of the reclassified layers was done in three steps: (i) weighting each reclassified factor; (ii) merging the factors by applying the weights and (iii) reclassifying the result of the weighted overlay and merging steps. It is evident that the suitability and constraint factors being considered have varied degrees of influence and importance in determining exclosure suitability. Thus, defining weights is crucial for detecting the importance or preference of each factor relative to the others (Saaty, 1980). Accordingly, we transformed all the factors into similar units, and used pairwise comparisons to determine the weights of each factor relative to the other factors (Saaty, 1980). We used a standard scale with values 1-7 to determine the weight of each factor. This was done for each scenario resulting in two outputs, one for each scenario based on two scenarios, one based on FAO criteria (Sheng, 1990) and the other on FDRE (MoA, 2016) criteria. Weighting of the factors was applied in two parts. The first part considered the reclassified factors of slope, soil depth and biomass with weights of 0.6, 0.2 and 0.2, respectively ( Table 3). The second part considered the reclassified factors of proximity to water bodies and degraded wetland, slope and biomass by applying weights of 0.6, 0.2 and 0.2, respectively.
The reclassified factors were then merged by applying the weights assigned to each class. This was done for the first and second parts separately. Applying the weights in the first part produced the aggregated layer WF S1 using the reclassified factors (RF) of slope, soil depth and biomass with weights (F W ) of 0.6, 0.2 and 0.2, respectively, as illustrated by Equation (1): Applying the weights in the second part produced the aggregated layer WF S2 using the reclassified factors (RF) of proximity to water bodies and degraded wetland, slope and biomass with weights (F W ) of 0.6, 0.2 and 0.2, respectively, as illustrated by Equation (2): WF S1 and WF S2 were merged to produce a single thematic layer by taking the maximum of the values from the two layers. The output layer can have decimal values between 5 and 0. It is not possible to estimate area coverage of the different classes using decimal values. As a result, potential areas for exclosures following the multicriteria model were further reclassified into three classes: very highly suitable (>4.5), highly suitable (4.5 to 4) and suitable (4 to 3.5). Areas with a score below 3.5 were excluded from analysis as establishing exclosures in those areas is not possible or productive (Fig. 1B).
Land-use adjustment to conservation measures
We conducted land-use adjustment analysis in relation to the proposed suitability classes by overlaying a present land-use layer on the suitability maps produced for each of the two scenarios (Fig. 2C). The present land-use layer was created using the Sentinel-2 land cover (Ramoino et al., 2016) and agricultural land cover (IWMI, 2015). The outputs generated by this step indicate whether the land identified as suitable for exclosure corresponds to current degraded agricultural land or to current natural or perennial vegetation. Such analyses provide support to practitioners or land managers to make informed decisions on future land use and the adjustments needed. It helps also to Renewable Agriculture and Food Systems S93 understand the potential tradeoffs and mitigation measures needed as well as the need to implement land restoration interventions. In this regard the key challenge is that the current landuse map aggregates agricultural land but does not separate grazing and crop areas. Therefore, this study does not address the specific competition or tradeoff between land conversion to exclosure and other agricultural land-use types.
Validation of the suitability mapping
As indicated earlier, a two-step procedure was followed to validate the analyses. First, we selected two basins-the Rift Valley lakes and Abay River basins (Fig. 1)-and nested therein three sample watersheds for validation: Koga and Gumera in the Abay basin and the Hawassa-Zeway lakes in the central Rift Valley lakes basin ( Fig. 1B and C). The suitability mapping analysis was validated using ground truthing and recent high-resolution images from Google Earth. The information we collected included the presence of existing exclosures, agricultural land-use data, slope, status of land degradation, soil depth and proximity to water bodies. The current land-use type was observed for each point on Google Earth imagery, which showed detailed features around the selected points. The slope of each point was obtained from the DEM at 30 m resolution. Field data were collected from accessible areas in the three selected watersheds which included a total of 385 ground truthing points (253 in Hawassa-Zeway and 132 in Koga and Gumera). Compilation of the survey data from each ground truthing point provided information on: (a) exclosure suitability as per field observations and (b) exclosure suitability according to the suitability map. The surveyed points were assigned a suitability class value corresponding to the classes of the multicriteria model 5 indicating suitability or 1 nonsuitability for exclosure-based intervention. Accuracy assessment was done for each of the two scenarios (FDRE and FAO) using the observations and the corresponding map values. The overall accuracy was calculated by dividing the sum of the suitability class matching samples by the total number of sample points (Liu et al., 2007). Accordingly, the validation analyses conducted in the two basins indicated a suitability class accuracy of 90% for the FDRE scenario and 95% for the FAO scenario (Table 4).
Ex-ante analysis of above-ground biomass and carbon stocks
The potential accumulation of AGB is one of the ecosystem services that exclosure-based land restoration promises. In this study, we estimated this ecosystem benefit using Ethiopia's Forest Reference Level (EFRL) submission to the UNFCCC 2017 (MEFCC, 2017). The suitability map of exclosures was overlaid on the biome and agroecology map of Ethiopia to derive the area coverage of exclosures in different biomes and agroecologies. The biome data classify the country into biomes including Acacia-Commiphora, Combretum-Terminalia, Dry Afromontaine, Moist Afromontaine and Other (water bodies). The area coverage of exclosures in each biome was multiplied with the EFRL (tons ha −1 ) to derive the AGB in tons. The coverage considered only the very highly and highly suitable exclosure areas. Carbon stock was estimated by multiplying the estimated AGB by a carbon fraction of 0.47. A carbon fraction of 0.47 has been applied in Ethiopia, which is the default value for wood in the tropical and subtropical domains (MEFCC, 2017).
Results
Values and spatial distribution of land found suitable for exclosure-based ecosystem service restoration Table 5 and Figure 3 show the empirical values and spatial distribution of land found to be suitable for exclosure-based ecosystem service restoration under the FDRE-based (MoA, 2016) and FAO-based (Sheng, 1990) scenarios. Except for the lowlands in the northeastern, eastern and southeastern parts of Ethiopia, most of the areas were suitable for exclosures, in particular the central, northern, eastern and southern highlands, which were found to be highly suitable. The total estimated potential land was 9 million ha for the FDRE-based scenario and about 11 million ha for the FAO-based scenario. This is equivalent to 8 and 10% of Ethiopia's surface area under the two scenarios, respectively. Table 5 also shows data on enclosure-suitable areas that are currently under agricultural and non-agricultural activities. For example, as per the FDRE (MoA, 2016) scenario, 0.45 million ha of the suitable land are currently under agriculture. That is 0.4% of Ethiopia's land area. Of this, 0.08 million ha are classified as very highly suitable and 0.37 million ha as highly suitable for exclosures. Under the FAO-based (Sheng, 1990) scenario, about 0.64 million ha (0.57% of Ethiopia's land area) of the potential land are currently under agriculture. A significant proportion of the land, however, is under non-agricultural land use, including degraded lands or bush land, which have relatively lower opportunity costs when converted to exclosure compared to agricultural land. These non-agricultural lands found suitable for exclosure are distributed around the western and eastern parts of the country (Fig. 3). Values and spatial distribution of enclosure-suitable land by river basins Table 5 presents data on suitability of land for exclosure-based interventions in the major river basins of Ethiopia. Table 6 disaggregates the same attributes by agroecology for the Abay and Rift Valley basins. Figures 4 and 5 show the spatial distribution of this land in these two basins respectively. The major purpose of this exercise was to assess the extent of area that is suitable Over all accuracy (%) 90 93 The % is calculated in relation to respective river basin areas.
Renewable Agriculture and Food Systems S95
for exclosure-based ecosystem service restoration in the river basins and to prioritize these areas for investment to protect water infrastructure (such as hydropower reservoirs) from sedimentation. The results demonstrated that the Tekeze, Rift Valley, Mereb and Abay river basins contain the largest proportion of area suitable for exclosure (Table 5). Based on the FDRE (MoA, 2016) scenario, the Abay river basin showed a potential of about 2.2 million ha of land (11.2% of the basin area) that is suitable for exclosure-based ecosystem service restoration, whereas the Tekeze river basin contains about 1.8 million ha (22% of the basin area). The second scenario, based on FAO (Sheng, 1990), demonstrated a similar trend but with slightly different values across the river basins. A major part of the area in the Abay basin that was found to be suitable for exclosure is not currently under agricultural land use ( Table 5). The Moist Mid-highland and Moist Highland agroecological zones host major parcels of land suitable for exclosurebased ecosystem service restoration (Table 6).
In the central Rift Valley basin, a total of 61.8 × 10 3 and 69.8 × 10 3 ha of land were estimated to be suitable for exclosure under the FDRE-based (MoA, 2016) and FAO-based (Sheng, 1990) scenarios, respectively (Table 5). However, a considerable proportion of this (Sheng, 1990). The white areas are either constraints or areas not suitable for exclosure measures.
area is currently under agricultural land use ( (Sheng, 1990). The white areas are either constraints or areas not suitable for conservation measures.
Renewable Agriculture and Food Systems S97
use. The Rift Valley lake basin has relatively fewer agroecological zones; only subtle differences were observed among these different zones in terms of suitability for exclosure interventions.
Discussion
Decision-support tools to target and prioritize exclosure-based interventions can help countries in delivering their commitments In the context of the Bonn Challenge, Ethiopia has voluntarily committed to restoring 15 million ha of degraded land by 2025 (Pistorius et al., 2016). Establishment of exclosures can be considered as one of the various approaches to restore lands and enhance ecosystem services (Mekuria 2011a(Mekuria , 2011b. However, decision-support tools to prioritize investment and target hotspot areas for exclosure-based ecosystem service restoration have been lacking. In principle, such tools could enhance the success of regional and national initiatives. For example, they can support the ongoing efforts to sustain the benefits of water infrastructure (e.g., hydropower dams). This approach also includes procedures to prioritize areas for intervention and provides information on changes in the aboveground carbon stock following the establishment of exclosures. Therefore, programs such as REDD + (Reduction of Emission from Deforestation and Degradation) and Ethiopia's voluntary commitment to the Bonn Challenge could benefit from this approach and its decision-support tools. The approach employs freely accessible data including Sentinel-2 land cover (Ramoino et al., 2016), DEM, EVI, water occurrence (Gorelick et al., 2017) and soil depth (Hengl et al., 2015).
Despite these contributions, four limitations were identified in this study. First, our analysis used a land cover data layer (IWMI, 2015) that aggregates all agricultural lands (e.g., crop and grazing land, large-scale specialized farms and smallholder mixed farms) as one class. This impeded an in-depth understanding of the competing factors and the socio-economic implications of land conversion to exclosure. Secondly, it is obvious that the type of land ownership, current economic value and the sociocultural attachment of the community determine land conversion to exclosure. Due to its limited scope, the current study did not explore this dimension. Thirdly, the agricultural land-use data used in this study had a resolution of only 250 × 250 m 2 . Use of medium-resolution land-cover maps would increase the accuracy of land adjustment maps. Given the high level of agricultural land fragmentation in Ethiopia, using medium-resolution data would greatly increase the precision of this tool. Fourthly, the study covered different agroecological zones, river basins and associated livelihoods whereas the validation points come from only two geographic areas. Validation points from more areas would be more appropriate and would lead to a higher level of accuracy.
Targeting and prioritizing river basins for exclosure-based ecosystem service restoration sustains the longevity of water infrastructure The government of Ethiopia has pledged to restore 15 million ha of degraded land by 2025-which is equivalent to one-sixth of the country's total land area. Our study showed that over 60% of the pledged area could be restored through establishing exclosures. The Ethiopia Highland Reclamation Study (FAO, 1986) estimated that about 14.4 million ha in the Ethiopian highlands are severely (Sheng, 1990). The white areas are either constraints or areas not suitable for conservation measures.
degraded and need rehabilitation. The difference between the estimates presented in FDRE (2015) and our study can be explained by the fact that we were concerned only with exclosure-based rehabilitation and therefore excluded all areas that needed restoration but were not suitable for exclosure. Recent estimates show that currently about 4 million ha of land are under exclosure in Ethiopia. This indicates that there is room for expansion of the exclosure area by 60%. Information generated in this study will help in targeting and prioritizing land for expansion. Our results support the view that a considerable proportion of land that is currently being used for agriculture and nonagricultural purposes can be converted to exclosures to restore degraded landscapes or maintain non-degraded landscapes and improve ecosystem services. Several studies have reported that exclosures are effective in improving ecosystem services. The many benefits include increased vegetation and biodiversity (Mekuria and Veldkamp, 2012); enhanced ecosystem carbon stock (Mekuria et al., 2011b(Mekuria et al., , 2015; reduced soil erosion (Mekuria et al., 2009); restoration of soil fertility (Mekuria and Aynekulu, 2013); decreased runoff and sediment load (Tefera et al., 2005;Girmay et al., 2009) and increased incomes and improved livelihoods of smallholder farmers over the medium to long term (Babulo et al., 2006;Tilahun et al., 2007;Mekuria et al., 2011a). However, converting agricultural lands into exclosures needs careful planning, and should include, for example, creation of alternative livelihoods for farm households currently using the land for agricultural and livestock production. Exploring mechanisms for productive use of exclosures including apiaries and fattening could help mitigate the potential tradeoffs. Areas under non-agricultural land-use systems have fewer opportunity costs and thus could be considered as priority areas for exclosure intervention.
In Ethiopia, high rainfall in the Moist Highland and Moist Mid-Highland agroecological zones, where large tracts of land are suitable for exclosure interventions, results in high rainfall erosivity. In the Abay basin, major parts of these agroecological zones overlap with the class of land characterized by severe soil erosion [30-50 tons ha −1 yr −1 (Haregeweyn et al., 2017)], and hence were selected by the applied model as highly suitable for exclosure. An additional reason why these zones have large areas suitable for exclosures might be related to the rugged topography and their long cultivation history resulting in severe land erosion and shallow soil depth (FAO, 1986). The primary drivers, according to FAO (1986), are related to anthropogenic interventions such as cultivation on the steep slopes. Given the favorable climate, these areas have been under predominantly cereal cultivation, which has resulted in depletion of soil organic matter and increased erosion leading to shallow soils.
Our results show that most of the lands found suitable for establishing exclosures fall in areas that have large-scale development projects (e.g., Tekeze dam) or freshwater bodies (e.g., Lake Tana and Ziway). This suggests that exclosure interventions in such areas can be a viable strategy to reduce siltation, thereby protecting large development projects and natural lakes which play a key role in the livelihoods of the local community. Welde (2016) reported that degraded watersheds in the Tekeze river basin generated a significant amount of sediment, posing a threat to the Tekeze dam. Similarly, Haregeweyn et al. (2017), in a comprehensive assessment of the soil erosion risk in the upper Blue Nile basin, reported a continued increase in erosion and sedimentation despite decades of soil and water conservation efforts. Berhane et al. (2016) showed that more than 61% of 92 micro dams, mainly located in the subcatchment of the Tekeze basin, suffer from sedimentation.
Studies of the upper Blue Nile basin have indicated that the Great Ethiopian Renaissance Dam (GERD) itself could be threatened by excessive sedimentation unless proper soil and water conservation measures are implemented upstream (Haregeweyn et al., 2017). Therefore, in the Abay basin, restoring degraded landscapes through establishing exclosures could be an option to reduce siltation and increase the service years of water infrastructure.
The Rift Valley basin too is in need of proper use and management of resources as it contains sensitive ecosystems that are threatened by siltation, which has already resulted in the shrinking of lakes (e.g., Lake Abjata, Fig. 5). Our study identified major parts of the eastern and western highland areas of the Rift Valley as suitable and the valley floor as not suitable for exclosure. The volcanic ash (Andosols) in the valley floor (dry kola) is a major contributor to soil degradation and siltation of the lakes. However, as these areas act as sediment sinks for eroded material transported from the upland areas, they have deeper soil profiles. Therefore, based on the criteria used in our framework, these areas were classified as areas not needing exclosure. This entails that future exclosure mapping efforts need to include attributes that capture these gaps.
A closer look at the distribution and current use of exclosuresuitable lands in the Rift Valley showed that management of exclosure areas there would have more opportunity costs compared to those in the Abay basin. Exceptionally, some of these areas in the Rift Valley basin are covered by natural forests, plantations and perennial crops. While these areas are already protected, FAO (Sheng, 1990) criteria recommend that all land >46% regardless of their land use land cover type should be under exclosure. These ecosystems are important for the livelihoods of the local communities and for the environment. For example, areas around Lake Langanao serve as the habitat for different wild animals including the native mountain nyala (Tragelaphus buxtoni), and as sources of timber and fuelwood for household consumption and commerce. Exclosure-suitable areas in the Hawassa lake catchment are similarly important for the livelihoods of local communities. Sites around the lakes were deemed suitable for exclosure due to a combination of their proximity to the water body and the level of degradation (as in the case of Lake Abjata) or the slope of the surrounding terrain (as in the case of Lake Shalla). The Rift Valley basin authority is currently developing a manual for buffer zone protection around these areas, which supports our findings. This study will contribute to the implementation of this policy.
Ex-ante analysis of impacts of exclosure-based ecosystem service restoration on above-ground biomass and carbon stock The contribution of exclosures to AGB and carbon stock varies with agroecology and biomes (Table 7). Exclosure-suitable areas located in the moist mid-highland, dry lowland, dry mid-highland and wet mid-highland zones contribute the largest share of AGB and carbon (Table 7). Our study found that Acacia-Commiphora (AC) forests cover the largest area in the dry Lowland zone and Dry Afromontaine (DA) forest in the Moist Mid-Highland zone, contributing 47.8 and 132.2 tons of above-ground carbon, respectively.
Approximately 418.5 tons of carbon could be protected in the areas found to be exclosure-suitable in our study (Table 7). In Renewable Agriculture and Food Systems S99 moist highland and moist midland areas of Abay basin, exclosures could significantly contribute to climate change mitigation in addition to playing a role in the regulation of ecosystem services. Our results demonstrate that establishing exclosures in degraded landscapes could be an option to restore and protect the AGB and carbon and mitigate climate change. This assessment is only about protection; the actual value of additional storage as a benefit of exclosure needs to be assessed with more ground monitoring. In undertaking scaling of exclosures, the concerns of local communities including tradeoffs such as the need for grazing land and fuelwood need to be addressed. Restoration of degraded landscapes through establishing exclosures should be oriented toward managing and improving the productivity of degraded land such that the need for conservation of biodiversity and environmental sustainability and the demands of local people for biomass resources are both achieved. This concern might to some extent be addressed by planting forage species in the exclosures and practicing a cut-and-carry system. Furthermore, to deal with local people's concerns, exclosures have to be integrated with income-generating activities, for example by planting high-value trees . Also, rewarding or compensating farmers for their investments in soil and water conservation practices is crucial to sustain the exclosures and their benefits . This way exclosure land can meet both mitigation and adaptation to climate change.
The exclosure policy has been the cornerstone of the dynamics of landscape recovery (Nyssen et al., 2014). Exclosure-suitability mapping done by this study identified non-agricultural land use including forests and remnants of forest areas. For instance, nonagricultural exclosure-suitable sites identified in the Hawassa watershed included the remnants of a natural forest. While sustainably excluding such areas from human influence and implementing improved protection and management of the remnant vegetation, the highest priority should be given to enhancement of access to alternative sources of urban energy and encouraging changes in cooking habits (Nyssen et al., 2014). The issue of whether benefits from carbon sequestration (e.g., through carbon credit) can meet the livelihood expectations of farmers needs to be explored further. In the context of exclosure-based ecosystem restoration carbon sequestration is only one of the multiple ecosystem services, and to sustain exclosure-based interventions, these benefits must be diversified and include incentives that enable adaptation to climate change.
Conclusion
In this study, we attempted to develop a GIS-based MCDA method that helps to identify suitable areas with potential for exclosure-based ecosystem service restoration. From these results we concluded that MCDA can be applied to support new and ongoing global and local initiatives related to exclosure-based ecosystem service restoration. But this exclosure bio-physical suitability information is just a first planning tool. Ground surveys and local community consultations are required to ensure that interventions are tailored to local socio-ecological conditions.
The estimated extent of land areas suitable for exclosures showed wide variations between scenarios, river basins and agroecology, which could provide guidance on prioritizing areas for future investment. Specifically, about 9-11 million ha of land in Ethiopia are suitable for establishing exclosures (depending on the scenario). A closer look at these lands shows that a significant proportion of them are currently under agricultural land-use systems. Implicitly, converting these lands to exclosure would entail opportunity costs. In view of this, we conclude that successful exclosure-based ecosystem service restoration must give due consideration to socio-economic tradeoffs.
From our ex-ante analysis of the impacts of exclosure-based ecosystem service restoration, we estimated that about 418.5 million tons of above-ground carbon could be protected-with further additional storage in the long run-if the suitable areas under non-agricultural land use are protected and managed properly. But targeting exclosure only for climate change mitigation measures (e.g., carbon sequestration) might not serve the shortand medium-term livelihood expectations of the local people. Conversion of land to exclosures and implementation of improved management practices must be accompanied by income diversification (e.g., apiary, livestock fattening). This would serve as an incentive for adapting to exclosures and achieve climate change adaptation and mitigation goals.
|
2020-02-13T09:22:27.724Z
|
2020-02-10T00:00:00.000
|
{
"year": 2020,
"sha1": "96a4171af67a7c476cc6d2dbc02d87f0ae243c58",
"oa_license": "CCBY",
"oa_url": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/503FFF1F6BFB3F8140E97071961D2302/S1742170520000034a.pdf/div-class-title-multicriteria-decision-support-system-to-assess-the-potential-of-exclosure-based-conservation-in-ethiopia-div.pdf",
"oa_status": "HYBRID",
"pdf_src": "Cambridge",
"pdf_hash": "5a5a70838f6f21a0e4f44d3d17e166e70f507e20",
"s2fieldsofstudy": [
"Economics"
],
"extfieldsofstudy": [
"Geography"
]
}
|
24687681
|
pes2o/s2orc
|
v3-fos-license
|
Influence of Constriction, Wall Tension, Smooth Muscle Activation and Cellular Deformation on Rat Resistance Artery Vasodilator Reactivity
This study investigated how vasoconstriction (tone), wall tension, smooth muscle activation, and vascular wall deformation influence resistance artery vasodilator reactivity. Resistance arteries, from two different regional circulations (splanchnic, uterine) and from pregnant and non-pregnant rats, were cannulated and pressurized, or mounted on a wire myograph under isometric conditions prior to being exposed to both endothelium-dependent (acetylcholine, ACh) and -independent (sodium nitroprusside, SNP) vasodilator agonists. A consistent pattern of reduced vasodilator sensitivity was noted as a function of extent of preconstriction for both agonists noted in pressurized arteries. A similar pattern regarding activation was noted in wiremounted arteries in response to SNP but not ACh. Wall tension proved to be a major determinant of vascular smooth muscle vasodilator reactivity and its normalization reversed this pattern, as more constricted vessels were more sensitive to ACh relaxation without any change in SNP sensitivity, suggesting that endothelial deformation secondary to vasoconstriction augments its vasodilator output. To our knowledge, this is the first study to dissect out the complex interplay between biophysical forces impinging on VSM (pressure, wall tension), the ambient level of tone (vasoconstriction, smooth muscle cell activation), and consequences of cellular (particularly endothelial) deformation secondary to constriction in determining resistance artery vasodilatory reactivity.
Introduction
Biophysical forces that normally impinge on the vascular wall (wall tension, shear stress) exert significant physiological effects through the process of mechanotransduction [1] . For example, many studies have investigated the interaction between flow-induced shear stress and vascular reactivity, especially with regard to endothelial secretion of nitric oxide [2][3][4][5].
The effects of transmural pressure are also wellcharacterized vis-a-vis myogenic reactivity, in which wall 884 tension secondary to transmural pressure induces vasoconstriction through a combination of mechanisms that include vascular smooth muscle (VSM) membrane depolarization, calcium entry, and the activation of enzymes that modulate calcium sensitivity, such as Rho kinase and protein kinase C [1,[6][7][8]. In spite of considerable research effort, the cellular elements involved in sensing and mediating flow-or pressuremediated mechano-transduction have not been defined. Candidate structures include the actin cytoskeleton [9][10][11], intermediate filaments [11,12], focal adhesions [13,14], caveolae [15,16], myoendothelial junctions [17] and intracellular-matrix linkages through transmembrane molecules such as integrins [1,10,[18][19][20][21] . The cytoskeleton in particular is viewed as an important scaffold for enzymes and regulatory molecules whose activity may be affected, or even regulated by the physical state of the cell which, in turn, determines the spatial relationship between functional components such as cytoskeleton, contractile filaments, enzymes, ion channels and intracellular organelles [22][23].
Although many studies on isolated, cultured cells grown on an elastic matrix have shown that cyclic stretch leads to activation of signal transduction pathways associated with contractility and secretion of vasoactive molecules [24][25][26], little consideration has been given to the effects of cellular deformation resulting from vasoconstriction on vascular behavior. More than 20 years ago, Greensmith and Duling [27] described morphological changes in small vessels as they underwent constriction, and noted the significant inward buckling of the intimal surface associated with the formation of numerous "endothelial ridges" that altered lumenal profile and resulted in a reorientation of VSM contractile filaments from the circumferential to the radial direction. These ridges induced pronounced deformation of the endothelium and were quite prominent in constricted arteries, ranging from 5-10 microns in height and extending several hundred microns along vessel length. This is an important physiological issue since small vessels normally possess intrinsic myogenic tone in vivo, and the level of tone may, in turn, be altered by physical stimuli such as pressure. Chronic hypertension is an excellent example of this, as it illustrates how an alteration in a physical stimulus (blood pressure) leads to a change in the vascular response (tone and reactivity) which, in turn, leads to vascular wall remodeling [14,[28][29][30].
Our hypothesis in this study was that vasodilator reactivity -as defined by pharmacologic sensitivity and efficacy -is significantly influenced by the level of arterial constriction and/or the extent of VSM activation. Two secondary hypotheses were that: (1) any observed differences in arterial reactivity could be related to changes in wall tension and/or vascular smooth muscle activation, and (2) endothelial deformation in and of itself is an important determinant of its vasodilatory influence.
By obtaining dimensional measurements of pressurized, cannulated vessels (lumen diameter and wall thickness) from two regional circulations (splanchnic and uterine), and by varying the level of tone and transmural pressure, we were able to evaluate the contribution of circumferential wall tension to any observed differences in reactivity. To segregate the effects of VSM activation from cellular deformation, and of VSM force from wall tension, separate experiments were conducted using one vessel type (mesenteric second order resistance arteries) mounted in an isometric (wire myograph) rather than isobaric (pressure arteriograph) preparation.
In addition, we used vessels from both pregnant and non-pregnant animals, since pregnancy is known to produce significant structural and functional changes in uterine and mesenteric resistance arteries. We reasoned that documentation of similar responses in vessels from both pregnant and non-pregnant animals (such as was obtained) strengthens the conclusion that the observed changes in vasodilator responsiveness as a function of vessel tone reflect fundamental properties of both endothelial and vascular smooth muscle cells, rather than behavior specific to vessel type or physiological state.
The results illustrate the complex interplay between wall tension, smooth muscle activation and endothelial deformation as complementary and significant determinants of arterial vasodilatory reactivity.
Animals and experimental design
Non-pregnant (NP) and late pregnant (day 20/22; LP) 12-14 wk old female Sprague-Dawley rats were purchased from Charles River Canada and housed at the University of Vermont Animal Facility for studies on pressurized vessels. Adult (12-14 wk old) female Sprague-Dawley rats were purchased from Charles River Canada and housed at the University of Alberta Animal Facility for studies with wiremounted mesenteric vessels (experiment 2, see below). The Institutional Animal Care and Use Committee approved the experimental protocols for use of the animals at each institution, and the provided guidelines were followed. Animals were provided with feed and water ad libitum and given three days to acclimate to their surroundings. On the day of an experiment, animals were injected with pentobarbital (50 mg/kg) intraperitoneally to induce a surgical plane of anesthesia. When no longer responsive to a hard pinch to the feet, rats were decapitated with a small animal guillotine. The uterus and mesentery were dissected out and placed in cold (4°C) HEPES (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid)buffered saline solution.
In vitro vessel methodologies Pressurized (isobaric) vessel preparation. Mesenteric arteries (100-200 µm unstressed diameter), and uterine arcuate arteries (120-200 µm) were dissected from surrounding tissue and cut in half to produce two functioning vessel segments. We used similarly-sized resistance arteries from two different regional circulations (splanchnic and uterine), and from pregnant (LP) and non-pregnant (NP) animals to evaluate whether observed differences were generalizable in a physiological sense; i.e. the occurrence of a similar pattern of reactivity regardless of vessel type and physiological state would suggest this to be a fundamental physiological property. Arterial segments were transferred to a dual chamber arteriograph (Living Systems Instrumentation, Burlington, VT). Each chamber contained two glass cannulaes and had its own independent superfusion system for drug delivery. All arteries were cannulated, and intraluminal pressure set to 50 mmHg prior to a 45-60 min equilibration at 37 o C using a servo-controlled pump and in-line pressure transducer (Living Systems Instrumentation). Following cannulation, the arteriograph was set on a mobile stage attached to an inverted microscope (Zeiss SR, Carl Zeiss, Thornwood, NY) with an attached monochromatic video camera that was connected to a video dimension analyzer and a television monitor as described in earlier publications [31,32].
Wire-mounted (isometric) vessel preparation. Mesenteric arteries were isolated and cleaned of all surrounding adipose and connective tissues, and mounted on two 25 µm wires attached to a wire myograph (DMT, Aarhus, Denmark) to allow isometric tension recordings. Vessels were normalized through a series of stepwise increases in diameter to determine their optimal resting tension, set to 0.8 x IC 100 (the internal circumference equivalent to a transmural pressure of 100 mmHg). Following a 30-minute equilibration period, endothelial integrity was confirmed by assessing relaxation with methylcholine.
Arterial imaging. Cannulated, pressurized (50 mmHg) mesenteric arterial segments from NP rats were constricted to 20% (low tone) or 60% (high tone) prior to fixation in 4% paraformaldehyde for 3-4 h. After rinsing in cacodylate buffer (lado Research, Williston, USA), the tissues were postfixed in 1% osmium tetroxide for 2 hours at 4 o C. The tissues were then rinsed in cacodylate buffer, dehydration through graded ethanols, cleared in propylene oxide and embedded in Spurr's epoxy resin. Semithin sections (1µm) were cut with glass knives on a Reichert ultracut microtome and stained with toluidine blue (Electron Microscopy Sciences, Fort, Washington PA). Images were obtained using an Olympus BX50 light microscope (at 40x magnification) coupled to a CCD camera.
Experimental protocols Experiment 1 (pressurized mesenteric and uterine vessels from non-pregnant and late pregnant animals) -effects of level of preconstriction (tone) on SNP and ACh vasodilator reactivity. All uterine or mesenteric vessels were given increasing concentrations of phenylephrine (Phe) to achieve low vs. high levels of constriction; specifically, one vessel was constricted 20-40% from its original diameter (defined as 'low tone') while the second vessel was constricted 60-80% from its original diameter ('high tone'). Axial length was adjusted as needed to remove any buckling resulting from changes in tone or pressure.
Once the desired levels of constriction were achieved, vessels were allowed 20 minutes to ensure that the level of preconstriction (lumen diameter) was stable. Incremental concentrations of SNP or ACh (endothelium-independent and -dependent vasodilators, respectively) were then administered until a maximal response was observed (defined as no further dilation with additional drug). Complete relaxation was then achieved using a mixture of papaverine (10 -4 M) and diltiazem (10 -5 M), and lumen diameters recorded.
The vasodilator efficacy of each drug was calculated as the percent dilation achieved by a vessel at the supramaximal concentration of vasodilator relative to full relaxation obtained in the presence of papaverine and diltiazem. Sensitivity was determined by best fit standard curve analysis (SigmaPlot), which allowed calculation of the concentration of vasodilator required to produce 50% of the maximal agonist response (EC 50 ).
Once data were obtained, regression analyses were conducted to relate vessel lumen diameter or tone to sensitivity (EC 50 value) to wall tension (T, which is defined as the product of pressure (P) and radius (r) by the LaPlace equation, T=Pr). In the first series of experiments, pressure was kept constant at 50 mmHg in all high-and low-tone vessels; hence, tension was directly proportional to diameter. The strength of relationship between tension and pharmacologic sensitivity was expressed by the correlation coefficient (r 2 value), which indicates how much of the variability in sensitivity could be attributed to the dependent variable (e.g. lumen diameter/wall tension, or level of tone). Experiment 2 (wire-mounted mesenteric vessels, SNP and ACh) -effects of varying smooth muscle force (activation) without changing wall tension or cellular (vascular wall) deformation on SNP and ACh reactivity. Although pressurized vessels are under more physiological conditions than wiremounted arteries, as they maintain a cylindrical shape and can change diameter, the wire technique is isometric, which permits the imposition of a fixed amount of stretch (wall tension), and thereby vary only smooth muscle activation (force production). The extent of cellular deformation is also similar, since all vessels are equally pre-stretched.
Mesenteric vessels were constricted with Phe to approximately 35% or 70% of their maximal tension (determined by an initial cumulative concentration-response curve to Phe, 0.1 to 30 µM) and allowed to stabilize for 10 minutes. Following preconstriction, a concentration-response to SNP (0.001 to 100 µM) or ACh (0.001 to 30 µM) was then obtained in vessels with high (≈70%) vs. low (≈35%) pre-activation to determine both sensitivity and efficacy using the standard pharmacological approach for isometric wire-mounted vessel preparations. Experiment 3 (pressurized mesenteric vessels) -effects of varying wall tension on SNP reactivity in vessels with similar levels of tone: Two vessel segments were pressurized to 50 mmHg, and constricted with Phe to induce ≈ 60% tone. Pressure was then increased from 50 to 100 mmHg in one vessel to double wall tension with a similar degree of preconstriction (tone). By slowly changing pressure (~1 mmHg/sec), vessel diameter remained unchanged due to myogenic activation. This procedure allowed us to directly evaluate of the effect of wall tension on vasodilator reactivity to SNP without any change in the level of constriction (tone).
Experiment 4 (pressurized mesenteric vessels) -effects of cellular (vascular smooth muscle, endothelial) deformation at different levels of constriction and similar wall tension on SNP and ACh reactivity: These experiments were aimed at dissociating the level of cellular deformation from wall tension by inducing different levels of constriction (low vs. high) in one pair of vessels, as above, and then normalizing the wall tension by increasing the intralumenal pressure in the highly constricted artery until the wall tension was similar to that of the less constricted artery. In practice, wall tension values of approximately 6,000 mN/mm (intermediate between the high and low tension values achieved in Experiment 3) were achieved in all vessels studied. This was achieved without changing the lumen diameter or tone in the more constricted vessels since vessels were able to maintain their diameter, presumably due to myogenic activation. Once diameter was stable, vessel responses to SNP and ACh were evaluated.
Drugs and solutions
The HEPES physiologic salt solution contained the following (in mmol/L): sodium chloride 141.8, potassium chloride 4.7, magnesium sulfate 1.7, calcium chloride 2.8, potassium phosphate 1.2, HEPES 10.0, EDTA 0.5, and dextrose 5.0. The solution was prepared in deionized water and titrated with sodium hydroxide to a physiologic pH of 7.4. Lphenylephrine hydrochloride, sodium nitroprusside and acetylcholine hydrochloride (Sigma, St Louis, MO) were administered from stock solutions prepared daily in deionized water. All chemicals were purchased from Fisher Scientific (Fair Lawn, NJ) and Sigma unless otherwise specified.
Calculations and statistics
All data are expressed as means ± SE. SNP and ACh efficacies are presented as a percentage of complete vessel dilation obtained in relaxing solution, while sensitivities are presented relative to the maximal extent of dilation obtained in response to each agonist. Sensitivity was determined by constructing a semi-logarithmic concentration response curve for each vessel, and then extrapolating to the concentration that produced 50% of the maximum effect to calculate the EC 50 . Data are shown as the average of the individual EC 50 values ± SE. The results were analyzed with Students t-test (paired or unpaired, as appropriate); p values of <0.05 were considered statistically significant.
Low vs. high levels of constriction (tone) in pressurized vessel
Vessels with "low tone" had a mean constriction of 30 ± 1.3% (n = 24; range = 20-41%). Vessels with "high tone" had a mean constriction of 68 ± 1.1% (n = 24; range = 60-79%). A cross-sectional view of vessels with low vs. high tone is shown in Fig. 1. Tone was similar in both vessel types (mesenteric and uterine) for each level, and high vs. low tone levels were significant (p<0.05) within each vessel type studied (Fig. 2).
Effects of pre-constriction on SNP and ACh vasodilator reactivity in mesenteric and uterine arteries (Experiment 1): Although the actual EC 50 values differed with vessel type, significantly higher vasodilator concentrations were consistently required to produce an equivalent extent of dilation in vessels with high tone (Tab. 1). Specifically, mesenteric vessels with high tone had an SNP EC 50 7-fold greater than those with low tone (p<0.05). LP uterine arcuate arteries displayed a similar trend, as high tone vessels had EC 50 values that were ≈15-fold greater than those with low tone (p<0.05). The effect of constriction (level of tone) on endothelial vasodilator influence was tested via the administration of ACh, which leads to the release of several vasodilator molecules including nitric oxide, prostacyclin and EDHF. The proportion of each may vary with vessel type, and is known to be affected by pregnancy [33]. Mesenteric high tone vessels were 3.4-fold less sensitive to ACh than low tone vessels (p<0.05); similarly, uterine arteries with high tone were 5-fold less sensitive to ACh than the arteries with low tone (p<0.05).
Graphs summarizing sensitivity of pressurized mesenteric and uterine vessels in response to SNP and ACh are shown in Fig. 3 and Fig. 4, respectively. There were also significant differences in three out of the four groups in efficacy (Tab. 2). At maximal drug concentrations, dilation in all vessels with low tone was virtually complete (≥ 98%), while vessels with high tone dilated >94% to ACh, but less to SNP, averaging 82 and 67%, respectively, for LP uterine and NP mesenteric arteries (p<0.05 vs. low tone counterpart in each vessel type).
A strong correlation between vasodilator sensitivity (EC 50 value) and wall tension or tone was evident for all vessel types and for both SNP and ACh stimulation, with coefficients of determination (r 2 values) ranging from 0.42 to 0.82 for sensitivity vs. wall tension or tone. The r 2 values indicate that wall tension accounts for approximately 40 to 80% of the variability in vasodilator sensitivity. By the Law of LaPlace, the circumferential force that favors dilation (wall tension) is greater in less constricted vessels; thus, less agonist is required to produce an equivalent effect regardless of vessel type or vasodilator. On the other hand, an effect of activation per se, rather than wall tension cannot be distinguished using the pressurized vessel methodology (see Discussion); alternatively, changes in cell deformation (VSM shortening) may be contributory as well.
Effects of vascular smooth muscle activation on mesenteric artery SNP and ACh vasodilator reactivity at similar levels of wall tension and wall deformation (Experiment 2).
In wire-mounted vessels in which the level of pre-stretch was normalized, the only difference was level of smooth muscle activation, as the isometric nature of the technique prevents VSM cell shortening (deformation). Levels of activation were approximately Table 1. Summary of SNP and ACh sensitivity (EC 50 values) for pressurized mesenteric and uterine arteries preconstricted to high vs. low tone with Phe. Data are expressed as mean ± SE, n = 6 for each experimental result; total vessels: n = 48. Table 2. SNP and ACh efficacy (% of maximal relaxation obtained in a relaxing solution containing papaverine + diltiazem) of pressurized mesenteric and uterine arteries. Data are expressed as mean ± SE, n = 6 for each experimental result; total vessels: n = 48. More activated vessels were significantly less sensitive to SNP relaxation, while there were no differences in sensitivity to ACh, (p=0.56, Tab. 3). Differences in SNP relaxation were most pronounced at low concentrations (Fig 5A), while the ACh concentrationresponse curves were virtually superimposed (Fig 5B).
When individual vessels were considered, a positive correlation between smooth muscle activation and SNP responses was noted, with a coefficient of determination (r 2 ) of 0.56, indicating that roughly half of the variability in sensitivity could be accounted for by the level of activation. The corresponding r 2 value for activation and ACh sensitivity was 0.01.
Effects of varying wall tension at similar levels of preconstriction (tone) and wall deformation on mesenteric artery reactivity to sodium nitroprusside (Experiment 3). Given that vessels with varying levels of tone also have different degrees of cellular deformation secondary to constriction, we next conducted a group of experiments using pressurized vessels in which the level of constriction (and, therefore, cellular deformation) was kept constant while varying wall tension. This was accomplished by constricting mesenteric vessels by about 60% and then slowly increasing the transmural pressure in one of the two vessel pairs from 50 to 100 mmHg prior to the administration of SNP. In this case, we were attempting to isolate the influence of wall tension on vascular smooth muscle vasodilator sensitivity. EC 50 values were derived for each vessel and used for statistical analysis as described in Methods.
This experimental protocol resulted in an approximate doubling of wall tension without changing the level of preconstriction or circumferential deformation (Tab. 4), as vessel diameter did not change in response to increased transmural pressure due to myogenic activation. Hence, this was an attempt to mimic the conditions experienced by a wire-mounted vessel in a pressurized preparation (a change in VSM force without any change in diameter/ VSM shortening/endothelial deformation). Under these conditions, there was a 3.6-fold difference in the SNP EC 50 values in equally constricted vessels experiencing high vs. low wall tension (Tab. 4). Efficacy values were similar in both groups (80% vs. 74% in high vs. low tension; p>0.05; Tab. 4)
Effects of equalizing wall tension at different levels of preconstriction and wall deformation on mesenteric artery reactivity to SNP and ACh (Experiment 4
). An influence of wall tension does not, however, preclude a separate effect of cellular (endothelial and vascular smooth muscle) deformation on its vasodilatory influence. A fourth and final experimental protocol was developed to address this question.
Vessels were constricted to high vs. low levels of tone, as in Experiment 1, and the transmural pressure was then varied until wall tension was similar in both vessels (Tab. 5) prior to the administration of SNP or ACh. To achieve wall tension values intermediate between those measured in Experiment 3, we aimed for a wall tension of approximately 6,000 mN/mm, which resulted in transmural pressures that averaged ≈56 vs. ≈35 mmHg in the high vs. low tone vessels.
As shown in Tab. 5, vessels with high tone, in which endothelial deformation was greatest, required 3-fold less ACh to achieve half-maximal relaxation, while sensitivity to SNP was equivalent. There were no differences in efficacy between high and low tone vessels (p>0.05).
Together, these results clearly show wall tension and level of activation, rather than shortening, to be the principal determinants of VSM vasodilator sensitivity; conversely, the release of endothelial vasodilator molecules is significantly augmented by its deformation secondary to VSM activation and shortening. The physiological implications of these findings are considered below. . A -SNP concentration-response curves (sensitivity) of wire-mounted mesenteric normalized to maximal vasorelaxation obtained at the highest concentration (35% force: n = 7; 70% force: n = 5).B -ACh concentrationresponse curves (sensitivity) of wire-mounted mesenteric arteries normalized to maximal vasodilation obtained at the highest concentration (35% force: n = 8; 70% force: n = 6). EC 50 values were derived for each vessel and used for statistical analysis as described in Methods. Table 4. Pressurized mesenteric arteries constricted to a similar level of tone but subjected to different wall tensions by increasing transmural pressure show differences in sensitivity (EC 50 ) but not efficacy to SNP. Data are expressed as mean ± SE, n = 4, * p<0.05. [Experiment 3] Table 5. Pressurized mesenteric arteries at different levels of constriction (high vs. low tone) but similar wall tensions accomplished by varying transmural pressure (mmHg), show differences in sensitivity (EC 50 ) but not efficacy to SNP and ACh. Data are expressed as mean ± SE, n = 8 (for ACh), n = 4 (for SNP), * p<0.05. [Experiment 4]
Discussion
As an artery constricts, the vascular wall undergoes considerable three dimensional deformation of both cells and matrix. VSM cells twist, shorten and thicken [34] and the internal elastic lamina and endothelium are buckled inwards to form pronounced axial ridges that encroach upon and deform the normally-cylindrical shape of the lumen in a relaxed vessel (see Fig. 1 and ref 27). The endothelium is also stretched axially as a vessel constricts, and radially compressed, as it is positioned between the activated smooth muscle and intraluminal pressure. While many studies have examined the effects of cyclic stretch using the Flexercell apparatus in which cultured VSM or endothelial cells are seeded onto an elastic substrate on cellular responses [35][36][37], the effects of compressive deformation on endothelial function have not been experimentally addressed.
The goal of this study was to use available smallartery methodology (pressurized vs. wire-mounted vessels) to understand how wall tension, constriction (tone), VSM activation (force production), and cellular deformation interact and influence resistance artery vasodilatory reactivity. SNP and ACh were chosen because they are relatively specific and commonly-used agonists for VSM vs. endothelial activation, respectively.
The results of the first experimental protocol show that more constricted vessels require more agonist to dilate, as evidenced by the 3-to 15-fold increases in EC 50 values to both SNP and ACh. This was clear and acrossthe-board in terms of the similarity in pattern observed in vessels from the gut vs. the uterus, and of those from pregnant and non-pregnant animals. The latter comparison was included to expand the scope to include a unique physiologic state (pregnancy) that is associated with a significantly altered vasodilatory reactivity, as demonstrated by the results of numerous studies. Our assumption was that, if this is a fundamental physiological response, it should be present in all small vessels irrespective of source and physiological condition.
The results support this line of reasoning. The practical implication is that comparison of in vitro vasodilator reactivity data from vessels that are unequally pre-constricted is not valid, as the level of constriction introduces significant bias. A corollary is that betweenstudy comparisons in which different levels of preconstriction were induced also cannot be inferred. Because vessel constriction results in a narrowing of the lumen, unless pressure is increased, wall tension decreases to an extent that is proportional to the level of constriction by the Law of LaPlace, which states that tension is a product of pressure and radius (T=Pr). Everything else being equal, more drug would be required to elicit an equivalent vasodilator response because of the reduced distending force (wall tension). The effect is further amplified by increased wall thickness, which reduces wall stress (defined as tension per unit wall thickness; δ = T/ ω, where T = tension and ω = wall thickness). This is an oversimplified view, however, since at least three other factors deserve consideration.
The first is that the relationship between tone and smooth muscle activation in terms of crossbridge dynamics is difficult to predict since changes in VSM length, crosssectional area and volume may alter actin-myosin interactions in ways that are not well understood (e.g. via altered actomyosin motor function and overlap).
Second, there is uncertainty regarding the counterforce produced by compression of intracellular structures; studies in isolated smooth muscle cells have provided evidence for intracellular compression having a spring-like effect that opposes 'sarcomere' shortening (also poorly defined in VSM), and facilitates the restoration of cell length (or, in this case, of arterial diameter) once the contractile stimulus is withdrawn [38]. This internal counterforce is difficult to evaluate experimentally, but the level of activation must balance the sum of pressureinduced circumferential distension and internal (VSM intracellular) compression.
Third, the potential existence of load-bearing but not force-producing elements [latch-bridges, 39] in partially constricted vessels further complicates interpretation, since it uncouples activation from diameter maintenance. Based on these considerations, it is not possible to determine the true level of activation (relative level of VSM force production) in a pressurized vessel with tone. For this reason, isometric wire-mounted experiments (Experiment 2) were carried out in an attempt to evaluate the effects of VSM activation without the complication of cellular shortening/deformation or altered wall tension. The resulting data show that activation is, in and of itself, an important factor in VSM responsiveness to a vasodilatory stimulus, as significantly more SNP was required to produce a proportional decrease in force in a more activated vessel. Relaxation to ACh, however, was unchanged, as might be expected since the endothelium was stretched to an equivalent extent.
It is worth noting that, while the SNP response predicts a blunting of the NO response as well, under these experimental conditions, ACh stimulates the release of multiple vasodilators, so the observed response is the summation of all endothelial vasodilators (NO, prostacyclin, EDHF, etc.) and, therefore, not specific to NO; precluding direct inference from SNP to ACh data. Thus, the data from the first two experimental series are congruent in showing that wall tension and level of VSM activation both significantly impacted VSM reactivity.
In Experiment 3, vessels were equally constricted (and, therefore, equally deformed), but set at different wall tensions by varying intralumenal pressure. The absence of changes in vessel diameter in spite of a 100% increase in transmural pressure points to enhanced VSM activation as well, presumably via a myogenic mechanism geared to maintaining constriction (diameter) in response to increased pressure. SNP was used to target VSM, and the significantly increased sensitivity to SNP (as reflected in 4-fold lower EC 50 values) in vessels exposed to a higher wall tension reflect the relative predominance of the wall tension effect over the influence of VSM activation, since an increase in the latter would, based on the result of Experiment 2, have been expected to have the opposite effect (to increase EC 50 values). Considering the 7-fold reduction in high-vs. low-tone mesenteric artery sensitivity to SNP in Experiment 1 and the 2.5-fold increase in sensitivity to a doubling of activation in Experiment 2, the 3.6-fold reduction in highvs. low-tension vessels in Experiment 3 is quantitatively consistent.
In the fourth and final experimental series, we were interested in determining the specific influence of endothelial deformation on its vasodilatory influence. To accomplish this, pressurized vessels were constricted to high vs. low tone, as in Experiment 1, although here both vasoconstrictor concentration and transmural pressure were adjusted so as to produce similar wall tensions. Under these conditions, more constricted vessels were significantly more sensitive to ACh vasodilation; i.e. without the influence of wall tension, endothelial vasodilatory influence was significantly augmented.
Since ACh releases multiple endothelial vasodilators (nitric oxide, EDHF, prostanoids, endothelin), we only observed a sum effect under these experimental conditions and do not know which is most affected. The fact that the EC 50 values for SNP were virtually identical confirms the importance of wall tension in regulating vascular smooth muscle dilator reactivity and eliminates the degree of cell shortening as a significant variable.
The most straightforward interpretation is that, under conditions where the distending force is normalized, endothelial deformation -most likely, circumferential and radial compression -produces a significant increase in vasodilator release in response to an identical stimulus. As seen in Fig. 1, the infolding of the internal elastic lamina pushes the endothelium into the lumenal compartment and clearly results in a three-dimensional deformation since vessels also lengthen as they constrict. Our results suggest that the influence of the endothelium becomes more potent as the cell deforms secondary to increased vasoconstriction. This effect could be due to any number of factors, e.g. changes in the spatial relationships between cytoskeletal and membrane structures that may alter enzyme or ion channel activity, and thereby affect endothelial secretory activity in favor of increased vasodilatory influence.
In conclusion, our findings indicate that both wall tension and the level of vasoconstriction/activation are important determinants of vascular smooth muscle vasodilator reactivity; conversely, endothelial function does not appear to be appreciably impacted by wall tension, although cellular deformation secondary to vasoconstriction does induces significant changes in vasodilator influence. The results of this study are provocative in their physiological implications since they provide an insight into the dynamic interactions between biophysical forces (pressure, wall tension) and fundamental cellular processes (VSM activation and endothelial deformation) in the determination of vascular reactivity. The real challenge for future studies is to understand how these processes are affected by vascular pathologies such as chronic hypertension (where differences intramural forces such as wall tension induce changes in tone and cellular deformation), or diabetes -a metabolic condition in which hyperglycemia and increased insulin levels induce changes in vascular smooth muscle contractility and endothelial function that may, in turn, alter the physical forces that impinge upon the vascular wall.
|
2018-04-03T04:52:36.405Z
|
2012-05-11T00:00:00.000
|
{
"year": 2012,
"sha1": "9ad3c374ff84de14fd1b7c09b1df7642f02c020c",
"oa_license": null,
"oa_url": "https://www.karger.com/Article/Pdf/178465",
"oa_status": "GOLD",
"pdf_src": "Karger",
"pdf_hash": "2ba021cccdd048f3eb6cbcaaecfed89c45a07c9f",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
238898923
|
pes2o/s2orc
|
v3-fos-license
|
Increased survival rate of hybrid grouper Epinephelus sp. after feeding with microalginate particles
This research aims to increase fish survival rate by providing oral feed containing formalin killed cells of Vibrio alginolyticus. It investigates whether it is effective or not. In the hatchery, the survival rate of grouper Epinephelus sp. is only 2-3 percent. The death of 4-6 g size grouper infected by Vibrio sp . is very quick. Fish dies only after 12 h of infection. The problem on larval stage is that the fish size is small. So, it is not possible to perform injection or immersion because it may result in high stress of the fish. The only way to overcome the problem is oral provision by incorporation of feed. Formalin-killed cells of Vibrio were coated with alginate and administered orally through feed. Alginate microparticles were prepared from colloidal complexes using sodium alginate gel as a core. All alginate solutions (2% w/v) were prepared by dissolving sodium alginate in deionized water. The survival rate of fish increased to 20–22 percent; the antibody titer was 64-256 and phagocytic rate was 32-60. The increased survival rate was due to higher antibody and superoxide production in fish. This study suggests that vaccination for small fish can be done in the future by making various improvements.
Introduction
Indonesian grouper export in 2016 was quite large. Central Bureau of Statistics (BPS) noted that in 2016 Indonesia exported groupers to various countries with the value which reached 32.18 million US dollars or about IDR 427 billion. However, the survival rate of grouper seeds Epinephelus sp. is only 2-3 percent after hatch. The death of 4-6 g-sized grouper seeds infected by Vibrio sp. occurs quickly [9] , no more than 12 hours after infection. The mortality rate may reach 80 percent of the population. There are two types of Vibrio bacteria that infect the grouper, V. alginolyticus and V. anguillarum [9] , while those infecting shrimp are Vibrio harveyi and Vibrio splendidus [14] , similar to those identified in the Philippines [6] . In Malaysia, Vibriosis is caused by bacteria and V. parahaemolyticus as well as V. alginolyticus. Whereas, in Japan, Vibrio sp. P. japonicus reportedly attacked shrimp [15] . Vibriosis found in China was mainly caused by V. parahaemolyticus in shrimp larvae [8] . Mass mortalities on larval stage is due to the small size of the fish. So, it is not possible to carry out injection or immersion because it may result in high stress on the fish. There is one possible way to overcome the infection, that is by incorporating Vibrio antigen into the feed. Vaccination can enhance specific and non-specific self-defense mechanisms. Specific defense can be enhanced by incorporating Vitamic C into feed [7] . This study aims to determine whether the incorporation of Vibrio antigen into the feed can enhance the survival rate of grouper larvae. This study reveals that increasing the survival rate of small grouper fish can possibly be done by feeding the fish with alginate particles containing formalin killed cells of Vibrio alginolyticus. This study helps the researchers uncover critical areas of vaccination that many researchers were unable to explore. Thus, a new theory on little fish vaccination may be invented.
Bacteria strain
Vibrio parahaemolyticus was grown on cellophane-covered NA for 24-48 hours [17] . Then, they were harvested and placed into a Petri dish. The bacteria were added with 2 ml of phosphate buffered saline (PBS), and washed 3 times for 10 minutes. The bacteria were killed with 3% formalin for 72 hours. Each FKC obtained was washed 3 times for 10 minutes by centrifugation. Bacterial density was made 50 mg/ml. It was then stored in -4 °C and dialyzed with PBS.
Preparation for alginate
Alginate microparticles were prepared from colloidal complexes using sodium alginate gel as a core. All alginate solutions (2% w/v) were prepared by dissolving sodium alginate in deionized water. Micronized prednisolone powder (2-4% w/v) plus Vibrio antigen was then dissolved in sodium alginate and stirred for 10 min, and sprayed directly on the solution of calcium chloride (0.5-1.0% w/v) containing 0.5-1.5% chitosan. Microparticles were left for 2 hours to harden before being washed 2 x with DW at room temperature.
Fish used in this study
Twenty fish (7 cm in length) in each aquarium were fed at libitum for 7 days. Fish that had been feed, were challenged with Vibrio alginolyticus at a dose of 7.1 x 104 CFU. The fish were kept in an aquarium for 14 days and their mortality rates were recorded.
Feeding fish with alginate
Another group of grouper of ±10 cm in length was injected intraperitoneally with 0.01 of glycogen to stimulate macrophage production. After 24 hours, the fish were anesthetized with MS-222. Blood was taken from the base of the tail to prevent the contamination of macrophage with red blood. A half milliliter of Leibovitz medium (L-15) containing 2% Fetal Bovine Serum (FBS) and 10 units/ml heparin was injected intraperitoneally into the grouper. The rest of the macrophage was washed again with 0.5 ml of L-15. Macrophage cells were then centrifuged at 400xg for 10 minutes and dissolved in L-15. Viable cells (living cells) were observed and counted by trypan blue staining. The cell density was calculated 2-5x10 6 cells/ml using H aemocytometer (Assistant, Karl Hecht KG Germany).
Phagocytosis
Bacteria and macrophages (vaccinated and control) were mixed and incubated at 25 ºC for 30 mins by shaking at 25 rpm, allowing contact between the macrophages and the bacteria. The macrophage was taken to be smeared on a slide glass that had been coated with albumin and stained with May-Grunwald Giemsa. Further, phagocytosis was calculated.
Intracellular O2
Head kidney macrophages that have been collected were poured into L15 medium (Sigma, USA). The macrophages were then filtered with 100 µm nylon. Percoll (2 ml of 30-51%, Sigma, USA) was added to the cell suspension and centrifuged at 400 xg for 40 min at 4ºC. Macrophages obtained were poured into the L-15 (10% fetal bovine serum) at concentration of 1 x 10 7 cells/ml. Macrophage was then adjusted to 1 x 10 6 cells/ml and poured into the wells. Superoxide anion production was conducted by nitroblue tetrazolium (NBT) [11] . NBT solution per well (100 ml, 1 mg of NBT and 1 mg phorbol 12-myristate 13 acetate in 1 ml of medium L-15) was added to the wells and incubated at 30ºC for 1 hour. Cells were fixed with 100% alcohol and then 120 ml of KOH (2M) and 140 ml dimethyl sulfoxide was added to the formation of formazan. Superoxide anion production was measured at OD 620 nm.
Results and Discussion
After fish were fed a diet containing whole cell bacteria (whole cell vaccine) the phagocyte rate was between 34 -53 and control 32 percent, while the survival rate (SR) was 20-22 and 52 control percent, this indicates an increase in life, because it usually does not up to 5 percent (normally 2-3 percent). This increase was related due to an increase in phagocytosis and antibody production ( Table 1) and an increase in intracellular O2 production (Figure 3), both of which increased significantly, this in turn would lead to an increase in the grouper survival rate. Phagocytosis is the process by which cells engulf some solid particles to form internal vesicles known as phagosomes. Phagocytosis is, in fact, a specific form of endocytosis involving vesicular interiorization of particles. Degradation may be oxygendependent or oxygen-independent. Oxygen-dependent degradation depends on NADPH and the production of reactive oxygen species. Hydrogen peroxide and myeloperoxidase activate a halogenating system, which leads to the creation of hypochlorite and the destruction of bacteria [3] . The antibody titer detects the presence and measures the amount of antibodies in fish blood. The amount and diversity of antibodies correlate with the strength of body's immune response. Superoxide dismutase is an important antioxidant defense in nearly all living cells exposed to oxygen. One exception is Lactobacillus plantarum and related lactobacilli, which use a different mechanism to prevent damage from reactive (O2-). Protection of vaccinated fish is produced from a high antigen concentration, but if the antigen level is too high, sometimes it does not confer proper protection. Optimum concentration between the highest and lowest dose and the same pattern will be followed by an increased antibody titer, which indicates a high degree of protection as well. Production of superoxide anion in the immune system is critical to the animal itself as a defense system against disease and significant as an immune function (Secombes, 1996;Crosby et al., 2010) [12,1] . Superoxide anion production varies between fish because the production of superoxide anion in this study tended to increase (Fig 3).
Vaccines are small amounts of microbes (killed or attenuated) or antigens administered to stimulate the immune system to produce memory cells against a specific disease. Protection of grouper varies and depends on the dose used. Memory cells are so effective that fish's immune system may fight off an infection before any symptoms occur. The main obstacle of oral feeding may be "bad taste" or the palatability of feed. The fish were fed twice a day and left hungry. It would increase the appetite of fish. Palatability has been a problem of using oral system. Antigens that have a Molecular Weight (MW) more than 5 kD are immunogenic, whereas antigens with smaller molecular weight are not. It is known that fish only have IgM, the main immunoglobulin specific against bacterial infection. However, this immunoglobulin affinity is very low. Antigens from whole bacteria provide good protection results. In this study, the mortality of control fish was 52 percent, because they only received Phosphate Buffered Saline, pH 7.0 (PBS) alone. The survival rate of vaccinated fish is between 20-22 percent. The potential immunogen is in the cell walls of the bacteria because most of the experimental vaccination with whole bacterial cells provides good protection. Vaccination with a low antigen concentration did not result in protection. The protection of vaccinated fish is produced from a high concentration of antigen. However, if the antigen level is too high, sometimes it does not confer good protection. An optimal concentration between the highest and lowest dose and the same pattern will be followed by an increased antibody titer which indicates a high degree of protection as well. Basically, protective antigen level is above 5 kD to 200 kD [4] . The result of tested grouper is presented in Table 1. Formalinkilled cells antigen provides protection against Vibrio challenged. Protection due to the large size and intact antigen has not been broken or damaged due to physical or chemical factors. Antibody production is dependent upon the age of the fish, e.g. trout after 4 weeks of age, but memory development takes at least 8 weeks.
An antigen that has a lower MW could not be recognized by lymphocytes of fish. There are no scientific explanations for this. Production of superoxide anion in the immune system is critical to the animal itself as a defense system against diseases and significant as immune function [1,11] . Superoxide anion production varies between the fish because the production of superoxide anion tends to increase in this study (Fig 3). Superoxide dismutase is an enzyme that alternately catalyzes the dismutation of the superoxide (O2-) radical into either ordinary molecular oxygen (O2) or hydrogen peroxide (H2O2) [17] . Superoxide is produced as a by-product of oxygen metabolism and, if not regulated, causes many types of cell damage [2] . Hydrogen peroxide is also damaging and is degraded by other enzymes, such as catalase. Thus, SOD is an important antioxidant defense in nearly all living cells exposed to oxygen. One exception is Lactobacillus plantarum and related lactobacilli, which use a different mechanism to prevent damage from reactive (O2-) [17] . Similarly, phagocytic rate (PR) increased after vaccination. This process is actually easy to understand because, when compared to controls, it varied enormously and this phenomenon occurred almost in every vaccination. Although this study has not obtained excellent results, at least, it has given a hope that the vaccination of small fish is possible in the future. The success of vaccination is expected to increase grouper production and reduce the poverty of fishermen.
As conclution, the survival rate of fish increased to 20 -22 percent; antibody titer was 64-256 and phagocytic rate was 32-60. The increase of survival rate was due to higher antibody and superoxide production in fish. This study gives a hope that vaccination for small fish can be done in the future by making various improvements.
|
2021-10-16T00:12:13.014Z
|
2021-07-01T00:00:00.000
|
{
"year": 2021,
"sha1": "5294ad5e841baf90a1187925d5728319d93d311f",
"oa_license": null,
"oa_url": "https://www.fisheriesjournal.com/archives/2021/vol9issue4/PartC/9-4-2-876.pdf",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "d98a63f7653e9c50be448522b0146994fd30a4d7",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"Environmental Science"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
15941620
|
pes2o/s2orc
|
v3-fos-license
|
Intensive Intravenous Infusion of Insulin in Diabetic Cats
Background Remission occurs in 10–50% of cats with diabetes mellitus (DM). It is assumed that intensive treatment improves β‐cell function and increases remission rates. Hypothesis Initial intravenous infusion of insulin that achieves tight glycemic control decreases subsequent insulin requirements and increases remission rate in diabetic cats. Animals Thirty cats with newly diagnosed DM. Methods Prospective study. Cats were randomly assigned to one of 2 groups. Cats in group 1 (n = 15) received intravenous infusion of insulin with the goal of maintaining blood glucose concentrations at 90–180 mg/dL, for 6 days. Cats in group 2 (n = 15) received subcutaneous injections of insulin glargine (cats ≤4 kg: 0.5–1.0 IU, q12h; >4 kg 1.5–2.0 IU, q12h), for 6 days. Thereafter, all cats were treated with subcutaneous injections of insulin glargine and followed up for 6 months. Cats were considered in remission when euglycemia occurred for ≥4 weeks without the administration of insulin. Nonparametric tests were used for statistical analysis. Results In groups 1 and 2, remission was achieved in 10/15 and in 7/14 cats (P = .46), and good metabolic control was achieved in 3/5 and in 1/7 cats (P = .22), respectively. Overall, good metabolic control or remission occurred in 13/15 cats of group 1 and in 8/14 cats of group 2. In group 1, the median insulin dosage given during the 6‐month follow‐up was significantly lower than in group 2 (group 1: 0.32 IU/kg/day, group 2: 0.51 IU/kg/day; P = .013). Conclusions and Clinical Importance Initial intravenous infusion of insulin for tight glycemic control in cats with DM decreases insulin requirements during the subsequent 6 months.
D iabetes mellitus (DM) is a frequent endocrine disease in cats. Approximately 80% of diabetic cats are believed to have a form similar to type 2 diabetes in humans, which is characterized by inadequate insulin secretion and impaired insulin action. 1 Current cornerstones of diabetes management in cats are twice daily injections of insulin and feeding a high-protein and low-carbohydrate diet. Ten to 50% of diabetic cats experience remission and no longer require exogenous insulin because of resolution of clinical signs and normalization of blood glucose and fructosamine concentrations. [2][3][4][5][6][7] A number of trials in human medicine evaluated short-term intensive administration of insulin in the management of newly diagnosed type 2 diabetic patients. [8][9][10][11][12] Treatment included either multiple daily injections or continuous subcutaneous infusion of insulin for 2-3 weeks. Early intensive insulin treatment resulted in improvement of b-cell function and prolongation of remission. 10 In addition, remission rates after 1 year were significantly higher in humans treated with intensive insulin treatment compared with those treated with oral hypoglycemic agents. [10][11][12] The effect of intensive insulin treatment in cats needs to be explored; to date, only two studies have investigated this treatment modality using the long-acting insulin analogs glargine and detemir. 13,14 Owners were required to follow a strict treatment protocol that included measurement of blood glucose several times per day. In both studies, remission rates of up to 64 and 67% were achieved, which might suggest that intensive insulin treatment is beneficial in cats. 13,14 Because intensive insulin treatment is associated with favorable results in human patients with type 2 diabetes and appears to be advantageous in diabetic cats, the aim of this investigation was to determine the effect of early and tight glycemic control using shortterm intravenous infusion of insulin in cats with newly diagnosed diabetes.
Animals
Cats with newly diagnosed diabetes were enrolled in the study from July 2008 to January 2011. Cats were excluded from the study if they had received insulin treatment for longer than 1 week before admission and if glucocorticoids or progestagens had been administered during the previous 4 months. All cats underwent a thorough evaluation including physical examination, complete blood cell count, serum biochemical profile, measurement of fructosamine and total T4 concentrations, urinalysis, including bacterial culture and urinary protein-to-creatinine ratio, blood pressure measurement, abdominal and thoracic radiographs and abdominal ultrasonography. Diabetic cats with ketoacidosis were included in the study, if acidemia resolved and their general condition improved within 48 hours of insulin treatment. Cats with concurrent diseases at diagnosis (eg, renal failure, gastrointestinal disorders, heart disease, other endocrinopathies, neoplasia) were not enrolled.
The study was approved by the Cantonal Veterinary Office of Zurich and conducted in accordance with guidelines established by the Animal Welfare Act of Switzerland (permission no.: 83/ 2008). Informed consent to participate in the study was provided by the owners.
Randomization and Treatments
Enrolled cases were hospitalized for 7 consecutive days. The cats were allocated to one of two treatment groups using a software a providing a partial minimization procedure to adjust the randomization probabilities between groups and to balance for covariates collected at baseline. The covariates sex, age, body weight, treatment with insulin before admission (ie, for <1 week), presence of ketoacidosis, blood glucose, and serum fructosamine concentrations were selected.
After the initial work-up, all cats were sedated with midazolam/butorphanol b,c and anesthetized with propofol d to place a central venous catheter e in the jugular vein and to insert the sensor of the continuous glucose monitoring system (CGMS) f in the subcutaneous tissue of the lateral chest wall, as described previously. [15][16][17] The CGMS measures glucose in the subcutaneous interstitial fluid every 10 seconds and displays the 5-minute average on the monitor. After a 2-hour period of initialization, the first calibration was carried out; thereafter the CGMS was calibrated after 6 hours and then every 10 hours. To calibrate the CGMS, capillary blood glucose was measured with a portable blood glucose meter (PBGM), g validated for the cat. 18 In addition to CGMS measurements, capillary blood glucose concentrations were also determined with the PBGM g every 4-6 hours to ensure reliability of the system.
Cats of group 1 received intravenous infusion of insulin for 6 consecutive days with the goal of achieving tight glycemic control by adjusting the insulin dosage as required. Insulin was infused via the central venous catheter. The insulin solution consisted of 12.5 IU rapid-acting insulin aspart h dissolved in 250 mL 0.9% NaCl. The solution was renewed daily, and the tube was flushed with the same solution to prevent insulin adsorption to the solid surfaces of the infusion sets. 19 The initial insulin dosage used in the infusion was 0.05 IU/kg/h (equal to 1 mL/kg/h). In addition, 0.9% NaCl infusion was given concurrently through the same catheter, and the infusion rate was adjusted to maintain the total amount of intravenous fluids at 2 mL/kg/h. The cats were monitored 24 hours a day. The infusion rate of insulin was adjusted to achieve a target glucose concentration range of 90-180 mg/dL. For this purpose, the infusion rate of insulin was adjusted in steps of 0.025-0.05 IU/kg/h, every 15-30 minutes, if necessary. The infusion of insulin was discontinued at 6:00 AM on day 7 of hospitalization, and at 8:00 AM, the cats were started on subcutaneous injections of insulin glargine. i To ensure that the same dosage conversion was used for each cat, an arbitrary calculation was used. The total amount of insulin infused IV on day 6 was divided by 4. This calculated dosage was administered twice daily on day 7, while the cats were still being monitored by the CGMS. The cats were discharged on day 8 with insulin injections prescribed twice daily. If glucose concentrations dropped below 72 mg/dL on day 7 or 8, the insulin dosage was reduced by 50%.
Cats of group 2 were treated during the same 7-day period of hospitalization as cats of group 1 and received subcutaneous injections of insulin glargine i ; tight glycemic control was not a goal in this group. The insulin dosage was 0.5-1.0 IU, q12h, in cats weighing <4 kg and 1.5-2.0 IU, q12h, in cats >4 kg. In addition, all cats received an infusion of 0.9% NaCl via the central venous catheter at a rate of 2 mL/kg/h. The subcutaneous dosage of insulin was not adjusted during the 7-day period unless the blood glucose concentration dropped below 72 mg/dL, in which case the insulin dosage was reduced by 50%. The insulin dosage was increased by 0.5 IU per cat, q12h, if hyperglycemia persisted after the 7 days of hospitalization.
All cats of both groups were fed a high-protein, low-carbohydrate diet j during hospitalization and after discharge from the clinic. The study was carried out by two internal medicine residents under the supervision of two board-certified specialists in internal medicine. Cats were continuously monitored during the 7 days.
Analyses During Hospitalization
To calculate the duration of time that glucose was within and outside the target range in cats of group 1, the glucose concentrations recorded by the CGMS and calibrated with the PBGM were divided into the following ranges: hypoglycemia (<90 mg/ dL), target range (90-180 mg/dL) and hyperglycemia (mild: 181-270 mg/dL; and moderate to severe: >270 mg/dL). This calculation included all glucose values recorded with the CGMS starting 12 hours after initiation of the insulin infusion (group 1) or after the first insulin injection (group 2), until day 7 at 6:00 AM. The percentage of glucose measurements within each glycemic range was calculated for each cat in both groups. On day 8, a blood sample was collected for determination of a complete blood cell count, serum biochemical profile, and fructosamine concentration in all cats. The central venous catheter was then removed, and bacterial culture of the catheter tip was carried out.
Serum potassium and phosphorus concentrations were measured at admission and discharge in all cats. At admission, if hypokalemia or hypophosphatemia was documented, supplementation was provided with potassium chloride or phosphate diluted in the 0.9% NaCl infusion and electrolytes were measured every 2-6 hours until the concentration normalized. During hospitalization, serum potassium, and phosphorus concentrations were measured only if clinical signs compatible with electrolyte abnormalities were documented.
Follow-up
Re-evaluations were scheduled 1, 2-3, 6-8, 12-16, and 24 weeks after discharge from the clinic and were carried out by the same internal medicine residents who cared for the cats during the first week of hospitalization. The insulin dosage was adjusted based on clinical signs and the results of physical examination, blood glucose curves and fructosamine levels 1 ; the goal was to resolve clinical signs of DM and to maintain blood glucose curves between 90 and 270 mg/dL and fructosamine concentrations <400 lmol/L. To exclude the development of concurrent diseases, diagnostic tests including a complete blood cell count, serum biochemical profile, fructosamine concentration, urinalysis, urinary protein-to-creatinine ratio, and blood pressure measurement were done at each re-evaluation. Abdominal ultrasonography and total T4 measurement were carried out 6-8 weeks and 24 weeks after hospitalization. A dexamethasone suppression test was done 6-8 weeks after discharge from the clinic to rule out hyperadrenocorticism. Remission of diabetes was defined as absence of signs of DM (eg, polyuria/polydipsia, polyphagia) with normal blood glucose (72-162 mg/dL) and fructosamine concentrations (<340 lmol/L) for at least 4 weeks after discontinuation of the insulin injections. 1,4,5 In cats in which remission occurred before insulin administration had been discontinued, the dosage was decreased in increments of 0.5 IU per dosage, once weekly. The last dosage before insulin was discontinued was 0.5 UI once daily, for at least 1 week. The onset and duration of remission were recorded. Good metabolic control was defined as absence of clinical signs of DM, fructosamine concentrations <400 lmol/L and blood glucose curve measurements ranging from 90 to 270 mg/dL. 1 In both groups, the median insulin dosage per kg per day was calculated over the 6-month study period; phases of remission were excluded from the calculation.
Statistical Analysis
Data are presented as median and ranges. Differences in rate of remission and rate of good metabolic control between groups were analyzed using Fisher's exact test. Differences in laboratory results, blood pressure measurements, glucose measurements within each glycemic range during hospitalization, onset of remission and daily insulin dosage given over the 6-month study period between groups were analyzed using the Mann-Whitney U-test. A commercial software k was used for all analyses. The level of significance was set at P < .05.
Animals
A total of 51 cats with newly diagnosed DM were admitted to our clinic during the study period. Thirty cats fulfilled the inclusion criteria and were enrolled in the study; each of the 2 groups consisted of 15 cats. In group 1, the median age was 11.0 years (range: 7.0-15.0) and median body weight was 5.8 kg (range: 3.4-9.0). Eleven were domestic shorthair or longhair cats, and 4 were purebred cats (Abyssinian, Burmese, Siamese and Norwegian forest cat). Nine cats were neutered males and six were spayed females. In group 2, the median age was 11.0 years (range: 8.0-17.0) and median body weight was 4.7 kg (range: 2.5-9.6). Fourteen were domestic shorthair or longhair cats and one was a Ragdoll cat. Eight cats were neutered males and seven were spayed females. There was no significant difference in sex, age, or bodyweight between the two groups. At the time of initial presentation, 2 of the cats enrolled in the study had ketoacidosis, which resolved within 1 day of intramuscular injections of insulin; one was allocated to group 1 and the other to group 2. In group 2, 1 cat died of pyelonephritis (3 months after admission) and another died of alimentary lymphoma (3 days before the end of the study). Only cats that were alive for the duration of the study were included in the analysis; the cat that died 3 months after admission was excluded.
Laboratory Results and Blood Pressure Measurement on Admission
In group 1, the median serum glucose concentration was 439 mg/dL (range: 203-581) and the median fructosamine concentration was 615 lmol/L (range: 528-783). In group 2, the median serum glucose concentration was 425 mg/dL (range: 248-560) and the median fructosamine concentration was 575 lmol/L (range: 449-778). The laboratory results and blood pressure measurements at the time of admission are shown in Table 1. There were no significant differences between groups.
Intensive Intravenous Infusion of Insulin
In group 1, blood glucose concentrations decreased to the target range within 12 hours of initiation of treatment in 13 of 15 cats. In those cats, it was necessary to reduce the insulin infusion rate to 0.02-0.03 IU/kg/h to avoid hypoglycemia. In 2 obese cats with a body condition score l of 9 of 9, the initial insulin infusion rate was not sufficient to achieve the target glucose concentration within 12 hours and it was therefore increased to 0.07 IU/kg/h. Within 24 hours, all 15 cats had blood glucose concentrations within the target range. In 2/15 cats of group 1, the glucose concentration decreased below 72 mg/dL during the 6-day infusion, and the insulin infusion was stopped temporarily. On day 7, 2 of the cats received 0.5 IU, 8 received 1 IU, 4 received 2 IU and 1 received 3 IU of insulin injected SC, q12h. Bacterial culture of the tip of the jugular catheter after removal revealed bacterial growth in 2 cats; one culture yielded Pseudomonas aeruginosa and the other Enterococcus spp.
Subcutaneous Injections of Insulin
In group 2, nine of the 15 cats weighed <4 kg; the initial dosage of insulin injected SC was 0.5 IU, q12h, in 1 cat and 1.0 IU, q12h, in the other 8. Of the 6 cats that weighed >4 kg, 4 received an initial dosage of insulin injected SC of 1.5 IU, q12h, and 2 received 2.0 IU, q12h. In 4 of the 15 cats of group 2, glucose concentrations decreased to the target range within 12 hours. During hospitalization, 4 of the 15 cats of group 2 had glucose concentrations <72 mg/dL, which necessitated a 50% decrease in the insulin dosage for the remainder of the study; the dosage was reduced after a median of 2.6 days (range: 0-4). One of those 4 cats had normoglycemia at the time of discharge (day 8), and insulin was therefore not prescribed. In 6 of the 15 cats, the insulin dosage was increased at the time of discharge because the amount given during hospitalization was considered insufficient for adequate glycemic control; the insulin dosage was increased to 1 IU in one cat, to 1.5 IU in two cats and to 2 IU in the remaining three cats, administered SC, q12h. Bacterial culture of the tip of the jugular catheter after removal revealed bacterial growth of Pantoea agglomerans in one cat and Acinetobacter spp. in one other.
Analyses During Hospitalization and Costs
To determine whether short-term intravenous infusion of insulin maintained blood glucose levels within the target range during hospitalization, the percentage of glucose measurements that fell into different concentration ranges was compared with those of cats that were started on subcutaneous injections of insulin without aiming at tight glycemic control. The percentage of glucose measurements for cats within the target range of 90-180 mg/dL was significantly higher for cats in group 1 than in group 2 (group 1: median 59% [range: 15- On admission, 3 cats presented with mild hypokalemia (range: 3.3-3.7 mEq/L; reference range: 3.8-5.4) and 5 with mild hyperkalemia (range: 5.5-6.5 mEq/L), 2 cats presented with mild hypophosphatemia (1.8 and 2.0 mg/dL, respectively; reference range: 2.8-5.6), and one with mild hyperphosphatemia (6.2 mg/dL). Abnormal serum potassium or phosphorus concentrations normalized within 24 hours from admission; potassium chloride or phosphate supplementation was provided in cats with hypokalemia or hypophosphatemia, respectively. None developed clinical signs suggestive of electrolyte imbalance during the period of hospitalization. At discharge, all cats had normal serum potassium concentrations and all but one had normal phosphorus. The cat with abnormal serum phosphorus had very mild hyperphosphatemia (5.8 mg/dL). There were no differences between groups at admission and discharge for either electrolyte. The cost of the 6-day intensive intravenous infusion of insulin was approximately US$1,700 per cat.
Diabetic Remission and Metabolic Control During
Follow-up The remission occurred in 10/15 in group 1 and in 7/14 in group 2, P = .462. Of the cats that went into remission, relapse occurred 4 weeks after insulin had been discontinued in one cat of group 1 and 7 weeks after discontinuation of insulin in one cat of group 2. Table 1. Laboratory results and blood pressure measurements in cats on initial intensive intravenous infusion of insulin with tight glycemic control achieved by insulin dosage adjustment (group 1) and in cats started on subcutaneous injections of insulin without the goal of initial tight glycemic control (group 2). ALAT, alanine aminotransferase; ALP, alkaline phosphatase; ASAT, aspartate aminotransferase; UPC, urine protein-to-creatinine ratio.
The remaining cats were in remission until the end of the study. In the majority of all cats, remission occurred within 16 weeks of discharge ( Table 2). The onset of remission did not differ between groups.
Good metabolic control was obtained in 3/5 cats that did not achieve remission in group 1, and in 1/7 cats that did not achieve remission in group 2 (P = .222). Overall, good metabolic control or remission was achieved in 13/15 cats in group 1 and in 8/14 cats in group 2. During the 6-month follow-up period, the median insulin dosage was significantly lower in group 1 cats compared with group 2 (group 1: 0.32 IU/kg/day [range: 0.13-0.53]; group 2: 0.51 IU/ kg/day [range: 0.05-1.52]; P = .013) (Fig 1).
Discussion
Intravenous infusion of insulin was effective in quickly reducing glucose concentrations to within the target range, and compared with subcutaneous injections of insulin, prevented long periods of moderate to severe hyperglycemia (ie, >270 mg/dL). Glucose concentrations were rarely below the reference range, and episodes of clinical hypoglycemia did not occur. These findings support the conclusion that our method of intensive intravenous infusion of insulin can be safely used to rapidly decrease hyperglycemia without adverse effects. In previous studies, we demonstrated that CGMS is clinically accurate for generating 12-and 24-hour glucose curves in diabetic cats. 15,16 The results of the present study showed that a CGMS is a valuable tool for long-term monitoring of glucose curves (ie, 1 week) in hospitalized cats and allows fine-tuning of the insulin dosage.
Remission occurred in 10/15 cats initially treated with intensive intravenous infusion of insulin compared to 7/ 14 cats treated with subcutaneous injections of insulin. When cats that went into remission were excluded, good glycemic control was achieved in 3/5 cats that had been initially treated with intensive intravenous infusion of insulin and in 1/7 cats treated with subcutaneous injections of insulin. However, differences between the 2 groups were not significant, possibly because of the relatively low number of cats in the study. It was interesting to note that cats started on intensive intravenous infusion of insulin required significantly less insulin than cats started with subcutaneous injections of insulin (approximately 40% less, based on differences between median values). Roomp and Rand carried out two intensive glucose monitoring studies in cats, one with insulin glargine and the other with insulin detemir. 13,14 In both investigations, the aim was to achieve tight glycemic control by frequent monitoring of blood glucose concentration and adjustment of insulin dosage. Owners were asked to measure blood glucose concentrations of their diabetic cats at least three times daily at home and to adjust the insulin dosage with the help of a web-based protocol. The rate of remission was 64% in diabetic cats treated with insulin glargine and 67% in diabetic cats treated with insulin detemir. 13,14 Their rates of remission were similar to our results in cats started on intensive intravenous infusion of insulin. These results suggest that strict blood glucose monitoring and frequent insulin dosage adjustments are as effective as intensive Table 2. Onset of remission in cats started on initial intensive intravenous infusion of insulin with tight glycemic control achieved by insulin dosage adjustment (group 1) and in cats started on subcutaneous injections of insulin without the goal of initial tight glycemic control (group 2). intravenous infusion of insulin. However, the study design of those previous investigations differed substantially from that of the present study making direct comparison difficult. In the study by Roomp and Rand, owners of cats with DM were responsible for maintaining tight glycemic control, 13 whereas in our study, veterinarians were in charge of insulin dosage adjustments. One would expect that treatment decisions made by veterinarians in a hospital setting are more consistent than those made by owners at home. In addition, it should be noted that in contrast to our study, a substantial percentage of the cats included in the study by Roomp and Rand had been treated with glucocorticoids before being diagnosed with DM. 13 In our study, cats with prior steroid treatment were excluded. The chance of remission in cats with glucocorticoid-induced DM is good, 13 which may explain the slightly higher remission rate found in the glargine-treated cats in their study compared with the cats of group 2, in which initial tight glycemic control was not a goal. Of note, by excluding glucocorticoid-treated cats from that study, 13 the remission rate would have decreased from 64% to 51%, yielding a remission rate similar to our group 2 cats as well as to other cats started on subcutaneous injections of insulin without initial tight glycemic control. 5 Clinical studies in human medicine demonstrated that initial intensive infusion of insulin has positive effects on metabolic control in patients with newly diagnosed type 2 diabetes compared with oral antidiabetic drug treatment. [10][11][12] This beneficial effect was attributable to improved b-cell function, evidence of which was seen as augmented glucose stimulated firstphase insulin secretion after intensive infusion of insulin. 12 The mechanisms involved in improved b-cell function after intensive insulin treatment in humans include reversal of glucotoxicity and lipotoxicity, reduction in islet inflammation, or both, all of which contribute to loss of b-cell function and mass. [20][21][22] Glucotoxicity is considered to be the major cause of loss of b-cell function and mass. 22 Similar to findings in humans, hyperglycemia has been shown to induce early and severe dysfunction and loss of b cells via apoptosis in healthy cats. 23,24 Thus, eliminating the detrimental effects of hyperglycemia using initial intensive infusion of insulin may improve b-cell viability in diabetic cats, ultimately decreasing insulin requirements for the maintenance of glycemic control. Evidence of this in the present study was the lower median insulin dosages required by cats in group 1.
This study had some limitations including the relatively low number of cats; studying a larger number of cats may or may not have shown significant differences in rates of remission or good metabolic control. The follow-up period was set at 6 months and it is possible that this period was too short for remission to occur in some of the cats. However, based on an earlier study from our clinic, more than 90% of cats with newly diagnosed diabetes achieved remission within 6 months, 5 and therefore the duration of the follow-up period presumably had little effect on the remission rate. The central venous catheters from 4 cats yielded positive bacteriological cultures, but because none of these cats had related clinical or laboratory abnormalities, it was assumed that the adverse effect of the recovered microorganisms was negligible. Finally, finetuning of the insulin dosage to maintain blood glucose levels within the target range was difficult in some cats; concentrations above the target range occurred despite strict 24-hour monitoring. It is possible that in some cats this interfered with the beneficial effects of intensive intravenous infusion on b cells.
In summary, cats that initially received intensive intravenous infusion of insulin with tight glycemic control had significantly decreased insulin requirements in the 6-month follow-up period. Remission rate and metabolic control did not differ significantly between the two groups, possibly because of the small number of cases. Intensive intravenous infusion of insulin is a demanding treatment, and owner compliance may not be satisfactory because of the high cost and the relatively long hospitalization period.
|
2018-04-03T04:53:22.545Z
|
2014-10-13T00:00:00.000
|
{
"year": 2014,
"sha1": "3238fab81e41934c8a9aea188c2a2aa2c4b09a8d",
"oa_license": "implied-oa",
"oa_url": "https://europepmc.org/articles/pmc4895636?pdf=render",
"oa_status": "GREEN",
"pdf_src": "PubMedCentral",
"pdf_hash": "3238fab81e41934c8a9aea188c2a2aa2c4b09a8d",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
268362109
|
pes2o/s2orc
|
v3-fos-license
|
Enhancing endoscopic measurement: validating a quantitative method for polyp size and location estimation in upper gastrointestinal endoscopy
Background Accurate measurement of polyps size is crucial in predicting malignancy, planning relevant intervention strategies and surveillance schedules. Endoscopists’ visual estimations can lack precision. This study builds on our prior research, with the aim to evaluate a recently developed quantitative method to measure the polyp size and location accurately during a simulated endoscopy session. Methods The quantitative method merges information about endoscopic positions obtained from an electromagnetic tracking sensor, with corresponding points on the images of the segmented polyp border. This yields real-scale 3D coordinates of the border of the polyp. By utilising the sensor, positions of any anatomical landmarks are attainable, enabling the estimation of a polyp’s location relative to them. To verify the method’s reliability and accuracy, simulated endoscopies were conducted in pig stomachs, where polyps were artificially created and assessed in a test–retest manner. The polyp measurements were subsequently compared against clipper measurements. Results The average size of the fifteen polyps evaluated was approximately 12 ± 4.3 mm, ranging from 5 to 20 mm. The test–retest reliability, measured by the Intraclass Correlation Coefficient (ICC) for polyp size estimation, demonstrated an absolute agreement of 0.991 (95% CI 0.973–0.997, p < 0.05). Bland & Altman analysis revealed a mean estimation difference of − 0.17 mm (− 2.03%) for polyp size and, a mean difference of − 0.4 mm (− 0.21%) for polyp location. Both differences were statistically non-significant (p > 0.05). When comparing the proposed method with calliper measurements, the Bland & Altman plots showed 95% of size estimation differences between − 1.4 and 1.8 mm (− 13 to 17.4%) which was not significant (p > 0.05). Conclusions The proposed method of measurements of polyp size and location was found to be highly accurate, offering great potential for clinical implementation to improve polyp assessment. This level of performance represents a notable improvement over visual estimation technique used in clinical practice.
Endoscopy is crucial for diagnosing gastrointestinal diseases and detecting early malignancies.In the management of gastrointestinal diseases, optical assessment during endoscopy is essential for following strategies like "diagnose and leave" and "resect and discard", as well as setting surveillance intervals and choosing resection techniques.Enhancing this assessment can optimise patient care, balance intervention risks and benefits, and manage healthcare resources effectively.One important factor in predicting the malignancy risk and growth rate is the size of detected lesions, such as polyps.Symptomatic hyperplastic polyps larger than 10 mm with pedunculated morphology should be resected [1].Polypectomy of fundic gland polyps of 10 mm or larger, and hyperplastic polyps of 5 mm or larger is recommended [2].The endoscopic mucosal resection (EMR) technique for gastric lesions smaller than 10 mm in size and the endoscopic submucosal dissection (ESD) technique for lesions larger than 10 mm in size are recommended [1].Larger gastric polyps require more frequent surveillance [3].Diminutive colorectal polyps less than 5 mm are targeted for "diagnose and leave", or "resect and discard" strategies [4].
In current clinical practice, both endoscopists and pathologists estimate polyp size.Studies show discrepancies regarding whether measurements by pathologists should influence clinical decisions [5][6][7] or those by endoscopists [8], with a majority favouring histopathological measurements.The visual estimation made by endoscopists in vivo is challenging because the size of an object, as viewed in the image of the monitor, changes depending on the distance between the endoscope camera and the object.Therefore, guessing the size visually is subject to error and variability.On the other hand, histopathological measurements are not applicable in strategies such as "diagnose and leave" or "resect and discard".Moreover, factors such as resecting the rim of the normal tissue, and shrinkage after polypectomy can manipulate the actual in vivo size.Misjudging the polyp size, either by overestimation or underestimation [9,10], may result in inaccurate decisions regarding the necessity for resection or surveillance [7,11].Consequently, the advised follow-up schedule may not correlate with the actual risk level, leading to either unnecessary, frequent surveillance or insufficient monitoring.Inaccurate polyp size estimation may also influence the selected resection technique, potentially reducing the rate of complete and curative resection [12].All these factors impact patient outcomes and healthcare system costs.
To overcome this challenge, several studies have proposed different approaches for in vivo polyp measurement, including the use of an endoscopic graduated device [13][14][15], applying deep learning models [16,17], and integrating specialised devices with conventional endoscopes [18][19][20][21][22].Among these, the integration of a specialised device seems to hold significant promise for providing a direct, objective, and accurate measurement.
In our prior study [23], we developed a novel quantitative method to provide polyp size and location measurements based on integrating an electromagnetic tracking sensor with a conventional endoscope evaluated in an upper gastrointestinal experimental model.Building on our previous work, the main aim of this study is to further investigate the quantitative method applicability by conducting ex vivo experiments to assess the system's performance under conditions more resembling real endoscopy procedures.
Materials and methods
The authors declare that no human subjects were involved in this study, and as such, IRB approval was not required.
System overview
In our previous research, we introduced a novel quantitative method, which consists of integrating electromagnetic tracking technology1 with a conventional endoscope (Pentax EPK-i), as illustrated in Fig. 1, and is complemented by a newly developed computer-vision-based algorithm.This algorithm can yield the 3D coordinates of a detected polyp's border in real scale.By fitting an ellipse to these 3D point coordinates, one can derive both the longest length of the polyp and the distance from its centre to a specific anatomical landmark, as detailed in our previous work [23].Evaluated in a simulated upper gastrointestinal model and an artificial rounded polyp, a root mean square error (RMSE) of less than 1 mm for size and 3 mm for location estimation were achieved.
The sensor accuracy test results indicated no significant degradation in sensor performance in endoscopic settings, underscoring its potential applicability.Moreover, through sensitivity analysis, we provided recommendations for capturing endoscopic images from a polyp to ensure optimal accuracy with our proposed method.This includes avoiding large relative movements (specifically, relative translation > 30 mm or relative rotation > 30 degrees), minimal movements (displacement < 3 mm), or pure forward-backward movements.
Following the promising initial results, the quantitative method was evaluated in this study, while endoscopy procedure was performed using ex vivo pig stomachs.While the core principles of the quantitative method remain intact for this evaluation, a few modifications were necessary to adapt to the unique characteristics of the new setting.These modifications related to the two steps of the computer visionbased algorithm including finding corresponding points and polyp segmentation.
While the movement of the polyp cannot be controlled, signs of its movement might be detected when multiple image pairs are used for a single polyp size determination.Therefore, here a simple approach of capturing multiple images and subsequently multiple estimations to compensate for potential polyp movements between taking image pairs were investigated.If the polyp remains stationary, the 3D coordinates of the polyp's centre would be expected to stay consistent across all measurements.Consequently, outliers in the estimated 3D coordinates of the polyp centres were identified using quartiles, and those image pairs were excluded from the final estimation, while the average of remaining estimations was considered as final estimation.
In the following sections first, two adapted steps including polyp segmentation and corresponding points will be described briefly, and subsequently, the experimental design for method evaluation will be explained.
Polyp segmentation
Given the diverse characteristics of polyps, which reflect the variability observed in actual polyps, here we employed a pre trained DL-based model called Polyp-PVT (Pyramid Vision Transformers) [24], for polyp segmentation purposes in our experiment.This model consists of a pyramid vision transformer [25] as an encoder to extract multi-scale feature maps where the image divided to smaller patches in each level.The model consists of three other modules, a cascaded fusion module (CFM) to fuse the high-level features through progressive integration, a camouflage identification module (CIM) to capture low-level features using an attention mechanism, and a similarity aggregation module (SAM) to integrate low and high-level features for final segmentation.The Polyp-PVT model aims to extract more powerful and robust features to improve polyp segmentation regarding accuracy and generalisation ability.The model achieved a Dice similarity coefficient (DSC) of 0.917 and 0.937 on the Kvasir-SEG [26] and CVC-ClinicDB [27] that consist of 1000 polyp images and 612 polyp frames, respectively.For our ex vivo experiment, we employed this model and subsequently applied some image post-processing techniques to further improve the segmentation quality including, basic morphological operations such as opening, and filling operations, to eliminate small detected regions of interest (ROI), and fill in small holes within the ROIs, or deformable segmentation based on active contour models [28] in situations where the model output covers the polyp partially.Figure 2 illustrates some examples of segmentation outputs on polyp images as explained above.
Corresponding points
Repetitive and indistinctive texture surfaces in endoscopic images can adversely affect the feature detection and matching algorithms.Consequently, in this paper, we explore a new approach for determining corresponding points based on shape context, which is unaffected by the texture-less nature of the endoscopic images.
An object's shape can be represented by a discrete set of points, denoted P = p 1 , p 2 , …, p n where p i ∈ R2 from the internal or external contours of the object.In the case of a polyp, the shape can be defined by a set of points located on its border.Given the set of points P, with n points of the polyp shape in one image and the set of points Q, with m points in the other image, the shape context method [29] can be used to find one-to-one correspondences between these two 2D sets of points, P and Q.The shape context is a vector of local geometric information that characterises the distribution of points around a reference point on the shape boundary.The shape context can be defined as a histogram using a log-polar coordinate system.The log-polar system Fig. 1 Right: Endoscope with a fixed electromagnetic sensor using a 3D printed cover.Left: Dimensions of the electromagnetic sensor represents a point based on the logarithm of its distance from the origin (log r) and its angle (θ) from a reference direction.This is illustrated in Fig. 3 with a log-polar coordinate system of 5 bins for log r and 12 bins for θ.
Two shapes of the polyp border in two endoscopic images are presented by sample points.Considering a log-polar coordinate system; each shape context is a log-polar histogram of the coordinates of the rest of the point set measured using a reference point as the origin of the log-polar coordinate system.As shown in Fig. 3, the shape context of relatively similar points (i.e.green and red) on the polyp border is visually more similar, while the shape context of another point (i.e.blue) is quite different.This concept was used to find corresponding points along the border of a polyp in the present study.The result of finding corresponding points for one polyp in the pig stomach experiment using this approach is illustrated in Fig. 4.
Experimental design
The sample size was determined based on the correlation coefficient and considered a two-tailed hypothesis: Null Hypothesis: The correlation coefficient is zero.Assuming a power of 0.8 (β = 0.2), a significance level of α = 0.05, and an anticipated correlation coefficient of C = 0.7, a preliminary sample size was calculated.To account for unforeseen circumstances, an additional 10% was added to this initial calculation, resulting in a final sample size of 15.
Five pig stomachs were utilised in one of which three polyps were created by tying off sections of the stomach wall, in various sizes representing the common real polyp sizes.For this validation, endoscopy procedure was simulated on ex vivo pig stomachs, which were placed in a container and secured at the oesophagus end and sealed at the pylorus end, as depicted in Fig. 5. Notably, the stomach remained unfixed, permitting potential random movements.This procedure, executed by an experienced expert, aimed to mimic real endoscopy in terms of navigation and environment.The stomach was inflated with air and actions such as water jet cleaning and suctioning were performed to replicate actual endoscopic conditions.
The extracted recommendations from model experiment in our previous work were used as a feedback mechanism to achieve the higher accuracy.Five different images of each polyp in a selected perspective made by the expert were taken and used for size and location estimation by the proposed quantitative method.The location of the oesophagus-gastro junction was also recorded, and the locations of the polyps with respect to the junction were determined using the proposed method.This procedure was repeated twice to compute the test-retest reliability for both size and location estimations.After each simulated endoscopy, polyps were measured with a digital calliper.Because the polyps are asymmetric, viewing perspective can affect length measurements.To reduce bias, we aligned the calliper perspective with the captured images.Although calliper measurements are not the most reliable for irregular and non-rigid polyps, we compared them to our method for validation, expecting general agreement between both techniques.
Test-retest reliability
The fifteen polyps examined in this study had an average size of approximately 12 ± 4.3 mm (mean ± SD), ranging Bottom row: Shape context for reference samples marked by green, red and blue, on the polyp border (Color figure online) Fig. 4 An example of finding corresponding points (down sampled) using shape context on a polyp border in an image pair in ex vivo experiment from 5-20 mm.The Intraclass Correlation Coefficient (ICC) for polyp size estimation between test-retest data was assessed using a two-way mixed effects model [30].The ICC value for absolute agreement was 0.991 (95% CI 0.973 to 0.997, p < 0.05).To explore the agreement of test-retest data for both size and location estimations, we utilised Bland & Altman (B&A) analysis [31].The differences in the B&A plot can be expressed as units or percentages (i.e.(test-retest)/average %).Given the standard deviation of the differences (s) and the mean of the differences (d), the two upper and lower agreement limits were computed as d ± 2s.Considering the normal distribution of the differences, 95% of differences are expected to lie between d + 1.96s and d − 1.96s.
The normal distribution of differences for test-retest data was confirmed using the Shapiro-Wilk test [32].Assuming a two-tailed test with a p-value of 0.05, the confidence intervals for the mean line were derived.
Figure 6 shows the Bland & Altman plots for both size and location estimations using the proposed method, presented in both unit and percentage forms.For size estimation, the mean difference stands at − 0.17 mm or − 2.03%.This difference is not statistically significant, given that the line of equality lies within the 95% confidence interval.Consequently, 95% of the differences between test-retest data lie between − 1.31 and 0.96 mm or from − 13 to 8.9%.
For the estimation of polyp location (i.e. the distance from the oesophagus-gastro junction), the Bland & Altman plot indicates a mean difference of − 0.4 mm or − 0.21% between test-retest data, which is not statistically significant.The limits of agreement, considering a 95% confidence interval, range from − 4.7 to 3.9 mm or from − 3.7 to 3.3%.
Comparison with clipper measurements
Figure 7 presents the Bland & Altman plot comparing polyp size estimations made by the proposed method and the calliper.According to the plot, no significant bias is evident (0.2 mm or 2.22%), with 95% of differences fall between − 1.4 and 1.8 mm, or − 13 to 17.4%.
Additionally, a comparison of polyp estimations from the proposed method with those from digital calliper measurements revealed a root mean square difference (RMSD) of 0.81 mm.A median absolute difference of 0.45 mm with an interquartile range (IQR) of 0.88 mm was observed.This difference was not statistically significant (p = 0.37).The mean absolute percentage of difference was recorded as 6 ± 5.8%.
Discussion
In this study, we validated a method developed in our previous work for determining the size and location of polypoidal lesions using a monocular endoscope combined with an electromagnetic tracking sensor through an ex vivo experiment.The test-retest results for size estimation yielded an ICC of 0.99 (p < 0.05).A non-significant mean difference of − 0.17 mm, with 95% of differences falling within the range of − 1.31 to 0.96 mm, indicates high reliability.This represents a significant enhancement in comparison to the inter-physician variability seen with visual estimation, which can deviate by more than 5 mm or yield an ICC as low as 0.13 [10].
Moreover, our proposed method demonstrated high reliability in location estimation, with a non-significant bias of − 0.4 mm and 95% agreement limits ranging from − 4.7 to 3.9 mm.In a recent study [8], visual estimations by 6 Bland and Altman plot comparing test-retest measurements by the proposed quantitative method.First row: polyp size estimations where differences are presented as units (left) and where differences are presented as percentages (right).Second row: polyp location estimations where differences are presented as units (left) and where differences are presented as percentages (right) Fig. 7 Bland and Altman plot comparing two measurements for polyp size estimation by the proposed quantitative method and digital Calliper where differences are presented as units (left) and where differences are presented as percentages (right) endoscopists were compared to post-resection before fixation measurements of the polyp, considered as actual value.Based on their Bland & Altman analysis, an agreement limit of − 2.5 to 2.8 mm was reported that was more accurate than pathologic report.When comparing our method with calliper measurements, an agreement range of − 1.4 to 1.8 mm was observed.This underscores that polyp size estimation, using our proposed method, can more accurately represent polyp size, offering an improvement compared to both visual estimations and pathological reports.
In comparison to other device-based methods, Table 1 encapsulates various recent studies focussed on device-based approaches for polyp size measurement.These studies are summarised in terms of their method, accuracy, subjectivity, and estimation time.
Acknowledging the varied evaluation settings across studies, our proposed method demonstrates better accuracy in endoscopic polyp size measurement.The method is advantageous due to its minimal impact on the endoscopy procedure as it involves only taking pictures of the polyp, which is a step already recommended in clinical guidelines.Nonetheless, it needs to be further validated by testing in human subjects during endoscopic procedures.Measurements can be estimated immediately after acquiring the endoscopic images of the polyps, aligning with the time required for visual estimation.The proposed method identifies the polyp border and corresponding points automatically, enhancing the level of objectivity as the estimations do not involve comparisons with a scale.Moreover, our method also explored the potential of polyp localisation showed good reliability without necessitating modifications to the existing equipment or procedure.
While our approach demonstrates promising results in the ex vivo environment, the findings of the current study are limited by the fact that a clinical trial on human subjects has not been carried out.It is important to note that the current setup, with the sensor attached externally, does not reflect the final intended design for clinical application.In the real-world application, the sensor will be fully incorporated in the endoscope's structure internally.Considering the sensor's small size (1.8 mm in diameter), we anticipate there will be minimal increase in the endoscope's diameter, post-integration, to allow the endoscope to be used normally describing a non-regular shape using a single-length measurement can introduce variability into the assessment.Using an alternative metric, such as volume, to evaluate asymmetric morphology may potentially reduce both inter and intra rater variability.Even when a method provides the ability to reconstruct the shape, the longest length is usually extracted from that shape because clinical guidelines primarily emphasise the longest length of polyps.To the best of the author's knowledge, clinical guidelines do not establish a clinical connection or cut-off for lesion volume in relation to clinical decisions.
Additionally, it is worth noting that shape reconstruction comes with certain challenges, including longer computation times and challenges when there are texture-less surface and restricted endoscope movement, which affect the shape reconstruction approach more than the chosen approach in this study.Nonetheless, it is important to highlight that the proposed quantitative method has the potential to be further improved in future research for full 3D reconstruction of the endoscopic view, shape reconstruction, and volume measurement.Furthermore, the applicability of the proposed quantitative method extends beyond its use in this study; it can be easily generalised to other endoscopic imaging modalities where assessing the size and location of abnormalities is crucial for evaluation and treatment planning, for example, to measure the length of Barrett's oesophagus.Additionally, the integration of this measurement system with an AI-based diagnosis system could potentially equip clinicians with more comprehensive analysis.Combining size measurement with other morphological and textural features with an AI-based algorithm could improve the risk stratification of polyps.This can help in reducing unnecessary biopsies or polypectomies for polyps that are identified to be benign.It will also help avoid missing potentially malignant polyps as a result of under-estimation of its size.The optimisation of endoscopic procedures will consequently benefit the patient as well as the healthcare system.
The integration of this technology undoubtedly increases the initial cost of the endoscopic device.However, this should be weighed against its potential to facilitate optimal clinical decision-making.By enhancing diagnostic accuracy and treatment precision, it is also plausible that the use of this technology could lead to a reduction in overall healthcare costs, such as those associated with suboptimal clinical decision-making.Further studies will be recommended to accurately quantify these potential cost savings.
In conclusion, this study validates a novel method for providing polyp size and location measurements during simulated endoscopy in an ex vivo setting.The results of the current study suggest that this approach has substantial promise in improving the clinical accuracy of polyp assessment.This research serves as a significant proof of concept study, paving the way for future clinical investigations and marking a significant advancement in developing endoscopic devices with quantitative capabilities.
Fig. 2
Fig. 2 Visualisation of ex vivo polyp images.Green: polyp segmentation results using Plop-PVT model followed by post-processing operations (Color figure online)
Fig. 3
Fig. 3 Concept of the Shape Context.Top row: Sample points on the border of one polyps in two endoscopic images and the diagram of a log-polar coordinate system with five bins for log r and 12 bins for ɵ.
Fig. 5
Fig. 5 Demonstration of the set up for performing ex vivo pig stomach experiment
Table 1
Comparison of device-based methods for endoscopic polyp size measurement
|
2024-03-13T06:17:56.250Z
|
2024-03-11T00:00:00.000
|
{
"year": 2024,
"sha1": "e63b23fadfb27e52f0e7f12389d423a18b16ec88",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s00464-024-10758-2.pdf",
"oa_status": "HYBRID",
"pdf_src": "PubMedCentral",
"pdf_hash": "0fc0479ea3cfb0c4e8bffe0bd97424450127d8a4",
"s2fieldsofstudy": [
"Medicine",
"Engineering"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
260303258
|
pes2o/s2orc
|
v3-fos-license
|
The impact of bank competition on bank stability in Vietnam: The moderating role of shadow banking
Abstract This paper aims to examine the moderating role of shadow banking in relation to the impact of bank competition on bank stability over a period from 2016 to 2021 in Vietnam. After building a bank stability index by combining the principal components of CAMELS through Principal Component Analysis (PCA), and the Lerner index as a measure of bank competition, this research uses panel corrected standard errors (PCSE) to analyze data of 20 Vietnamese commercial banks over a period from 2016 to 2021. As a result, the research shows that shadow banking reduces the positive impact of bank competition on bank stability in Vietnam despite it being considered a competitive strategy of banks. Furthermore, the research also indicates the positive role of bank size, equity to total assets, state ownership, and banking sector development for enhancing bank stability, while the opposite impact can be seen in the case of inflation. These results can help authorities in the banking sector and commercial banks in Vietnam to take appropriate measures to actively supervise or carefully implement shadow banking services in order to ensure bank stability.
Introduction
Bank stability can be defined as a state in which banks operate effectively in terms of resource distribution, risk dispersal, and income distribution (Jahn & Kick, 2012), helping them cope well with both internal and external issues, particularly economic shocks. By contrast, instability occurs when banks assets drop and they cannot repay their debts or the market value of their assets is less than their total liabilities (Jokipii & Monnin, 2013). Bank stability is strictly associated with the stability of the financial system, and GDP growth (Bayar et al., 2021;Jayakumar et al., 2018;Jokipii & Monnin, 2013).
There are different determinants of bank stability, such as bank characteristics (Nyangu et al., 2022;Tabak et al., 2012), financial structure (Goetz, 2018;Mirzaei et al., 2013), and macroeconomic factors (Bashir, 2022;Goetz, 2018). As an important factor, the degree of market power of banks, called bank competition, has inconsistent impacts on bank stability. To be precise, the positive impact of bank competition on bank stability is supported by the theory of market position of Boyd and De Nicolo (2005), while the theory of charter value of Marcus (1984) and Keeley (1990) tries to explain the negative aspect of this relationship. Boyd and De Nicolo (2005) develop the theory of market position based on the ideas of "too big to monitor" of Boyd et al. (1993) and "too big to fail" of Mishkin (1999). Authors argue that banks in centralized or concentrated markets seem to be, almost without exception, bailed out by governments in case of bankruptcy since they are too large and too important to the economy. Banks are willing to take excessive risks or acquire riskier assets on portfolio to generate a lot of profit, which when successful increases the stability of the bank. In other words, the power of a high position in the market promotes financial stability in banks. Besides this, Supangkat et al. (2020) argue that as banks becomes more competitive, the operational efficiency of banks rises, leading to an increase in bank stability.
By contrast, Marcus (1984) and Keeley (1990) base their research on the charter value in order to explain the relationship between bank competition and bank instability. Charter value is defined as the difference between the market price and the book value, meaning capital gains for the bank's shareholders. In the case of concentrated or monopolistic markets, the market power of banks increases (Berger et al., 2008;Mateev et al., 2021), then banks will get higher returns on the lending market by offering higher lending rates (or higher charter values), leading to an increase in the likelihood of bankruptcy of borrowers, and consequently the risk of banks' bankruptcy or financial instability. Jiménez et al. (2013) conclude that less bank competition encourages banks to protect their high charter value using capital adequacy strategies, thus making banks more stable. However, this hypothesis is rejected by Beck (2008) who argues that more competitive banks are also less likely to the bank system distress. Negative or positive impacts of bank competition on bank stability are mostly affected by non-traditional or non-interest products offered by banks. Shadow banking is an example of such a product. Shadow banking is defined as activities providing capital to businesses through credit while breaking certain regulatory restrictions and constraints on lending by applying non-standard accounting methods (Sun, 2019). Shen et al. (2020) show that shadow banks involve unregulated credit intermediaries of banks, which are created "inside the bank". In a competitive market, developing shadow banking is a way to increase bank competition (Tan et al., 2022). However, shadow banking can lead to considerable risk for banks, thus increasing the banking system's vulnerability according to some (Ding et al., 2020;Ilesanmi et al., 2019), or not according to others (Demirgüç-Kunt & Huizinga, 2010;Rose, 1989).
For many years, Vietnam has been considered as a banking-based economy. However, commercial banks have been accused of carrying out insider activities, called shadow banking, which could damage the stability of banks and the banking system (T. A. Pham, 2022). Some banks own controlling stakes in several companies which conduct business separately and independently, while some enterprises own banks. Recently, banks introduced corporate bonds to their clients (even to those with low credit ratings) but mislead them into believing that all corporate bonds were guaranteed by the bank, whereas in fact banks only played the role of brokers (T. A. Pham, 2022). Consequently, bank clients had to face all risks when businesses went bankrupt in 2022, causing a decrease in confidence in the banking system Considering the current situation in Vietnam and the inconsistent findings observed in previous studies, the study aims to investigate the moderating role of shadow banking in the relationship between bank competitiveness and bank stability. This focus sets this study apart from previous studies of scholars, such as Vo and Dang (2016), T. T. , and Son (2022) who only focus on the relation between bank competition and bank stability, Tran (2016) who investigate the effect of shadow banking activities on the financial conditions of Vietnam securities company or Nguyen (2018) who are interested in shadow banking separately. This paper investigates 20 Vietnamese commercial bank-level datasets which are collected from audited financial statements of banks over a period from 2016 to 2021 and uses the Hierarchical Multiple Regression Approach using Panel Corrected Standard Errors (PCSE) to achieve the research objectives. This paper makes the following two contributions. First, this paper adds to the research on the link between bank competition and bank stability as well as the moderating role of shadow banking. This research provides evidence of the positive impact of bank competition on bank stability among Vietnamese commercial banks, providing support for the theory of market position. In addition, the empirical results also show that shadow banking reduces the positive impact of competition on stability instead. Furthermore, this paper makes policy recommendations dealing with the relationship between bank competition and bank stability while considering the role of shadow banking. In terms of policy implications, this study suggests that authorities should actively supervise shadow banking activities. Moreover, commercial banks should be aware of the trade-off between benefits and drawbacks brought by shadow banking, in order to implement appropriate strategies of risk management. This paper proceeds in six sections. Section two presents a related literature review on the link between bank competition and bank stability and the moderating role of shadow banking in this relationship. Section three explains variables, the data collection, and the data analysis. Section four describes the analysis results before discussing findings and providing conclusions in section five and section six.
Literature review and developing hypotheses
It can be seen that empirical results on the impact of bank competition on bank stability follow two theoretical perspectives, including "competition-stability" and "competition-fragility". The positive impact of bank competition on bank stability can be found in many studies, such as those of Schaeck and Čihák (2012), Amidu and Wolfe (2013), Noman et al. (2017), Goetz (2018), Noman et al. (2018, and Islam et al. (2020). These authors argue that competition is necessary to enhance financial stability and overcome the limitations of monopoly markets (Caminal & Matutes, 2002;Schaeck et al., 2009). In an era of the rapid growth of financial institutions, GC and Sharma (2020) show a positive relationship between greater bank competition and more excellent financial stability in Nepal, providing evidence for the "competition-stability" view. However, the positive impact of bank competition on bank stability depends on the extent to which banks can benefit from subsidies because they are deemed "too large to fail" (Mishkin, 1999;Soedarmono et al., 2013). By contrast, the "competition-fragility" view can be seen in many studies executed in different regions of the world, such as the East Asian countries (Phan et al., 2019), the Sub-Saharan African countries (Sarpong-Kumankoma et al., 2020), the MENA countries (Albaity et al., 2019;Moudud-Ul-Huq et al., 2022), and the BRICS countries (Moudud-Ul-Huq, 2021). López-Penabad et al. (2021) argue that competition motivates banks to take risks, which increases the fragility of banks (Leroy & Lucotte, 2017;Schaeck et al., 2009).
In particular, Martinez-Miera and Repullo (2010) show that the likelihood of bankruptcy increases but then decreases after a certain level of banking competition occurs, which results in a bellshaped non-linear relationship between competition and financial stability. According to the authors, when the level of competition does not exceed the optimal, the hypothesis of "competition-stability" holds, and the hypothesis of "competition-fragility" is accepted when bank competition is beyond optimal (Dutta & Saha, 2021;S. Hou, 2023;Liu et al., 2013). Similarly, inverted U-shaped relationships can be seen in the studies of Tabak et al. (2012) in Latin America, Jiménez et al. (2013) in Spain, Jeon and Lim (2013) in Korea, González et al. (2017) in MENA, and Yuan et al. (2022) in America. However, the non-linear relationship between bank competition and bank stability is not found in the European market (López-Penabad et al., 2022) or in the East African Community (Nyangu et al., 2022).
The relationship between competition and bank stability in Vietnam is also of interest to researchers. Vo and Dang (2016) and T. T. support the "competitionstability" nexus, while Son (2022) indicates that banks tend to take on more risks when having to face an increase in competition among banks. This research proposes the first hypothesis as follows: Hypothesis 1 Bank competition impacts on bank stability in Vietnam.
In terms of shadow banking, it has interactions with both bank competition and bank stability (Tan et al., 2022;Wu & Shen, 2019;Zhang et al., 2022;Zhu, 2022). The non-zero-sum game theory of Brandenburger and Nalebuff (1995) explains that commercial banks co-conduct conventional credit business with shadow banking to maximize profits, avoid scrutiny from authorities, and enhance the innovation and competitiveness of banks. Zhang et al. (2023) conclude that the greater the proportion of investment in shadow banking, the higher the bank's profitability, which motivates banks to carry out shadow banking activities. Tan et al. (2022) also find that the extent of shadow banking can be affected by the intensity of competition in the banking market and vice versa.
Moreover, the risk-return trade-off theory proposed by Rose (1989) indicates that diversifying operations into non-traditional sectors will mitigate the overall risk of the bank if the rate of return or cash flow of these activities is lower than the return rate or general cash flow of the bank. According to Demirgüç-Kunt and Huizinga (2010), the more diverse the bank's non-interest income index is, the lower the lending risk, leading to improved bank stability. Using a variety of measuring of bank stability, Tahir et al. (2016) demonstrated that through diversification strategies, banks' risks can be minimized, improving their stability in South Asian countries. By contrast, shadow banking increases credit risk (Bashir, 2022) or liquidity risk since shadow banking investments are hardly ever liquidated in volatile market conditions. The interconnectedness of shadow banks and traditional banks easily leads to a risk transformation from shadow banking activities to large commercial banks (Ilesanmi et al., 2019;Shen et al., 2020), which results from a lack of regulations (Isayev & Bektas, 2022). Ding et al. (2020) claim that the non-transparency of shadow banks detailed in financial reports and their hard-to-identify underlying assets make it difficult to assess their quality or make provisions for default. Furthermore, Ouyang and Wang (2022) confirm a negative impact of shadow banking on bank stability in large-scale banks, but this effect is not considerable in small and medium-sized banks. It can therefore be seen that the relationship between bank competition and bank stability can be moderated by shadow banking. Accordingly, this paper proposes the following hypothesis: Hypothesis 2 Shadow banking activities moderate the impact of bank competition and bank stability in Vietnam.
Dependent variable: bank stability
For years, the Z-score has been largely used to measure bank stability by scholars, including Beck et al. (2013), T. T. , and Nyangu et al. (2022), while CAMELS is considered a benchmark criterion for most financial institutions to assess the strength of banks and financial institutions. Comparing these two indicators, Permata and Purwanto (2018) criticize the Z-score indicator for two reasons, including: (i) It is said that the reliability of scale largely depends on financial reports of quality; (ii) The Z-score only reflects bank risks but ignores spillover or interconnectedness in system risk or non-financial factors. That's why, this paper uses CAMELS which is described in Table 1, as indicators of bank stability.
Since each financial ratio only explains a specific aspect of bank stability, this paper tries to build a bank stability index by combining the principal components of CAMELS through Principal Component Analysis (PCA) in the same way as Creel et al. (2015), Horváth and Vaško (2016) and Jayakumar et al. (2018) did. The process of calculating the bank stability index is shown in the following steps: Step 1: Collect bank-level data on CAMELS-related-financial ratios.
(ii) Step 2: Normalize data by using the Min-Max method proposed by Moesen and Cherchye (1998) to convert component ratios to the same unit of calculation so that they can be compared as well as combined. The normalization process can be expressed as following: In which: Source: Authors.
(iii) Step 3: Use the PCA approach to assign weights to all six bank stability dimensions and combined them to create a composite index called the Bank Stability Index. The process for calculating the weights is shown in Appendix 1.
Independent variable: bank competition
In a non-structured approach used to measure bank competition, there are different indicators, such as H-statistic (Panzar & Rosse, 1987), Boone indicator (Boone, 2008), the Lerner index (Lerner, 1934) and efficiency-adjusted Lerner. However, H-statistic and Boone indicator have major drawbacks. To be precise, H-statistic is strictly based on the assumption of an equilibrium market while this is unrealistic (Claessens & Laeven, 2004). For the Boone indicator, it exploits the reallocation from inefficient units to efficient ones, while the Lerner index is a competitive measure rooted in banking optimization problems (Maudos & Solís, 2011 The ratio of bank assets and bank system assets (0.5) S1 The ratio of bank assets and bank system assets -Difference between interest-sensitive assets and interestsensitive liabilities to total assets (0.5)
Lerner index has a solid theoretical basis and can identify different patterns of behavior in the same market and/or between years as well as better capture the market power of banks. Hence, this paper uses the Lerner index to measure bank competition. The Lerner index is determined by the ratio of the difference between the output price and the marginal cost to the output price, through the following formula: In which: o Price it is the output price of the bank i at time t, which is calculated as total revenue on total assets. o MC it is the marginal cost of the bank i at time t.
The value of the Lerner index ranges from 0 to 1. The lower Lerner index value implies a weaker level of competition between banks whereas equal to 1, the market becomes a complete monopoly, i.e. the bank has absolute market power. The Lerner index is presented in Appendix 2.
Moderating variable: shadow banking
Although the task of extracting this activity-related-data from banks' financial statements is quite difficult, this paper tries to measure the moderating variable of shadow banking in Vietnamese commercial banks based on previous studies, the availability of data, and results from deep interviews with experts in the banking sector. Firstly, based on the research of Ding et al. (2020), and Zhang et al. (2023), and available data, we can calculate interbank loans, entrusted loans, financial products or investment receivables through "loans to other credit institutions" (LCI), payment on behalf of customers (PBC), and contingent liabilities (CL) which are detailed on notes of banks' financial statements. Moreover, data on financial conglomerates which have recorded their total investment in subsidiaries, investment in associate companies, and other longterm investments (ILT), can also be utilised. Accordingly, this paper measures shadow banking in banks by using the following formula: In which: o SB i;t is the shadow banking of the bank i at time t. o LCI i;t is the loans to other credit institutions of the bank i at time t. o PBC i;t is the payment on behalf of customers of the bank i at time t. o CL i;t is the contingent liabilities of the bank i at time t. o ILT i;t is the investment in associate companies, and other long-term investments of the bank i at time t. o TA i;t is the total assets of the bank i at time t.
Secondly, another banking activity which Vietnamese financial experts have been interested in for many years in Vietnam, is the services of agency and brokerage for corporate bonds. In fact, the income from fees and services is considered to be a representative indicator of shadow banking activity in a corporate sense based on regulatory loopholes. Therefore, the ratio of income from agency and brokerage services to total income of banks will be used to measure bank services of agency and brokerage for corporate bonds.
In which: o SB i;t is the shadow banking of the bank i at time t. o IABS i;t is the income from agency and brokerage services of the bank i at time t. o TI i;t is the total income of the bank i at time t.
Control variables
In order to ensure the stability of the estimated results, this paper uses some control variables, including: (i) bank characteristics; (ii) financial structure; and (iii) macroeconomic conditions.
In terms of bank characteristics, this paper refers to bank size, equity to total assets, and ownership structure. Firstly, large banks are able to dominate the market and then generate higher revenues than small banks, resulting in a more stable state (Laeven et al., 2016). However, Tabak et al. (2012), Mirzaei et al. (2013) and Fu et al. (2014) argue that large-scale banks tend to engage in more high-risk activities than small banks, leading to risk for their financial stability. As a result, the relationship between bank size and bank stability is still undetermined. Secondly, equity to total assets (ETA) is used to maintain bank stability in the case of losses due to bad debts or financial shocks (Diaconu & Oanea, 2014;Tabak et al., 2012). Finally, the ownership structure (STO) which in this paper, refers to state ownership, is a factor influencing banks' innovation business strategies (Bashir, 2022;X. Hou et al., 2018).
Concerning financial structures, this paper refers to banking sector development (BSD) and stock market development (SMD). The first indicator which is measured by the ratio of banking sector assets to GDP, means the level of development of the banking sector (Goetz, 2018;Mirzaei et al., 2013). Moreover, stock market development (SMD) measured by the ratio of listed companies' market capitalization to GDP, refers to an efficient capital market, lowering moral hazard and adverse selection (Mirzaei et al., 2013).
Finally, Nyangu et al. (2022), and Bashir (2022) argue that macroeconomic conditions can affect the quality of assets and financial stability of banks. This paper uses GDP and the inflation rate (INF) as factors of macroeconomic conditions, as per the studies of Goetz (2018), and Zhang et al. (2023).
In brief, all variables are summarized in Table 2.
Data collection
The panel bank-level data used in this paper is sourced from audited financial statements and annual reports of 20 commercial banks over a period from 2016 to 2021 which are extracted from FiinPro Platform. The research period starts at 2016, which marks the second restructuring phase of the banking system in Vietnam. Moreover, 20 commercial banks examined by the research are able to represent the whole of the Vietnamese banking sector because their total assets account for a major part of total bank industry assets. Further data related to financial market structures and macroeconomic factors such as GDP growth rates and inflation rates are collected from the IMF and the World Bank.
Data analysis
To test the impact of independent variables and moderating roles, the hierarchical regression method is perfectly suitable. This is because this framework for model comparison allows for statistical control by assessing the contribution of added variables in the model with previous variables and as a means of checking for increased validity. There are three regression models to be tested. The regression Equation 1 expresses the effect of the independent variable (BC it ) on the dependent variable (BSt it ). The regression Equation 2 shows the effect of the independent variable (BC it ) on the dependent variable (BS it ) but the moderated variable (SB it ) is introduced into the model as an independent variable. The regression Equation 3 represents the effect of the independent variable (BC it ) and the interaction variable (BC it � SB it ) on the dependent variable (BSt it ). If the interaction variable has statistical significance, then that proves that the variable SB it acts as the regulatory variable.
Hierarchical regression in this study is approached in three steps with two proxy variables for shadow banking control variables. In the second step, shadow banking is treated as an independent variable, whereas in the third step, there is an interaction variable between bank competitiveness and shadow banking. Furthermore, this research uses the PCSE and FGLS estimators as base regression techniques to deal with issues of cross-sectional data, such as serial correlation, heteroskedasticity, cross-sectional dependence, autocorrelation (Le & Nguyen, 2019;Parks, 1967), and the problem that there are more cross-sections than intervals (N>T). In fact, to account for this disparity, the FGLS estimator generates undervalued standard errors, while the PCSE estimator generates accurate standard error estimates with no loss in efficiency compared to FGLS. This means that the FGLS estimator is better suited for temporal-dominant data in panel data. Table 3 reports the data description with 120 total observations for each variable. It can be seen that the average value and the standard deviation of bank stability (BS) is 0.352 and 0.199, respectively. The Lerner Index which represents the independent variable of bank competition (BC) has an average value of 0.410, a standard deviation of 0.085, and the min-max values of 0.205 and 0.613, respectively. Furthermore, the volatilities of two proxies of the shadow banking variable, including SB1 and SB2, are Table 4 shows the estimates of pairwise and multilinear correlations for variables in this sample. It can be seen that the correlation values reported had no absolute values greater than 0.8, ruling out the possibility of high multicollinearity between the study variables. Furthermore, the variance inflation factor values (VIF) are all as small as 10, so it adds to the certainty that no multilinear phenomena occur with the selected estimators.
Multicollinearity analysis and other data diagnosis
The results of the variance change test which are detailed in Table 5 show that FEM and REM estimation methods experience variable variance, while the OLS method does not. However, the p-values of F and Chi 2 are smaller than 0.05 (Appendix 3), meaning that either FEM and REM models are more efficient than OLS or there is a heteroskedasticity phenomenon. Secondly, the Wooldridge test for autocorrelation in panel data is also performed (Appendix 4). The results obtained statistical values F of 37.304 and p-value of 0.0000 which is smaller than 0.05, meaning that there is an autocorrelation in the panel data. Finally, Breusch-Pagan LM, Pesaran scaled LM, Bias-corrected scaled LM, and Pesaran CD tests are applied with estimated p-values of less than 5% to ensure the existence of a cross-sectional dependence defect in most variables (Appendix 5). Based on the above test, a PCSE regression estimator is used in the panel data settings with the smaller time interval (T) and large cross-section (N), while an FGLS estimator is also executed to ensure the robustness of results. Table 6 shows hierarchical PCSE regression results. Firstly, the Lerner index has a positive effect on the bank stability index at statistically significant levels of 1%, 5%, and 10%. This indicates that the higher bank competition is, the higher bank stability is. Therefore, hypothesis (H1) is accepted.
Moderating role of shadow banking for the impact of bank competition on bank stability in Vietnam
The shadow banking variable is represented by two proxies, SB1 and SB2. The PCSE estimate indicates that SB1 positively affects BS at a significant 1%. Furthermore, the interaction variable (BC×SB1) was statistically significant at 1%, meaning a negative impact on BS. In particular, the more banks engage in shadow banking activities, the less BC impacts BS. In other words, shadow banking activities moderate the impact of bank competitiveness and bank stability, and thus hypothesis (H2) is accepted. As regards to SB2, there is no relationship with BS.
Furthermore, most of the variables related to bank characteristics affect bank stability. To be precise, SIZE and ETA positively impact BS at a statistically significant level of 1%, while the inverse influence can be seen in STO at a level of 1%. For factors of financial structure, a positive relationship between BSD and BS is found at a level of 1%, which is not the case in respect of SMD. In addition, there is a negative relationship between INF and BS, while there is no evidence of a relationship between GDP and BS.
To check the model robustness, Feasible Generalized Least Squares Regression (FGLS) is substituted to test the sensitivity of the generated results (Appendix 6). There are some differences in the results. Firstly, bank competition has a positive impact on bank stability at statistically significant levels of 1% and 5%, but there is no impact on the participation of interactive variables (BC×SB1). Shadow banking (SB1) still has a positive influence on bank stability at 5%, while shadow banking's interactive role in the competition-stability relationship has not been detected as PCSE estimates. Moreover, the group of bank characteristics retains the same dimension of impact as the main result. Meanwhile, banking sector development (BSD) has a positive influence on bank stability in model (1) with a statistically significant level of 10%. By contrast, the remaining controlled variables have no correlation with bank stability.
1.98
Notes: The sign * represents significance at 5% level of significance.
Source: Authors extracted information from Stata 17.
Discussions
The research results indicate that bank competition, shadow banking with SB1-proxy, bank size, equity to total assets, and banking sector development have a positive impact on bank stability, while impacts of state ownership and inflation on bank stability are negative. Furthermore, there is no evidence of a relationship between bank stability and shadow banking with SB2-proxy, stock market development, or GDP growth. In particular, shadow banking, which is measured as the ratio of loans to other credit institutions, payment on behalf of customers, contingent liabilities, and long-term investments to total assets, reduces the impact of bank competition on bank stability.
Firstly
This paper indicates that shadow banking reduces the bank competition-stability nexus. This result is completely consistent with the conclusion of Zhang et al. (2022) who argue that in the long run, unsupervised shadow banking poses relatively high risks in the investment and financing sector, causing losses for banks or damaging bank stability. In Vietnam, shadow banking activities, which rely heavily on trust between organizations, easily create moral hazard, weakening the financial stability of banks.
(i) Vodova (2011) argues that reliance on capital from the interbank market increases liquidity risk when banks have to borrow at high-interest rates, increasing debt ratios. Moreover, the fact that commercial banks borrow externally to meet the borrowing needs of customers can increase the debt-to-equity ratio, thereby affecting banks' efforts to maintain an optimal capital structure (Arif & Anees, 2012). In addition, Shen et al. (2020) argue that shadow banking carried out by commercial banks converts risky corporate loans into riskweighted interbank loans that carry much smaller risk weights in calculating capital ratios and liquidity measures, resulting in skewed bank ratings. In Vietnam, banks have close relationships with each other through transactions in the interbank market. Data from the Vietnamese banking system show that the figures for 2017 and 2021 are VND 231,438.73 billion and VND 279,212.28 billion, recording growth rates of 31.38% and 59.45%, respectively. The speed and scale showed signs of slowing down from 2018 to 2020 due to the impact of the COVID-19 epidemic. During this period, many businesses disappeared and the number of non-performing loans increased rapidly, causing credit and liquidity risks for banks and the whole banking sector.
(ii) Entrusted loans are viewed as bank off-balance sheet business or complex funding sources. In fact, since this fiduciary activity often goes through many parties and is recorded offbalance sheet, it is difficult to access information, leading to an asymmetry of information between lenders and borrowers (Zhang et al., 2023). Consequently, the wrong selections can be made, and moral hazard can easily occur. In Vietnam, accounts payable on behalf of customers experience a gradual growth over the period from 2018 to 2020 and a decrease in 2021. Starting at VND 371.88 billion in 2018, the figures increased to VND 782,345 billion in 2019 and VND 1186,175 billion in 2020, which resulted largely from losses of business during the COVID-19 pandemic. There were also some cases in which commercial banks, on behalf of their clients, had to fulfil obligations to a third party. For instance, AGRIBANK pays VND 38.5 billion on behalf of Cao Truong Son Co., Ltd for Industrial and Construction Equipment Joint Stock Company or SEABANK must fulfil the obligation to guarantee the payment of Vina Megastar bonds worth VND 150 billion to VVF financial company.
(iii) Data from 20 examined banks show that financial products or investment receivables are the largest share of shadow banking in commercial banks. To be precise, latent debt growth reached a peak of 43.35% in 2017, followed by a growth rate of 22.59% in 2021. Moreover, liabilities experienced a significant growth with the figure of VND 986,334.09 billion in 2021, compared to only VND 668,036.57 billion in 2017, leading to credit risks, where such credit risk from contingent liabilities has a negative effect on bank profitability (Aktan et al., 2013), damaging bank stability (Kashian & Tao, 2014).
Secondly
Bank competition increases bank stability, which is totally consistent with the market position theory proposed by Boyd and De Nicolo (2005) as well as the conclusions of Noman et al. (2017), Goetz (2018), and Islam et al. (2020). In the context of international economic integration, Vietnamese banks have had to deal with an increase in competition pressure from both domestic and foreign financial institutions, which has required them to implement various strategies not only to maintain their market share and power but also to get higher profits and improve their performance or financial stability (Fiordelisi & Mare, 2014).
Thirdly
The research result provides evidence of a positive impact of shadow banking on bank stability only when shadow banking is measured by the ratio of loans to other credit institutions, payment on behalf of customers, contingent liabilities, and long-term investments to total assets, which is consistent with findings of Ding et al. (2020), and Zhang et al. (2023). This finding also supports the Trade-off theory proposed by Rose (1989). In theory, shadow banking is a kind of portfolio diversification activity outside traditional banking activities. This is a clever strategy in which banks can cut loan rates in typical company operations to grow their market share thanks to noninterest income (Maudos & Solís, 2009). However, this result is totally different from the conclusions of Shen et al. (2020), Ding et al. (2020), and Bashir (2022), which argue that shadow banking activities significantly reduced bank stability.
However, the shadow banking scale of income from agency and brokerage services to total income of banks does not affect bank stability. This result can be originated from the fact that this research can not access detailed data about income from sales of corporate bonds specifically, and the activity of selling corporate bonds can have lagged impacts on bank stability. In fact, various problems related to corporate bonds have only appeared since 2022, while the study period is from 2016 to 2021, causing a decrease in confidence in the banking system in Vietnam.
Finally
All variables related to bank characteristics influence bank stability in different aspects. Firstly, bank size and equity to total assets have a positive effect on bank stability, which is totally consistent with the findings of Laeven et al. (2016), and Diaconu and Oanea (2014). Secondly, state ownership reduces bank stability, which is totally different from the conclusions of Bashir (2022). Quoc Trung and Abdul Wahab (2021) argues that in general, state-owned banks have lower profits and higher costs than private banks, due to asymmetric information, state-owned banks ineffectively control agency costs, affecting their operational efficiency and financial stability. Related to banking system factors, banking sector development has a positive impact on bank stability, which confirms the findings of Mirzaei et al. (2013) and Goetz (2018). However, there is no evidence of an impact of stock market development on bank stability, while Mirzaei et al. (2013) claim that a higher stock market development ratio indicates a more efficient capital market, in which banks may obtain perfect information about firms, reducing moral hazard and adverse selection risks, and engender greater bank stability. Furthermore, inflation has an inverse relationship with bank stability, while GDP growth has no effect on competitive stability, which does not support the findings of Goetz (2018) or Bashir (2022).
Conclusions and recommendations
By analyzing panel data of 20 commercial banks in Vietnam using a Hierarchical Multiple Regression Approach using PCSE regression, this paper shows a positive effect of bank competition on bank stability. It also shows that shadow banking plays a role in reducing this positive relation. To the best of our knowledge this is the first quantitative study on the moderating role of shadow banking in the relationship between bank competition and bank stability, and therefore this research potentially has important theoretical and practical contributions.
In terms of theory, this paper provides support to theories of competitive-banking stability such as the charter value theory (Keeley, 1990;Marcus, 1984), and the market position theory (Boyd & De Nicolo, 2005), by giving evidence on the negative impact of shadow banking on the bank competition-stability nexus in Vietnam. Furthermore, the study tries to build a bank stability index based on CAMELs ratings and Principal Component Analysis (PCA), as well as indicators measuring shadow banking in emerging countries like Vietnam.
As regards to practical aspects, this paper argues that although shadow banking is a type of non-traditional banking service that contributes to reducing competitive pressure among banks, it can still lead to serious risks or bank instability in Vietnam. Therefore, bank managers should be careful when developing non-traditional banking activities in general, and shadow banking services in particular. For authorities like the State Bank of Vietnam, it is necessary to strictly supervise these activities.
In spite of the above-mentioned contributions, this paper still has certain limitations. The first one stems from the proposed model, which does not account for the non-linear relationship between bank competition and bank stability. The second concern is that the endogenous problem of bank competition's impact on bank stability has not been addressed. Finally, the data is not as up to date as it could be, as the audited consolidated statements will not be released until April 2023. These gaps are expected to be filled in future studies.
|
2023-07-30T15:18:32.295Z
|
2023-07-28T00:00:00.000
|
{
"year": 2023,
"sha1": "18c39b36d9961ee198744e3bbf66c95804c32fb7",
"oa_license": "CCBY",
"oa_url": "https://www.tandfonline.com/doi/pdf/10.1080/23311975.2023.2241208?needAccess=true&role=button",
"oa_status": "GOLD",
"pdf_src": "TaylorAndFrancis",
"pdf_hash": "4bc767357a8123893260e9f2095f1637860f221e",
"s2fieldsofstudy": [
"Economics"
],
"extfieldsofstudy": []
}
|
202537703
|
pes2o/s2orc
|
v3-fos-license
|
An application of medial limits to iterative functional equations
Assume that $(\Omega,\mathcal A,P)$ is a probability space, $f\colon[0,1] \times \Omega\to[0,1]$ is a function such that $f(0,\omega)=0$, $f(1,\omega)=1$ for every $\omega\in\Omega$, $g\colon[0,1]\to\mathbb R$ is a bounded function such that $g(0)=g(1)=0$, and $a,b\in\mathbb R$. Applying medial limits we describe bounded solutions $\varphi\colon[0,1] \to \mathbb R$ of the equation \begin{equation*} \varphi(x) = \int_\Omega \varphi(f(x,\omega)) dP(\omega)+g(x) \end{equation*} satisfying the boundary conditions $\varphi(0)=a$ and $\varphi(1)=b$.
We say that a function ϕ : [0, 1] → R is a solution of equation (E g ) if for every x ∈ [0, 1] the function ϕ • f (x, ·) is measurable and (E g ) holds.
The main purpose of this paper is to describe all solutions of equation (E g ) in some classes of bounded functions h : [0, 1] → R such that h(0) = a and h(1) = b. We are also interested under which assumptions any bounded solution ϕ : [0, 1] → R of equation (E g ) with a certain property can be expressed in the form ϕ = Φ + ϕ * , where Φ is a solution of the equation having the same property as ϕ and ϕ * is a specific solution of equation (E g ). This problem seems to be easy to answer, but the difficulty is that the classes considered in this paper are not linear spaces. It is even not clear when the existence of a solution with a certain property of one of the equations (E g ) and (E 0 ) implies the existence of a solution with the same property of the other of these equations. Such a problem is quite natural in the theory of functional equations and it has been studied several times by many authors for different functional equations in various classes of functions; mainly in cases where the class of considered functions forms a vector space. Functional equations (E 0 ) and (E g ), as well as their generalizations and special cases, are investigated in various classes of functions in connection with their appearance in miscellaneous fields of science (for more details see [18,Chapter XIII], [19,Chapters 6,7] and [4,Section 4]). As emphasized in [19, section 0.3] iteration is the fundamental technique for solving functional equations in a single variable, and iterates usually appear in the formulae for solutions. In most cases such formulas are obtained by taking the limit of sequences in which iterates are involved. In this paper we make use of this fundamental technique, but the goal is to apply a subclass of Banach limits instead of the limit. The idea of replacing the limit by a Banach limit seems to be clear, because we do not need any additional assumption guaranteeing the existence of a Banach limit of a bounded sequence, in contrast to the case when we want to calculate the limit of such a sequence.
This paper is organized as follows. Section 2 contains the notation and basic tools required for our considerations. In sections 3 and 4 we describe bounded solutions ϕ : [0, 1] → R with ϕ(0) = a and ϕ(1) = b of equations (E 0 ) and (E g ), respectively. Finally, in section 5, we formulate some consequences of the main results obtained and we present a few examples of the possible applications of those results.
Preliminaries
Denote by B( Note that T is linear and continuous with T = 1. Moreover, equation (E g ) can be written now in the form To the end of this paper we fix a subspace B( To describe solutions of equation (E g ) in the case where A = 2 Ω we need the concept of Banach limits, established in [1]. However in the general case, when integration is required, we need the concept of medial limits, established in [25] (cf. [24]) as a very special class of Banach limits.
Denote by l ∞ (N) the space of all bounded real sequences equipped with the supremum norm and by B the family of all Banach limits defined on l ∞ (N). Recall that B ∈ B if B : l ∞ (N) → R is a linear, positive, shift invariant and normalized operator. It is easy to see that any B ∈ B is continuous with B = 1. It is known that the cardinality of B is equal to 2 c (see [14]), and even that the cardinality of the set of all extreme points of B is equal to 2 c (see [11], cf. [29]); here c is the cardinality of the continuum.
As it was mentioned above, in the general case we need to integrate the pointwise Banach limit of a bounded sequence of measurable functions. However, the problem is that there is no guarantee that the pointwise Banach limit of a bounded sequence of measurable functions is a measurable function (see [31, page 288]). Fortunately, it is known that there are Banach limits possessing exactly the required property. More precisely, a Banach limit B is called a medial limit if Ω B((h m (ω)) m∈N )dP (ω) is defined and equal to B(( Ω h m (ω)dP (ω)) m∈N ) whenever (h m ) m∈N is a bounded sequence of measurable real-valued functions on Ω. It is also known that the continuum hypothesis implies the existence of medial limits. More results on the existence and non-existence of medial limits can be found in [15,Chapter 53] and in [20].
It is also clear that if we determine all solutions of equation (E g ) in the class B b a , then we can easily describe all solutions of this equation in the class B([0, 1], R). Now we are in a position to begin describing solutions of equation (E g ) in the class B b a . Our first lemma is a simple consequence of (1) and (3).
The functions B h plays a crucial role in this section as well as in this paper. So, we need some fact about them.
a . Applying properties of medial limits we obtain for every x ∈ [0, 1].
We now want to find conditions under which
This observation suggests the following definition.
We say that the class To prove the conclusion let us put B([0, 1], R) = C x0 ([0, 1], R); this is possibly according to Example 2.5, because (H 3 ) yields the continuity of f (·, ω) at x 0 for every ω ∈ Ω. Let B ∈ B and let x ∈ A, and by an easy induction, we obtain sup Finally, (H 1 ) jointly with properties of Banach limits implies that both the functions B h1 and B h2 are increasing.
We are now in a position to formulate the main results of this section. To simplify their statements, let us denote by sol b a (E 0 ) the family of all functions from B b a satisfying equation (E 0 ).
The opposite inclusion follows from Lemma 3.1.
Solutions of equations (E 0 ) was investigated in [26,27], basically in almost the same classes of bounded functions. However, Theorem 3.2 is incomparable with the results obtained in the papers mentioned, in which the existence and the uniqueness problems have been considered as well as properties of the unique solution have been studied.
Solutions of equation (E g )
In this section we describe all functions belonging to the class B b a which are solutions of equation (E g ). We also give the formula for these solutions showing that each of them can be written in the form Φ + ϕ * , where Φ ∈ B b a is a solution of equation (E 0 ) and ϕ * ∈ B 0 0 is a particular solution of equation (E g ). To find ϕ * we need define a certain family of functions generated by g ∈ B 0 0 ; recall that g ∈ B 0 0 is a necessary condition for equation (E g ) to have a solution in the class B b a by Lemma 2.1(ii). If g ∈ B 0 0 , then Lemma 2.1(i) yields {T l g : l ∈ N} ⊂ B 0 0 . Therefore, given g ∈ B 0 0 and k ∈ N we can define a function g k : [0, 1] → R by putting As in the previous section, denote by sol b a (E g ) the family of all functions from B b a satisfying equation (E g ). Proof. Fix ϕ ∈ sol b a (E g ). Then Lemma 2.1 implies that G ⊂ B 0 0 . Applying (2) we obtain The above lemma shows that boundedness of the family G is a necessary condition for equation (E g ) to have a solution in the class B b a . This also demonstrate, that B g k is well defined for all k ∈ N and B ∈ B whenever equation for every x ∈ [0, 1]. Now, it only remains to see that B g k = kB g .
If g ∈ B 0 0 and G is bounded, then for every B ∈ B we define a function B * : [0, 1] → R by putting for every x ∈ [0, 1].
We now want to find conditions under which B * ∈ B 0 0 . The situation is similar to that for B h ∈ B b a . Namely, to prove that B * ∈ B 0 0 , we would have to show that T B * ∈ B 0 0 , but by Lemma 4.3 we have T B * = B * − g. This leads us to the following definition.
We say that a function g ∈ B 0 0 is admissible for B ∈ B, if the family G is bounded and B * ∈ B 0 0 . Note that the assumption on boundedness of G in the admissibility definition is not restrictive, because if the family G is unbounded, then B * can not be a solution of equation (E g ) by Lemma 4.1.
Before we give examples of conditions guaranteeing admissibility of a given function under a Banach limit, let us recall the definition of almost convergence of sequences. Namely, a bounded sequence (x m ) m∈N of real numbers is said to be almost convergent to a real number x if B((x m ) m∈N ) = x for any B ∈ B. The sequence (0, 1, 0, 1, 0, 1, . . .) is a simple example of a non-convergent sequence which is almost convergent. However almost none of the sequences consisting of 0's and 1's are almost convergent (see [12]). It is proved in [22] that a sequence (x k ) k∈N is almost convergent to x if and only if lim n→∞ Let us note that condition (5) is not very far from a necessary condition for g derived in Lemma 4.2, which says that B (T m g(x)) m∈N = 0 for all x ∈ [0, 1] and B ∈ B.
We now formulate the main result of this paper.
Theorem 4.4.
(i) Assume that a (E g ) and B ∈ B. Obviously, B ϕ is well defined. From Lemma 4.1 we conclude that B * is also well defined. Applying induction to (2) we get (7) ϕ = T k ϕ + g k for every k ∈ N, and hence ϕ( a , and making use of (7) we obtain that sup k∈N g k ≤ 2 ϕ and B * = ϕ − B ϕ ∈ B 0 0 .
a . Then Lemmas 3.1, 4.1 and 4.3 give which means that B h + B * ∈ sol b a (E g ). (ii) It suffices to apply assertion (i). (iii) Fix ϕ ∈ sol b a (E g ). Lemma 4.3 jointly with the admissibility of g implies that a (E 0 ). Then again Lemma 4.3 jointly with the admissibility of g implies that Φ + B * ∈ sol b a (E g ). a (E g ) = ∅, then it may happen that there is no B ∈ B for which B * is well defined; see e.g. the equation ϕ(x) = ϕ(x)+ 1. Therefore, assumption (6) can not be omitted in assertion (i) of Theorem 4.4. The above exemplary equation also shows that the admissibility assumption in assertion (iii) of Theorem 4.4 is necessary.
Consequence of the main results
In this section we formulate some exemplary consequences of the main results, making use of the presented examples and applying some know results on equation (E g ). We begin with the case where A = 2 Ω . p n ϕ(f n (x)) + g(x), which is discussed in more details in [18,Chapter XIII] and in [19,Subsections 6.3 and 6.7]).
Purely bounded solutions of equation (E g ) are considered rather rarely. Usually some additional property is requited, such as monotonicity (see e.g. [16,17,28]), Borel measurability (see e.g. [2,6]), continuity at a point (see e.g. [5]). The next two corollaries concern just such cases. To formulate the first one we need some notion. Namely, following [8] (cf. [13]) we define iterates of a function h : for all x ∈ [0, 1], ω = (ω 1 , ω 2 , . . .) ∈ Ω ∞ and n ∈ N. Note that if h is an rvfunction, then all its iterates are also rv-functions defined on the product space (Ω ∞ , A ∞ , P ∞ ). The next example is in the spirit of the idea of the manuscript [30] with the use of Corollary 5.3.
Lipschitzian solutions of equation (E g ), in a more general setting than in this paper, were recently examined in [3,7,9,10]. However, the next Corollary gives a general formulae for a wide class of Lipschitzian solutions of equation (E g ), in contrast to the papers mentioned, in which assumptions made force uniqueness or uniqueness up to an additive constant of Lipschitzian solutions of the equation considered. The next corollary gives a formulae for the general solution of equation (E g ) in the space BV ([0, 1], R), and hence, partially solves the problem considered in [23] for a very spacial case of equation (E 0 ). Before we formulate the last corollary of this paper let us to extend the main result of [23] to equation (E 0 ). Proposition 5.6. Assume (H 1 ). If Φ ∈ BV ([0, 1], R) satisfies (E 0 ), then also Φ + and Φ − satisfy (E 0 ).
|
2019-09-05T07:38:31.000Z
|
2019-09-05T00:00:00.000
|
{
"year": 2019,
"sha1": "8f48571cca29b5a1c14cafacb7b58b752fee5f1e",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s00025-020-01229-w.pdf",
"oa_status": "HYBRID",
"pdf_src": "Arxiv",
"pdf_hash": "8f48571cca29b5a1c14cafacb7b58b752fee5f1e",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
245117285
|
pes2o/s2orc
|
v3-fos-license
|
Maximal Super-Yang-Mills at Six Loops via Novel Integrand Bootstrap
We construct the complete (planar and non-planar) integrand for the six-loop four-point amplitude in maximal $D\le10$ super-Yang-Mills. This construction employs new advances that combat the proliferation of diagram contributions and state sums when evaluating multi-loop $D$-dimensional unitarity cuts. Concretely, we introduce two graph-based approaches to evaluating generalized unitarity cuts in $D$ dimensions: 1) recursively from lower-loop cuts, or 2) directly from known higher-loop planar cuts. Neither method relies on explicit state sums or any sewing of tree-level amplitudes. These methods are based on identities that we expect to hold for a broad family of theories, including QCD and Einstein gravity.
INTRODUCTION
Gauge theory and general relativity are centerpieces in the theoretical framework that describes modern physics. Tremendous theoretical and experimental effort has gone into improving our understanding of non-abelian gauge theory. In comparison, much less is known about gravity, at the quantum level. A primary reason for this disparity is the lack of experiments that probe relevant energy scales. A secondary reason is the technical challenge of working with the highly non-linear behavior of gravity. General relativity, as a quantum theory, is famously known to be both non-renormalizable and ultraviolet (UV) divergent [1], and the question of its UVcompletion is an all-important open problem.
A longstanding well-behaved theory, that extends general relativity with a finite number of fields without higher-derivative interactions, is the maximally supersymmetric N = 8 supergravity, discovered by Cremmer, Julia and Scherk [2]. It is expected to be UV finite in four dimensions up to at least six loops, with the first counterterm compatible with known symmetries appearing at seven loops [3][4][5][6][7][8][9][10][11][12][13][14][15][16]. In the absence of a proof of finiteness, the only available means of showing the presence or absence of this divergence is by an explicit seven-loop calculation.
Progress in maximal supergravity hinges directly on developments in N = 4 super-Yang-Mills (SYM) theory, first written down by Brink, Schwarz and Scherk [17]. The theories are related by the double copy at the classical [18,19] and quantum level [20], Consider the asymptotic spectra: N = 4 SYM has 8 bosonic plus 8 fermionic states, and the N = 8 supergravity states are the square of these (16) 2 = 256. The double copy, in combination with the unitarity method [21,22], has been successfully used to determine the N = 8 supergravity amplitudes through five loops [23][24][25][26][27][28][29]. However, the progress relied, in part, on the powerful constraints implied by color-kinematics duality [19,20], which obtains the complete integrand from only a small subset of all diagrams and cuts [20,26,30].
A generalized double copy [27,28] was introduced to ameliorate challenges with manifesting color-kinematics duality at five loops, and while it was sufficiently powerful to complete the calculation [29], new methods and refinements are clearly needed to tackle the six-and seven-loop calculations. At high loop orders, apparently mundane tasks can become impenetrable walls, such as evaluating unitarity cuts. In N = 4 SYM, one faces the problem of exponentially growing state sums with (16) L terms at L loops. Alternatively, state sums can be handled covariantly in ten-dimensional notation, but the fermionic Dirac traces resist evaluation; for L = 6, one expects traces of up to 28 gamma matrices. Additionally, the approach obscures the enormous cancellations due to supersymmetry.
In this Letter, we introduce two new and complementary bootstrapping methods capable of overcoming obstacles encountered in ascending towards seven loops. The methods circumvent the need for laborious state sums when evaluating multi-loop unitarity cuts, and they manifest symmetries (supersymmetry, Ddimensional Poincaré symmetry) that may otherwise rely on non-trivial cancellations. The general idea is to relate unknown cuts to known simpler cuts, and recursively bootstrap the integrand from tree level. The two approaches indirectly exploit that gauge-theory amplitudes can be stratified in two complementary ways: the loop and 't Hooft expansions [31].
The second relationship, called the H identity, similarly relates a degeneration limit of the factorized four-point amplitude to two disconnected planar lines. Applied to a multi-loop cut, the H-identity allows for the bootstrap of cuts in terms of known lower-loop cuts, while maintaining the planarity properties (crossing count) of graphs.
As a demonstration of the usefulness of the new bootstrapping techniques, we apply them together with the method of maximal cuts [24,26,29,33,46], to construct the complete six-loop four-point D-dimensional integrand in N = 4 SYM. This constitutes the first nonplanar computation at six loops in N = 4 SYM, thus extending the planar six-loops results [35,47,48]. We anticipate the new methods introduced here to be highly useful in many other theories relevant to modern physics, including QCD and Einstein gravity.
DERIVING THE X-ING AND H-ING MOVES
X identity: Take 1 a , 2 b |S|3 c , 4 d to be a tree-level Smatrix element in some Yang-Mills gauge theory, where the states are distinguished by their momenta p 1 + p 2 = p 3 + p 4 as well as Lorentz (little group) representations, which we indicate using generalized indices a, b, c, d. The particles also transform in representations of the gauge group, but the details are not needed for this argument. We only need to assume that the S-matrix elements can be expanded in terms of partial amplitudes that have a notion of planarity, analogous to purely-adjoint theories.
Thus for our purposes, it is convenient to strip off most factors (color, coupling, momentum-conserving delta functions, phases) and consider a planar tree-level amplitude A(1 a , 2 b |3 c , 4 d ). We note that this amplitude, which is a dimensionless function that is covariant in the generalized indices, must have a simple behavior in the kinematic limit p 3 → p 1 , which has but one scale, s 12 = (p 1 + p 2 ) 2 . In a suitably chosen normalization, we expect it to either evaluate to 0 or 1. It is not difficult to see in explicit examples that the correct answer is where δ ab simply identifies the states on diagonally opposite legs. We will call this equation the X identity, and it is graphically represented in Fig. 1. Let us illuminate eq. (2) by considering D-dimensional pure SYM theory, described by the adjoint fields {A µ , ψ} and the Lagrangian Tr[− 1 4 (F µν ) 2 + i 2ψ / Dψ]. In the given limit, the non-vanishing partial amplitudes are where ε µ and χ are dimensionless polarizations of respective states (for convenience, the fermion wavefunctions are normalized asχχ = 1 here). All other amplitudes either vanish, or are trivially related to the above ones by permutations of legs, in accordance with eq. (2). The simple result for the four-fermion amplitude relies on the same Fiertz identity that is responsible for supersymmetry in the theories under consideration. In our case, we have D = 3, 4, 6, 10 as valid dimensions where the X identity holds in term of the SYM fields {A µ , ψ}. Dimensional compactifications of these theories will also obey eq. (2), which will introduce scalars and possibly mass terms in the SYM spectrum. In D = 4, it follows that the N = 0, 1, 2, 4 SYM theories obey the X identity (with possible mass terms compatible with spontaneous symmetry breaking or dimensional compactification). The X identity (2), simple as it is, has profound consequences. It is sufficient to take a kinematic limit in order for the planar tree amplitude to collapse into a non-planar identity operator. The fact that this can hold for arbitrary states in large families of gauge theories presages powerful promise for loop calculation.
Assume that we know the planar L-loop n-point amplitude integrand in a gauge theory. Using the X identity, we can directly compute non-planar n-point cuts with more edge crossings, but lower loops via taking appropriate kinematic limits on planar cuts with quartic vertices. Using g to refer to the number of non-planar crossings, we can write a higher-loop cut for a graph γ X that contains an isolated quartic interaction as where the four-point tree is factored out, repeated indices are summed over, and external states are not shown. Applying the X identity (2) on this cut yields where in the final step a generic cut is obtained, with two on-shell loop legs sewn in a non-planar fashion. This final sewing lowers the loop order at the cost of introducing a new crossing. (Equivalent relations apply if some of the legs of the four-point sub-amplitude are external, but the details of the crossing and loop counting may differ.) To reach an L-loop cut with g edge crossings, we simply start with an (L + g)-loop cut with g fourpoint sub-amplitudes and apply the X-identity to each of those sub-amplitudes. In this way, any physical cut in a gauge theory can be straightforwardly computed from the planar cuts, assuming they are known.
H identity: Since the loop-level X identity requires knowledge of higher loop (planar) integrands, there is a non-trivial threshold to pass before using it. However, we now consider a similar identity, derivable from the X identity, that admits a bootstrap of integrands from lower loops without necessarily adding non-planar crossings. At tree-level the H identity takes the form where, as before, repeated indices are summed over, and momentum is conserved for each three-point amplitude (p 1 −p 3 = k = p 4 −p 2 ). The diagrammatic interpretation of the H identity is given in the right-side pannel of Fig. 2.
The left-side pannel of the same figure indicates how it is derived from the X identity. It is obtained from applying the X identity to a BCJ relation [19] lim p3→p1 s 13 A tree 4 (1, 2, 4, 3) = s 23 A tree 4 (1, 2, 3, 4) . (7) Here the left hand side evaluates to the residue on the s 13 = (p 1 + p 3 ) 2 pole, yielding the factorized three-point amplitudes, and using eq. (2) the right-hand-side amplitude becomes the product of delta functions. The H identity is straightforward to apply to any cut that has a connected pair of three-point sub-amplitudes yielding a constraint equation satisfied by the cut where C L−1 γH\ℓ5 is a lower-loop cut that is assumed to be known. While the H identity does not calculate the cut C L γH for general momenta, the full set of such conditions is often powerful enough for a complete determination, as will be discussed below for the six-loop N = 4 SYM calculation. One might also worry that the set of cuts with two connected tree-point sub-amplitudes is too special. However, in the maximal cut method most of the needed cuts are of this form, and for N = 4 SYM these should be sufficient to determine the full amplitude.
Note that while the H identity is reminiscent of the rung rule of planar N = 4 SYM [49,50], here we stress that the H identity is not a heuristic rule, and it applies both to non-planar amplitudes, and a multitude of theories (including those theories mentioned above for the X identity). Its generality can be traced back to the universality of soft factorization [51,52], which is a general feature of gauge [53,54] and gravity theories [55] (see also [56,57]). Indeed, the H identity has a direct extension to a multitude of gravitational theories, simply by replacing s 12 → s 2 12 in eq. (6).
SIX LOOP INTEGRAND CONSTRUCTION
With the X and H identities established, we set out to demonstrate their usefulness by constructing the complete six-loop four-point integrand in N = 4 SYM. To assemble the integrand from cuts, we follow the generalized unitarity method [21,22,[58][59][60] in the refined form known as the method of maximal cuts [24,26,29,33,46].
The construction can be parsed into three steps. The first step is to enumerate all candidate diagrams entering the six-loop integrand, and grade them by the cut level k. Cut level zero corresponds to the maximal-cut diagrams, obtained by attaching four external legs to the edges of all cubic six-loop vacuum diagrams, which are known [61]. By the N = 4 SYM "no-triangle rule" [62], we exclude one-loop triangles, bubbles, or tadpoles subdiagrams, as their maximal cuts are zero. Cut level k diagrams are then obtained by contracting an internal edge (by merging the edge's vertices) in the cut level (k−1) diagrams, in all possible ways, modding out by graph isomorphisms. (If an edge only connects to one vertex, i.e. tadpole, contraction is avoided.) This recursively constructs the (next-to) k -maximal-cut (N k MC) diagrams, and a typical such graph is below called γ (k) . The final counts are given in Table I, first row. [26,30,46]; (c) require non-vanishing numerator contribution.
The second step is to assign physical expressions to each diagram, specifically two non-trivial expressions for each graph: a diagram's cut C γ (k) is a rational and gauge invariant (unique) function that corresponds to a physical factorization process in the theory; a diagram's numerator n γ (k) is a non-unique (gauge dependent) polynomial that describes local interactions similar to numerators of Feynman diagrams. The unique C γ (k) are determined by the X and H identities. From the numerators, the amplitude is where S γ (k) and d γ (k) are a graph's symmetry factor and product of propagator denominators, respectively. K is a crossing-symmetric kinematic factor that captures the external state dependence, K = s 12 s 23 A tree SYM (1, 2, 3, 4). Because the first sum runs over both cubic (k = 0) and contact (k > 0) diagrams, it is convenient to make the numerators n γ (k) functions of both kinematic and color data, hence we use no separate color factors. Similar to previous constructions of the L-loop four-point amplitude in N = 4 SYM, the cut level k = L−2 is the last one that contains new data, hence we anticipate no contributing diagrams beyond k > 4 (also confirmed by our cuts).
In the third step, we construct the n γ (k) by appropriately matching to the unique C γ (k) , on the kinematic support where every factor in d γ (k) vanish, denoted by d γ (0) → 0. For maximal cuts, the matching is simply The matching procedure on a N k MC diagram γ (k) can be expressed as (with d γ (k) → 0 implicit from now on) where C γ (k) is the unique cut we need to match, and R γ (k) is a rational function obtained by summing over previously-determined lower-k diagrams (numerator over non-vanishing denominator factors) that shares the same poles as C γ (k) . By construction R γ (k) matches all lower-k cuts, hence P γ (k) must be a local polynomial. We would like to identify P γ (k) with the numerator n γ (k) ; however, first we must promote P γ (k) to an off-shell quantity respecting the automorphism symmetry of the graph γ (k) (recall that S γ (k) equals the order of the automorphism group). This can be done in several ways, such as explicit symmetrization, or using an Ansatz. Here we use an approach that directly identifies special combinations of generalized Mandelstam variables that are automorphism invariants [63]. Hence we can promote P γ (k) → n γ (k) . Let us return to the new identities. The X identity directly supplies C γ (k) expressions from higher-loop planar integrands, currently known explicitly up to L = 10 [36]. However, in the following we use the H identity, and show that it alone is sufficiently powerful to determine the complete six-loop integrand. Recall that the H identity does not directly calculate C γ (k) , but rather provide a set of constraints, where C γ (k) \ℓm is a known lower -loop cut with edge ℓ m removed, and its prefactor 2ℓ i · ℓ j corresponds to the momenta of the two edges that ℓ m was attached to. Consider a pictorial example: with a slight shuffe, eq. (13) becomes where the pictures are representative 6-and 5-loop diagrams producing the functions P γ (k) +R γ (k) and C γ (k) \ℓm , respectively. The edges labeled by i, j are on-shell, and the momentum of the dashed line vanishes, ℓ m = 0. Repeatedly applying this procedure to every edge of γ (k) provides all of the information required to uniquely determine P γ (k) . Note that, if the lower-loop cut happens to vanish, eq. (13) still provides valuable information, in a similar vein to how vanishing cuts were used to triangulate the Amplituhedron in Ref. [43].
In the six-loop construction, only mild additional power-counting assumptions on the integrand are required to fully constrain the numerators. For each candidate diagram we use eq. (13) to construct a P γ (k) (n γ (k) off shell) that satisfies: 1) every monomial term is proportional to either s 12 or s 23 ; 2) if there are any terms left unconstrained that violate the no-triangle power counting, we safely drop them. Condition 1 holds for all lowerloop integrands [20,26,28] as well as in the planar sixloop amplitude [35]. Condition 2 is needed because there are instances of particularly simple ladder-type diagrams that are not uniquely constrained by the H identity, but subdiagram power counting rules out such freedom. We defer an in-depth discussion of the implementation of the method and details to upcoming work [64]. Both assumptions are validated by the consistency checks of the final integrand. The third row of Table I provides the final count of non-zero diagrams that contribute to the integrand. The complete N = 4 SYM six-loop integrand, with explicit numerators for all diagram topologies, is provided in Ref. [65], including additional user-manual details.
By construction, the maximal cut method gives an integrand that satisfies all independent unitarity cuts. However, cross-checks are always desirable. We have subjected our six-loop integrand to two independent consistency checks. The first is verifying that it correctly produces all so-called generalized box cuts [26,30,46]. This class of cuts targets graphs with lower-loop (6 > L ≥ 1) four-point sub-diagrams, which are factorized into the product of two known four-point cuts, one at L loops and one at 6−L loops. Cuts verified with this method constitute a large proportion of each cut level k, as shown in the second row of Table I. Second, we have verified that all contact-contributing non-multigraph N 4 MC are con-sistent with the n-point BCJ amplitude relations [19], as applied to sub-amplitudes within the cuts. This second check is a highly nontrivial verification that no diagrams or terms were lost in each cut, as all contributions must delicately conspire between different k ≤ 4 diagrams to satisfy the relations.
CONCLUSIONS
In this Letter, we have presented two new identities which have extensive applications in the recursive calculation of unitarity cuts in many theories (including QCD), avoiding laborious state sums and manifesting desirable symmetries. We demonstrated the power of the H identity by using it alone to compute the fourpoint six-loop D-dimensional integrand in N = 4 SYM, providing the first non-planar result at this loop order (see refs. [35,47,48] for planar results). Maximal SYM has a rich integrand structure both in the planar [36,44,45,47,[66][67][68] and non-planar [69][70][71] cases, with analogous multiloop string perspectives [72][73][74][75][76][77], and we provide here an ample source for further studies [65].
The current result constitutes major progress towards determining the UV behavior of N = 8 supergravity at seven loops. However, before getting there further sixloop studies are motivated. First, the current understanding of (generalized) double-copy construction starts from a cubic integrand representation [27]. Second, previous studies [24][25][26][27][28][29] have demonstrated the significant simplification of gravity calculations that comes from using a gauge theory integrand that manifests the expected sub-diagram power counting. Thus, the current integrand can be improved upon by efficiently constructing a cubic representation, perhaps with improved manifest sub-diagram power counting.
The six-loop integrand provides an essential stepping stone towards the corresponding seven-loop SYM calculation, using the new recursive methods presented in this Letter. The combinatorial growth of complexity when reaching seven loops will demand significantly streamlined tools, and we anticipate that the new identities will play a key role.
|
2021-12-13T02:15:20.550Z
|
2021-12-09T00:00:00.000
|
{
"year": 2021,
"sha1": "30258e21e6c9d89f1b11af488e9861c9f985ee22",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "30258e21e6c9d89f1b11af488e9861c9f985ee22",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
25094534
|
pes2o/s2orc
|
v3-fos-license
|
Applying a potential across a biomembrane: electrostatic contribution to the bending rigidity and membrane instability
We investigate the effect on biomembrane mechanical properties due to the presence an external potential for a non-conductive non-compressible membrane surrounded by different electrolytes. By solving the Debye-Huckel and Laplace equations for the electrostatic potential and using the relevant stress-tensor we find: in (1.) the small screening length limit, where the Debye screening length is smaller than the distance between the electrodes, the screening certifies that all electrostatic interactions are short-range and the major effect of the applied potential is to decrease the membrane tension and increase the bending rigidity; explicit expressions for electrostatic contribution to the tension and bending rigidity are derived as a function of the applied potential, the Debye screening lengths and the dielectric constants of the membrane and the solvents. For sufficiently large voltages the negative contribution to the tension is expected to cause a membrane stretching instability. For (2.) the dielectric limit, i.e. no salt (and small wavevectors compared to the distance between the electrodes), when the dielectric constant on the two sides are different the applied potential induces an effective (unscreened) membrane charge density, whose long-range interaction is expected to lead to a membrane undulation instability.
I. INTRODUCTION
Biomembranes are thin fluid films composed mostly of lipids. In cells they help separating different cellular environments and compartments. Biomembranes are typically "soft", i.e., the typical energy required to bend them is of the order the thermal energy and membrane tension is often quite small. Softness implies that membrane geometry can become sensitive to different perturbations, such as alteration of the electrostatic configuration. Much effort has for instance been devoted to calculation of the electrostatic contribution to tension and bending rigidity for membranes with fixed charges or surface potentials in an electrolyte solution, see [1] for a review; in general the presence of fixed (and screened due to the electrolyte) charges tend to increase bending rigidity and hence make the membrane stiffer. However, it has also been found that when the membrane charges are not fixed but free to rearrange themselves on the surface and no electrolyte solution is present to screen the interaction, a long-wavelength undulation instability can occur [2], somewhat similar to DNA condensation [3]. Electric fields can also be present across intrinsically neutral membranes. An example would be a nerve cell, where ion pumps create a potential difference between the two sides of the nerve cell membrane [4]. Another example is provided in laboratories by the routine formation of liposomes in a process known as electroformation [5], during which lipid membranes are swelling from electrodes under the application of electric fields.
In this study we investigate the electromechanical coupling of a membrane and an applied potential. In particular, we solve the Debye-Hückel and Laplace equations for the electrostatic potential for a non-conductive, incompressible membrane between two flat electrodes (kept at fixed potentials). On either side the membrane is surrounded by different electrolyte solutions. From the solutions for the potential we quantify how the corresponding induced membrane charges change the free energy for the membrane and identify electrostatic contributions to membrane mechanical parameters. In the presence of an electrolyte (in the small screening length limit, see below) we find that the electrostatic contribution to membrane bending rigidity is positive. In the absence of added salt (the dielectric limit) the membrane becomes unstable against long wavelength undulations (in a somewhat similar fashion to the behaviour of the interface between two immiscible fluids, see Ref. [6]) if the two fluids surrounding the membrane have different dielectric constants. For the symmetric dielectric case, as well as for the small screening length limit, the membrane tension receives a negative contribution; for a sufficiently large applied potential this contribution would lead to a membrane stretching instability.
Several studies (see [1] and references therein) have investigated the electrostatic contributions to tension and bending rigidity for membranes having a fixed surface charge density or fixed potential at the membrane. However, less work has been dedicated to the effect of induced charges due to an applied potential, as considered in the present study. The results found here complement previous results given in [7] (using coupled hydrodynamicalelectric field equations, similar to [8])) for the symmetric case of a membrane surrounded by identical electrolytes (in the small screening length limit), to the asymmetric case and by giving an explicit expression for the bending rigidity; in the limit of identical electrolytes on the two sides of the membrane our expression for the tension agrees with that derived in [7]. Also, our formulation allows us to investigate the dielectric (no salt) limit, which was not considered in [7]. However, unlike [7] we do not consider any dynamical effects. In contrast to the study [9] based on electrolyte conductivities, our approach through the Debye-Hückel equation allow us to study the effect of a finite Debye screening length. In Ref. [9] a non-zero membrane conductivity was considered, whereas we consider non-conductive membranes. Similar to the results in [9] we find a negative contribution to the membrane tension in the presence of an electrolyte, although the two results are difficult to compare because of the different mathematical formulations.
This work is organized as follows: In Sec. II we give the general equations governing the electrostatic response of a non-conductive membrane of any shape; the membrane region is described by the Laplace equation and the electrolyte solutions on either side satisfy the Debye-Hückel equations. In a standard fashion, the boundary conditions are that the potential and the displacement fields should be continuous. In Sec. III these equations are solved for the case of a flat membrane in an external potential. In Sec. IV corrections to the flat case solutions are derived for a weakly curved incompressible membrane. In Sec. V the forces acting on the membrane, as well as the corresponding electrostatic contribution to the membrane free energy, are obtained. Assuming that membrane fluctuations occur on a time scale slower than the relaxation time for the electrostatic potential we use the expressions for the forces for the weakly curved membrane in order to obtain the renormalized membrane mechanical parameters, such as tension and bending rigidity, (in terms of a power series expansion in the wavevector) as a function of the applied potential, the salt concentrations (entering through the Debye screening lengths) and the dielectric constants of the membrane and the solvents. We investigate three different limits: (1.) the small screening length limit, where the Debye screening length is smaller than the distance between the electrodes; (2.) the dielectric limit, i.e., no salt; (3.) the symmetric case, where the salt concentration and dielectric constants on the two sides of membrane are equal. The results for the membrane mechanical parameters in the three limits above (the main results in this study) are given in Eqs. (41)-(51). Finally, in Sec. VI a summary and discussion are given.
II. GENERAL FORMULATION
We are interested in how biomembrane mechanical parameters (and thereby, for instance, membrane fluctuations) are effected by an applied potential. Two parameters characterizing a membrane in the absence of an applied potential are the tension σ and the bending stiffness K [10,11]. As an external potential is applied there will in general be electrostatic contributions (σ el and K el ) to both of these quantities so that σ → σ + σ el and K → K + K el in the presence of the applied potential. An aim of this paper is to calculate σ el and K el . In a standard fashion we consider a small perturbation from a flat membrane, characterized by a height undulation h(x, y) (where x and y are coordinates in the plane of the flat membrane) and solve the electrostatic equations (via a Fourier-transformation in the x-and y coordinates) for this weakly perturbed geometry. Through a power series expansion in wavevector q (q = q 2 x + q 2 y , where q x and q y are the Fourier-transform variables of x and y respectively) of the free energy G one may identify σ el and K el (see chapter 2 in Ref. [11]). We find it convenient to, rather than utilize the free energy directly, consider the electrostatic contribution to the "restoring" force, from which we identify σ el (K el ) as the prefactor in front of the −q 2h (−q 4h ) term in a small q-expansion of the force, whereh =h(q x , q y ) is the Fourier-transform of h(x, y). Notice that for obtaining the tension and bending rigidity it suffices to keep terms linear inh in the restoring force expression. In general there may be other terms in the power series expansion in q. For instance, as noted in the introduction, for the asymmetric dielectric case there is a membrane undulation instability which mathematically arises due to the presence of a negative term linear in q in the series expansion.
The approach described above relies on a "quasistatic" approximation, i.e. we assume that membrane fluctuations occur on a time scale t mem slower than the time scale t el over which the electrostatic configuration adjusts itself (t el ≪ t mem ). This assumption allows us to solve the electrostatic problem for a fixed, weakly curved (but otherwise arbitrary) geometry. Let us estimate the time scales t el and t mem : To estimate t el for an electrolyte we assume that this time scale equals the time for an ion to diffuse the distance of the order the Debye screening length, i.e. t el ≈ κ −2 /D, where D is the ion diffusion constant and κ is the inverse Debye screening length introduced below (one may more realistically assume that t el = min{κ −2 , q −2 }/D, since for a wavelength perturbation of the order 1/q the ions need only diffuse a distance 1/q for the ion cloud to relax). From the Einstein relation and Stoke's law, we have D = k B T /(6πηR), where η is the viscosity, k B is the Boltzmann constant, T the temperature and R the ion Stoke's radius, using k B T = 4 · 10 −21 J, η = 10 −3 Ns/m 2 and R ≈ 0.1 − 0.3 nm [12], we find D ≈ 10 −9 m 2 /s. Furthermore, taking κ −1 = 10 nm we obtain: t el ≈ 10 −7 s. For the membrane relaxation time we estimate [13] t mem ≈ η/(q 3 K), and assuming q −1 > 100 nm, K = 10 −19 J we find t mem 10 −5 s. Therefore, indeed, we have t el ≪ t mem in general. We note from the expressions above that the longer the wavelength perturbation (smaller q) the better justified is our quasi-static approximation. In the dielectric limit there are no ions and the relevant relaxation time t el is instead that of water relaxation (hydrogen-bond rearrangement time), which typically is of the order 10 −12 s (see Ref. [14]) at room temperature, again certifying that t el ≪ t mem .
We are now set to consider the the effect of an applied potential on the mechanical properties of a biomembrane within the quasi-static approximation. The explicit problem we consider is depicted in Fig. 1: an incompressible membrane of thickness 2d is placed with its center-ofmass at positions z = 0 between two flat electrodes (at z = ±(L + d)) which are kept at potentials ∓∆φ/2; the distance between the membrane surface (for a flat membrane) and the electrodes are hence L. Regions 1 and 3 are electrolyte solutions, and in general these two regions are of different composition (different concentration of ions and different dielectric constants). As noted above the first stage towards calculating electrostatic forces on the membrane and thereby the membrane free energy in the presence of the applied potential is to obtain the electrostatic potential Φ( x). In this section we give general equations determining Φ( x) for any membrane shape. In the subsequent sections we analyze in detail: (i) the flat membrane case, see Fig 1a); all quantities for this case carry a superscript (0); (ii) for the weakly curved situation (since we assume the membrane to be incompressible we only consider undulation deformation modes), illustrated in Fig. 1b), there will be corrections of all quantities compared with the flat case; all such corrections carry the superscript (1).
For all three regions the electrostatic potential satisfies the Poisson equation (using SI units [15]) We will henceforth use a subscript γ (=1,2 or 3) to distinguish quantities in the three different regions. Above ε 0 is the permittivity of vacuum, ε γ is the dielectric constant for region γ, and ρ γ ( x) is the free charge density (the bound charges are, in a standard fashion, taken into account through ε γ ). We require i.e, that in each region we have charge neutrality (the membrane is assumed to be impermeable to ions). Let us now consider the explicit expression for the charge density in each of the three regions. Region 2. In region 2 we assume i.e. there are no free charges in this region. Region 1.
In region 1 we assume that there are z=d+h(x,y) z=h(x,y) z=−d+h(x,y) b)
Region 2
FIG. 1: Cartoon of the problems considered in this study: a membrane of width 2d is placed between two electrodes. Region 1 and 3 are characterized by dielectric constant ε1 and ε3 and Debye screening lengths κ1 and κ3 respectively. The membrane is assumed non-conductive (and non-compressible) and characterized by a dielectric constant ε2. Solving for the electrostatic potential Φ( x) for (a) a flat membrane, and (b) a weakly curved membrane (and utilizing the stress-tensor) allows us to obtain the electrostatic contribution to the membrane mechanical parameters.
ions of two types (positively and negatively charged) which are taken to be Boltzmann-distributed, i.e. (i distinguishes the different ionic species) where q i is the charge of ionic species i, c i bulk,1 is the concentration of ions in region 1, and β = 1/(k B T ), with k B the Boltzmann constant and T the temperature, as before. The constant φ 0 1 is determined through the charge neutrality condition. Making a linear approximation, i.e. assuming βq i (Φ 1 ( x) − φ 0 1 ) ≪ 1, and using the charge neutrality condition in the absence of the external potential, i q i c i bulk = 0, we find that the charge density can be written: with where κ 1 is the inverse Debye screening length for region 1. Region 3. For region 3, with the same approximations as above, the charge density becomes: where is the square of the inverse Debye screening length and c i bulk,3 is the concentration of ions in region 3.
Let us now consider the boundary conditions supplementing the equations above. At the electrodes we have: In addition we have that the potential and the normal component of the displacement fields are continuous across the region 1-region 2 and region 2-region 3 boundaries, i.e. [15] and wheren is the normal to the respective interface. Eqs. (8), (9) and (10), together with the boundary conditions Eqs. (11), (12) and (13) completely determine the electrostatic potential Φ( x). From the solutions for Φ( x) one can calculate other quantities. For instance, one may obtain the induced The total electric field is given by E( x) = − ∇Φ( x), the applied electric field is E appl = − ∇Φ appl and the induced field is E ind = − ∇Φ ind = E − E appl . In the next section we solve the equations given in this section for the case of a flat membrane. In the section after we find corrections to the flat case solutions for a weakly curved membrane.
III. POTENTIAL FOR A FLAT MEMBRANE
Below we obtain the electrostatic potential in and around a flat membrane. We use a superscript (0) to indicate the flat case quantities.
For a flat membrane the solutions depend only on z, and explicitly the solutions to Eqs. (8) and (9) are where φ 0 2 and A 2 are constants determined by the boundary conditions below. Also, and where φ 0 1 , φ 0 3 , A 1 and A 3 are constants, and we used the charge neutrality condition, Eq. (10).
We now use the boundary conditions (together with the fact that the boundary surfaces are at z = ±d for the flat case considered here, see Fig 1a) in order to determine the unknown constants above. From Eq. (11), the condition that the potential is continuous, Eq. (12) and the fact that the displacement field is continuous, Eq. (13), we get 6 equations for the 6 constants φ 0 γ and A γ (γ = 1, 2, 3). Solving these equations leads to and where and Γ = 1/[g(κ 1 )l 1 + g(κ 3 )l 3 + d/ε 2 ], and we introduced the "rescaled" Debye screening lengths l 1 = (ε 1 κ 1 ) −1 and There are three limits of particular interest: 1. "small" screening length, κ 1 L, κ 3 L ≫ 1. For this case we have g(q) → 1 in Eqs. (17) and (18). Also the prefactors for A 1 and A 3 simplify.
2. we define the dielectric limit as the limit of no salt, i.e. c i bulk,1 , c i bulk,3 → 0; within the Debye-Hückel approximation this is the equivalent to κ 1 , κ 3 → 0. Expanding the exponentials in Eqs. (15) and (16), and using the explicit form for A 1 and A 3 above one straightforwardly show that the solutions in all three regions take the form Φ( x) = az + b where a and b are constants independent of κ 1 and κ 3 , as it should since in the dielectric limit the potential satisfies Laplace equation, ∇ 2 Φ( x) = 0, see Eq. (9).
3. the symmetric case, ε 1 = ε 3 and κ 1 = κ 3 . In this limit we find that A 1 = −A 3 , φ 0 1 = −φ 0 3 and φ 0 2 = 0. The potential and charge densities are illustrated in Fig. 2, using the flat membrane results in Eqs. (14), (15) and (16). The electric field in the z-direction (the electric field components in the x-and y-direction are zero for a flat membrane) is also illustrated. We notice that the potential is continuous as it should and that the free charges tend to build up close to the membrane and electrodes (for κ 1 , κ 3 = 0). Since the normal component of the displacement field is continuous across the boundaries, the relative jump in the electric field as the the boundary between region 1-2 (region 2-3) is crossed equals ε 2 /ε 1 (ε 2 /ε 3 ), see Fig. 2 (bottom).
IV. POTENTIAL FOR A WEAKLY CURVED MEMBRANE
We now consider a weakly curved membrane, see Fig 1b): the center of the membrane is slightly displaced from the flat (z = 0) case, according to z = h(x, y) with membrane surfaces at z = ±d + h(x, y) [16]. We write the solution for the electrostatic potential according to (γ = 1, 2, 3 as before): where Φ (0) γ (z) is the potential for region γ for the flat case given in the previous section and Φ (1) γ ( x) is a correction to the potential due to the perturbed geometry. In this section we will calculate Φ (1) γ only to first order in the perturbation h(x, y); this suffices for obtaining membrane mechanical parameters such as tension and bending rigidity (see discussion at the begininning of Sec. II). In each of the regions Eqs. (8) and (9) has to be satisfied. Φ γ ( x) satisfies these equations for h(x, y) ≡ 0, and we therefore require that the corrections satisfy: and (γ = 1, 3) Let us now consider the boundary conditions (the boundary surfaces are at z = ±d + h(x, y) for the curved membrane considered here). Since Φ (0) γ (z) satisfies Eq. (11) we require for the perturbation: Before imposing the conditions that the potential and the displacement fields are continuous, Eqs. (12) and (13), we note that for "small" h any scalar quantity g( x) may be expanded, to first order in h, according to: we briefly discuss the quantitative meaning of "small" h at the end of this section. Eqs. (12) and (13) can then be written h ∂Φ (24) and where we used the fact that Φ We proceed by introducing the Fourier-transform in the x-and y-direction (not for z-direction) of Φ (1) γ ( x): and similarly we denote byh(q x , q y ) the Fouriertransform of h(x, y). Eqs. (21) and (22) can then be written: and (γ = 1, 3) where q = q 2 x + q 2 y and The boundary conditions Eqs. (23), (24) and (25) remains the same in Fourier-space (using the fact that Φ 27) is with q-dependent coefficients C 2 (q) and D 2 (q). Using the boundary condition Eq. (23) we find that the solutions to Eq. (28) are: The unknown coefficients C 2 (q), D 1 (q), D 2 (q) and D 3 (q) are determined through the boundary conditions Eqs. (24) and (25). We find where we introduced the short-hand notations (γ = 1, 3) with A γ given in Eq. (17). The remaining coefficients are We notice that C 1 (q), D 1 (q), D 2 (q) and D 3 (q) are all proportional toh as they should be. The full solution for the electrostatic potentials Φ γ ( x) for a weakly curved membrane is given by Eq. (20), where Φ Again, we point out that in order to obtain membrane mechanical parameters, such as tension and bending rigidity, it suffices to know the restoring force (calculated in the next section) to first order inh, i.e. it is enough to consider "small" fluctuation amplitudes. Therefore, even thought the results given above formally assume that h is smaller than all other length scales in the problem (d, κ −1 γ , q −1 and L), they are sufficient for the purpose of calculating the tension and bending rigidity. A different matter is whether the Helfrich form of the electrostatic contribution to the restoring force (in terms tension and bending rigidity), derived in the next section, describe well "large" membrane fluctuations in the presence of an applied potential. To address this question we must clarify the meaning of "small" h, i.e. make clear what is the relevant dimensionless expansion parameter, in the context of calculating the electrostatic contribution to the membrane restoring force or membrane free energy. The electrostatic contribution to the membrane free energy is determined by the interactions between induced charges at or in the vicinity of the membrane, with characteristic interaction distances of the order d and κ −1 γ (here we consider the small screening length limit, where κ γ is non-zero). Therefore, whenever dq ≪ 1, κ −1 γ q ≪ 1 and hq ≪ 1 all interactions are effectively local on a locally flat membrane and the free energy must take the Helfrich form [10]. Note that this argument is valid independent of particular values of hκ γ and h/d, and the free energy expansions given in the next section (in the small screening length limit) is therefore an expansion in the (small) parameters hq, dq, κ −1 γ q, but when these parameters are small there is no restriction on the values of hκ γ and h/d. In consistency with the discussion above, we point out that in Ref. [17] a method that formally avoids the assumptions hκ γ ≪ 1 and h/d ≪ 1, by utilizing the geometrical transformation z ′ = z − h(x, y), was found to give results consistent with results obtained by the flat membrane perturbative approach (of the kind used in this paper) for a charged membrane in an electrolyte.
V. FORCES, FREE ENERGY AND ELECTROSTATIC CONTRIBUTION TO MEMBRANE MECHANICAL PARAMETERS
In this section we derive general expressions for the membrane forces (within the quasi-static approximation) and the corresponding electrostatic contribution to the membrane free energy, using the results from the previous two sections. In particular, we obtain the electrostatic contribution to the membrane free energy for three cases: (1.) the small screening length limit, where the Debye screening length is smaller than the distance between the electrodes; (2.) the dielectric limit, i.e., no salt, and; (3.) the symmetric case, where the salt concentrations and dielectric constants on the two sides of the membrane are equal.
A. Membrane forces via a stress-tensor calculation
The forces acting on the membrane are obtained using the relevant stress-tensor, T ij . Following the derivation in Appendix A we have (in the Debye-Hückel regime considered here) where x 1 = x, x 2 = y and x 3 = z. The first two terms are the usual Maxwell stress-tensor [15,18], the third term is an osmotic contribution for the ions being "confined" by the electric potential (see appendix A), and the last term incorporates pressures p 0 γ for each of the three regions (γ = 1, 2, 3). The discontinuities of T ij at the region boundaries will produce forces on the membrane which will have to be balanced by other forces in the system, such as for example viscous forces within the membrane or from the surrounding bulk fluids. We are only interested in calculating the electromechanical contribution at a given (x, y) to this total force balance here. To do this we note that the force (per unit area) in the i-direction on a region boundary from the stress in a given region is ± j n j T ji evaluated at the boundary, where the membrane normal n j is taken to point towards positive z and the plus (minus) sign applies when the region is at larger (smaller) z than the boundary. Defining where z = −d (z = d) is to be used for forces acting on the interface separating region 1 and 2 (region 2 and 3), and using that to first order in h: n z = 1, n j = −∂ j h (j = x, y), we find that the z-component of the total force acting on the surface separating regions 1 and 2 is f 1−2 = f 2 − f 1 . Using the explicit expressions for the potentials from the previous two sections we find where f 1−2 is the force on a flat membrane interface, and f (1) 1−2 is the first order correction for a weakly curved membrane, here expressed explicitly in terms of its Fourier-transformf where A γ are given in Eq. (17) and C 2 (q), D γ (q) are given in Eqs. (33) and (35). The results given in Eqs (38) and (39) completes the calculation of the forces acting on the membrane interfaces. In appendix B we utilize these results in order obtain results for the total net force on a flat membrane in some detail. In the next subsection the membrane free energy and the electrostatic contribution to the membrane mechanical parameters in different limits are investigated.
B. Contribution to the free energy of the membrane
Let us now investigate the electrostatic contribution to the membrane free energy. We first note that if the fluids surrounding the membrane are incompressible, then the pressures p 0 γ [occurring in the zero order terms in Eqs. (38) and (39)] adjust such that there is no net force (and hence no net movement of the membrane) in the z-direction; we will here consider such incompressible fluids and also assume the membrane to be incompressible. Nevertheless the investigation of the net force f (0) = f fromf (1) one can obtain the work on the membrane under an undulation deformation of the shape, and thereby the free energy and electrostatic contribution to membrane mechanical parameters (through a power series in q, i.e. a long wavelength expansion). In particular we want to compare the results of such an expansion to the corresponding result for a "free" membrane: the free energy G for a membrane is described by the Helfrich form [10,11] G = dA 2KH 2 + σ where H is the mean curvature, dA the area element on the membrane, σ is the tension and K is the bending rigidity. The restoring force is then obtained as f rs = −δG/δh giving in q-spacē This type of expansion requires only that the expectation value of (∇h(x, y)) 2 is small [11], i.e. that the characteristic fluctuation amplitude is small compared to 1/q. In the presence of an applied potential there will be electrostatic contributions σ el and K el to the tension and bending rigidity, so that σ → σ + σ el and K → K + K el . Below we proceed by expandingf (1) , using the results in Eqs. (38) and (39), in a power series in q for different limits in order to obtain σ el and K el (note, however, in the expansion for the dielectric limit, for qL ≫ 1, we also find terms odd in q). Note that since the tension and bending rigidities are identified through terms in the restoring force expansion which are proportional toh, second and higher order terms inh (see discussion at the end of the previous section) do not contribute to σ el and K el .
1. In the "small" screening length limit , κ 1 L, κ 3 L ≫ 1, a straightforward but lengthy expansion off (1) in a power series in q assuming that the wavelength of the perturbation (= 2π/q) is larger than the membrane thickness and the Debye screening lengths, qd, q/κ 1 , q/κ 3 ≪ 1, gives The explicit expression for the electrostatic contribution to the tension is where, as before, the "rescaled" Debye screening lengths are l 1 = (ε 1 κ 1 ) −1 and l 3 = (ε 3 κ 3 ) −1 . We also introduced being the potential difference between the main parts of the two bulk fluids. Notice that σ el gives a negative contribution to the tension. The fact that σ el < 0 originates from the fact that that the applied potential creates a net charge density on either side of the membrane surfaces, see Fig. 2; since ions of equal charge repel each other the system would, for a compressible membrane, be able to decrease the free energy by separating the charges through a stretching of the membrane (i.e., by increasing the membrane area). For an incompressible membrane (as assumed here) the membrane is likely to respond to the electrostatically induced negative tension by an opposite increase in the membrane elastic contribution to the tension. If the magnitude of σ el exceeds the membrane elastic strength (tensile strength) a stretching instability occur. For the symmetric case (κ = κ 1 = κ 3 and ε = ε 1 = ε 3 ) Eq. (42) becomes: where we introduced the ratioε m ≡ ε 2 /ε between the membrane and surrounding medium dielectric constant (typicallyε m ≈ 1/40, see [1,7]). We notice that when s =ε m (κd) −1 ≪ 1, i.e. for an effectively small screening length compared to the membrane width, the tension approaches σ 0 el ; the dimensionless parameter s is commonly appearing in membrane electromechanical problems, see Refs. [1,7]. From Eq. (42) we notice that for the asymmetric case we similarly have σ el ≈ σ 0 el for s γ = (ε 2 /ε γ )(κd) −1 ≪ 1, where γ = 1, 3. Since σ 0 el only depend on the membrane dielectric constant, membrane width, and the applied potential, σ el is for large salt concentration independent on the properties of the surrounding medium (i.e. independent on κ 1 , κ 3 , ε 1 and ε 3 ). The origin of this result is discussed below.
The electrostatic contribution to the bending rigidity is with coefficients We point out that K el > 0, i.e., the applied potential tends to make the membrane more rigid towards bending. During a bending deformation the induced charge density on one side of the membrane gets compressed, whereas the charge density on the opposite side gets expanded. The free energy changes of compression and expansion has different signs, but are in general of different magnitude. It is only for the case that all Debye screening charges are collapsed onto the surfaces (κ → ∞) and zero membrane thickness d → 0 that the expansion and compression free energies are identical and K el = 0 [see Eq. (45)]. Thus, loosely speaking, the smaller the "effective" membrane thickness (the membrane thickness including the Debye screening layer thicknesses) the smaller is the bending rigidity. For the symmetric case (κ = κ 1 = κ 3 and ε = ε 1 = ε 3 ) we write Eq. (45) according to We note that in the limit of small relative membrane dielectric constantε m → 0 as well as for small screening length compared to the membrane thickness, (κd) −1 → 0, we have K el → K 0 el . Notice that, in practice,ε m is always small ≈ 1/40, and therefore we have K el → K 0 el also for "not too small" values for (κd) −1 . Similarly, we note from Eq. (45) that for the asymmetric case we have K el → K 0 el in the ε 2 /ε γ → 0 and in the (κ γ d) −1 → 0 (γ = 1, 3) limits. For the above considered limits the major part of the potential drop is across the membrane (since ε 2 /ε γ or (κd) −1 is small), and hence the electric field is essentially zero everywhere except for the membrane region. Therefore, the membrane parameters play the dominant role in the expression for K el and σ el (notice that K 0 el and σ 0 el depend only on ε 0 ε 2 , d and ∆φ m ), provided that the interactions of the induced surface charges do not occur through the surrounding medium (again, certified if ε 2 /ε γ or (κd) −1 is small). Given that K el and σ el can only depend on ε 0 ε 2 , d and ∆φ m in the limits considered above one may use dimensional arguments to argue that the limiting results, K 0 el and σ 0 el , for the bending rigidity and tension must (up to a constant prefactor) take the forms given in Eqs. (44) and (47). Fig. 3 illustrates the electrostatic contribution to the tension and membrane bending rigidity and its dependence on salt concentration (Debye screening) for the symmetric case for simplicity. We see that the absolute value of electrostatic contribution to the membrane tension increases with increasing salt concentration, i.e. for increasing κ. We attribute this to an increase of screening charges (in a layer of decreased thickness) next to the membrane; the increased amount of charges will result in larger electrostatic repulsion between ions in the screening clouds [see discussion following Eq. (43)]. In contrast to the effect on tension the electrostatic contribution to the bending rigidity decreases with increasing salt concentration (with K el approaching K 0 el as κ → ∞). The reason behind this is that for bending properties the thickness of the Debye screening layers plays a role -a larger "effective" membrane thickness gives a higher bending rigidity [see discussion following Eq. (45), and the spontaneous curvature calculation in Appendix C]. If we choose the potential difference between the membrane sides ∆φ m = 100 mV, ε 2 = 2 and d = 2.5 nm we find that K 0 el = 0.018k B T (for room temperature, k B T = 4 · 10 −21 ). From Fig. 3 we see that K el /K 0 el can become quite large for small κ and, therefore, for sufficiently small salt concentration the bending rigidity K el can exceed the thermal energy k B T ; we therefore expect that the increase of the bending rigidity in the presence of an applied electrostatic potential predicted in this study can indeed be experimentally observed for sufficiently large ∆φ m and small salt concentrations (note however, the below restriction on salt concentration due to assumptions in the Debye-Hückel approximation). Let us compare the results above for the symmetric case to the results obtained in [7]. We find that the result for σ el given in Eq. (44) agrees with the finite bilayer thickness, non-conductive membrane tension (there denoted by Σ in + Σ out ) obtained in [7] (note that in [7] the membrane thickness is denoted by d, whereas we denote by d the size of a lipid monolayer, so that in our case the membrane thickness is 2d. Also note that, due to different boundary conditions at the electrodes, the V in [7] should be equated with our ∆φ m ). Concerning the bending rigidity result, we note that no explicit expression for K el was given in [7], only a numerical value for a specific set of parameter values. We choose the same parameter values (∆φ m = 50 mV, d = 2.5 nm,ε m = 1/40, and 2dκ = 7.4), but note that in order to get the expression for K el one must also choose the actual value of ε 2 (not just the ratio ε 2 /ε), which was not specified in [7]. We choose the standard value ε 2 = 2 [1,9] and then find that K el = 0.00467 k B T , which is a bit less than half the value found in Ref. [7]. Since no explicit expression for K el was given in [7] it is difficult to comment on the nature of this discrepancy. Finally, using the approximate expression K el ≈ K 0 el for the parameters above we find that this approximation underestimates K el by merely 1 %.
Here, a few words on the validity of the Debye-Hückel approximation, used throughout this study, are in place. This approximation should work when the quantity I = βq i (Φ γ ( x)−φ 0 γ ), where γ = 1, 3, is very small, i.e. I ≪ 1 (see Sec. II), but in practice the Debye-Hückel approximation works well whenever I < 1, see Ref [19]. The maximum of I occurs at the membrane surfaces, see Eqs. (15), (16) and Fig. 2, and we find that for the κ γ L ≫ 1 limit considered here we have I = βq i ∆φl γ /[2(l 1 +l 3 +d/ε 2 )]. Using d ≈ 2.5 nm, ε 2 ≈ 2 and ε γ ≈ 80, the denominator in the expression for I above is dominated by the d/ε 2 term whenever κ −1 γ < 50 nm (in this limit also ∆φ ≈ ∆φ m ); for such ion concentrations we have the following criterion for the validity of the Debye-Hückel approximation again involving the parameter For a large potential difference ∆φ m = 100 mV we find that I < 1 when κ −1 γ < 50 nm (using q i = 1.6 × 10 −19 C and assuming room temperature). This means that the Debye-Hückel approximation, which was made in Sec. II in the main text, works surprisingly well in general; the reason for this is the small value of ε 2 /ε γ (≈ 1/40) guaranteeing that the major part of the potential drop occurs across the membrane and that, therefore, the potential drop across the electrolytes, for which we applied the Debye-Hückel approximation, is modest, see Fig. 2. We point out that the dielectric limit, c i bulk,γ → 0, does not rely on a Debye-Hückel approximation and results to be given below are therefore valid for any value of the applied potential. We have shown above that our results for the small screening length limit (with the above explicit restriction) and in the dielectric limit are usually valid. We note, however, that there is in general an intermediate salt regime where the Debye-Hückel approximation breaks down (for sufficiently large applied potentials). We leave the investigation of this intermediate regime for future studies.
We finally point out that there are alternative ways of computing the membrane mechanical parameters. For instance one can calculate the tension by finding the integral of the deviation of the pressure profile from the value of the pressure far from the membrane. This approach is demonstrated in Appendix C, giving the same result as Eq. (42) for the tension. In that appendix the same type of approach is also used in order to obtain the electrostatic contribution to the membrane spontaneous curvature, see Eq. (C7).
2. in the dielectric limit, κ 1 , κ 3 → 0, one can again perform a power series expansion in q using Eqs. 44) and (47). We express K el in terms of the (room temperature) thermal energy kBT and σ el in terms of its infinite κ value σ 0 el . The parameters used are listed in the figure. Notice that increasing salt concentration, i.e. increased κ, leads to a decrease in the electrostatic contribution to the bending rigidity. In contrast, for increasing κ the magnitude of the electrostatic contribution to the tension increases.
(38) and (39) and assuming qd ≪ 1. In addition, there are two limits of interest depending on whether the wavelength perturbation is smaller or larger than L.
For (i) qL ≫ 1 we find that where and and the higher order terms in q are more complicated functions of the dielectric constants and d.
The term linear in q is negative and is expected to cause an instability for long wavelengths [20]. We notice that the prefactor a [see Eq. (50)] is proportional to the membrane polarization charge density (for a flat membrane) squared, a ∝ (ρ s ) 2 , see Eq. (B4). The linear, non-analytic, q term may be interpreted as follows: for the asymmetric dielectric case the external potential induces an effective net membrane polarization charge density ρ s . The membrane charges interact via the unscreened (there are no ions in the present limit) Coulomb interaction, giving rise to the (ρ s ) 2 proportionality for a; the non-analyticity of the free energy arises due to the long-range character of the Coulomb interaction (which decays as 1/r, where r is the distance between charges). We point out that the linear q term does not depend on the membrane parameters, d and ε 2 , which means that this instability should exist for any interface between coexisting fluids (fluid interface instabilities of a rather similar nature has been investigated previously, see for instance [6]). For the symmetric case ε 1 = ε 3 we see that a = 0, but the third order term above is still present, b = 0, in general (see discussion below).
For the case (ii) qL ≪ 1, i.e. the wavelength perturbation is longer than the distance between the electrodes, the first order force takes the form f (1) = −[c + σ el q 2 + K el q 4 + O(q 6 )]h. In this limit the interactions between the membrane polarization charges become short-range due to effectively small (compared to the wavelength of the perturbation) distance of the membrane to the electrodes. In this limit we thus do not have any odd q terms the effect of the applied potential is simply to give an electrostatic contribution to the tension and bending rigidity. The explicit expressions become somewhat complicated, but can be produced straightforwardly by a small q expansion using a symbolic mathematical software like Mathematica or Maple together the expressions forf 1−2 andf 2−3 given in Eqs. (38) and (39).
3. For the symmetric case (ε = ε 1 = ε 3 , κ = κ 1 = κ 3 ) the lowest order term in the q expansion is the tension term (q 2 term), both in the small screening length limit and for the dielectric limit. For both these limits we have that the electrostatic contribution to the tension σ el is negative. Thus when |σ el | becomes sufficiently large, i.e. large applied potential, a membrane stretching instability can occur. We also find that in the symmetric dielectric limit there is a q 3 term present (for qL ≫ 1), where explicitly we find [see Eq.
We notice that b ∝ P 2 z , where P z is the membrane polarization per unit area, see Eq. (B8). For the symmetric dielectric case the effect of the applied potential is to polarize the membrane and the q 3term is expected to originate from induced, unscreened, dipole-dipole interactions (which decays as 1/r 3 , where r is the distance between the dipoles [15]) within the membrane.
VI. SUMMARY AND DISCUSSION
We have in this paper derived expressions for the electrostatic contributions to biomembrane mechanical parameters (such as tension and bending rigidity) in the presence of an static applied potential across a membrane. The membrane was assumed non-compressible, non-conductive (membrane region described by Laplace equation) and surrounded by electrolyte solutions (described by the Debye-Hückel equation). By solving the equations for the electrostatic potential and using the stress-tensor the forces acting on the membrane were obtained, which in turn were used to obtain the free energy and the electrostatic contribution to the membrane mechanical parameters as a function of the applied potential, the salt concentrations (entering through the Debye screening lengths) and the dielectric constants of the membrane and the solvents. Results of particular interest, that are found in this study, include: for (1.) the small screening length limit, where the Debye screening length is smaller than the distance between the electrodes, the screening certifies that all electrostatic interactions are short-range, leading to a free energy expansion of the form ∼ σ el q 2 + K el q 4 + O(q 6 ) (where q is the wavevectors), the main effects of the applied potential are to decrease the membrane tension and increase the bending rigidity; explicit expression are given in Eqs.
(42) and (45). Our expression for the tension for the symmetric case reproduces the result in [7]. In [9] it was also found that an applied electric field gives a negative contribution to the tension. However in that study the medium surrounding the membrane was characterized by conductivities rather than Debye screening lengths, and it is therefore difficult to directly compare our results to theirs. For sufficiently large applied potentials the magnitude of the electrostatic contribution to the tension will exceed the maximum tension the membrane can sustain, leading to a membrane stretching instability. Possibly, this instability can result in the formation of pores and flow of ions through the membrane (in fact, the membrane tension is one of the key parameters in the modeling of membrane electroporation dynamics [21]). For (2.) the dielectric limit, i.e. no salt (for small wavevectors q compared to the distance between the electrodes), when the dielectric constants on the two sides are different, the applied potential induces an effective (unscreened) membrane charge density, whose long-range interaction causes a membrane undulation instability if the dielectric constants of the two bulk fluids are different; this effect is characterized by a negative term linear in q in the free energy expansion, see Eqs (49) and (50), i.e. this term is of lower order in q than the tension term. Previous similar results include: In [6] the interface between two immiscible fluids of different dielectric constants was found to be unstable in the presence of a perpendicular electric field. The case of stiff (charged) DNA with bound, but mobile, counter ions was investigated in [3] and a shape instability found (supposedly leading to a DNA condensation).
If the two dielectrics on each side of the membrane are identical, we found that then the applied electric field will give a negative contribution to the tension. Hence, if the applied potential is sufficiently large a membrane instability occurs also for the (dielectric) symmetric case.
We quantified the validity of the Debye-Hückel approximation (used throughout this study) and showed that our results are in general valid in the small screening length limit as well as in the dielectric limit. However for "small", but non-zero, salt concentration and large applied potential the Debye-Hückel approximation is no longer valid and one needs to consider the full Poisson-Boltzmann equation. It remains a future challenge to solve the full Poisson-Boltzmann problem in order to find expressions for the free energy for arbitrary salt concentration, and, in particular, to investigate in more detail the nature of the onset of the predicted membrane instability (via the negative term linear in q in the free energy) as salt concentration is lowered.
Possible applications of the result for the small screening length limit above would be to lipid membranes where a potential difference across the membrane is enforced by ion pumps incorporated in the membrane. Changes in membrane rigidity might then be observed in micropipette or video microscopy experiments, if the screening length and the membrane potential are large, see Fig. 3. Another possible experiment would be to observe the structural change of domains in a multi-component membrane using fluorescence correlation spectroscopy (FCS) as the membrane potential in a patch clamp experiment is altered; one might be able to observe a change from a phase with caps to one with stripes or buds as the effective bending rigidity is changed by the applied potential [22].
For the cases discussed above where a membrane instability occurs, the system is expected to be driven from the quasi-flat shape into a new equilibrium configuration. Our perturbation analysis cannot in general say anything about this new configuration. However, for the small screening length limit we above speculated that the negative electrostatic contribution to the tension (for large applied potentials) could lead to electroporation and a corresponding flux of ions through the membrane. It may also be speculated (similar to the studies in Refs. [2] and [9]) that the electrostatically induced new equilibrium configuration under certain conditions could be a spherical membrane (a vesicle); it is known that vesicles can be created in laboratories in a process known as electroformation [5] under the application of electric fields. We hope that the present theory will stimulate further work directed towards the controlled estimation of vesicle sizes as a function of electrostatic parameters. e.g. potential differences and electrolyte concentration.
|
2018-04-03T06:06:55.071Z
|
2006-10-13T00:00:00.000
|
{
"year": 2006,
"sha1": "e134bbc640c0d5bb4919cc28e33e3927511e968a",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/cond-mat/0610355",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "56e61bcd4b583c0593ab3f5d86f1cb7abda305f9",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Materials Science",
"Medicine",
"Physics"
]
}
|
254359605
|
pes2o/s2orc
|
v3-fos-license
|
CONSTITUTIONAL FEATURES OF THE OCCURRENCE OF BENIGN AND MALIGNANT SKIN TUMORS (ANALYSIS OF SCIENTIFIC LITERATURE)
Annotation. The purpose of the work is to analyze the scientific literature on the constitutional features of the occurrence and course of benign and malignant skin tumors. Increasing incidence of skin tumors, in particular, melanoma of the skin requires the creation of simple and effective mechanisms to predict the occurrence of this category of diseases. Foreign authors have been actively working in recent decades to find relationships between body structure and the occurrence and course of many skin diseases, and oncological pathologies have not become an exception. The analysis of scientific literature from the scientometric databases Google Scholar, MedLine, Web of Science, Scopus for 2008-2019 (the overwhelming majority of which - over the past 5 years), has shown promise of research in this direction, and due to the lack of such domestic works, there is an urgent need to conduct research on the Ukrainian population.
Oncological disease is still one of the leading causes of morbidity and mortality in the world. In the structure of this category of diseases a special place is occupied by skin tumors, which firstly have greater contact with carcinogenic factors -ultraviolet radiation, infectious diseases caused by viruses and bacteria, mechanical damage, etc., secondly difficult to diagnose (differential diagnosis of skin melanoma is one of the most difficult in pathomorphology) and thirdly have a high mortality (melanoma of the skin has some of the highest rates of metastasis and mortality).
The purpose of the work -to analyze the scientific literature on the constitutional features of the origin and course of benign and malignant skin tumors.
Skin cancers include many types of tumors, which according to the most common classification are divided by source: keratinocyte tumors, melanocyte tumors, skin appendage tumors, soft tissue tumors of the skin, neural tumors and tumors of the subcutaneous layer [13]. Accordingly, each of these species is represented by malignant and benign species.
The prevalence of malignant skin cancers in people over 50 years of age increases by 0.6 % every year. In 2016, 76,380 new cases of melanoma were recorded [4] and 3.5 million new cases of other skin cancers [8]. In Iran, from 2003 to 2008, 49,740 cases of skin cancer were recorded, among which the most common was basal cell carcinoma (65.40 %) [23]. Benign skin tumors account for 3 % of all skin samples sent for histological examination. The most common benign tumors are in people 20-40 years; in women, benign skin tumors are 2 times more common than in men [21].
The prevalence of skin cancer varies in different regions, even within one country. Thus, in the United States in the state of Hawaii, the prevalence of skin cancer is 422 cases per 100 thousand population, while in Minnesota -146 cases per 100 thousand population [8].
In Ukraine, as of 2015-2016, the number of people with malignant neoplasms of the skin was 919 093, of which 174,848 patients had non-melanoma malignant neoplasms, and 2796 people were diagnosed with melanoma [2].
This epidemiological situation requires the search for an effective, simple and inexpensive tool to predict the risk of occurrence and course of these pathologies. An effective tool in this case can be clinical anthropology, the reliability of which has been tested in various fields of medicine (sports medicine [1], forensic medicine [11], dentistry [18], cardiology [20], etc.). In particular, both domestic and foreign studies have found correlations between anthropometric indicators and the risk of occurrence and course of skin diseases such as acne [10], atopic dermatitis [3] and others and the occurrence of various cancers [5].
U.S. researchers conducted a long-term study on 58,213 people to determine the relationship between height, weight, body mass index and the risk of basal cell carcinoma of the skin, which was detected in 2,291 people. Statistical analysis revealed that the risk of basal cell skin cancer increased with increasing height, and decreased with increasing weight and body mass index in both sexes and even after adjustment according to the factor of susceptibility and exposure to ultraviolet radiation [9].
A group of Australian scientists measured the body weight and height of 1,171 men and women to determine the relationship between anthropometric values and the incidence of various skin cancers. The study considered exposure to ultraviolet radiation, skin phenotype and other factors. In 16 years of follow-up, in 550 people developed skin cancer. Of all the indicators, only a person's height was significantly associated with the development of squamous cell carcinoma and melanoma in men and basal cell carcinoma in women [15]. Zhou D. et al. [28] conducted a meta-analysis of epidemiological studies to find a correlation between body mass index and non-melanoma skin cancer. As a result, the authors considered only 9 publications in which a total of 971,795 people were examined, of which skin cancer was detected in 50,561 people. Nonlinear feedback between body mass index and non-melanoma skin cancer has been reported.
A similar study conducted by Karimi K. et al. [12] using meta-analysis of literature sources confirmed the correlation between the risk of melanoma and non-melanoma skin tumors and obesity. 71,645 postmenopausal women were assessed for body mass index, waist-to-hip ratio, and ultraviolet radiation from 1993 to 1998. During this period, 18.6 % of women developed non-melanoma skin cancer. Statistical analysis of the obtained data showed that the body mass index ?25 kg/m2 or the ratio of waist circumference to the thigh ?0.80 is associated with lower rates of non-melanoma skin cancer [6].
Norwegian researchers surveyed 292,851 people during 1972-2003. During the 27 years of the study, 3,000 people developed skin melanoma. All subjects underwent anthropometric studies. All data were adjusted for exposure to ultraviolet radiation, age, and smoking. Statistical analysis of the data revealed a positive relationship between the occurrence of melanoma of the skin and body mass index, body area, height and weight (p<0.001) [24].
Meyle K. D. et al. [19] investigated the association of the risk of melanoma in adulthood and anthropometric indicators of persons in childhood (7-13 years). The study analyzed data from 372.636 Danish children born between 1939 and 1989. During the observation in 2.329 of the subjects developed skin melanoma. There was a significant association between growth rates at the age of 7-13 years and the risk of melanoma -children who were tall at 7 and 13 years were more likely to develop melanoma than children who grew between 7-13 years. Birth weight was also positively associated with the risk of skin melanoma.
Korean researchers conducted a large sample of more than 22 million people over a 7-year period to find a correlation between wrist circumference and the risk of 23 cancers. During the observation period, 2963 subjects developed melanoma of the skin. Statistical analysis of the data revealed a relationship between wrist circumference and the risk of skin melanoma [16].
Zhang Y. at al. [27] studied anthropometric parameters in 377 people with basal cell skin cancer and 389 people with benign skin tumors (control group). The body mass index was inversely related to the early development of this type of cancer.
The team of scientists conducted a meta-analysis of 12 studies with a total of almost 5 million people surveyed, including 20.049 patients with skin melanoma. Statistical analysis of anthropometric data revealed that individuals with the highest growth category had a higher risk of melanoma compared with individuals with the lowest growth category (RR = 1.46, 95 % CI 1.24 to 1.73; p<0.001) [26].
Similar results were obtained in a meta-analysis performed by Vena G. A. et al. [25], who established a correlation between human body growth and the risk of melanoma.
Analysis of anthropometric data of 98,995 women in the French cohort born in 1925-1950 revealed a positive relationship between the risk of melanoma and human growth, and a negative relationship between body area in menarche and the risk of melanoma [14].
There is a higher risk of melanoma in overweight women compared to not overweight women (OR=1.64). In the group of menopausal women, the OR rate increases to 2.50 [7].
Analysis of data from more than 1 million Jewish men aged 16-19 collected between 1967 and 2005, among whom 1.562 were diagnosed with melanoma, revealed correlations between a person's background and the risk of melanoma. Thus, Jews of European and Israeli descent have higher rates of morbidity than Jews of African and Asian descent [17].
1.
The analysis of foreign literature sources convincingly proves the existence of a relationship between constitutional indicators and indicators of skin cancer morbidity. Unfortunately, the work performed by domestic scientists in this direction was not detected in the analysis of scientometric databases.
Search, selection and further use of anthropometric markers as genetic markers of benign and malignant skin tumors among the population of Ukraine is a promising area for domestic medicine, as it will find and form risk groups among different segments of the population. Individuals in these groups will be able to avoid or reduce contact with carcinogenic factors (primarily -excessive ultraviolet radiation), which in turn will reduce morbidity and mortality, especially to improve the epidemiological situation of the most dangerous type of tumors in this groupmelanoma of skin. Obesity as a risk factor for malignant melanoma and nonmelanoma skin cancer.
|
2022-12-07T20:29:35.731Z
|
2020-10-12T00:00:00.000
|
{
"year": 2020,
"sha1": "6cbd86a104a01bd3a5c05e9a1fc4bc69b6159870",
"oa_license": "CCBY",
"oa_url": "https://reports-vnmedical.com.ua/index.php/journal/article/download/1058/1013",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "c2b45239274fee42f66f747203107377fb6d6ed5",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": []
}
|
216401551
|
pes2o/s2orc
|
v3-fos-license
|
Changes in Assimilation Area and Chlorophyll Content of Very Early Potato ( Solanum tuberosum L.) Cultivars as Influenced by Biostimulants
: This paper analyses the effects of foliar application of the seaweed extracts Bio-algeen S90 ( Ascophyllum nodosum ) and Kelpak SL ( Ecklonia maxima ), as well as the humic and fulvic acids ini HumiPlant (leonardite extract), on the assimilation area and chlorophyll content of very early potato cultivars (‘Denar’, ‘Lord’, Mi ł ek’). The field experiment was carried out in central-eastern Poland over three growing seasons, using Luvisol. The biostimulants were applied according to the manufacturers’ recommendations. The use of biostimulants resulted in enlargement of the assimilation area, but had no effect on the specific leaf area (SLA) or chlorophyll content (Soil Plant Analysis Development (SPAD) value). The assimilation area was larger, on average, by 0.0505 m 2 and leaf area index (LAI) was higher by 0.30 compared with the plants from the control group without a biostimulant. The SLA and SPAD depend on the cultivar and weather conditions, or nitrogen and magnesium content in soil, to a greater extent. The biostimulants enhanced abiotic stress tolerance and increased marketable tuber yield (diameter above 30 mm) 75 days after planting (the end of June), on average by 2.15 t ∙ ha –1 . Bio-algeen S90 and Keplak SL produced better results in a warm and very wet growing season, whereas HumiPlant produced better results in a year with lower air temperature and with drought periods during potato growth. No correlations were found between the tuber yield and assimilation area or between the tuber yield and SPAD value, although a significant negative correlation was found between the tuber yield and SLA.
Introduction
In recent years, the growth and productivity of crop plants have been greatly influenced by abiotic stresses. Periods of high temperature and drought are becoming more frequent in regions with extensively crop production, such as Central Europe, South-Central Asia, south-eastern South America and the south-eastern United States [1]. Under climate change conditions, biostimulants play an important role in sustainable crop production. These natural products (seaweed extracts, humic substances, hydrolysed proteins, and amino acids containing products or microorganism) contain a bioactive substance which enhances nutrition efficiency, abiotic stress tolerance, and/or crop quality traits, regardless of its nutrients content [2][3][4][5]. In recent years, the use of seaweed extracts and humic substances as plant growth stimulants has been increasing. Seaweed extracts and humic acids can promote plant growth, enhance abiotic stress tolerance as well as increase nutrient use efficiency [6][7][8][9][10].
Many plant growth-stimulating compounds (auxins, cytokinins, gibberellins, betaines, polysaccharides, polyamines, abscisic acids, brassinosteroids, and minerals) have been identified from seaweed. The chemical composition of seaweed extracts depends on the algae species and on the method of extraction. Brown algae (Phaeophyta) are most commonly used for the manufacture of extracts used as biostimulants of plant growth, including Ascophyllum nodosum and Ecklonia maxima [7,8,11]. An increase in leaf area and chlorophyll content are common plant responses to seaweed extract treatment. Cytokinins present in the seaweed extracts stimulate cell division, resulting in enlarged leaf area, and also stimulate chlorophyll biosynthesis, whereas betaines slow chlorophyll degradation and delay leaf senescence [7,8]. Ascophyllum nodosum extracts applied on foliage or to soil caused an increase in the leaf chlorophyll content of French bean, tomato, barley, maize, wheat, pepper, and strawberry [8,11,12]. A one-year study carried out in Iraq showed an increase in chlorophyll content in potato following the application of brown seaweed Sargassum extracts [13]. Foliar application of seaweed extracts Ascophyllum nodosum and Ecklonia maxima increased potato yield [14][15][16]. Biostimulants based on seaweed extracts improved plant growth and yield of wheat, barley, maize, potato, tomato, pepper, onion, and carrot [7,8,10,11].
The biological activity of humic substances depends on their source, chemical structure, and concentration. Humic substances may influence both respiration and photosynthesis. One of the effects of humic substances applied to growing plants was an increase in chlorophyll content, which can affect photosynthesis [17]. Leonardite is the most common commercial source of humic substances. Leonardite humic acids stimulate melon and soybean growth and chlorophyll synthesis [6]. A one-year study carried out in Iraq showed an increase in chlorophyll content in potato following the application of humic and fulvic acids in HumiMax [13]. A one-year study carried out in Egypt showed that the application of humic acid under water stress conditions enhanced the leaf chlorophyll content of very early potato cultivars [18]. Application of humic substances originating from leonardite increased potato yield and nutrient uptake [19]. In most experiments, foliar or soil application of humic and fulvic acids increased potato yield [13,20,21], but one study showed no clear effect of humic and fulvic acids on the potato yield [22]. Humic and fulvic acids improved plant growth and yield quality of wheat, maize, tomato, pepper and cucumber [2,[22][23][24]. The effect of humic acids on plant growth depends of their source and concentration, and on the date and method (foliar or soil) of application, as well as the plant species and environmental conditions [9,17].
There is a relationship between leaf chlorophyll content and Soil Plant Analysis Development (SPAD) index [25]. Leaf SPAD values is related to nutrient plant status, especially nitrogen [26,27]. There was a relationship found between SPAD value and potato yield. A higher SPAD does not always guarantee a higher potato yield [28][29][30][31]. Plant-based biostimulants increased SPAD index and marketable yield of tomato and rocket [32][33][34].
To date, few studies have been focused on the effect of seaweed extract and humic acid application in early crop potato culture. The aim of the study was to determine the effect of foliar application of brown seaweed extracts and humic acids on the asssimilation area and chlorophyll content of very early potato cultivars. In the current study, it was hypothesised that seaweed extracts and humic acids could contribute to increasing assimilation area and chlorophyll content and, as a result, increase the early crop potato yield. The assumption was also made that the response to the application of these biostimulants depends on the cultivar and environmental conditions.
Experimental Site and Season
The study was carried out in central-eastern Poland (52°03ꞌ N, 22°33ꞌ E), over three growing season 2012-2014, on Luvisol with a low total nitrogen content, a high content of available phosphorus, a medium-to-high content of potassium and a low-to-medium content of magnesium, with an acidic-to-slightly-acid reaction. Spring triticale was grown as a potato forecrop. Farmyard manure was applied in autumn, at rate of 25 t•ha -1 , and mineral fertilizers were applied at rates of 80 kg N (ammonium nitrate), 35 kg P (superphosphate) and 100 kg K (potassium sulphate) per hectare in spring. Potato cultivation was carried out according to common agronomical practice.
The thermal and moisture conditions during the potato growth period were different ( Table 1). The mean air temperatures were above or similar to the long-term average. In 2012, total precipitation was similar and, in 2013 and 2014, above the long-term average, although it was unevenly distributed during the potato growth period. The most favourable hydrothermal conditions for early crop potato culture were in the warm and moderately wet growing season of 2012. The next year, 2013 was warm and with heavy rainfall, whereas 2014 was cool with heavy rainfall after plant emergence and a drought in the period of tuber growth.
Plant Material and Experimental Design
The field experiment was established in a split-plot design with three replications. The experimental factors were: (1) plant biostimulant; and (2) cultivar. The potato plants were treated with three biostimulants: Bio-algeen S90 and Keplak SL containing seaweed extracts, and HumiPlant based on humic and fulvic acids. Bio-algeen S90 is an extract from Ascophyllum nodosum which contains amino acids, vitamins, alginic acids and other active components of seaweeds, as well as macronutrients (N, P, K, Ca, Mg) and micronutrients (B, Fe, Cu, Mn, Zn, Se, Co). Kelpak SL is an extract from Ecklonia maxima containing auxin (11 mg•dm -3 ) and cytokinin (0.031 mg•dm -3 ). HumiPlant is an extract from leonardite which contains humic acid (12%) and fulvic acid (6%) as well as macronutrients (K, Ca, Mg, S) and micronutrients (Fe, Mn, B, Mo, Zn, Cu). The biostimulants were applied according to the manufacturers' recommendations: Bio-algeen S90-2 dm 3 •ha -1 at the beginning of leaf development stage (BBCH 10-11) and 2 dm 3 •ha -1 two weeks after the first treatment, Kelpak SL-2 dm 3 •ha -1 at the leaf development stage (BBCH 14-16) and 2 dm 3 •ha -1 two weeks after the first treatment, HumiPlant-2 dm 3 •ha -1 at the leaf development stage (BBCH 14-16) and 2 dm 3 •ha -1 one week after the first treatment. Potato plants sprayed with water were used as a control without a biostimulant.
The most popular very early potato cultivars (Denar, Lord and Miłek) in the research area were grown. In successive years, 6-weeks pre-sprouted seed potatoes were planted on April 12, April 18 and April 7 with a row spacing of 0.25 m and 0.675 m between rows. The plots were six rows wide and 4 m long (96 plants per plot). Potatoes were harvested 75 days after planting (the end of June).
Determination of Assimilation Area, Chlorophyll Content and Tuber Yield
At the tuber formation stage (BBCH 41-43), the assimilation area, leaf area index (LAI), specific leaf area (SLA), and chlorophyll content (SPAD value) were determined. The measurements were made on four successive randomized plants per plot. The assimilation area was measured by the weight method [36]. SLA was calculated as the ratio of assimilation area/weight of leaves [37].
The chlorophyll content was estimated with non-destructive methods using a portable SPAD-502 chlorophyll meter (Minolta, Osaka, Japan). The measurements were made on the youngest fully expanded leaf, i.e., the fourth or fifth leaf from the top.
The total and marketable tuber yield were determined. The marketable tuber yield constituted tubers with a transverse diameter above 30 mm, excluding cracked and deformed tubers. The marketable tuber yield was determined on the basis of the total tuber yield of ten successive plants per plot using a hand calibrator with a square hole.
Statistical Analysis
The results of the study were analysed statistically with an analysis of variance (ANOVA) for the split-pot design. The significance of differences between the compared averages was verified using Tukey's test at the significance level p ≤ 0.05.
Assimilation Area
The effect of biostimulants on the assimilation area depended on the weather conditions during potato growth ( Table 2). In the year with the highest air temperature and heavy rainfall after plant emergence (2013), the greatest enlargement of the assimilation area was caused by Kelpak SL, whereas in the year with the lowest air temperature and heavy rainfall after plant emergence (2014), the greatest enlargement of assimilation area was caused by Bio-algeen S90. The assimilation areas were larger, on average, by 0.0624 m 2 (11.5%) and 0.0941 m 2 (10%) respectively, and the leaf area index (LAI) was higher by 0.37 and 0.56 compared with the plants from the control group without a biostimulant. Regardless of the biostimulant applied, the assimilation area was largest in the year with the highest air temperature and moderate rainfall at the end of May (Table 3). The potato cultivars tested showed different responses to the biostimulants applied ( Table 2). The type of biostimulant had a greatest effect on the assimilation area of the 'Lord' cultivar. The greatest enlargement of the assimilation area of 'Lord' was caused by Bio-algen S90. Following the application of this biostimulant, the assimilation area of 'Lord' was larger, on average, by 0.1583 m 2 (24.5%) and the LAI value was higher by 0.94 compared with the plants from the control without biostimulant. The differences were highest in the year with a low air temperature and heavy rainfall after the plant emergence (2014). Despite the biostimulant applied, the assimilation area was higher for 'Miłek' than for 'Denar' and 'Lord' (Table 3).
Only in a warm and moderately wet growing season (2012), following application of Bio-algeen S90 and HumiPlant, was the specific leaf area (SLA) higher, on average, by 0.29 m 2 •kg -1 compared with the plants from the control group without biostimulant ( Table 2). With the use of Kelpak SL, the difference was smaller and not statistically confirmed. The SLA depended to a greater extent on the weather conditions during potato growth. Irrespective of the treatment (with or without biostimulant), the SLA was highest in the year with the highest air temperature and heavy rainfall after plant emergence ( Table 3). The type of biostimulant and cultivar interaction effect on SLA was not statistically confirmed ( Table 2). Regardless of the treatment, the SLA values of the potato tested cultivars were similar (Table 3).
Chlorophyll Content (SPAD Value)
The biostimulants used in the experiment had no significant effect on the chlorophyll content in leaves ( Figure 1). The SPAD value depended to a greater extent on the cultivar and weather or soil conditions during potato growth. Irrespective of the treatment (with or without biostimulant), the SPAD values were higher for 'Denar' and 'Lord' than 'Miłek'. The SPAD was highest in the warm and wet growing season (2013) and, at the same time, the highest content of total nitrogen and available magnesium in soil. (Figure 2). Means followed by the same letters do not differ significantly at p ≤ 0.05.
Relationship between Tuber Yield, Assimilation Area and Chlorophyll Content (SPAD Value)
The biostimulants used in the experiment had no effect on the weight of leaves [38], but caused enlargement of the assimilation area (Table 4). Over the three years of the study, the assimilation area was larger, on average, by 0.0505 m 2 (7%) and the LAI was higher by 0.30 compared with the plants from the control group without a biostimulant. The biostimulants had no significant effect on the SLA and SPAD (Figure 1). The biostimulants used in the experiment had a significant effect on the tuber yield [38]. The yield-increasing effects of biostimulants were comparable (Table 5). In the three years of the study, the total tuber yield was higher, on average, by 2.64 t•ha -1 (7.7%) and marketable tuber yield (diameter above 30 mm) by 2.15 t•ha -1 (6.5%). The yield-increasing effect of biostimulants depended on weather conditions during the potato growing season. Bio-algeen S90 and Kelpak SL caused the highest increase in tuber yield in the warm and very wet growing season (2013), and HumiPlant in the year with a low air temperature and a drought in the period of tuber growth (2014). The tuber yield was not significantly correlated with the weight and assimilation leaf area or LAI (Table 6). A significant negative correlation was found between the marketable tuber yield and SLA. No significant correlation was found between the marketable tuber yield and SPAD value.
Effect of Experimental Factors on Assimilation Area, Chlorophyll Content and Tuber Yield
The effect of the experimental factors and their interactions on potato assimilation area and chlorophyll content (SPAD value) are presented in Table 7. Table 7. Effect of experimental factors on assimilation area, chlorophyll content (SPAD value) and tuber yield.
Assimilation
Leaf Area LAI SLA SPAD
Discussion
In sustainable crop production, biostimulants play an important role in improving plant growth and crop quality. Assimilation area and chlorophyll content are important parameters of assessment plant growth. The biostimulants used in the experiment caused enlargement of assimilation area, but had no effect on the chlorophyll content (SPAD value) in leaves of very early potato cultivars. SPAD value depended on the cultivar and weather or soil conditions to a greater extent. The effect of foliar application of seaweed extracts on potato assimilation area was comparable to humic and fulvic acids. In the three years of the study, following biostimulant application, the average leaf area index (LAI) was 4.64, being higher by 0.30 compared to the average for the untreated control group. Potato cultivars showed different responses to the applied biostimulants. Studies have shown the highest light absorption efficiency values at the LAI value of 3, which corresponded to maximum ground cover. If potato LAI exceeds 3, the intercepted photosynthetically active radiation value changes very little [39,40]. According to Howlader and Hoque [41], irrespective of potato cultivars, LAI increased progressively over time, reaching a peak at 60 days after planting and thereafter declining. The rate of assimilation area expansion showed the interaction between genotype and environment and varied by year [42], which was confirmed in the present study. The effect of seaweed extracts on potato assimilation area depended on the weather conditions after plant emergence. In the year with the highest air temperature and heavy rainfall after plant emergence, the assimilation area was larger after the application of Kelpak SL (Ecklonia maxima), whereas in the year with the lowest air temperature and with heavy rainfall after plant emergence, the assimilation area was larger after the application of Bio-algeen S90 (Ascophyllum nodosum). Potato plants are very sensitive to heat stress. In general, heat stress increases plant height, reduces leaf size, increases leaf chlorophyll content, and severely reduces tuber mass [43]. Kelpak SL contains auxins and cytokinins in a ratio of 350/1. Exogenous auxin plays an important role in plant stress resistance. The action of auxin depends on its concentration, the light conditions and carbohydrate content in the plant [44]. Exogenous cytokinins also play an important role in plant adaptation to environmental stresses [45]. Cytokinins present in the seaweed extracts stimulate cell division, resulting in enlarged leaf area [7,8], which was confirmed in the present study.
The leaf area index describes the growth of lowland fields, whereas the growth of individual plants is characterized by the specific leaf area (SLA). Biostimulants caused enlargement of the assimilation area, but had no effect on the SLA. The SLA for potato depends on the cultivar and growth stage, and temperature [42], which was confirmed in the present study. Early foliar expansion of potato is associated with a strong increase in SLA [41].
Foliar or soil application of Ascophyllum nodosum extracts caused an increase in the chlorophyll content of some agriculture (barley, wheat, maize) and horticulture (French bean, tomato, pepper, strawberry) plants [8,11,12], which was not confirmed in the present study. A study carried out in Egypt showed that the application of humic acid under water stress conditions enhanced the chlorophyll content of very early potato 'Spunta' grown on sandy soil [18], which was not confirmed in the present study with very early potato cultivars grown on loamy soil (Luvisol). A one-year study carried out in Iraq showed that foliar application of humic and fulvic acids caused an increase in the chlorophyll content of medium-early potato cultivar [13]. The effect of humic acids depends on their source and concentration, and on the date and method of application, as well as the plant species and cultivar [9]. The increase in chlorophyll alone does not necessarily result in higher yields [17,26].
The biostimulants used in the experiment enhanced tolerance to abiotic stress and improved crop quality. In the three years of the study, the marketable tuber yield (diameter above 30 mm) was higher, on average, by 2.15 t•ha -1 . Bio-algeen S90 and Keplak SL containing seaweed extracts produced better results in a warm and very wet growing season, whereas HumiPlant based on humic and fulvic acids produced better results in a year with lower air temperature and with drought periods during potato growth.
A correlation between the tuber yield and assimilation area was not found. Li et al. [46] found a significant positive correlation between LAI and tuber yield, which suggests that the enlargement of leaf area could enhance the export of photosynthetic products and cause an increase in tuber yield. According to Ascione et al. [47], the tuber growth rate is only slightly correlated with LAI, and still less so with SLA, which was not confirmed in the present study. A significant negative correlation was found between the total and marketable (diameter above 30 mm) tuber yield and SLA.
No correlation was found between the tuber yield of three very early potato cultivars and SPAD value measured on the fourth or fifth leaf from the top at the tuber formation stage (BBCH 41-43), which suggest that the biostimulants used in the experiment had no effect on the plant nitrogen status. Bărăscu et al. [30] found a significant negative correlation between SPAD measured on the fourth and fifth leaves from the top and the tuber weight of two mid-early potato cultivars, which could have been associated with oxidative stress [29]. SPAD index as an indicator of crop nitrogen status may be used for the prediction of the potato yield, however a higher SPAD does not always guarantee a higher tuber yield [26,28,31]. SPAD value is a useful indicator for selecting the high yield cultivars in the early period, however, no single threshold leaf SPAD value can be used for all potato cultivars. The SPAD value can predict the level of tuber yield if the value is calibrated for a particular potato cultivar [28,31]. Establishing threshold SPAD value is quite difficult due to the influence of climate and technical factors. SPAD values can be affected by leaf age and position, as well as, time of the day [26,27]. As a rule SPAD measurements are carried out on the third-fifth leaf from the top.
Recently it was demonstrated that there is a significance difference in SPAD values between the upper and lower leaves among potato cultivars. It was shown that cultivar affects the SPAD values of the fourth and eighth leaf, but does not affect SPAD value of the fourth-eighth leaves and the difference between SPAD of the fourth and eighth leaf. Therefore the SPAD values of the fourtheighth leaves could be applied as a general index of nitrogen status across different potato cultivars [27].
Conclusions
In conclusion, the foliar application of seaweed extracts Ascophyllum nodosum (Bio-algeen S90) and Ecklonia maxima (Kelpak SL), as well as humic and fulvic acids from leonardite (HumiPlant), resulted in enlargement of the assimilation area of very early potato cultivars, but had no effect on the SLA or chlorophyll content (SPAD value). The assimilation area was larger, on average, by 0.0505 m 2 (7%), and LAI was higher by 0.30 compared with the plants from the control group without a biostimulant. The SLA and SPAD depend on the cultivar and weather conditions, or nitrogen and magnesium content, in soil to a greater extent. These biostimulants enhanced abiotic stress tolerance and increased marketable tuber yield (diameter above 30 mm) 75 days after planting (the end of June), on average, by 2.15 t•ha -1 . Bio-algeen S90 and Keplak SL containing seaweed extracts produced better results in a warm and very wet growing season, whereas HumiPlant based on humic and fulvic acids produced better results in a year with lower air temperature and with drought periods during potato growth. No correlation was found between the tuber yield and assimilation area or between the tuber yield and SPAD value, although a significant negative correlation was found between the tuber yield and SLA.
Author Contributions: Conceptualization, W.W. and T.D.; methodology and formal analysis, W.W.; investigation, W.W. and T. D.; writing-original draft preparation, W.W. and T.D.; writing-review and editing, W.W. All authors have read and agreed to the published version of the manuscript.
Funding: This research was financed from the science grant granted by the Polish Ministry of Science and Higher Education, research theme number 218/05/S.
Conflicts of Interest:
The authors declare no conflict of interest.
|
2020-03-19T10:52:46.396Z
|
2020-03-12T00:00:00.000
|
{
"year": 2020,
"sha1": "e5979ad6f055b80cf7e71dd8ed58d0503b12ef22",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2073-4395/10/3/387/pdf",
"oa_status": "GOLD",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "9ae0370000da53736367a0c94eb8d0c3f3d6ee22",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"Environmental Science"
],
"extfieldsofstudy": [
"Biology"
]
}
|
79640171
|
pes2o/s2orc
|
v3-fos-license
|
Laparoscopic nephropexy: Treatment outcome and quality of life
MATERIALS AND METHODS: For a period from March 2014 to March 2015, a total of 8 women at an average age of 54 years were operated on in our clinic . Four of the patients were with nephroptosis of the left kidney, three of the right one and one had bilateral nephroptosis. Most of the patients complained of pain and discomfort in active movement, only one individual had complaints consisting of intermittent macroscopic hematuria. One patient had been operated on in the past by a classic open method of the same kidney. Preoperatively, for all patients, intravenous urography in supine and standing position was performed. All patients were operated on trans-peritoneally through 3 ports: 1x10 mm and 2x5 mm. The kidney was completely mobilized and kidney fat was dissected. The upper and middle pole of the kidney were fixed to musculus psoas major, using a single non-absorbable suture and intracorporeal technique for tying.
INTRODUCTION
Nephropexy, which is supposed to be a popular surgical procedure, has been abandoned for many years. Surgical fixation or suspension of mobile kidneys was first described in 1881 by Eugene Hahn (1). Although more than 150 different surgical techniques and modifications of nephropexy have been described, there are no definitive treatments and almost all urologists have been hesitant to treat mobile Deyan Anakievski, Alexander Hinev, Rostislav Marinov et al.
all patients. Patients were operated on under general anesthesia and placed in a lateral position of 20 to 30 degrees with just one patient with bilateral nephroptosis on her back. All patients were operated transperitoneally. With the help of a Veress needle insufflation we created pneumoperitoneum to 14mm/Hg and then placed one 10-mm port for the camera. Additional ports were placed under optical control: one 5-mm port subcostally in the middle clavicular line and another 5-mm one in the anterior axillary line. The line of Toldt was incised between the liver and colon ascendens, the kidney was mobilized fully at all sides by blunt and sharp dissection. The kidney was fixed to the front lateral side with 2/0 non-absorbable suture starting the stitching from the upper to the lower pole of the renal capsule. Each suture was passed through the renal parenchyma itself kidney due to the high invasiveness of the available methods compared to the severity of the symptoms (2)(3)(4)(5). The procedure requires a large incision in the skin and division of multiple muscle layers. Postoperatively, the patients usually have no pain caused by renal ptosis, but suffer from pain due to the wound or dysesthesia caused by treatment. However, there is a certain group of patients suffering from severe symptoms due to renal ptosis, although the number of such patients is very low (2). The laparoscopic procedure is ideal for nephropexy as it produces less invasiveness and minimal post-operative pain (6,7). However, after the introduction of laparoscopic procedures for nephropexy by Urban in 1993 (4), this type of surgery became the first treatment of choice for nephroptosis being a minimally invasive procedure. We have described here our experience and subsequent quality of life of patients treated with laparoscopic nephropexy and discussed its indications and surgical technique.
Patients:
For the period from March 2014 to March 2015 a total of 8 women at an average age of 54 years were operated on in our clinic. Four of the patients had left nephroptosis, three had right one and one had bilateral nephroptosis. Their BMI ranged from 18 to 30, with an average of 22 kg/m 2 . The most frequent symptom was pain in the lumbar region during active movement, when standing upright, or in some cases -a palpable mass in the lower abdomen. Moreover, one of the eight patients had a history of intermittent gross hematuria and another was operated on in the past by the classic open method of the same kidney. Although various studies were performed to determine the cause of symptoms no reason except for nephroptosis was discovered. Seven patients (86%) had previously undergone abdominal surgery including laparoscopic cholecystectomy, hysterectomy and classic nephropexy. Preoperatively, for all patients, intravenous urography and ultrasound in supine and upright position was performed. A descent of the kidney at distance of 2 or more vertebrae was found in all patients on the IVU (Table 1).
Surgical technique:
Preoperative bowel preparation with glycerin enema and low dose of heparin was administered to on the one side and the fascia of the quadratus lumborum muscle on the other side, and the knot was intracorporeally tied. In all patients we used four threads. We placed a contact drain. The skin incision was closed with absorbable sutures (8).
RESULTS
The perioperative data and the characteristics of the patients are shown in Table 2. There were no perioperative complications that required conversion in any of the cases. All operations were performed with minimal blood loss and an average operating time of 45 min. The oral intake of fluids and light food began at 10 hours after surgery in all patients, and parenteral analgesia was achieved with NSAIDs. The average hospital stay was 4 days. Patients returned to their daily activities on the second postoperative week, and in four weeks returned to work ( Table 2). Postoperatively, the patients were tracked and monitored by ultrasound examination. At 3 months we did IVP, which showed the correct location of the kidney. All patients remained asymptomatic for an average of 11 months after surgery.
DISCUSSION
Nephroptosis is a common finding in routine IVU, with an incidence of 20% in healthy patients (9). The majority of patients are asymptomatic and nephroptosis does not require treatment. However, if symptoms occur, such as pain, hematuria or recurrent urinary tract infection, medical treatment is necessary. Although some patients are treated conservatively, such as with dieting or wearing special abdominal support bandages, usually these methods are not effective. There is no consensus on when and what type of surgery is an indication for the treatment of nephroptosis. Boeminghaus (10) differentiates patients with nephroptosis into three groups: (I) ptosis without symptoms that do not require treatment; (II) patients with symptoms but no functional changes; and (III) symptomatic patients with symptoms, as well as functional and morphological changes occasionally. He recommends performing nephropexy only in the third group of patients. There are three different operational techniques, which offer high success rate: 1. fixation using the renal capsule; 2. fixation with the use of various foreign materials; 3. fixation with the use of the patient's own tissues, such as muscle or fascial flap. All three have the same technique steps. These are the immobilization of the kidney in the possible via the cephalic retroperitoneal position, release of urine obstruction (retention) due to nephroptosis, fixation of the kidney on its axis, so that the lower pole is positioned laterally in order to avoid traction on the vascular pedicule or ureter (11). Whichever technique is used, all authors report good results with very high success rates (12,13). Rassweiler et al. (14) provide an indication of laparoscopic nephropexy in patients who after completion of IVU experience a descent of the kidney with at least two vertebrae and when there is existence of two objective symptoms or an objective and a subjective symptom alone. Also, the use of extraperitoneal access offers fewer adhesions between kidney, colon and peritoneum, although many authors prefer the transperitoneal access and report that it is better than the extraperitoneal one because of the opportunity to work in a larger space and the opportunity for intracorporeal tying (6).
CONCLUSION
Laparoscopic transperitoneal nephropexy is a safe and effective procedure and a promising method for correction of symptomatic documented nephroptosis.
Median estimated blood loss, ml 50
Conversion rate, n 0 No. of sutures for nephropexy 4
|
2019-03-17T13:07:54.573Z
|
2017-06-28T00:00:00.000
|
{
"year": 2017,
"sha1": "dbd0ccfc2757f83c9adf56663e159a8ce7b9de85",
"oa_license": null,
"oa_url": "http://journals.mu-varna.bg/index.php/ssm/article/download/1945/3510",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "6b8374a3d35a60ad609ca55ccaa7f3206a7c3be1",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
249642742
|
pes2o/s2orc
|
v3-fos-license
|
Proton and neutron exchange as a prelude to fusion at near-barrier energies
Systematic examination of fusion for $^{39,41,45,47}$K + $^{28}$Si and $^{36,44}$Ar + $^{28}$Si provides insight into the impact of neutron and proton exchange on fusion for nuclei at and near the N=20 and N=28 shells. Comparison of the reduced excitation functions reveals a marked difference between the behavior of open-shell and closed-shell systems. While coupled channels calculations provide a good description for the closed-shell nuclei they significantly under-predict the fusion cross-section for open-shell nuclei. The observed trends are examined in the context of a potential energy surface, including shell effects, and multi-nucleon exchange with consideration of Pauli-blocking.
I. INTRODUCTION
Nuclear reactions of neutron-rich nuclei play a key role in nucleosynthesis both in astrophysical environments [1] as well as terrestially in accelerator-based experiments [2,3]. One topic of particular interest both theoretically as well as experimentally is the question about the enhancement or suppression of the fusion cross-section for neutron-rich nuclei [4][5][6][7]. For extremely neutron-rich nuclei, as a result of their weakly bound valence neutrons, one might observe reduced spatial coupling of the neutron and proton distributions and the emergence of novel neutron dynamics which enhance the fusion crosssection. At energies near the fusion barrier the fusion process is particularly interesting as the timescale of the collision is sufficiently long for collective dynamics of the neutron and proton density distributions to influence the fusion process. It is presently unclear how this dynamics is impacted by the shell structure of the initial nuclei. * desouza@indiana.edu Although it is well established that inelastic excitation of the two nuclei as they approach [8] and transfer of one or more nucleons [9,10] can modify the fusion probability in particular systems, a more comprehensive understanding is presently lacking [10].
Theoretical calculations of the fusion using a densityconstrained TDHF approach found an enhancement of fusion for the asymmetric system 24 O + 16 O as compared to 16 O + 16 O [11]. This enhancement is understood as resulting from neutron transfer which modifies the potential between the nuclei, lowering the barrier. For even more neutron-rich nuclei-at the limit of stabilitynamely 24 O + 24 O, fusion is suppressed relative to 24 O + 16 O. This suppression of fusion for symmetric neutronrich systems has been attributed to a repulsive Pauli potential arising from the overlap of the neutron-rich tails [12]. However, these calculations are one-body and neglect many-body correlations which could enhance correlated transfer. Moreover, they are limited in that they only reflect the average behavior of the system. Experimental evidence of fusion enhancement for neutron-rich nuclei also exists. Neutron exchange of va-lence neutrons in Ni + Ni systems were proposed as possibly responsible for an observed increase in the sub-barrier fusion cross-section [13]. Recent measurements provide further evidence of fusion enhancement due to the presence of a one-neutron halo ( 15 C) [14] or an unpaired neutron ( 19 O) [15]. However, experimental measurements confirm that the neutron-richness of the colliding nuclei alone is not the only factor impacting the fusion probability as indicated by examination of Ca+Ca collisions. While fusion of a 48 Ca projectile with a 40 Ca target nucleus is enhanced as compared to a 40 Ca projectile [16], fusion of 48 Ca + 48 Ca is suppressed below the barrier [17]. It has recently been observed that, at above-barrier energies, after accounting for systematic size and Coulomb effects, the fusion cross-section for open shell nuclei near the N=20 and N=28 shells is larger than that of the closed-shell nuclei [18]. This result has been interpreted as enhanced binding of the closed-shell nuclei as compared to open-shell nuclei as they merge.
In the present work, motivated by these prior abovebarrier results, we examine for the first time fusion in 39,41,45,47 K and 36,44 Ar + 28 Si and explore the role of shell structure and N/Z equilibration on the fusion crosssection.
II. EXPERIMENTAL DATA
Radioactive beams of K and Ar ions were produced by the coupled cyclotron facility at MSU-NSCL and thermalized in a linear gas stopper before being re-accelerated by the ReA3 linac [18]. The re-accelerated beam was transported to the experimental setup where it impinged upon the 28 Si target. Details on the experimental setup have been previously published [19].
Contaminants in the radioactive beam were identified and rejected on a particle-by-particle basis by performing a ∆E-TOF measurement [18,19]. The target composition was characterized using Rutherford Backscattering measurement (RBS) and confirmed using X-ray Photoelectron spectroscopy [20]. This RBS measurement revealed a 28 Si thickness of 258 ± 10 µg/cm 2 and an oxygen thickness of 98 ± 4 µg/cm 2 . The experimental resolution allowed reaction products from the fusion of the beam with 28 Si and 16 O to be distinguished [19]. The intensities of the K and Ar beams incident on the target ranged between 1.0 x 10 4 ( 44 Ar/s) and 4.5 x 10 4 ( 39 K/s). Fusion of the incident K and Ar ions with the oxygen nuclei has been previously published [18].
Fusion of K (Ar) ions with the 28 Si target results in a compound nucleus (CN) of As (Ge). De-excitation of the CN via neutron, proton and α emission deflects the resulting evaporation residue (ER) from the beam direction. The ER was detected in annular Si(IP) detectors (1.0 • < θ lab < 7.3 • ) and distinguished from scattered beam using the energy/time-of-flight (ETOF) technique [19].
Extraction of the fusion cross-section, σ F , is achieved by measuring the yield of ERs and utilizing the relation where N ER is the number of evaporation residues detected, N I is the number of beam particles of a given type incident on the target, t is the target thickness, and ǫ ER is the detection efficiency. The number of detected residues, N ER , is determined by summing the number of detected residues identified by the ETOF technique. Uncertainty in identifying an ER associated with fusion on 28 Si is reflected in the error bars presented. Beam particles with the appropriate identification in the ∆E-TOF map provided the measure of N I . A statistical model was employed to describe the de-excitation of the fusion product. Together with the geometric acceptance of the experimental setup this provided the detection efficiency, ǫ ER . which varied between ≈ 78-84% over the entire energy range. An effective means of comparing the fusion excitation function for different systems is the use of the reduced excitation function [7]. Comparison of fusion for an isotopic chain allows utilization of the simplest scaling prescription. The systematic increase in size with increasing mass number A is accounted for by scaling the fusion cross-section σ F by the quantity (A Differences in the Coulomb barrier for the different systems are considered for by examining the dependence of this reduced cross-section on the incident energy relative to the Coulomb barrier. The Coulomb barrier, V C , is taken ). This simple accounting of the Coulomb barrier suffices as significant interpenetration of the charge distribution does not occur outside the fusion barrier.
Presented in Fig. 1a are the reduced fusion excitation functions for 39,41,45,47 K + 28 Si. For all systems the reduced fusion cross-section above the barrier is similar. Below the barrier however, significant differences are apparent between the different systems. The data clearly organize into two groups: one associated with 39 K and 47 K (closed neutron shells at N=20 and N=28 respectively) and the other with 41 K and 45 K (open neutron shells). This similarity of the reduced fusion cross-section for 39 K and 47 K projectiles indicates that the density distributions, relevant to fusion, for the two closed-shell K isotopes are similar when scaled by A 1/3 . In marked contrast, a larger reduced fusion cross-section is evident for the open-shell 41 K (N=22) and 45 K (N=26), beyond the systematic A 1/3 scaling. This enhancement of the fusion cross-section for the open-shell nuclei increases with decreasing energy below the barrier. The same enhancement at sub-barrier energies is observed for the open-shell 36,44 Ar nuclei as compared to the closed-shell K isotopes in Fig. 1b. This present observation of the difference in the fusion of open-shell and closed-shell was confirmed by reexamining the literature. Enhancement of the fusion cross-section for an open-shell nucleus ( 124 Sn) as compared to a closed-shell nucleus ( 132 Sn) is also evident for 40,48 Ca targets [21].
Presented in Fig. 2 are the reduced fusion excitation functions grouped by their proximity to the N=20 and N=28 shells. In Fig. 2a one observes that 36 Ar and 41 K exhibit similar excitation functions with a marked enhancement of the reduced fusion cross-section as compared to the closed-shell 39 K (N=20). This result indicates that the presence of two holes below the closedshell ( 36 Ar) is effectively the same as the presence of two particles above the closed-shell ( 41 K) in determining the reduced fusion cross-section. A similar enhancement in the reduced fusion cross-section is observed at the N=28 shell for the presence of two holes in 44 Ar and 45 K as compared to 47 K.
III. COMPARISON WITH THEORETICAL MODELS
The simplest description of fusion involves the interaction of the density distributions of the two interacting nuclei. For a non-adiabatic interaction (sudden approximation) consideration of the ground-state density distri- butions suffices. For adiabatic collisions, collective modes in the colliding nuclei can be excited and also need to be considered. Inclusion of these modes in a coupled channels (CC) formalism results in an increase in the fusion cross-section at energies near and below the Coulomb barrier [22,23]. To investigate whether the observed fusion excitation functions can be described by the interaction of the density distributions of the projectile and target nuclei, the São Paulo model was used. The São Paulo potential (SPP) [24] is a local equivalent double folding of the projectile and target matter densities on the zero-range interaction. Prior work demonstrated the sensitivity of the fusion cross-section to accurate ground-state density distributions [18]. To provide reasonably accurate matter density distributions, which include two-body correlations, we performed Dirac-Hartree-Bogoliubov (DHB) calculations [25]. The correlations in the DHB calculations of the present work are limited to surface-pairing correlations. These correlations can make subtle modifications to the nuclear surface, extending and modifying the nuclear density. The details of these mean field calculations using an axially-symmetric self-consistent approximation are reported in Ref. [26].
Using the ground-state DHB matter distributions for both the projectile and 28 Si target nuclei, the SPP was generated and used to calculate the fusion cross-section. The theoretical predictions, represented by the dashed lines, are compared with the experimental data in Fig. 3. Comparison of these one-channel (DHB-OC) theoretical predictions with the experimental excitation functions is revealing. For the closed neutron shell isotopes 39,47 K, the DHB-OC calculations provide a reasonable prediction of the excitation function over the entire energy interval measured although the model calculations lie slightly below the experimental data particularly in the sub-barrier regime. In the case of the open neutron shell 41 K, 45 K and 36 Ar, the model dramatically underpredicts the measured cross-sections, particularly at subbarrier energies. This under-prediction for the case of the open-shell nuclei suggests that the ground-state configu- rations alone are insufficient in describing the measured cross-sections. In the case of 44 Ar insufficient data exists at low energy to draw a definitive conclusion.
As coupling to low-lying collective modes acts to increase the fusion cross-section we have performed coupled-channels (CC) calculations to investigate the extent to which the presence of low-lying states increases the fusion cross-section. The 1.779 MeV, 2 + and 4.618 MeV, 4 + first states of the target were considered. The coupling to the low-lying projectiles states does not produce a considerable effect on the fusion cross-section. To account for the couplings between the low-lying states the transition probabilities were taken from Ref. [27].
The results of the CC calculations are shown in Fig. 3 as solid lines. In the case of the closed-shell nuclei, 39 K and 47 K, inclusion of the excitations considered provides a good description of the fusion cross-section. However, in the case of the open-shell nuclei the experimental data are significantly enhanced relative to the CC calculations with inclusion of low-lying excitations. It is particularly interesting to note that the magnitude of the enhancement is much larger than the increase due to the inclusion of inelastic excitation in the CC calculations. This enhancement suggests that transfer might be occurring prior to fusion.
Neutron transfer prior to fusion is often proposed as responsible for an enhancement in the fusion cross-section [13,28,29]. For a system with zero Q-value for two neutron transfer, 60 Ni + 58 Ni, inelastic excitations dominate and neutron transfer plays a negligible role [8]. When one of the colliding nuclei is neutron-rich relative to its collision partner, as in the case of 40 Ca + 96 Zr positive Q-value neutron transfer channels act to increase the fusion cross-section at sub-barrier energies as compared to 40 Ca + 90 Zr [30][31][32][33]. We present the relevant Q-values in Table I [34]. With the exception of 39 K the Q-value for two-neutron transfer in the other K isotopes is positive. Transfer of one neutron from 39 K to 28 Si is -4.604 MeV, while for 47 K it is slightly positive (+0.1 MeV). Nonetheless, the fusion excitation function for these two nuclei with 28 Si is comparable. The Q-value for neutron transfer for the open-shell cases 41,45 K lies between that of 39 K and 47 K yet the fusion excitation functions of the open-shell cases differ from those of the closedshell. Clearly the observed behavior of the experimental fusion excitation functions cannot be understood simply by consideration of the Q-value for neutron transfer.
The consideration of the Q-value for neutron transfer ignores the role of protons during the fusion process. Description of fusion using a density-constrained time- dependent Hartree-Fock (DC-TDHF) model allows the neutron and proton density distributions to evolve as the collision proceeds while incorporating all of the dynamical entrance channel effects such as neck formation, particle exchange, internal excitations, and deformation effects [35]. Such calculations for the system 132 Sn + 40,48 Ca clearly indicate the correlated flow of neutrons and protons. Unfortunately, for nuclei with unpaired nucleons the DC-TDHF calculations are considerably more challenging with a significant sensitivity to the inclusion of pairing [36]. We therefore consider qualitatively how initial nucleon exchange could impact fusion at near and sub-barrier energies using a conceptually simple physical framework. When the two colliding nuclei are within the range of the strong force nucleon exchange is allowed. This exchange of protons and neutrons is governed by a potential energy surface (PES). Flow of nucleons between the two nuclei is stochastic and allows equilibration of mass, charge, and energy [37]. The differential flow of neutrons and protons between the colliding nuclei results in both a net change in the atomic and mass numbers as well as excitation of the system. The nucleon flow is mitigated by Pauliblocking of scattering into occupied states. Independent of the gradient of the potential, proton exchange is initially disfavored relative to neutron exchange because of the Coulomb barrier between the two nuclei. This physi-cal picture was largely successful in explaining the charge and mass distributions associated with strongly damped collisions along with the characteristic dissipation of kinetic energy [38]. A key factor driving the equilibration of N/Z in strongly-damped heavy-ion collisions is the gradient of the PES in the vicinity of the entrance channel [39]. A stochastic mean field approach utilizing this nucleon exchange framework successfully explained the dispersion of the mass distribution in 58 Ni + 60 Ni for damped collisions [40]. It was hypothesized that for slightly more central collisions that resulted in fusion such a physical picture should still be valid. Unfortunately, the diffusion approach employed does not allow a description of the transition from multi-nucleon transfer to fusion [40]. We emphasize that in the present work we only utilize this physical picture to understand the factors influencing the initial neutron and proton exchanges prior to fusion.
To assess the factors influencing the initial nucleon exchanges, the PES was calculated for all binary combinations of a colliding system. The PES calculated corresponds to the liquid drop energy modified by shell corrections as well as a proximity interaction [38]. The surface was calculated at the strong absorption radius (approximately 10 fm in all cases shown) for zero impact parameter. As our aim is a qualitative description for these near and sub-barrier collisions and the systems considered are similar in mass asymmetry, ignoring the role of angular momentum in modifying the surface is justified.
The PES for each of the four K + 28 Si systems is displayed in Fig. 4. Arrows indicate the gradient of the potential in the NZ plane with the initial projectile-target combination indicated by the solid (red) symbol. The magnitude of the gradient is indicated by the numbers (in MeV) adjacent to selected arrows.
To begin we examine the cases of extremes in neutronrichness which nonetheless exhibit the same reduced fusion excitation function. In the case of 39 K, the projectile-target combination already lies along the valley of the PES in the NZ plane. Therefore, correlated neutron and proton exchange is required in order to maintain N/Z equilibrium. While any initial proton transfer is disfavored because of the Coulomb barrier, proton transfer from 28 Si to 39 K is additionally suppressed by Pauli blocking [38]. This suppression of initial proton exchange suppresses the neutron exchange.
In the case of 47 K, the PES is quite different. The initial system has a significant gradient to decrease the neutron number and increase the atomic number of the 47 K nucleus. While neutron transfer out of the K nucleus is favored, proton pickup from the 28 Si is also favored due to the large N/Z asymmetry of the system. Pauli blocking of initial proton transfer limits the ability of the system to follow the gradient of the PES and attain N/Z equilibrium in an effective manner.
For the open neutron shell nuclei, neutron transfer is not hindered by the energy cost of breaking the neutron shell. For 41 K, as indicated by the PES, transfer of a neutron from 41 K to 28 Si can occur without any driving force for proton transfer. Net transfer of one neutron in this physical picture corresponds to multiple neutron exchanges. These multiple neutron exchanges excite the K nucleus which lessens the Pauli-blocking of subsequent proton exchanges. Subsequent proton transfer into or out of the K nucleus are equally energetically favorable as indicated by the PES. One might hypothesize that these initial neutron exchanges, not just the net transfer of one neutron, by reducing the Pauli-blocking act to increase the fusion probability.
The case of the 45 K is intermediate between that of 41 K and 47 K and more difficult to interpret. While pickup of a proton by the 45 K is favored along with loss of a neutron, the magnitude of the gradient is less than in the 47 K case. The smaller driving force for proton pickup relative to 47 K suggests a lesser role of proton transfer on the fusion cross-section.
IV. CONCLUSIONS
Comparison of the fusion excitation functions for 39,41,45,47 K + 28 Si and 36,44 Ar + 28 Si reveals that at sub-barrier energies the open neutron shell nuclei of 41,45 K manifest a significantly larger reduced fusion crosssection as compared to the closed neutron shell isotopes 39,47 K.
For the closed-shell nuclei, the use of Dirac-Hartree-Bogoliubov (DHB) ground state densities in the São Paulo fusion model provided a reasonable description of the data -one that was improved by inclusion of lowlying states of the 28 Si. For the open-shell nuclei, use of the DHB densities, even with the inclusion of the excited states, significantly under-predicts the measured crosssections, particularly below the barrier. Q-value calculations of neutron transfer alone are unable to explain the similarity in cross-section for the closed-shell nuclei and the enhancement for the open-shell nuclei. If transfer is the reason for the enhancement, a slightly more expansive perspective is required.
Consideration of the energetics involved with both proton and neutron exchange, along with Pauli-blocking, provided insight into the difference between the closedshell and open-shell nuclei. A more quantitative description of the observations requires development of a more complete theoretical description, one which properly accounts for multi-nucleon transfer and Pauli-blocking in the initial stages of the collision.
|
2022-06-15T01:15:49.522Z
|
2022-06-13T00:00:00.000
|
{
"year": 2022,
"sha1": "8faf0735dd24ae3fd58a337c73cb0cf7818c388e",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/2206.06436",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "8faf0735dd24ae3fd58a337c73cb0cf7818c388e",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
22749777
|
pes2o/s2orc
|
v3-fos-license
|
Numerical simulation of a rare winter hailstorm event over Delhi , India on 17 January 2013
This study analyzes the cause of the rare occurrence of a winter hailstorm over New Delhi/NCR (National Capital Region), India. The absence of increased surface temperature or low level of moisture incursion during winter cannot generate the deep convection required for sustaining a hailstorm. Consequently, NCR shows very few cases of hailstorms in the months of December-January-February, making the winter hail formation a question of interest. For this study, a recent winter hailstorm event on 17 January 2013 (16:00–18:00 UTC) occurring over NCR is investigated. The storm is simulated using the Weather Research and Forecasting (WRF) model with the Goddard Cumulus Ensemble (GCE) microphysics scheme with two different options: hail and graupel. The aim of the study is to understand and describe the cause of hailstorm event during over NCR with a comparative analysis of the two options of GCE microphysics. Upon evaluating the model simulations, it is observed that the hail option shows a more similar precipitation intensity with the Tropical Rainfall Measuring Mission (TRMM) observation than the graupel option does, and it is able to simulate hail precipitation. Using the modelsimulated output with the hail option; detailed investigation on understanding the dynamics of hailstorm is performed. The analysis based on a numerical simulation suggests that the deep instability in the atmospheric column led to the formation of hailstones as the cloud formation reached up to the glaciated zone promoting ice nucleation. In winters, such instability conditions rarely form due to low level available potential energy and moisture incursion along with upper level baroclinic instability due to the presence of a western disturbance (WD). Such rare positioning is found to be lowering the tropopause with increased temperature gradient, leading to winter hailstorm formation.
Introduction
An unusual winter hailstorm occurred over National Capital Region (NCR)/New Delhi, India on 17 January 2013 (16:00-18:00 UTC). Figure 1 shows satellite image of exten-25 sive cloud cover over the north Indian region during this time. Heavy precipitation was 6034 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | observed over north Indian region including some parts of western Himalayas. In NCR and surrounding regions along with severe rainfall there was freak hail incidence also. Table 1 corresponds to the summary of hailstorm cases over NCR (NNDC-CDO, 2013). Primarily, the Indian climate is divided into four seasons; pre-monsoon (March-April-May), monsoon (June-July-August-September), post-monsoon (October-November) 5 and winter (December-January-February) (Attri and Tyagi, 2010). With such a classification, a pattern is observed, where most hailstorms cases over NCR dominate during the warmer periods of pre-monsoon and monsoon months. It is seen that no hailstorm cases occurred during months of February and October whereas a very few hailstorm cases occur during the winter months. Out of the 33 hailstorm cases only 5 10 occur during the colder/winter period; with January showing only 2 cases. Thus, winter hailstorms can be considered a rare occurrence over NCR. With such a distinction between summer and winter hailstorms, it can be hypothesized that the mechanisms of these two kinds of hailstorms are different, which are deliberated in the following paragraphs. 15 Over north Indian region, pre-monsoon storm events occur during convectively unstable atmospheric conditions culminating due to transient disturbances observed in the air mass due to the surface heating. Storms may be categorized as severe storms if it is associated with hail, thunder, lightning, high winds etc (Houze Jr., 1981). These severe storms occur during strong vertical wind shear, which are ideal conditions for 20 hail formation (Orville and Kopp, 1977). During monsoon, the precipitation is caused due to development of deep convection because of higher surface temperature and vertical wind shear of south-westerly flow and associated moisture incursion over the Indian subcontinent. The convective precipitation is also facilitated by surface terrain and the Himalayan orography. These periods of precipitation occurrences during the 25 monsoon are associated with rainfall and in extreme cases hail (Koteswaram, 1958;Ramage, 1971;Sikka, 1977;Lau et al., 2012). In the context of hailstorms, it can be concluded that convection due to high surface temperatures and moisture laden flow are important for hailstorm formation. Hail is precipitation in the form of hard, rounded 6035 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | pellets of irregular lumps of ice. In cross section, hail shows concentric shells of different densities and degrees of opaqueness (Pruppacher and Klett, 2010). These multiple layers of ice within hail are formed due to continuous deposition and shedding of ice over the condensation nuclei as the hailstones cycle through strong convective clouds having multiple updrafts and downdrafts (Chatterjee et al., 2008). 5 The winter months are associated with absence of increased surface temperatures or low level of moisture incursion, making it a cold and dry season (Attri and Tyagi, 2010). These conditions are not conducive for generation of deep convection required for sustaining a hailstorm. In month of January the average temperature of Delhi is 13.5 • C (NNDC-CDO, 2013). With such low mean temperature, it is difficult for a storm 10 to attain the wind shear required for hail formation. Studying such unusual/abnormal meteorological condition conducive for winter hail formation is of interest. Thus, this study has a new and unique perspective to understand the dynamics of a winter hailstorm occurrence. Though hailstorms are a rare phenomenon due to very specific conditions of formation and subsequent occurrence (Table 1), they are quite hazardous. 15 Hailstones destroy crop, infrastructure, property and in extreme cases may cause injuries to humans (De et al., 2005). Due to the perspective of hailstorms happening only in summer, hailstorms are not expected in winter months. This may lead to unpreparedness in early warning systems for hail prediction and mitigation of damage due to hailstorms, which may lead to devastating consequences. 20 Numerical weather prediction technique is utilized for understanding the above discussed storm event. This storm is simulated with Weather Research and Forecasting (WRF) model with Advanced Research WRF (ARW) dynamical core. As hail formation is a microphysical process, the study mainly focuses on the use of Goddard Cumulus Ensemble (GCE) (Lin et al., 1983;Rutledge and Hobbs, 1984;Tao and Simpson, 25 1993) microphysics scheme to simulate hail formation. This scheme simulates six different hydrometeors or water particles: water vapour (WV), cloudwater (CW), cloudice (CI), rainwater (RW), snow and a third class of ice. The third class of ice can be graupel or hail as per the options used during simulation. Graupel is relatively smaller and less dense hydrometeor ice particle when compared with hail. These variations cause differences in microphysical properties of the model parameterization. This study includes analysis of simulations with both options of GCE microphysics for the hailstorm event, which will hereafter referred to as hail option or graupel option.
With these considerations, the objectives of this study are listed below:
5
-To understand and describe a winter hailstorm event over NCR, with discussion on the large scale flow and precipitation associated during such storms.
-To simulate the hailstorm event using GCE microphysics scheme's hail and graupel options for a comparative analysis in the two options.
-To study in detail the various thermodynamic and microphysical processes asso-10 ciated with the hailstorm formation using the GCE microphysics with hail option.
This paper is divided into the following sections with Sect. 2 describing experimental design and data, Sect. 3 showing results and discussion and summary and conclusion provided in Sect. 4.
2 Experimental design and data 15 In this study, the WRF model (version 3.0) with the ARW dynamic solver is used to simulate the hailstorm event. 2013, to understand the meteorological processes contributing to the storm development and propagation. In the current study, not only model performance in simulating the precipitation event is evaluated for verification, but the large scale atmospheric processes leading to the weather event are also analyzed. With such an aim, a three day simulation is deemed necessary for the study.
5
NCEP Final analysis data (FNL), at 1 • × 1 • spatial resolution, is used as the initial and lateral boundary conditions for the study (NCEP/NOAA/US-DoC, 2000). Initial conditions for the model are extracted from the FNL dataset and are interpolated to the model domain. NASA's Modern Era Retrospective-analysis for Research and Applications (MERRA) (Rienecker et al., 2011) 6 hourly analysis dataset with a spatial 10 resolution of 0.5 • × 0.7 • is used as corresponding observational data. Tropical Rainfall Measuring Mission (TRMM) multi-satellite precipitation analysis (real time data) 3B42 V7 (Huffman et al., 2007) at 0.25 • × 0.25 • spatial resolution is used as the observation for the precipitation fields. The daily OLR (Liebmann and Smith, 1996) dataset provided by Earth Systems Research Laboratory (ESRL), National Oceanic and Atmo- 15 spheric Administration (NOAA), USA, available at 2.5 • × 2.5 • spatial resolution is used for validation purposes.
As discussed previously the GCE microphysics scheme (Tao and Simpson, 1993) is used in the model simulation experiments. This parameterization scheme is based on the multi-dimensional, non-hydrostatic microphysics cloud resolving model (Simpson 20 and Tao, 1993). The microphysics scheme simulates six different hydrometeors: WV, CW, CI, RW, snow and the third class of ice (which can be either hail or graupel, as specified). The two options are different due to different microphysical processes considered for hail/graupel formation, which in turn impact the hydrometeor population. As both the hydrometeors are quite alike, their formation processes considered in both options of the parameterization scheme are similar; accretion of RW, aggregation of snow, rimming of CI, sublimation and melting (Lin et al., 1983). But with graupel option, two other processes are included for graupel formation; autoconversion of snow and deposition of WV (Rutledge and Hobbs, 1984). In this study to understand the microphysical process of hail formation and its complexities, simulation is conducted using the hail option specifically. It is to be mentioned that both simulations 3 and 1 km resolution nests were simulated with explicit representation of cumulus parameterization scheme. As model for simulations at horizontal resolutions smaller than 3 km, estimates the precipitation by the cloud microphysics scheme itself (Gomes and Chou,5 2010).
Analysis of large scale flow and precipitation
The model simulated large scale flow patterns at 500 hPa and its corresponding MERRA observation is depicted in Fig. 3. The large scale circulation showed a deep trough being formed over the western Indian and Pakistan region in model fields and corresponding MERRA observational analysis. Figure 3 depicts that both the model simulations capture the wind and geopotential height patterns with slight over estimation. The depression observed over the region corresponds to an incoming western disturbance (WD). WDs are eastward moving synoptic scale extra-tropical cyclonic 15 systems in the sub-tropical westerly jet. These originate in the Mediterranean Sea and cause precipitation over northern India mainly during the months of December-January-February, due to their interaction with the Himalayan orography (Dimri, 2004). This migratory disturbance in the mid-troposphere along with the stationary surface low over western India is called the WD, and generates the instability necessary for 20 winter precipitation (Dimri and Chevuturi, 2013). The presence of this system can be the cause for the enhanced the instability over the region that influences the storm formation.
Daily mean outgoing longwave radiation (OLR) for 17 January 2013 is depicted in Fig. 4
25
values is seen over whole of northern India including the NCR in both model options 6039 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | and observations. OLR can be considered as an indicator to describe the clouding condition or precipitation globally (Xie and Arkin, 1998). In the presence of increased extent and depth of cloud cover and lower cloud top temperatures in high clouds, low OLR values are recorded. Thus, the figure represents the cloud extent during the time period of the hailstorm. On evaluating the satellite image of Fig. 1, similar cloud extent is ob-5 served over the same region. The low values of OLR over the whole of northern Indian region depict extensive cloud cover and the possibility of precipitation over the region. The advance of WDs and interaction with Himalayas is associated with the development of mesoscale cloud formation (Puranik and Karekar, 2009) as seen during this event. Moisture flux transport and divergence integrated over the vertical atmospheric 10 column for observation and model simulations is shown in Fig. 5. In model simulated outputs, a strong region of moisture convergence is seen over NCR and some parts of Himalayan region north of NCR. Corresponding observations show a similar extent of moisture convergence zone, but the strength of moisture convergence is lower. According to the figure, it is observed that Arabian Sea and Bay of Bengal are both sources of 15 moisture for the convergence zone. The moisture laden flow from these sources flows over the Indian subcontinent, congregates and flows towards NCR. This moisture is important for the maintenance of the winter storm over northern India and its role will be discussed later in detail.
Analysis of three hourly precipitation rate from model simulations and its correspond- 20 ing observation is shown in Fig. 6. The figure suggests that there was a storm having maximum intensity over NCR between 15:00 to 18:00 UTC of 17 January 2013. This storm showed a localized formation over NCR and its surrounding region. When compared with the observation, model simulated output shows a similar spatial extent of precipitation patterns. The axis of the storm evolution is observed to be oriented along 25 north-east to south-west direction. On evaluating the two model simulations, it is observed that hail option shows closer precipitation intensity with TRMM observation than the graupel option. The spatial extent of precipitation pattern observed in hail option is also better match when compared with the observational analysis. Figure also depicts that storm cluster moves eastward as the storm progresses. The analysis shows localized cluster of storm that is the cause of the heavy precipitation event. But when the point specific precipitation intensity at NCR is compared between the different options, overestimation is observed in model simulation. Inter-comparison in temporal variation in half hourly precipitation intensities of both model simulations is shown in Fig. 7 for 5 all four horizontal model resolutions. The figure suggests peak concentration of precipitation around NCR region was observed around 17:00 UTC on 17 January 2013. Here it is also clearly noted that hail option shows higher precipitation estimates than the graupel option. Deliberations on reasons for variation in precipitation intensities within the options of microphysics are provided in next section by comparative analyses of 10 the flux in the different hydrometeors in different options.
Comparative analysis between hail and graupel options of GCE microphysics
Figure 8 depicts the mixing ratios of different hail and graupel outputs over spatial and temporal scales for both hail and graupel options. Increased concentrations of mixing 15 ratios of all six hydrometeors are observed over the NCR region around the peak of the storm (Supplement; Figs. S1-S5). These hydrometeors are indicators of cloud formation over the region. All the hydrometeor mixing ratio concentrations show higher values over NCR around 16:30 UTC, which is just before storm peaks representing storm evolution and/or build-up. In hail option, the increased formations of hydrometeors are 20 specifically observed over NCR, whereas some displacement is seen in graupel option output. In addition, the spatio-temporal variation of these mixing ratios shows an eastward movement of the storm. The increase of WV in the atmospheric column is due to the moisture incursion from the Arabian Sea and Bay of Bengal. As the WV rises in the atmosphere, formation of CW and CI begins by condensation and deposition pro-25 cesses due to decreasing temperatures, which form the clouds. The CW droplets are suspended in the atmospheric column in the mid-tropospheric layers, whereas CI particles are found in upper-troposphere where much colder conditions prevail. The RW 6041 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | droplets formation is enhanced by the processes of collision and coalescence (Oriville and Kopp, 1977), and ultimately sediment out as rain. Increased RW mixing ratios are an indicator of increased precipitable water availability in the atmospheric column. With the development of stronger storm conditions, there is formation of other hydrometeors like snow, hail and graupel. With the main focus of the paper on the third class of ice, 5 we will be discussing hail and graupel in more detail. Hail and graupel are both ice particles seen during storm formation. Hail is an ice particle which is formed due to consecutive processes like soft hail processes (dry growth and raindrop freezing) or hard hail processes (wet growth and shedding) (Wisner et al., 1972). These continuous and repeated processes form a differentially layered ice pre-10 cipitation type called hail (Pruppacher and Klett, 2010). The hailstones grow as the ice particles cycle through multiple cells of convective clouds (Chatterjee et al., 2008). Graupel are similar but smaller ice particles with a diameter less than 5 mm, formed exclusively by soft hail processes (Pruppacher and Klett, 2010). In the model simulation peak hail mixing ratios are observed over NCR around 16:00-16:30 UTC, whereas 15 graupel mixing ratios maxima is seen east of NCR. In the vertical profile hail mixing ratios span from 800-200 hPa whereas graupel mixing ratios show extent from 750-300 hPa. It is interesting to note that when hail and graupel mixing ratios are compared, the hail mixing ratios are lower than graupel mixing ratios. But graupel precipitation or sedimentation is not observed in the model simulation but hail precipitation is observed 20 in the hail option around NCR between 16:30-17:00 UTC as shown in Fig. 9. This observation can be attributed to the fact that graupel have high number concentrations in the atmospheric column as these small ice particles are formed much quicker than hailstones. But due to their small size the graupel particles also melt quicker due to the temperature conditions that are not as low as seen over the snowline. Thus, sedimentation of graupel is not observed over NCR but hail precipitation is seen during this hailstorm. With this discussion it can be concluded that to understand hail formation over NCR, microphysics with hail option needs to be studied in depth. Subsequent section focuses on understanding the cause of winter hailstorm formation.
Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper |
Detailed analysis of winter hailstorm formation
For a focused analysis of the winter hailstorm formation the time period of peak precipitation 17:00 UTC 17 January 2013 is considered (Fig. 10). Vertical cross sections of various parameters analyzed along the axis of core precipitation zone of the storm region as demarcated by a green line Fig. 10. Some of the model simulated fields are also 5 evaluated by considering their area averaged values over the 1 • × 1 • grey box around NCR as drawn in Fig. 10. The region over and around 77.2 • E and 28.6 • N would be considered the NCR or the region of study/interest. The analysis of the dynamical properties pertaining to the winter hailstorm along the axis demarcated in Fig. 10 is provided in Fig. 11. Model simulates a region of high 10 equivalent potential temperature (EPT) over the NCR, as seen in the Fig. 11a. This represents higher temperatures and moisture content in the mid-troposphere whereas in lower troposphere decreased EPT is noticed. This represents the region of instability spanning the atmospheric column which is the cause of the storm. The instability in the mid-tropospheric level is caused by the WD depression as observed at 500 hPa. 15 In this region, vertical wind (Fig. 11a) shows cells with updraft and downdraft conditions. Hailstorms forming processes are known for multiple cellular structure showing regions of updrafts and downdrafts within the cloud, with dynamic movement and reflectivity patterns. The updrafts originate from the air inflow from the surface towards the cloud in the opposite direction of storm movement (Chalon et al., 1976). Whereas, 20 the downdraft cells are caused by the air that ascends in the updrafts, and ultimately flows as outflow of air towards the front and back of the storm movement (Frankhauser, 1976;Strauch and Merrem, 1976). Relative vorticity and divergence in Fig. 11b, show alternative cells of positive and negative relative vorticity over NCR, which correspond to cyclonic and anti-cyclonic circulation respectively. These cells represent a gradient 25 of vorticity, depicting a positive vorticity advection which culminates in rising air. Associated to these cells are regions of divergence and convergence. The convergence zone observed in the mid-tropospheric level corresponds to the multiple cells of the 6043 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | hailstorm. These define the movement of hydrometeors through the associated updrafts and downdrafts. As discussed above, the multiple cells formed in the clouds promote the formation of hailstones. Further, positive specific humidity anomaly over NCR is observed in the mid-troposphere (Fig. 11c). This indicates that the atmospheric column over NCR upto 200 hPa shows a high moisture zone. This increased level of 5 moisture is required for not only growth of rain drops but also for diffusional and accretional growth of the hailstones. With the evolution of the storm, there is increase in different hydrometeors as discussed above. This indicates increase in reflectivity during storm formation (Fig. 11c). The storm shows increase of RW droplets, which congregate in the lower levels of atmosphere, showing high reflectivity values in the same To understand the reason for development of instability driving the winter hailstorm, 15 the geopotential height over the region is studied (Fig. 11d). Over NCR the geopotential height anomaly shows an increase around 400-200 hPa. This increase is associated with the dipping in the perturbation geopotential height contour lines. These changes are due to the tropopause fold penetrating the troposphere. The dip in the tropopause height values is also observed in the station data over New Delhi (Source: university 20 of Wyoming, Sounding data). This tropopause lowering is associated with baroclinic instability occurring over the region (Bush and Peltier, 1994). The increase storm intensity over the region is caused by the baroclinic instability due to the passing WD and the development of cyclonic circulation. The mid-latitude migratory WD attains higher intensities in form of a baroclinically unstable disturbance specifically over In-25 dian region (Rao and Rao, 1971;Singh and Agnihotri, 1977). This instability in the mid-to upper tropospheric levels generates the turbulent convective energy required for the development of updrafts during storm occurrence. With the availability of moisture in the atmospheric column the instability leads to heavy precipitation. But a WD over northern India does not always lead to hail formation during winter. Hail precipitation in model simulation is seen from 16:00-17:00 UTC 17 January 2013 (Fig. 9). The updrafts driven by the instability developed over the region, cycles the hail through the cloud. Hail particle successively moving through the vertical column grows in the upward motion and melts/sheds in the downward movement. With each cycle the hail-5 stone grows a new layer of ice forming the concentric circles seen in a hailstone cross section as discussed above. Thus, vertical wind velocity is an important factor for the hail formation. Heymsfield et al. (2005) describes that strong convective updrafts (with vertical wind speed greater than 5-10 m s −1 ) suppress homogenous nucleation to form ice particles which grow to form hail. Whereas, lower wind speeds would not attain 10 enough energy to develop a strong hailstorm. The model simulated vertical wind updraft speeds over NCR show a magnitude of 4-6 m s −1 which provide sufficient time for ice particle growth by dry or wet growth. The instability developed in the mid-tropospheric levels due to the WD develops propensity for baroclinic atmosphere in the upper half of troposphere. When the tem-15 poral variation of temperature profile of the region is analyzed, a dip is observed in the −60 • C isotherm around 16:00-17:00 UTC (Fig. 12). This lowering corresponds to the troposphere fold discussed before. The lowering of tropopause causes incursion of colder stratospheric layers into warmer troposphere. This in turn causes development of a steep temperature gradient as seen in the figure, which enhances upper level in-20 stability. Still the reasons for instability in the lower layers are yet not clearly discussed.
To understand the lower layer instability, temporal variation in area averaged convective available potential energy (CAPE) and specific humidity are represented in black and blue contours of Fig. 12 respectively. In this figure, an increase of moisture over NCR in the lower levels of atmospheric column is observed along with development of CAPE 25 from around 13:00 UTC. The source of this low level moisture incursion is majorly from Arabian Sea and on a lesser extent from Bay of Bengal (Fig. 5). The moisture convergence develops buoyancy which enhances the propensity of increase of CAPE in the atmospheric column. There is reduction in CAPE values in subsequent time periods 6045 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | after 13:00 UTC. The increase of CAPE defines the potential energy that is available to drive a storm and release of CAPE in form of kinetic energy promotes storm development. Along with this the low level moisture incursion provides the buoyancy required for the air parcel to rise. Thus, upper level instability is predominantly by the existing WD embedded with the low level instability due to moisture incursion and development 5 of CAPE lead to instability spanning the troposphere which makes it conducive conditions for hailstorm formation. Muller et al. (2008) and Rosenfeld and Lensky (1998) described the "continental" clouds having three zones in the vertical direction based on temperature variation: diffusional droplet growth zone (upto −10 • C), mixed phase zone (−10 to −38 • C) and glaciated zone (above −38 • C). The isotherm of −38 • C is termed 10 as homogeneous freezing isotherm beyond which homogenous nucleation occurs. As per Fig. 12, it can be concluded that the mixed phase zone for the current hailstorm can be considered within 600-350 hPa. The mixed phase zone is the region for development of hail particles due to growth by deposition of water particles. This zone in the model simulated output coincides with region having strong convergence influence 15 and higher specific humidity, promoting hail formation along with raindrop growth. The region of glaciation in the upper levels of troposphere is imperative for the development of small ice particles through homogenous nucleation. These particles further grow to form the various ice precipitation forms, in this case hail. The problem with winter storms is that instability is not sufficient enough for the cloud to extend to this height or 20 form an anvil. But in the 17 January 2013 hailstorm, the instability extending from low to upper levels of troposphere discussed above allows the formation of ice nuclei in the glaciations zone leading to winter hail formation. The clarity of two different zones of instability can be described in the sounding data of NCR for 17 January 2013. Figure 13 represents the model and observed sounding data in a graphical format. The development of instability at 12:00 UTC is an indicator for the storm development. This instability is measured by the gap between the air parcel lapse rate and environmental lapse rate. There is increased instability around mid-tropospheric level in both model and observation due to the approaching WD and development of baroclinic instability. Another region of higher instability seen in the observation is around 700 hPa, which is underestimated in the model simulation. This region of instability corresponding to the lower level convection described due to development of CAPE and the moisture incursion. There is augmented moisture content in the atmospheric column as seen in the relative humidity profile provides buoyancy 5 to the air parcel. Combination of low level available potential energy and moisture incursion along with upper level baroclinic instability due to the presence of WD, led to sufficient instability for winter hail formation.
Summary and conclusions
This study investigates the cause of hailstorm during winter over NCR. Winter months of December-January-February, specifically over north India, are considered a cold and dry season. But high surface temperature and supplementary moisture availability are the prerequisite for hailstorm development. And winter storms are not able to attain the intensity and depth of a convective storm. Thus, the question of hail being formed during the winter season is fascinating due to its uniqueness. 15 The comprehensive analysis of the results has led the authors to describe the mechanism of the winter hailstorm formation over NCR. These conclusions have been summarized in the form of an illustration or conceptual model as given in Fig. 14. On the day of the hailstorm, 17 January 2013, high moisture availability is seen over NCR. Arabian Sea and Bay of Bengal are identified as the source of this moisture in-flow. 20 This moisture incursion along with the release of CAPE in the lower levels leads to development of the lower level instability. But this instability is not sufficient for a winter storm. On the other hand winter precipitation is experienced over north India due to WDs. These migratory disturbances in the sub-tropical westerly jet cause baroclinic instability in the mid/upper troposphere. Thus, with two different sources of instability 25 this winter storm is able attain conditions similar to deep convective storms favorable for hail formation. The lower level instability formed due to release of kinetic energy and buoyancy of air parcel due to high relative humidity. This mid/upper level instability develops due to lowering of the tropopause which intensifies the temperature gradient. Thus, a rising air parcel gets an enhanced push to continue rising, which denotes the formation clouds having larger vertical extents. Ice nucleation, which develops hail nuclei, is enhanced 5 in such clouds as they attain the glaciated zone. These clouds are also associated with cells having turbulent upward and downward wind movement. Development of these conditions supports cycling of hail through the clouds and promotes its formation through shedding, deposition and other hail processes. As the hailstones grow, they ultimately sediment or precipitate out to cause the hailstorm during winter.
6065
Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Figure 13. Sounding data at NCR at 12:00 UTC 17 January 2013 in graphical output for (a) observation and (b) 1 km horizontal model resolution with hail option.
|
2017-11-19T12:42:12.294Z
|
2014-12-19T00:00:00.000
|
{
"year": 2014,
"sha1": "cb579875a3c4d66111f357db6d85f000644c1bf5",
"oa_license": "CCBY",
"oa_url": "https://nhess.copernicus.org/articles/14/3331/2014/nhess-14-3331-2014.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "73fbf8d9888aca28506e9c931d5f03e1dc5df375",
"s2fieldsofstudy": [
"Environmental Science",
"Physics"
],
"extfieldsofstudy": [
"Environmental Science"
]
}
|
239621915
|
pes2o/s2orc
|
v3-fos-license
|
FORMULATION AND EVALUATION OF MELOXICAM MICROSPHERES FOR COLON TARGETED DRUG DELIVERY
Objectives: The objective of this research was to organize and evaluate colon specific microspheres of the meloxicam for the treatment of colorectal cancer.
Methods: Sodium alginate microspheres were prepared by ionotropic gelation method using different ratios of meloxicam and sodium alginate (1:1, 1:2, and 1:3). Eudragit-coating of meloxicam, sodium alginate microspheres was performed by coacervation phase separation technique.
Results: The microspheres were characterized by shape, particle size, size distribution, incorporation efficiency, in vitro drug release studies and stability studies. The outer surfaces of the core and coated microspheres, which were spherical in shape, were rough and smooth, respectively. The size of the core microspheres ranged from 20 to 52 μm, and therefore the size of the coated microspheres ranged from 107 to 178 μm. The core microspheres sustained the discharge for 10 h during a pH progression medium mimicking the condition of gastrointestinal tract. The release studies of coated microspheres were performed during a similar dissolution medium as mentioned above. In acidic medium the discharge rate was much slower, however the drug was released quickly at pH 7.4 and their release was sustained up to 24 h.
Conclusions: It’s concluded from this investigation that Eudragit-coated sodium alginate microspheres are promising controlled release carriers for colon-targeted delivery of meloxicam.
INTRODUCTION
Targeted drug delivery into the colon is highly desirable for local treatment of variety of bowel disease such as ulcerative colitis, cirrhosis disease, amoebiasis, colonic cancer, local treatment of colonic pathologies and systemic delivery of protein and peptide drugs. The colon specific drug delivery system should be a capable of protecting the drug in route to the colon, that is, drug release and absorption should not occur in stomach as well as in the small intestine, and neither the bioactive agent should be degraded either of the dissolution sites, but only released and absorbed once the system reaches the colon. A colonic targeted approach found to be effective in minimizing side effects [1,2].
Epidemiological studies demonstrated that 40-50% reduction in the risk for colorectal cancer following prolonged use of non-steroidal antiinflammatory drugs. Recent studies show that cyclooxygenase (cox)-2 levels are increased in 85% of colorectal adenocarcinoma. Meloxicam is an non-steroidal anti-inflammatory drug drugs shows cox-1 IC50 of 3.27 μm and cox-2 IC50 of 0.25 μm, 13.1 times more preferential for cox-2 inhibition. Evaluation of selective cox-2 inhibitors for effects on colorectal cancer are currently an area of intense investigation and preclinical studies have clearly shown potent antitumor properties and several studies are reported in the application of meloxicam for the prophylaxis of colorectal cancer [3,4].
Various approaches were tried to achieve colonic delivery of drugs include use of prodrugs, pH-sensitive polymer coatings, time dependent formulations, bacterial degradable coatings, time/pH-controlled delivers, and intestinal luminal pressure-controlled colon delivery capsules [5]. A well-designed controlled drug delivery system can overcome some of the problems of conventional therapy and enhance the therapeutic efficacy of a given drug. To obtain maximum therapeutic efficacy, it is necessary to deliver the agent to the target tissue in the optimal amount in the right period of time there by causing little toxicity and minimal side effects. There are various approaches in delivering a therapeutic substance to the target site in a sustained and controlled release fashion. One such approach is using microspheres as carriers for drugs. Microspheres are characteristically free flowing powders consisting of proteins or synthetic polymers which are biodegradable in nature and ideally having a particle size <200 μm [6,7].
Mainly colon-specific polymer-based matrix tablets are reported for meloxicam for colon targeting. However, due to variations in transit throughout the colon, the drug release is often impaired when the colonspecific tablet matrix isn't readily disintegrated, and treatment will remain ineffective. This problem might be circumvented by reducing the dimensions of the delivery carrier, since it's been reported that gastrointestinal retention depends on the dimensions of the carrier, meaning that smaller carriers will cause longer residence within the colon. Hence within the present investigation we are aimed to develop a colon-specific microsphere delivery system of meloxicam using natural and enteric polymer as a carrier and to develop the colon-specific delivery that has potential to be used as an adjuvant therapy for colorectal cancer.
METHODS
The meloxicam was supplied by yarrow chemicals; Sodium alginate was supplied by Thomas Baker; Eudragit S-100 Research lab; Calcium chloride and Hydrochloric Acid; n-hexane, Disodium hydrogen phosphate, Potassium di-hydrogen phosphate, Ethyl acetate Span 80 was supplied by S.D. Fine Mumbai.
Design and formulation of multiparticulate system of meloxicam
Preparation of drug loaded sodium alginate microspheres study In the present microspheres were prepared using sodium alginate as polymer by ionotropic gelation technique. The meloxicam was dispersed in a solution of 4% w/v sodium alginate with stirring to supply a viscous form. Then polymer drug solution was added drop wise using syringe of twenty-two G in diameter from a height of about 5 cm into a beaker containing 4% w/v solution of salt with continuous stirring by magnetic stirrer for 10 min. Then the answer containing the gel formed microspheres was filtered using Whattman paper no1. The microspheres were allowed to dry at about 30-40°C and stored in well closed container for further use.
Preparation of Eudragit S-100 coated sodium alginate microspheres
Sodium alginate microspheres were coated with Eudragit S-100 using coacervation phase separation technique. Sodium alginate microspheres (50 mg) were dispersed in 10 ml of coating solution prepared by dissolution of 200 mg of Eudragit S100 in ethanol: acetone (2:1) to offer 1:1, 1:2, and 1:3 (coat: core ratio) and containing 0.2%w/v Span 80. This mixture was agitated for 5 min at 600 rpm. Subsequently 50 ml n-hexane (as the non-solvent) was poured into the polymeric solution containing the core material with the speed of 1 ml/min. The medium was stirred for 60 min to finish the method of micro particles coating. Coated microspheres were then washed with a more than n-hexane, filtered and dried at room temperature [9].
Characterization of beads
The developed microspheres beads were studied for compatibility studies by Fourier-transform infrared (FTIR) and subjected for various characterizations such as size and shape analysis, FTIR studies, percentage yield, incorporation efficiency, micrometric properties, degree of swelling, in vitro wash off test for microspheres.
FTIR studies
IR spectroscopic studies were carried out for prepared by meloxicam powder, using Shimadzu FTIR 1700 model to determine the integrity of the drug in the formulation.
Compatibility studies
Compatibility of the meloxicam with sodium alginate and Eudragit S-100 used to formulate microspheres was established by FTIR. Spectral analysis of meloxicam, sodium alginate and Eudragit S-100 and combination of the meloxicam with sodium alginate and Eudragit S-100 was carried out to investigate any changes in chemical composition of the drug after combining it with the excipients (Figs. 1-5).
Size and shape analysis
Microscopic analysis was performed to work out the typical size of microcapsules. The microcapsules prepared were dispersed in liquid paraffin and a drop of above dispersion was placed on a glass slide and observed under a microscope. The diameter of 100 microcapsules decided using calibrated eyepiece micrometre and stage micrometre. The typical diameter was calculated using the subsequent formula.
Average diameter nd n c f u ¦ .
Where n = Number of micro beads. d = Diameter of the micro beads, C.F = Calibration factors.
Micrometric properties of alginate microspheres
The flow properties of coated microspheres were investigated by determining the angle of repose, bulk density, and tapped density. The angle of repose decided by the fixed base cone method. Bulk and tapped densities were measured in 10 ml of a graduate. The sample contained within the cylinder was tapped mechanically by means of a constantvelocity rotating cam. The tapped volume was noted down when it showed no change in its value and bulk density and tapped density was calculated. Each experiment was performed 3 times [5].
Determination of percentage yield of microspheres
Thoroughly dried microspheres were collected and weighed accurately. The percentage yield was calculated using formula [10].
Incorporation efficiency
In 100 ml volumetric flask 20 mg of crushed microspheres were taken and dissolved with small quantity of methanol of the volume is made up to mark with pH 6.8 and stirred for 12 h. After stirring the solution was filtered through Whatman filter paper and from the filtrate appropriate dilutions were made and absorbance was measured at 365 nm using UV-spectrophotometer 1700 (Shimadzu) [11].
Where "a" is the theoretical drug content and "b" is the drug entrapped. The incorporation efficiency of coated microspheres was determined as
In vitro drug release studies of coated meloxicam microspheres
The release of meloxicam microsphere was carried out using USP basket-type dissolution apparatus at a rotation speed of 100 rpm, and a temperature of 37±0.5°C. For Microsphere simulation of gastrointestinal transit conditions was achieved using different dissolution media. Thus, drug release studies were conducted in simulated gastric fluid (SGF, pH 1.2) for the first 2 h as the average gastric emptying time is about 2 h. Then, the dissolution medium was replaced with enzyme-free simulated intestinal fluid (SIF, pH 7.4) and tested for drug release for 3 h, as the average small intestinal transit time is about 3 h, and finally enzyme-free (SIF, pH 6.8) was used up to 8 h to mimic colonic pH conditions. About 900 ml dissolution media was used. Medium 5 ml of sample was withdrawn every time and replaced with fresh medium, samples withdrawn at various time intervals were analyzed spectrophotometrically at 365 nm.
Kinetics of drug release experiments were treated with the different release kinetic equations [14,15].
RESULTS AND DISSICUSION
The aim of this project was to design and develop an oral, controlled release multi particulate drug delivery system of meloxicam proposed for colonic targeting. In this regard formulation studies were carried out. The results for the experiment conducted are as follows,
Meloxicam sodium alginate microspheres
Sodium alginate microspheres of meloxicam were successfully prepared by ionotropic gelation method. Uniform and almost spherical microspheres were obtained. The mean diameter of sodium alginate microspheres varied with varying drug: polymer ratio (AI, AII, III, BI, BII, BIII, CI, CII, and CIII). The results were shown in Table 1.
Compatibility studies
From the FT-IR spectra of the pure drug and therefore the combination spectra of drug with the polymers, it had been observed that each For finding out the mechanism of drug release from tablets, the dissolution data obtained from the above experiments were treated with the various release kinetic equations [14,15].
Morphology, size of microspheres, and micrometric properties of microspheres
The sodium alginate microspheres were discrete and spherical in shape with a rough outer surface due to the surface-associated crystals of the drug. Table 2 indicates that a better ratio of drug and polymer is related to increase in microsphere size decrease within the alginate concentration resulted within the clumping of microspheres, whereas a better sodium alginate concentration resulted within the formation of discrete microspheres with a mean diameter of 78 μm. This might flow from to higher viscosity at a better concentration and formation of larger microspheres. This might be explained by the very fact that more of the calcium ions became available for cross-linking guluronic acid units of sodium alginate, leading to the formation of more crosslinked alginate, which successively could increase the viscosity of the formulation, resulting in the formation of larger microspheres. The diameter of the core microspheres was in the range of 20-52 μm.
A scanning microscopy photograph of coated alginate microspheres showed that the microspheres were discrete and spherical in shape, with a smooth outer surface. The dimensions of coated microspheres ranged from 107 μm to 178 μm. All formulations showed excellent flowability, as represented in terms of angle of repose (Table 3).
In vitro drug release studies for uncoated microspheres
The in vitro release profile of different core alginate microsphere formulations in there was no significant difference in rate and extent of drug releases from formulations and statically significant difference within the rate and extent of drug release was observed in formulation compared to different ratio of formulations. This might be attributed to a rise within the density of the polymer matrix and therefore the diffusional path length that the drug has got to transverse. The discharge of meloxicam was characterized by a burst release followed by a moderate, slow release. The biphasic pattern of drug release is characteristic of matrix diffusion kinetics. The burst release are often reduced by increasing the polymer concentration, leading to better incorporation efficiency, as discussed earlier, and a decrease in surface associated drug. The results indicate that rate and extent of drug release decreased significantly.
In vitro drug release studies for coated microspheres
The second a part of the formulation focused on coated alginate core microspheres. The cores were microencapsulated by the coacervation phase separation technique. The coating polymer, Eudragit S-100, dissolves above pH 7.0, thereby protecting the drug from releasing from the alginate core before reaching the colonic region. Once the enteric coating dissolves, it's expected that drug release would then be controlled by alginate within the target. The in vitro release behaviour of encapsulated microspheres was very dramatic needless to say, no drug release occurred at gastric 1.2 for 2 h. As shown in Fig. 1, no drug release occurred below the pH of polymer solubility. After this lag time, drug release and therefore the time for the entire drug varied counting on the core-to-coat ratio. The discharge of meloxicam bogged down because the concentration of coating polymer increased. The in vitro release studies data were fitted into various release equations to elucidate the kinetics of drug release from these microspheres. The kinetic models used were first-order, zero-order, Higuchi and Peppas release models. Linear regressions are shown for maximum drug release batch. The examination of the determination R2 coefficient indicated that drug release followed the diffusion control mechanism from the core and coated microspheres. To explore the kinetic behavior, in vitro release results were further fitted into the subsequent Kors Meyer and Peppas equation.
Mt M∞ = K tn
Where Mt/M∞ is the fraction of drug time t, k is a kinetic constant, and n is a release exponent that characterizes the drug transport.
In vitro drug release for coated microspheres The equation in terms of coded factors can be used to make predictions about the response forgiven levels of each factor. By default, the high levels of the factors are coded as +1 and the low levels of the factors are coded as -1. The coded equation is useful for identifying the relative impact of the factors by comparing the factor coefficients.
|
2021-09-27T20:08:22.117Z
|
2021-08-07T00:00:00.000
|
{
"year": 2021,
"sha1": "a62679d0c075b0772bd6d3b24dc5d75146ac6eb9",
"oa_license": "CCBY",
"oa_url": "https://innovareacademics.in/journals/index.php/ajpcr/article/download/38482/25226",
"oa_status": "GOLD",
"pdf_src": "Adhoc",
"pdf_hash": "cecc5778962701065892863f9dd9ebd22f7cd023",
"s2fieldsofstudy": [
"Medicine",
"Materials Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
239050295
|
pes2o/s2orc
|
v3-fos-license
|
Stochastic Learning Rate Optimization in the Stochastic Approximation and Online Learning Settings
In this work, multiplicative stochasticity is applied to the learning rate of stochastic optimization algorithms, giving rise to stochastic learning-rate schemes. In-expectation theoretical convergence results of Stochastic Gradient Descent equipped with this novel stochastic learning rate scheme under the stochastic setting, as well as convergence results under the online optimization settings are provided. Empirical results consider the case of an adaptively uniformly distributed multiplicative stochasticity and include not only Stochastic Gradient Descent, but also other popular algorithms equipped with a stochastic learning rate. They demonstrate noticeable optimization performance gains, with respect to their deterministic-learning-rate versions.
Introduction
Machine learning models today range from simple linear regression to deep neural networks. These models try to capture the underlying behavior of systems which produce outputs y when given inputs x. The performance of these models is usually measured via a loss function (h(x; θ), y) where h(.; .) is the predicted output of the model for an input x for given model parameters θ ∈ R d . Then, if the joint distribution P of x and y is known, the estimates θ are the minimizers of the expected loss E x,y∼P [ (h(x; θ), y)]. Nonetheless, complete knowledge of P is absent in most practical scenarios, and a limited set of training data is available instead. In that case, an estimate of the expected loss is minimized, known as the empirical loss function, giving rise to the empirical loss minimization (ERM) problem: where the sequence {(x i , y i )} n i=1 denotes the available training data. Then, if ξ represents a random sample from the the available training data, let the loss function as a composition of (., .) and h(.; .) be denoted by L(θ; ξ). Also let f (θ) := E ξ [L(θ, ξ)]. This yields a stochastic optimization setting where the computed gradient g(θ, ξ) is a function of the true gradient ∇f (θ). If the former is assumed to be an unbiased estimator of the latter as in this work, then: where ξ t and θ t are independent. Due to its simplicity, cost-effectiveness and well studied properties, one of the most commonly used algorithms for ERM is stochastic gradient descent (SGD) originally developed in Robbins and Monro [1951], which iteratively gives an estimate of θ in (1) via the rule: for t = 1, 2, · · · and where g t is short for g(θ t , ξ t ). This rule will be referred to as original SGD or simply SGD throughout the paper.
Even though commonly the learning rate a t is taken to be deterministic, the novelty is that it introduces the setting where the learning rate in SGD becomes stochastic, by equipping it with multiplicative stochasticity. That is, the learning rate now becomes a t = η t u t (where e.g., η t = a/ √ t or η t = a, with a a positive constant; with some abuse of terminology, η t will be referred to as step size) and u t = f (v b , ..., v t ) is a function of past random variables v i , i = b, ..., t, referred to as stochastic factors (SF) in the rest of the paper. This work considers two settings, firstly one where the learning rate has memory of all previous SFs, i.e., b = 1, and secondly a setting where the learning rate has no memory of past SFs, i.e., b = t. The latter is analyzed theoretically in the Appendix. Furthermore, a Multiplicative-Stochastic-Learning-Rate is referred to as MSLR. An example of MSLR with memory applied on SGD for f (u 1 , ..., u t ) = t i=1 u i is shown in Algorithm 1. In general, it has been observed that minimization performance can be noticeably improved under appropriate learning schemes e.g., under deterministic adaptive learning rate schemes with memory (such as memory of past gradients, updates etc). Examples include ADAM Kingma and Ba [2015] and some variants (e.g., AMSGrad Reddi et al. [2018] or ADAMW Loshchilov and Hutter [2019]) or precursors (e.g. McMahan and Streeter [2010], Duchi et al. [2011]). The stochastic learning rate in this work possesses memory by taking into account SFs of previous timesteps to update the parameters of the current timestep. This work demonstrates that performance can be significantly improved if instead of just adaptive, the learning rate becomes both adaptive and stochastic, and possesses memory. In this case, the memory concerns past values of the SF.
Convergence analysis in expectation but for deterministic learning rates has been studied in Bertsekas and Tsitsiklis [2000], Nemirovski et al. [2009], Moulines and Bach [2011], Ghadimi and Lan [2013], Bottou et al. [2018]. This work provides convergence results in-expectation for SGD using stochastic learning rate with memory as well as results under the online learning framework (introduced in Zinkevich [2003]). It shows that with stochastic-learning-rates with memory optimization performance is significantly improved, while providing similar technical results to the deterministic-learningrate case. Concerning memoryless stochastic learning rates, theoretical convergence results in the stochastic approximation and online learning settings are given in the Appendix. Convergence in the almost-surely setting using stochastic learning rate schemes without memory, and in specific MSLR, for algorithms including SGD has been done in Mamalis et al. [2021].
Moreover, the analysis will allow for resetting SF distributions. Such distributions yield resetting learning rate schemes, which have been shown to improve optimization performance (O'Donoghue and Candes [2012], Loshchilov and Hutter [2017]). A memoryful Resetting-MSLR scheme will be referred to as RMSLR with memory. A memoryless Resetting-MSLR scheme will be referred to as RMSLR. If, moreover, the SFs are uniformly distributed, the resetting learning rate scheme will be referred to as RUMSLR with memory or RMSLR for memoryful and memoryless learning rates respectively. If, moreover, its SFs are uniformly distributed, the resetting learning rate scheme Figure 1: Plot that illustrates a resetting SF v t , used in the RMSLR-with-memory scheme considered in the experiments, For the case depicted in the plot, the value of the resetting parameter is β = 100, i.e. the distribution of v t resets every 100 iterations to U(0.8, 1.2).
will be referred to as RUMSLR with memory or RUMSLR respectively. An example of a uniformly distributed SF is shown in Fig. 1.
The remainder this work is organized as follows. Section 2 provides the stochastic optimization setting of this work and provides the in-expectation convergence results. The online learning setting along with the respective convergence results is given in Section 3. Section 4 provides a discussion on the RUMSLR scheme used to obtain the empirical results.
In Section 5, the experimental results for various optimization algorithms using this memoryful RUMSLR scheme are presented and compared to the deterministic-learning rate case. Section 6 concludes the paper with paths for future work. The Appendix contains results for stochastic learning-rate schemes without memory in the stochastic approximation and online learning rate settings.
Stochastic Optimization Setting
This section will provide convergence results for stochastic-learning-rate SGD under the stochastic optimization setting, briefly discussed in Section 1. These results hold for both MSLR and RMSLR schemes. The following assumptions will be used in the theorems. Firstly, function f (θ) is L-smooth, i.e.: for all θ , θ ∈ R d . Furthermore, for all θ ∈ R d , it satisfies: where f min is the global minimum of f (θ), and 0 < c ≤ L. It is noted that this assumption only requires the function to attain a global minimum (more detailed discussion in Karimi et al. [2016]).
Algorithm 1 MSLR-with-memory SGD
Require: initial point x 1 , stepsize η t , stochasticity factor v t while x t not converged do for some nonnegative scalars M and M G ≥ 1. Moreover, for the stochastic learning rates it is assumed that the following holds: These assumptions are the stochastic-learning-rate equivalents to the typical assumptions made in the deterministic-learning-rate case-commonly referred to as Robbins-Monro conditions (Robbins and Monro [1951])-namely, j η j = ∞ and j η 2 j < ∞. The theorem that follows provides a bound on the optimality gap of the SGD algorithm when using a stochastic-learning-rate with a constant step size: Theorem 1. Assume (4-7) are satisfied. Furthermore, assume a positive learning rate of: where η t ∈ R is either constant or decreasing, and u t : , for some positive λ ∈ R. If, moreover: then the iterates of the SGD algorithm in (3) satisfy: The guarantee in (10) is the same as when SGD uses a decreasing deterministic-learning-rate. However, the stepsize's upper bound 1 Eu[ut]LM G is larger than the 1 LM G in the deterministic-learningrate case. Moreover, the guarantee in (10) holds for arbitrary decreasing step sizes and does not depend on the initial point unlike decresing stepsizes in previous works (see, e.g., Bottou et al. [2018] for all of the above). In addition, the optimality gap for the stochastic-learning-rate with constant stepsize is zero, whereas for the deterministic-learning-rate case is nonzero, i.e., ηLM 2c . This is also true for a stochastic learning rate scheme without memory in the Appendix which yields a nonzero optimality gap for a constant stepsize, whereas Theorem 1 is able to obtain a zero optimality-gap asymptotically.
Proof. Taking expectations with respect to ξ t in (4), and from (3) substituting θ t+1 for θ and θ t for θ : The second inequality followed from (2), the third from (6), and the last equality from (8). Then, taking expectations with respect to u t and denoting the result of E ξt [E ut [.]] as E t [.]: where the equality followed from using Then: ut] . This also means thatc ≤ ϕ t < 1 wherec = 1 − λ. Then, adding and subtracting the same terms to create appropriate factor terms: Then rearranging terms: By unrolling the sequence it is: (1 − cϕ n E a [a n ]).
Then, for the second term in (22) it is that: from the assumptions of the theorem. Then let the algorithm run for long enough, so that the last two terms in (22) are close enough to their limits so that for their tails after some timestep K the following holds: (1 − cϕ n E a [a n ]) ≈ 0.
Then, partitioning the last two terms in (22) at K, and using (24), it is that: (1 − cϕ n E a [a n ]) (1 − cϕ n E a [a n ]).
(25) Therefore, from (25), and from E a [a t+1 ] = η t+1 E[u t+1 ] it follows that: which means: Concluding this section, from Theorem 1 it is thus observed that for the stochastic optimization setting the optimality gap bounds for SGD using a stochastic-learning-rate are improved from the ones when using a deterministic-learning-rate. This is because, the optimality gap is zero regradless of whether a constant or a decreasing learning rate is used, whereas for a deterministic learning rate, the optimality gap when using a determinisitc learning rate is nonzero (e.g. see Bottou et al. [2018] for more details).
Online Learning Setting
Now, an alternative analysis for stochastic-learning-rate SGD which holds for both MSLR and RM-SLR is given by considering the following optimization problem, referred to as online optimization: where θ * = arg min θ∈X T t=1 f t (θ), functions f 1 through f T are convex, g t := ∇f t (θ t ), and X ⊆ R d is a convex and closed set. The function R T is called the regret up to timestep T . This optimization framework was introduced in Zinkevich [2003]. Then, let the online algorithm: where Π X (y) = arg min θ∈X x − y is the projection of y on X. Studying an online algorithm with vanishing average regrets amounts to studying a stochastic algorithm for ERM as discussed in Cesa-Bianchi et al. [2004].
The following assumptions are used in the online minimization setting. The first is that functions f t (θ) are convex, i.e.: Furthermore, if X is a set with bounded diameter D then: for all θ , θ ∈ X. Moreover, if X ⊆ R d is a convex and closed set, then for any θ ∈ X and y ∈ R d , it is that: for θ = Π X (y) = arg min θ∈X y − θ . It is noted, that the regret bounds for the deterministiclearning-rate version of SGD are obtained by following the proofs of the next Theorem 2 but using a t = a √ t as the learning rate. Then, it will be shown that stochastic-learning-rate SGD has vanishing average regret for diminishing step size η t = a √ t . The theorem for the online learning setting follows: Theorem 2. Assume that X is convex, closed, and has bounded diameter D, and that the gradient of f t is bounded, i.e. ∇f t (θ) ∞ ≤ G. Furthermore, assume a positive learning rate of a t = a √ t u t with a = D 2G . Then for all T ≥ 1 it is that: where the subscript a denotes the expectation of R T with respect to the stochastic learning rate sequence {a t } T t=1 . Therefore, R a,T /T → 0 for T → ∞.
Proof. Firstly, it is noted that: where the first inequality followed from (32) replacing θ by θ * and y by θ t −a t g t . Taking expectations with respect to the learning rates a 1 · · · a t and denoting the results as E a [.], it is that: Rearranging with respect to g T t (θ t − θ * ) and using (30), it is: Then, summing over all steps: Using the diameter of the convex set D: Expanding the telescoping sum: This means that: where the third line followed by using the diameter D of the convex set X, and the fourth line by using the telescoping sum of the middle term. Using a t = a √ t u t it is: Then, minimizing the right hand side of the third inequality yields a = D/(2G) for which: thus concluding the proof. Q.E.D.
In conclusion, from Theorem 2 it is thus observed that for the online optimization setting the expectedwith respect to the learning rate-regret bound yield the same expected regret rate O( 1 √ T ) as the deterministic-learning-rate SGD algorithm (e.g., in Duchi et al. [2011]).
Discussion on the Stochasticity Factor
In this work, the stochastic learning rate scheme considered for the empirical results is a RMSLR-withmemory scheme and specifically and β ≥ 1, shown in Fig. 1. In this case the distribution of v t resets every β number of iterations to U(c 1 , c 2 ), named β−RUMSLR-with-memory because the SF is uniformly distributed. Moreover, it has been observed that choosing low values for c 1 or large values for c 2 slows down the algorithm significantly or destabilizes it. On the other side, they should be kept adequately apart so that the bursts are significant enough.
The distribution of v t is chosen to be reset after it is adequately close to unity and enough, β timesteps have passed, e.g., as shown in Fig. 1, where β = 100, indicated by the bursts that occur every 100 iterations. Furthermore, it can be shown that the condition of Theorem 1 that Vu[ut] E 2 u [ut] ≤ λ holds with λ = 1 3 for all c 1 , c 2 . This is because since v t is uniformly distributed. This means that: E 2 u [ut] ≤ 1 3 , for t = 1, which also holds for all t since 1 3 + 1 t − 1 is increasing.
Additionally, it can be checked numerically that the condition E 2 u [v t ] ≤ 1 as well as that the stochastic Robins-Monro conditions in Assumption 7 are satisfied for the RUMSLR scheme used to derive the empirical results in Section (5). It is noted that, since the aforementioned conditions are sufficient and not necessary, stochastic learning rates that do not satisfy these conditions but still improve performance might exist.
Experimentally, it will be demonstrated that β−RUMSLR-with-memory SGD yields noticeable improvements in minimization performance for datasets such as CIFAR-10, and CIFAR-100 (Krizhevsky [2009]) when compared to SGD without stochastic learning rate. This scheme was equipped to the learning rates of other algorithms as well such as ADAM Kingma and Ba [2015], AMSGrad Reddi et al. [2018], ADAMW Loshchilov and Hutter [2019], and two SGD with momentum algorithms originally presented in Polyak [1964] and Nesterov [1983] (subsequently referred to as SGD with momentum first and second versions respectively), all of which, even though not theoretically studied in this work, yielded similar improved performance when using β−RUMSLR with memory.
Results
Experimental results using the β−RUMSLR-with-memory scheme described in 4 applied to SGD, SGD with momentum versions first and second, ADAM, AMSGrad, and ADAMW are presented in this section. For β−RUMSLR-with-memory with memory the plots show the average running loss every 20 mini-batches for CIFAR-10 and CIFAR-100, and every 200 mini-batches for MNIST. The batch size was 128 in all experiments. The rest of the hyperparameters, i.e., learning rates, momentums (when applicable) and SF constants, are chosen via grid search. The datasets were standardized before use. For CIFAR-10 and CIFAR-100, the network architecture was Pytorch's ResNet18. For MNIST, logistic regression was used. For the former two datasets a constant stepsize was used, and a diminishing stepsize for the latter, for all algorithms.
The MNIST dataset was used to validate the online framework theoretical guarantees for β−RUMSLR-with-memory SGD using a convex loss function, in specific logistic regression as aforementioned. Indeed, the results in Fig. 2 verify the convergence of the MNIST β−RUMSLRwith-memory scheme in a convex setting as predicted by Theorems 2. Moreover, CIFAR-10 and CIFAR-100, the β−RUMSLR-with-memory scheme demonstrated significantly improved convergence than when using a deterministic learning rate as shown in Figs. 3 and 4 respectively. As can be seen by the plots, β−RUMSLR-with-memory algorithms convergence much faster than their deterministic-learning-rate counterparts.
Conclusion and Future Work
This work introduced learning-rate schemes that make the learning rate stochastic by multiplicatively equipping it with a function of random variables, which are coined stochastic factors. Convergence of the SGD algorithm employing stochastic learning rate schemes with memory of past Figure 3: Plots that illustrate the β-RUMSLR-with-memory scheme (in orange) and the deterministic learning rate (in blue) schemes for β = 100, c 1 = 0.8 and c 2 = 1.2. It is observed that for the CIFAR-10 dataset algorithms with β-RUMSLR yield noticeable gains in minimization performance. Figure 4: Plots that illustrate the β-RUMSLR (in orange) and the deterministic learning rate (in blue) schemes for β = 100, c 1 = 0.9 and c 2 = 1.1. Both use the same initialization. AMSGrad uses c 1 = 0.85, c 2 = 1.15. For CIFAR-100, algorithms with β-RUMSLR-with-memory yield noticeable minimization performance gains. Where necessary, epochs are omitted for clarity at the start. stochastic factors was theoretically analyzed and compared with known results for the algorithm's deterministic-learning-rate version. The discussion that followed suggested how the hyperparameter values introduced by the two stochastic-learning-rate schemes should be chosen. Empirical results on popular algorithms demonstrated noticeable increase in optimization performance, presenting stochastic-learning-rate schemes as a viable option for enhancing performance. In-depth generalization performance studies, convergence analysis of algorithms besides SGD, and investigating the effect of various stochastic factor distributions and hyperparameters on algorithm performance are some paths for future work.
A APPENDIX
The appendix contains the proofs for stochastic learning-rate schemes without memory in the stochastic approximation and online learning rate settings.
A.1 Stochastic Optimization Setting for Memoryless Stochastic Learning Rates
The theorem that follows provides a bound on the optimality gap of the SGD algorithm when using a memoryless stochastic-learning-rate with a constant step size: Theorem 3. Assume (4-6) are satisfied. Assume a bounded stochasticity factor c 1 ≤ u t ≤ c 2 . Furthermore, assume a positive learning rate of: and: then the iterates of the SGD algorithm in (3) satisfy: The guarantee in (47) is the same as when SGD uses a deterministic-learning-rate, e.g., as given in Bottou et al. [2018]. The step size is c1 c 2 2 times smaller than in the deterministic-learning-rate case. This is pragmatically nonrestrictive for convergence speed as supported by experimental results, since c 1 and c 2 are adequately close in practice (see discussion in Section A.3).
Then, using ηc 1 ≤ a t ≤ ηc 2 in (48), ξ i up to time t and denoting the result as E ξ [·]: or equivalently: This further results in: By unrolling the inequality: Furthermore, from (46) it follows that: which for M G ≥ 1, c ≤ L yields: Therefore from (52) it is that: or equivalently: Substituting back c1 c 2 2 a for η proves the theorem. Q.E.D.
Next, a convergence theorem for diminishing step size is provided. Theorem 4. Assume (4-6) are satisfied. Assume a bounded stochasticity factor c 1 ≤ u t ≤ c 2 . Furthermore, assume a positive learning rate of a t = η t u t , Then, if η t is decreasing, ∞ t=1 η 2 t < ∞ and: the iterates of the SGD algorithm in (3) satisfy: The guarantee in (58) provides the same convergence result as in the deterministic-learning-rate case, it holds for arbitrary decreasing step sizes and does not depend on the initial point unlike previous works (cf. Bottou et al. [2018]).
Proof. The proof for diminishing step size is identical to the proof of Theorem 3 until equation (48), except that the fifth inequality in (48) now follows from a t = η t u t ≤ η 1 c 2 , and the sixth from (57) which gives η 1 c 2 M G L − 2 ≤ −1 and therefore η t c 2 M G L − 2 ≤ −1, since η t ≤ η 1 for t ∈ N. Then, taking expectations with respect to all ξ i up to time t: or: Taking expectations with respect to all learning rate random variables a t , denoting the result E a [·] and also denoting E a [E ξ [·]] as E[·], gives: By unrolling the sequence it is: (1 − cE a [a n ]) (1 − cE a [a n ]).
Then, partitioning the last two terms in (63) at K, and using (66), it is that: (1 − cE a [a n ]) (1 − cE a [a n ]).
Concluding this section, from Theorems 3-4, it is thus observed that for the stochastic optimization setting the optimality gap bounds for SGD using a stochastic learning rate do not differ from the ones when using a deterministic learning rate.
A.2 Online Learning Setting for Memoryless Stochastic Learning Rates
Now, an alternative analysis for SGD in the case which holds for both MSLR and RMSLR is given for the memoryless case by considering the online learning setting, similarly in Section 3. It will be shown that the memoryless-stochastic-learning-rate SGD has vanishing average regret for diminishing step size η t = a √ t . The first theorem for the online learning setting follows: Theorem 5. Assume that X is convex, closed, and has bounded diameter D, and that the gradient of f t is bounded, i.e. ∇f t (θ) ∞ ≤ G. Furthermore, assume a positive learning rate of a t = a √
|
2021-10-22T01:15:30.870Z
|
2021-10-20T00:00:00.000
|
{
"year": 2021,
"sha1": "a324c9a887d73ae4da2f463205497d096eb55467",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "a324c9a887d73ae4da2f463205497d096eb55467",
"s2fieldsofstudy": [
"Computer Science",
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics",
"Computer Science"
]
}
|
703128
|
pes2o/s2orc
|
v3-fos-license
|
Empirical evaluation of scoring functions for Bayesian network model selection
In this work, we empirically evaluate the capability of various scoring functions of Bayesian networks for recovering true underlying structures. Similar investigations have been carried out before, but they typically relied on approximate learning algorithms to learn the network structures. The suboptimal structures found by the approximation methods have unknown quality and may affect the reliability of their conclusions. Our study uses an optimal algorithm to learn Bayesian network structures from datasets generated from a set of gold standard Bayesian networks. Because all optimal algorithms always learn equivalent networks, this ensures that only the choice of scoring function affects the learned networks. Another shortcoming of the previous studies stems from their use of random synthetic networks as test cases. There is no guarantee that these networks reflect real-world data. We use real-world data to generate our gold-standard structures, so our experimental design more closely approximates real-world situations. A major finding of our study suggests that, in contrast to results reported by several prior works, the Minimum Description Length (MDL) (or equivalently, Bayesian information criterion (BIC)) consistently outperforms other scoring functions such as Akaike's information criterion (AIC), Bayesian Dirichlet equivalence score (BDeu), and factorized normalized maximum likelihood (fNML) in recovering the underlying Bayesian network structures. We believe this finding is a result of using both datasets generated from real-world applications rather than from random processes used in previous studies and learning algorithms to select high-scoring structures rather than selecting random models. Other findings of our study support existing work, e.g., large sample sizes result in learning structures closer to the true underlying structure; the BDeu score is sensitive to the parameter settings; and the fNML performs pretty well on small datasets. We also tested a greedy hill climbing algorithm and observed similar results as the optimal algorithm.
Introduction
Bayesian networks are compact graphical models for representing uncertain relationships among the random variables in a domain. Often, the relationships are unknown and must be learned from data. A popular approach called score-based learning [1] is to assign a score to each Bayesian network structure according to a scoring function and find the structure that optimizes the score. There are many scoring functions for Bayesian networks, such as minimum description length (MDL) [2] (or equivalently, Bayesian information criterion (BIC) [3]), Akaike's information criterion (AIC) [4], Bayesian Dirichlet equivalence score (BDeu) [5,6], factorized normalized maximum likelihood (fNML) [7], and others [8,9].
The score-based approach to learning Bayesian networks has been shown to be NP-hard [10]; both the running time and memory usage of exact learning are exponential in the number of variables in the worst case. Therefore, early research mainly focused on developing approximation methods [1,[11][12][13][14]. Recently, however, optimal learning algorithms such as dynamic programming [15][16][17], branch and bound [18], admissible heuristic search [19][20][21], and mathematical programming [22,23] have been developed to learn optimal Bayesian networks with several dozens of variables.
Because of the different theoretical underpinnings of these scoring functions, they typically result in different "optimal" networks. Once a scoring function has been selected, though, all optimal algorithms learn equivalent networks; they only differ in running time and memory usage. A major mystery surrounding Bayesian network learning is which scoring function to use given that there are so many choices. Several empirical investigations have been carried out on the performance of various scoring functions in learning Bayesian networks, e. g. [24][25][26]. These studies, however, have drawbacks in their evaluations because they used local search methods such as K-2 [1] and Greedy Thick Thinning algorithm [27] to select network structures, or even used randomly generated network structures [26]. These suboptimal structures may affect the reliability of their conclusions regarding the model selection capability of the scoring functions. Furthermore, these studies often generate random synthetic networks as the test cases; experimental data thus generated may not share similar properties as real-world data.
In this study, we use an optimal dynamic programming algorithm [16] to learn Bayesian network structures; any other optimal algorithm would yield the same results, however, because only the choice of scoring function affects the learned networks. We study the capability of four scoring functions, MDL, AIC, BDeu, and fNML, to recover the underlying Bayesian network structures. We generated artificial datasets from a set of gold standard Bayesian networks created based on real-world data, learned optimal Bayesian networks for them using different scoring functions, and compared the learned models with the gold standard models based on various evaluation measures. For comparison, we also included the results of a greedy hill climbing algorithm.
Our results offer new insights into the scoring functions in addition to confirming some other common beliefs. In contrast to the results of existing work, a major finding of our study suggests that the MDL/BIC score consistently outperforms AIC, BDeu, and fNML in recovering the underlying Bayesian network structures across various sample sizes. Other findings of our study support existing work. Our results confirm that the structural Hamming distance gives a more reliable measure of the distance between Bayesian net-work structures. We also observed that a parameter selection greatly affects the BDeu score. Finally, it is confirmed that fNML has good performance when the sample sizes are relatively small. Our results using the greedy hill climbing algorithm are similar to those of the optimal learning algorithm, although with higher variances, so our conclusions also hold for the greedy algorithm.
The remainder of this paper is structured as follows. We first review several prior empirical studies of scoring functions. We then provide an overview of Bayesian network and structure learning. After that, we introduce four scoring functions which we will compare. We follow that with a description of the experimental design of this study. Finally, we present the empirical results and discuss our findings.
Prior work
Several researchers have empirically evaluated the various scoring functions for learning Bayesian networks. In [26], Van Allen and Greiner compared the performance of three different model selection criteria, AIC, BIC, and cross-validation, in finding the right balance between the complexity of the model and the goodness of fit to the training data. First, they randomly generated the gold standard Bayesian network structures as well as the probability parameters. Second, they generated datasets with different sample sizes from the networks. For each dataset, they again randomly constructed a set of hypothesis structures and evaluated their quality based on the scoring functions. They found that AIC and cross-validation perform better in avoiding over-fitting in the model selection. While BIC may still work for large sample sizes, it can perform arbitrarily worse than other functions for small datasets. However, they did not use a learning algorithm to try to find good hypothesis structures; they also randomly generated their gold standard networks. It is unclear whether their results stem from the scoring functions or their random model selection technique, or whether the results can be generalized to real-world datasets.
In Yang and Chang's study [24], they compared the performance of five different scoring functions: uniform prior score metric (UPSM), conditional uniform prior score metrics (CUPSM), Dirichlet prior score metric (DPSM), BDe, and BIC. They restricted their experimental evaluations on random networks with three or five nodes as well as a benchmark network called Alarm. Then they generated random datasets from the networks. They used a K2like search method [1] to learn Bayesian networks. Their greedy structure learning algorithm assumes an ordering over the variables. Then, it greedily adds parents consistent with that ordering to maximize the likelihood of the structure and data set. Because of the ordering assumption and the greedy approach to adding parents, it does not guarantee finding the globally optimal structure. For evaluation, they use the cross-entropy (KL-Divergence) to measure the difference between the learned networks and the true networks. Their results indicated that UPSM, CUPSM, DPSM and BIC are able to correctly identify the true networks. Meanwhile, BDe and DPSM's performance are very sensitive to the a value. They may fail to find the true network if the a value is not set properly. This study shares the shortcoming of Van Allen and Greiner's study: their gold standard networks are randomly generated, so they may not accurately reflect real-world datasets. Furthermore, their K2-like search method requires an ordering of the variables; in real-world applications, an ordering is often not known a priori. Therefore, it is again unclear how their results generalize to real-world situations.
Another related empirical work by de Jongh and Druzdzel [25] investigates structural evaluation measures for Bayesian networks rather than scoring functions. They generated random datasets with different sizes from four benchmark Bayesian networks. Then for each combination of the network and sample size, they ran a local search algorithm called Greedy Thick Thinning [27] to learn Bayesian network structures and calculated the distances between the learned networks and the gold standard networks based on structural Hamming distance, Hamming distance, and other measures. They concluded that the structural Hamming distance is especially useful when looking for the causal structures.
All of these studies have drawbacks in their empirical evaluations. In particular, the conclusions of Van Allen and Greiner are drawn based on randomly generated network structures. Therefore, it is unclear how reliable their conclusions are regarding the model selection capability of the scoring functions. Additionally, the two studies which evaluate scoring functions rely on randomly generated gold standard networks; these may not accurately reflect real-world datasets. The work of de Jongh and Druzdzel only investigates structural evaluation measures using a single scoring function; other scoring functions may behave differently. The current study is designed to address these concerns.
Bayesian networks
A Bayesian network encodes a joint probability distribution over a set of random variables V = {X 1 , ..., X n }. We consider only discrete variables in this work. Formally, a Bayesian network B is a pair {G, Θ}, where G is a directed acyclic graph (DAG) in which each node corresponds to one of the random variables. The edges or lack of them encode the conditional independence relationships among the variables. The parents of X i are denoted P A i ; X i is independent of its non-descendant variables given its parents. Θ specifies the conditional probability distributions P (X i |P A i ) for each X i . Thus, the joint probability distribution of all of the variables is given as Given a dataset D = {D 1 , ..., D N }, where D i is an instantiation of all the variables in V, Bayesian network structure learning is the problem of learning a network structure from D. Assuming D is complete and discrete, Θ is maximized using frequency counts from the data [7]. Consequently, finding the optimal Bayesian network reduces to finding the optimal structure. Score-based learning is a commonly used technique to identify the optimal structure. In this approach, a scoring function is used to measure the goodness of fit of a structure to the data. The goal of the learning problem is then to find the optimally scoring structure. The score typically approximates the probability of the structure given the data and represents a tradeoff between how well the network fits the data and how complex the network is. In this work, we assume the scoring function is decomposable [6]. That is, the score for a network structure B can be calculated as the sum of scores for the individual variables, where the score for a variable is calculated based solely on the variable and its parents. Therefore, and the learning problem is to find B*, where A Bayesian network structure can represent a set of joint probability distributions. Two network structures are said to belong to the same equivalence class if they represent the same set of probability distributions [28]. A scoring function which assigns the same score to networks in the same equivalence class is score equivalent [6].
Unfortunately, the number of possible structures is super-exponential in the number of variables; learning an optimal Bayesian network from D is shown to be NP-hard [10]. Solving the learning problem exactly becomes impractical if the number of variables is too large. Consequently, much early work focused on approximate algorithms, such as greedy hill climbing approaches [1,11], tabu search with random restarts [13], limiting the number of parents or parameters for each variable [14], searching in the space of equivalence classes of network structures [29], and the optimal reinsertion algorithm (OR) [12]. These algorithms use local search to find "good" networks; however, they offer no guarantee to find the one that optimizes the scoring function. Recently, exact algorithms for learning optimal Bayesian networks have been developed based on dynamic programming [15][16][17]30,31], branch and bound [18], linear and integer programming (LP) [22,23], and heuristic search [19][20][21]. These algorithms have enabled us to learn optimal Bayesian networks for datasets with dozens of variables.
Given a scoring function, all optimal learning algorithms learn equivalent networks; hence, the choice of which optimal algorithm is used does not affect the learned network. Consequently, these algorithms make it possible for us to study the behavior of different scoring functions in structure learning without needing to consider the confounding factors resulting from the choice of structure learning algorithms.
Scoring functions
Many scoring functions are in the form of a penalized log-likelihood (LL) functions. The LL is the log probability of D given B. Under the standard i.i.d assumption, the likelihood of the data given a structure can be calculated as where D ij is the instantiation of X i in data point D j , and PA ij is the instantiation of X i 's parents in D j . Adding an arc to a network never decreases the likelihood of the network. Intuitively, the extra arc is simply ignored if it does not add any more information. The extra arcs pose at least two problems, though. First, they may lead to overfitting of the training data and result in poor performance on testing data. Second, densely connected networks increase the running time when using the networks for downstream analysis, such as inference and prediction.
A penalized LL function aims to address the overfitting problem by adding a penalty term which penalizes complex networks. Therefore, even though the complex networks may have a very good LL score, a high penalty term may reduce the score to be below that of a less complex network. Here, we focus on decomposable penalized LL (DPLL) scores, which are always of the form There are several well-known DPLL scoring functions for learning Bayesian networks. In this study, we consider MDL, AIC, BDeu and fNML. These scoring functions only differ in the penalty terms, so we will focus on discussing the penalty terms in the following discussions. In terms of memory and runtime, all of the scoring functions incur similar overhead [32].
Minimum description length (MDL)
The MDL [3] scoring metric for Bayesian networks was defined in [2,33]. MDL approaches scoring Bayesian networks as an information theoretic task. The basic idea is to minimally encode D in two parts: the network structure and the unexplained data. The model can be encoded by storing the conditional probability tables of all variables. This requires log N 2 * p bits, where log N 2 is the expected space required to store one probability value and p is the number of individual probability values for all variables. The unexplained part of the data can be explained with LL(D|B) bits. Therefore, we can write the MDL penalty term as where p i is the number of parameters for X i . For MDL, the penalty term reflects that more complex models will require longer encodings. The penalty term for MDL is larger than that of most other scoring functions, so optimal MDL networks tend to be sparser than optimal networks of other scoring functions. As hinted at by its name, an optimal MDL network minimizes rather than maximizes the scoring function. To interpret the penalty as a subtraction, the scores must be multiplied by -1. The Bayesian information criterion (BIC) [3] is a scoring function whose calculation is equivalent to MDL for Bayesian networks, but it is derived based on the asymptotic behavior of the models, that is, BIC is based on having a sufficiently large amount of data. Also, BIC does not require the -1 multiplication.
Akaike's information criterion (AIC)
Bozdogan [34] defined the AIC [4] scoring metric for Bayesian networks. It, like BIC, is another scoring function based on the asymptotic behavior of models with sufficiently large datasets. In terms of the equation, the penalty for AIC differs from that of MDL by the log N term. So the AIC penalty term is Because its penalty term is less than that of MDL, AIC tends to favor more complex networks than MDL.
Bayesian Dirichlet with score equivalence and uniform priors (BDeu) The Bayesian Dirichlet (BD) scoring function was first proposed by Cooper and Herskovits [1]. It computes the joint probability of a network for a given dataset. However, the BD metric requires a user to specify a parameter for all possible variable-parents combinations. Furthermore, it does not assign the same score to equivalent structures, so it is not score equivalent. To address the problems, a single "hyperparameter" called the equivalent sample size was introduced, referred to as a [6]. All of the needed parameters can be calculated from a and a prior distribution over network structures. This score, called BDe, is score equivalent. Furthermore, if one assumes all network structures are equally likely, that is, the prior distribution over network structures is uniform, a is the only input necessary for this scoring function. BDe with this additional uniformity assumption is called BDeu [6]. Somewhat independently, the BDeu scoring function was also proposed earlier by Buntine [5]. BDeu is also a decomposable penalized LL scoring function whose penalty term is where q i is the number of possible values of PA i , r i is the number of possible values for X i , D ijk is the number of times X i = k and PA i = j in D, and a ij is a parameter calculated based on the user-specified a. The original derivations [5,6] include a more detailed description. The density of the optimal network structure learned with BDeu is correlated with a; low a values typically result in sparser networks than higher a values. Recent studies [35] have shown the behavior of BDeu is very sensitive to a. If the density of the network to be learned is unknown, selecting an appropriate a is difficult.
Factorized normalized maximum likelihood (fNML)
Silander et al. developed the fNML score function to address the problem of a selection in BDeu based on the normalized maximum likelihood function (NML) [7]. NML is a penalized LL scoring function in which regret is the penalty term. Regret is calculated as where the sum ranges over all possible datasets of size N. Kontkanen and Myllymäki [36] showed how to efficiently calculate regret for a single variable. By calculating regret for each variable in the dataset, the NML becomes decomposable, or factorized. fNML is given by where C r i N ij are the regrets. fNML is not score equivalent.
Methods
Our empirical evaluation of the scoring functions consisted of four phases. First, we created a set of Bayesian networks from real datasets as the gold standard networks. Next, we generated a variety of datasets from each of those gold standard networks by logic sampling.
After that, we learned optimal Bayesian networks from the sampled datasets using both an optimal algorithm and a greedy hill climbing algorithm. Finally, we calculated a number of evaluation metrics by comparing the learned networks with the gold standard networks.
Creating gold standard networks
We need a set of gold standard Bayesian networks as the basis for our empirical evaluations. It is possible to use randomly generated Bayesian networks like several existing studies did, but we want to use models that resemble Bayesian networks that are created for realworld applications. There are many benchmark Bayesian networks available, such as Alarm, CPCS, Hepar, etc., but these benchmark models contain too many variables and are intractable for the current optimal learning algorithms. Therefore, we chose to create the gold standard networks by learning optimal Bayesian networks for a set of UCI machine learning datasets [37] with fewer than 25 variables. This section describes our data processing method for the reproducibility of the results.
The raw UCI datasets contain both continuous and discrete data, as well as missing values. Table 1 describes the detailed information for all the datasets used in this study. Continuous values were discretized using the minimum description length (MDL) discretization technique [38]. MDL discretization recursively partitions a dataset S with a single variable A by segmenting it into two distinct sets based on a boundary value T. The entropy between the two sets is minimal. The entropy between the two sets is defined as where S 1 and S 2 are the segments of S based on partitioning at T and Ent(·) is the entropy of the single set.
The recursion stops when the information gain of adding another partition does not exceed the cost of encoding the two new separate classes, given as where k i is the number of distinct values of A in S i . Although the MDL discretization technique has the same theoretical basis as the MDL scoring function, it is otherwise unrelated. That is, using the MDL discretization does not favor the MDL scoring function over the others in any way.
We used a k nearest neighbors (kNN) algorithm to impute missing values [39]. The kNN algorithm computes a missing value X p for record D i by finding the k closest D j s (out of those records which are not missing any values) to D i (using Euclidean distance, for example), excluding X p . If X p is a continuous variable, the value of X p is averaged for each of the D j s, and that value is assigned to X p for D i . If categorical, it is replaced by a majority vote among the k closest neighbors for X p . We set k = 5.
After processing the datasets, we applied an optimal learning algorithm based on the MDL scoring function [17] to learn optimal Bayesian networks. Again, the use of MDL score here does not affect the conclusions of this study, as other scoring functions yielded similar results. We used the maximum likelihood estimation method to learn the parameters of the networks. We took the learned networks as the gold standard networks and generated datasets from them.
Generating datasets from gold standard networks
After we created the gold standard networks, we generated datasets for each of these Bayesian networks with different numbers of data points ranging from 200 and 1000 (with increments equal to 200) and from 1,000 and 10,000 (with increments equal to 1,000), for a total of 18 sample sizes for each gold standard network. Each data point in a dataset corresponds to one random sample drawn from the joint probability distribution of a Bayesian network using logic sampling [40]. The basic idea is to sample the value for each variable according to the conditional probability distribution of the variable given its parents. The sampling is performed in a topological order of all the variables in order that all the parents already have sampled values before the child variable is sampled.
Learning from the sampled datasets
After generating datasets from the gold standard networks, we learned optimal networks for all the datasets by using the aforementioned scoring metrics. MDL, AIC and fNML are parameterless, so we learned one network for each combination of scoring function and dataset. All optimal learning algorithms would learn an equivalent network, so our choice of optimal learning algorithm does not affect the learned network. We tried the following a values, 0.1, 0.5, 1, 5, 10, 20, 50, 80, 100, for the hyperparameter a of BDeu and learned a network for each combination of a value and dataset. Thus, in total, we learned 12 "optimal" networks for each dataset and sample size. For comparison, we also tested a greedy hill climbing algorithm with random restarts and a tabu list in the same experiments.
Evaluating the learned networks
We used several structural evaluation metrics to compare the performance of the different scoring functions. Three of the evaluation metrics operate directly on the gold standard and learned DAG structures: accuracy, sensitivity, and average hamming distance (AHD). The formulas for those metrics are where a TP is an edge in the correct direction in the learned network, a TN is an edge in neither the learned nor the gold standard network, a FP is an edge in the learned network but not in the gold standard network, and a FN is an edge in the gold standard but not in the learned network. Note that an edge in the wrong direction in the learned network counts as both a FP and a FN.
We also used an evaluation metric called structural Hamming distance (SHD). As mentioned earlier, multiple structures with edges in different directions may belong to the same equivalence class. Intuitively, the distance between Bayesian networks in the same equivalence class should be zero. To accommodate this, SHD first identifies the equivalence class to which a Bayesian network belongs using an algorithm given by Chickering [28]. An equivalence class is represented by a partially directed graph (PDAG) in which some edges are directed and some undirected. The undirected edges can be orientated arbitrary as long as no new V structure in which multiple variables share a child is introduced. SHD then counts the number of directed and undirected edge additions, deletions, reversals and changes in direction to transform one PDAG into the other as the distance between two corresponding Bayesian networks. Tsamardinos et al. [41] provide a more formal algorithm for computing the SHD metric.
Results
In this section, we present the results of our empirical study. We first compared the evaluation metrics in order to select one metric for further analysis. We next looked into the effect of the hyperparameter a on the BDeu score. We then compared the capability of the scoring functions in recovering the Bayesian network structures from the sampled datasets generated from the gold standard Bayesian networks. After that, we compared the effect of sample sizes on the performance of the scoring functions in learning from the datasets when using both an optimal learning algorithm and a greedy hill climbing algorithm.
Comparison of evaluation metrics
We first compared the robustness of the evaluation measures as the sample size increases in the datasets. Theoretically, as the number of data points increases, the bias introduced by the penalty term in a scoring function has decreasing effect, and the learned model should gradually converge to the equivalence class of the true underlying model [29]. Figures 1 and 2 show the convergence results for the scoring functions on the optimal networks learned for the Statlog (Australian CreditApproval) and Cleveland Heart Disease datasets respectively. We consider an evaluation measure to have converged when adding more data points does not change the value of the metric. Our results show that the SHD metric converges for most of scoring functions with a small number of data points. In contrast, AHD, accuracy and sensitivity still fluctuate when there is a large number of samples. We only show the results on two datasets, but the results on the other datasets are similar. SHD exhibits better convergence behavior because it operates on the equivalence classes of networks rather than directly on the specific DAGs in question. As a simple example, suppose the gold standard network is X Y, but the learned network is X Y. The two networks represent the same conditional independencies, and SHD gives a distance of 0. However AHD, accuracy, and sensitivity all consider the arc incorrect because the arcs are oriented in different directions. We therefore only use SHD for the rest of our analysis.
BDeu parameterizations
We also investigated the effect of the hyperparameter a on BDeu. We focused on both the convergence behavior and the effect of a on recovering the gold standard networks. The results are shown in Figure 3 and Table 2. While some a values give good recovery results, it is clear that selecting either too low or too high of an a can dramatically impact the quality of the learned networks. BDeu was similarly impacted by a on other datasets as shown in the Additional File 1 S1.xls (sheet = results . optimal). On some of the networks, a poorly chosen a value may prevent convergence of the algorithms even when the sample size is large. As mentioned earlier, low as tend to result in sparser networks than higher as. Unfortunately, if the density of the gold standard network is unknown, selecting a is difficult. Consequently, BDeu is only a good scoring function if an expert can appropriately estimate a. Otherwise, the learned network is either too sparse (if a is too low) or too dense (if a is too high). This analysis supports previously published results [35].
Gold standard network recovery
We studied the capability of each scoring function in recovering the gold standard network based on the SHD metric. In the case of BDeu, we show the behavior of the best performing a value. Figure 4 shows that most of the scoring functions can recover the gold standard network on four of the datasets given a large enough sample size and appropriate parameters (a for BDeu). Other datasets exhibit similar behavior as shown in Table 3 and the Additional file 1 S1.xls (sheet = results . optimal). In particular, we consider the minimum distance of each scoring function and dataset. A minimum distance of 0 means that the gold standard network was recovered for the dataset. Small distances indicate that the scoring function guided the learning algorithm to find close to optimal networks. In contrast to the results reported by several previous studies, we found that MDL was able to recover the gold standard network more quickly than other scoring functions. We observe these differences both because we use an optimal learning algorithm and because we use gold standard networks representing real-world datasets. Given an appropriate a value, BDeu also converged to the gold standard networks within the sample sizes we tested. In some of the datasets, fNML converged to the gold standard network very quickly, but sometimes it converged to a different network. In contrast, AIC's behavior was much more erratic. It found the gold standard network on 8 of the datasets. But because of its high standard deviation, we infer it never completely converged. Figure 4 supports this conclusion. In light of these results, we conclude that MDL is a good scoring function because it often converges to the gold standard network. BDeu also exhibits good behavior if a suitable a is known before learning.
Convergence behavior
Next, we studied the convergence behavior of each scoring function. We did not consider whether the scoring function converged to the gold standard network; rather, we only focused on whether the scoring function converged to any network. In essence, this part of our study investigates the effect of the size of a dataset on the scoring functions. We again consult Figure 4 and Table 3 but this time look for convergence of the scoring functions; that is, we look to see at what point increasing sampling size does not change SHD anymore. As the figure shows, most of the scoring functions converged. To look for convergence in the table, we consider the mean, minimum, maximum, and standard deviation for the SHD statistics. We expect that if the scoring function converged quickly, its standard deviation will be small. This loose interpretation is robust in that it allows us to conclude that a scoring function converged even if SHD changes slightly from one sample size to the next. Figure 1 Comparing the evaluation measures for the optimal networks learned from the Austra datasets with different sizes. In this figure, we compare the performance of the four evaluation metrics (SHD, AHD, accuracy, and sensitivity) for the Australian Credit Approval dataset. The y-axis label indicates which evaluation metric that graph displays. We display the results for a = 1 for BDeu for all measures because it had the best convergence behavior for this dataset. We used the behavior of each of the curves to evaluate the convergence of the corresponding scoring function. We consider a scoring function to have converged for an evaluation metric when increasing the dataset size does not change the value for that scoring function and evaluation metric. Thus, we look for "flat lines" in the graphs.
As previously shown [7], fNML converges with fewer samples than the other scoring functions. Because the mean SHD is typically small, we conclude that the network to which it converges is often close to the gold standard network. MDL converged somewhat more slowly, but often converged to the gold standard network. BDeu with an optimal a value tends to converge quickly to a network close to the gold standard networks; however, with a sub-optimal a value, BDeu often neither converges nor comes close to the gold standard networks as shown in Table 2. Because AIC has a very low penalty term, more data encourages it to add more edges. Thus, it tends to overfit the data on large sample sizes and rarely converges. The SHD of AIC does tend to decrease as the sampling size increases, but that trend is somewhat inconsistent. Based on these results, fNML seems to be a good scoring function when data is limited, while MDL is superior when more data is present.
Comparison to greedy hill climbing
Finally, we compared the network recovery and convergence ability of a greedy hill climbing learning algorithm to those from the optimal algorithm. We performed this analysis because, as mentioned, optimal learning algorithms are limited to datasets with several dozens of variables. While some biological datasets (such as the Breast Cancer, Cleveland Heart Database, Diabetes, Statlog (Heart), Hepatitis and Iris datasets included in this study) are within this limit, many others, such as gene expression datasets, include hundreds or thousands of variables. Greedy hill climbing algorithms have been shown to scale to datasets of this size [14]. This part of our study verifies that our conclusions on scoring functions apply to this algorithm, as well.
We first evaluated the network recovery ability of the scoring functions on the greedy hill climbing algorithm. Table 4 shows that, much like the optimal learning algorithms, the hill climbing algorithm typically either adds extra edges or misses necessary edges. On the other hand, as the small values in the Reverse and Compelled columns show, the directionality of the edges is typically correct. The Total SHD follows a similar trend among the greedy hill climbing and optimal algorithms. That is, scoring functions that performed well for the optimal Figure 2 Comparing the evaluation measures for the optimal networks learned from the Cleve datasets with different sizes. In this figure, we compare the performance of the four evaluation metrics (SHD, AHD, accuracy and sensitivity) for the Cleve dataset.
algorithm also performed well for the hill climbing algorithm. We observed similar results on the other datasets as shown in the Additional file 1 S1.xls (sheet = results . greedy). These results confirm that the scoring functions have a similar impact on structure recovery regardless of whether an optimal or greedy algorithm is used. In almost all cases, though, the optimal algorithm finds a structure closer to the true gold standard networks, so its Total distance is always lower. This highlights the benefit of using optimal algorithms when possible.
We then evaluated the convergence behavior of the scoring function on the greedy hill climbing algorithm. As shown in Figure 5, the picture is not as clear as the convergence behavior of the optimal algorithm in Figure 4. Nevertheless, we still see similar trends. Of the scoring functions, fNML typically converges the quickest, though often to a worse network than MDL. On the Breast Cancer and Car Evaluation datasets, MDL converges to the gold standard network, except for a few perturbations caused by the uncertainty of the greedy search algorithm. BDeu also converges except for a few spikes, but it typically converges to a worse network than MDL. As with the optimal algorithm, AIC does not converge. These results also mirror those of the behavior we observed in the optimal algorithm, though a bit noisier. They again suggest that the conclusions we drew from the optimal algorithms apply to the greedy algorithm, albeit with some noise. We also see that the optimal algorithm gives more consistent behavior, both in terms of quality and consistent convergence, and should be used when possible.
Conclusion
In this work, we have empirically investigated the ability of four Bayesian network scoring functions (MDL, AIC, BDeu and fNML) to recover the generating distribution of a dataset; a gold standard Bayesian network represents Figure 3 The effect of the hyperparameter a on the BDeu score. This figure plots the SHD between the networks learned by BDeu and the gold standard networks for six values of a for the Breast, Glass, Diabetes, and Hepatitis datasets. We used the behavior of each curve to evaluate both the convergence and the recovery ability of each value of a. We evaluate the recovery ability by considering both the smallest SHD for the scoring function, the size of the dataset which gives that SHD, and whether the scoring function converged to the smallest SHD, some other SHD or did not converge. GoldNet gives the name of the network. We have used abbreviated names from Table 1, but the order of the datasets is the same in both tables. Min, Mean, Max and STD give the particular statistic for SHD for all sample sizes for the given network and a value. The a value with the lowest mean for each dataset is shown in bold and marked with "*". This table shows SHD statistics about the networks learned using the sampled datasets for all scoring functions that we analyzed. For BDeu, we used the value of a that gave the lowest mean SHD. GoldNet gives the name of the network. We have used abbreviated names from Table 1, but the order of the datasets is the same in both tables. Min, Mean, Max and STD give the particular statistic for SHD for all sample sizes for the given network and scoring function. The scoring function with the lowest mean for each dataset is shown in bold. this distribution. We used an optimal structure learning algorithm to ensure approximation algorithms did not affect the learned networks. All optimal learning algorithms would learn an equivalent network, so our choice of optimal algorithm did not affect our results or conclusions. Then, we controlled scoring function and sample sizes to test their effect on the quality of the learned networks. We also considered four different evaluation metrics: accuracy, sensitivity, AHD and SHD. In addition, we evaluated a greedy hill climbing algorithm to ensure that our conclusions are valid for algorithms which can learn networks with hundreds or thousands of variables. As a result of our investigation, we discovered that SHD is more well-behaved than the other evaluation metrics because it considers equivalence classes when comparing structures rather than the specific DAGs. Our most surprising result was that MDL was better able to recover gold standard networks than other scoring functions given sufficient data. As expected, BDeu's performance was highly dependent on the selected a parameter, which can be difficult to estimate a priori.
We also confirmed that fNML converges even with few samples. Throughout our analysis, we found AIC's behavior erratic and unpredictable. The greedy hill climbing algorithm exhibited similar behavior, so we conclude that our results hold for this algorithm, as well.
We plan to extend this work in several ways. We can use synthetic networks to more carefully control the properties of our gold standard networks. Unlike previous studies, though, we will not rely on random network generation; instead, we will handcraft a variety of networks to reflect a variety of real-world datasets. We will also incorporate other scoring metrics, such as MIT [8], and objectives, such as prediction [9], into our study.
Additional material
Additional file 1: Detailed empirical results and free software packages. The file (S1.xls) contains detailed empirical results from testing the various combinations of the scoring functions, sample sizes, and learning algorithms (sheet = results . optimal, results . greedy). It also contains a list of free software packages used in this study (sheet = Software).
Figure 5
Plot of structural Hamming distance of the networks learned by the sub-optimal learning algorithm from datasets with different sample sizes. This figure plots the SHD of the networks learned by each of the scoring functions for the Breast, Crx, Car, and Diabetes datasets. We display the results for a = 0.5 for BDeu for all datasets because it had the best behavior in terms of SHD.
|
2024-02-27T01:03:25.941Z
|
0001-01-01T00:00:00.000
|
{
"year": 2012,
"sha1": "51bfd9b02307f7fd66502e124b4601246b7e61e1",
"oa_license": "CCBY",
"oa_url": "https://bmcbioinformatics.biomedcentral.com/track/pdf/10.1186/1471-2105-13-S15-S14",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "39311a1b6a504e680243d5b824171ed8abd8ea23",
"s2fieldsofstudy": [
"Computer Science",
"Mathematics"
],
"extfieldsofstudy": []
}
|
15308089
|
pes2o/s2orc
|
v3-fos-license
|
Uropathogenic Escherichia coli Modulates Immune Responses and Its Curli Fimbriae Interact with the Antimicrobial Peptide LL-37
Bacterial growth in multicellular communities, or biofilms, offers many potential advantages over single-cell growth, including resistance to antimicrobial factors. Here we describe the interaction between the biofilm-promoting components curli fimbriae and cellulose of uropathogenic E. coli and the endogenous antimicrobial defense in the urinary tract. We also demonstrate the impact of this interplay on the pathogenesis of urinary tract infections. Our results suggest that curli and cellulose exhibit differential and complementary functions. Both of these biofilm components were expressed by a high proportion of clinical E. coli isolates. Curli promoted adherence to epithelial cells and resistance against the human antimicrobial peptide LL-37, but also increased the induction of the proinflammatory cytokine IL-8. Cellulose production, on the other hand, reduced immune induction and hence delayed bacterial elimination from the kidneys. Interestingly, LL-37 inhibited curli formation by preventing the polymerization of the major curli subunit, CsgA. Thus, even relatively low concentrations of LL-37 inhibited curli-mediated biofilm formation in vitro. Taken together, our data demonstrate that biofilm components are involved in the pathogenesis of urinary tract infections by E. coli and can be a target of local immune defense mechanisms.
Introduction
It has been recognized that bacteria in their natural milieu seldom grow as non-differentiated, single cell organisms. Instead, they form multicellular communities, biofilms, showing coordinated behavior [1]. Classically, biofilm formation includes surface adherence, cell-cell interactions, and production of extracellular matrix [2]. The extracellular matrix contributes to the development of higher-ordered three-dimensional structures that offer advantages to the bacteria, such as increased resistance to antimicrobial substances, mechanical forces and to nutrient depletion [3][4][5]. During urinary tract infections (UTI), the role of bacterial biofilms has previously been established in the presence of indwelling catheters [6]. However, uropathogenic E. coli also forms biofilm-like structures on and inside host cells in the absence of a foreign body [7][8][9], and the ability to form biofilms has been related to persistence of bacteria in the urinary tract [10].
Curli belong to a class of fibers known as amyloids [11] and are involved in adhesion to surfaces, cell aggregation and, finally, biofilm development. Functionally and genetically, curli are linked to cellulose [12], another extracellular matrix component of biofilms formed by bacteria from the family Enterobacteriaceae. Bacterial cellulose has mostly been investigated in soil bacteria of the family Rhizobiaceae, where this polysaccharide is required for the firm adherence and aggregation of bacteria at the root hair tip of plants [13]. Although the production of cellulose is common among many bacterial species, its biological function and role during infection is not entirely clear. When cellulose is expressed together with curli, the two substances produce a highly inert, hydrophobic extracellular matrix around the bacteria [14].
Biofilms built from curli and cellulose have widely been investigated on abiotic surfaces [15,16] and in commensal intestinal E. coli isolates [17,18]. Less information is available about the role of curli and cellulose during E. coli infection of the urinary tract [10,19].
Recently, we demonstrated that epithelial cells of the urinary tract up-regulate the production of the human antimicrobial peptide LL-37 upon infection with uropathogenic E. coli [20]. Thus, the cathelicidin LL-37 plays an important role in the protection against infections of the urinary tract. The proform of LL-37, hCAP-18, is mainly produced by epithelial cells and neutrophils [21,22]. After processing [23], the active LL-37 peptide is released and exhibits its bactericidal activity by interaction with the bacterial cell membrane [24].
In the current project, we sought to study the presence of curli and cellulose in E. coli isolated from uncomplicated communityacquired UTI and their impact on early UTI pathogenesis. In addition, we here investigate the influence of LL-37 on curlimediated biofilm formation in E. coli. We suggest that curli and cellulose protect the bacterium from immune defense mechanisms and in addition modulate the immune response of the host. We furthermore demonstrate an interaction of curli and LL-37, especially that LL-37 inhibits the polymerization of CsgA, the major subunit of curli.
Results
Uropathogenic E. coli isolates have higher adhesion capacity and produce more biofilm than commensal E. coli A total of 99 E. coli isolates were collected from urine of patients with UTI and 77 isolates were obtained from fecal samples of healthy individuals. Each isolate was assessed for biofilm formation using a standard microtiter assay (see Materials and Methods). On average, uropathogenic bacteria adhered significantly better and formed more biofilm as compared to fecal isolates (P,0.0001, Figure 1A).
Uropathogenic and commensal E. coli isolates produce extracellular matrixes with similar composition Curli and cellulose production by all isolates was monitored on Congo red and Calcofluor containing plates. To better mimic the host environment, we chose to analyze bacteria grown at 37uC. Based on the uptake of Congo red and fluorescence after the exposure of Calcofluor plates to UV light, we could identify that approximately half of the uropathogenic and commensal E. coli isolates expressed curli (54% and 45%, respectively), 30% of the uropathogens and 16% of the commensals expressed curli and cellulose together. This morphotype was significantly associated to uropathogenic E. coli (P = 0.032). The expression of cellulose alone was rarely detected in either collection (5% and 10%, respectively). Nearly all isolates were positive for expression of type 1 fimbriae, irrespective of their origin (99% of the uropathogenic and 92% of the fecal isolates).
Curli are expressed in isolates from fresh urine samples
To confirm the expression of curli in vivo, we collected fresh urine samples from patients with community-acquired E. coli UTI. Bacteria were analyzed directly from the urine by dot blot analysis and electron microscopy. Ten of seventeen investigated clinical E. coli isolates (59%) reacted with antibodies towards CsgA. This was in line with their Congo red/Calcofluor phenotype and the overall prevalence of curli in uropathogenic E. coli investigated here (54%). Bundles of curli expressed during UTI were visualized by electron microscopy and their identity was confirmed by goldlabeled antibodies ( Figure 1B).
Expression of curli increases the propensity of E. coli to cause infection
The relevance of curli and cellulose expression on bacterial adhesion and immune induction in target cells was investigated by the interaction of bacteria with human cells in vitro ( Figure 2). Bladder (UROtsa, T24) and renal (A498) epithelial cells were infected with the uropathogenic E. coli strain No. 12, producing curli and cellulose; and its isogenic mutants lacking curli and/or cellulose. The wild-type strain and its mutants also expressed type 1 fimbriae to similar extent. The total number of bacteria after 30 min of cell infection was determined. Curli expression resulted in an increased number of cell-associated bacteria in the presence or absence of cellulose (P,0.0001, Figure 2A). Likewise, levels of IL-8 were significantly higher in supernatants of cells infected with curliated E. coli compared to those induced by the respective non-curliated strain (P = 0.001 and P,0.0001 for celluloseexpressing and lacking strains, respectively, Figure 2B). Cellulose, on the other hand, reduced the ability of bacteria to adhere (P,0.0001 and P = 0.001 in the presence and absence of curli, respectively, Figure 2A). In curliated bacteria, cellulose expression significantly reduced the induction of IL-8 (P = 0.001, Figure 2B). The role of curli and cellulose on adherence and IL-8 induction was confirmed by complementation of the curli and cellulosedeficient mutants which restored the wild-type phenotype ( Figure 2C+D).
To confirm the role of curli and cellulose during the initial infection steps, mice were infected with the isogenic E. coli strains. After 1 h of infection, the expression of curli increased the number of bacteria significantly only in the absence of cellulose (P = 0.026, Figure 2E), whereas the comparison between the wild-type strain and the curli-lacking mutant was not significant. However, comparing the pair of curliated strains with the pair of noncurliated mutants, the effect of curli on adherence in vitro was supported (P = 0.007). Similar to the cell culture experiments, the
Author Summary
Most infections of the urinary tract are caused by uropathogenic E. coli. On abiotic surfaces, these bacteria are able to form biofilms, which protect them from various adverse environmental conditions. In this study, we sought to investigate whether two E. coli biofilm components, curli fimbriae and cellulose, provide a similar protection against innate immune defense mechanisms of the urinary tract. We put special emphasis on the interaction with the human antimicrobial peptide LL-37, which plays a crucial role in the protection against uropathogenic E. coli. We demonstrate that curli expression specifically reduces bacterial sensitivity to LL-37 by binding the peptide before reaching the bacterial cell membrane and exhibiting its bactericidal activity. A more general protection is mediated by cellulose, possibly by hiding immunogenic surface structures of the bacterium. In addition to providing protection, curli are also targeted by the immune system. The formation of new curli fibers is inhibited in the presence of LL-37. Moreover, curliated bacteria show higher immunogenicity than their non-curliated counterparts. Cellulose expression, on the other hand, appears to impair initial host colonization. In conclusion, our findings demonstrate an example of the tight interplay between bacterial virulence factors and the host immune defense.
expression of curli alone increased the induction of MIP-2 (P = 0.001, Figure 2F). Interestingly, the inhibitory effect of cellulose was even more pronounced in vivo, and was also observed in the absence of curli (P,0.0001 and P = 0.001 in the presence and absence of curli, respectively, Figure 2F). Moreover, none of the strains expressing cellulose induced MIP-2 levels significantly higher than levels in control mice inoculated with sterile PBS (32-52 pg/ml).
Bacterial cellulose influences neutrophil recruitment and elimination of E. coli from the kidney In the initial stages of UTI, curli promoted colonization ( Figure 2A+E). We further investigated the later course of UTI. Mice were infected with isogenic strains expressing curli and/or cellulose, and kidneys were analyzed 48 h post infection (p.i.). MIP-2 is the major neutrophil chemoattractant in the urinary tract [25]. Corresponding to immune induction ( Figure 2F), the curliated mutant was more efficiently eliminated after 48 h p.i. than the wild-type strain with cellulose (P = 0.011, Figure 3A).
To further investigate the role of cellulose in this process, we induced neutropenia in mice prior to infection. Neutrophildepleted and control mice were infected with curliated bacteria with or without cellulose. Clearance by neutrophils was more efficient for bacteria lacking cellulose ( Figure 3B). In neutrophildepleted mice, the number of cellulose-deficient bacteria after 48 h was as high as those of the wild-type strain.
Curli increase resistance to the antimicrobial peptide LL-37 To understand the mechanism underlying the more efficient infection by curliated bacteria, we specifically investigated the antimicrobial activity of bladder and renal epithelial cells on adhered bacteria. For this purpose, bacteria were coincubated with cells for 30 min and adherent bacteria were then subjected to a staining procedure allowing the discrimination between live and dead bacteria. Dependent on the expression of curli and cellulose, 19% to 67% of cell-associated bacteria were killed. Curli but also cellulose protected bacteria from antimicrobial activities of the cells ( Figure 4A, P,0.001 for curli and P = 0.024 and P = 0.003 for cellulose, respectively), most efficiently when both structures were expressed together.
Bladder and renal epithelial cells are known to produce cathelicidins in response to E. coli infection, in particular LL-37 in humans and mCRAMP in mice, respectively [20]. To relate the observed antimicrobial activity of uroepithelial cells to this peptide, bacteria were exposed to conditioned medium from cells stimulated with phenylbutyrate to enhance LL-37 production [26]. After 30 min of incubation, the number of curli-producing bacteria was almost unchanged (99% and 102% of the inoculated concentration), whereas bacteria lacking curli were reduced to 93% and 85% in the presence or absence of cellulose, respectively ( Figure 4B). The most pronounced difference was observed due to curli in the absence of cellulose (102% versus 85%, P = 0.006). Hence, we chose these two mutant strains for neutralizing experiments. Prior to inoculation, the activity of LL-37 in the Adhesion to renal epithelial cells A498 was measured after 30 min. Curliated strains adhered significantly better to renal epithelial cells than strains lacking curli, independent of the expression of cellulose ( ### P,0.0001, t-test). Cellulose expression decreased the number of cell-associated bacteria in curliated ( *** P,0.0001, t-test) and non-curliated strains ( ** P = 0.001, t-test). Results from three independent experiments in quadruplicates are shown as mean and standard deviation. Similar results were obtained for bladder epithelial cells (data not shown). (B) Induction of IL-8 was measured in culture supernatants of renal epithelial cells A498 stimulated with E. coli for 24 h. Curliated bacteria induced a significantly stronger IL-8 response than the mutants lacking curli in the presence ( ## P = 0.001, t-test) and absence ( ### P,0.0001, t-test) of cellulose. In curliated bacteria, the expression of cellulose reduced IL-8 induction ( ** P = 0.001, t-test). Results from three independent experiments in quadruplicate are shown as mean and standard deviation. Similar results were obtained for bladder epithelial cells (data not shown). (C+D) The phenotype of E. coli No. 12 could be restored by complementation of its mutants. Cellulose expression in strain B23 is inducible by aTc (left panels) and reduces adherence and IL-8 induction ( *** P,0.0001 and ** P = 0.007, respectively, t-test). The curli subunits CsgA and CsgB are expressed from pWSK29-csgBA in strain WE11_1 (right panels) and increases adherence and IL-8 induction compared to WE11_1 carrying the vector pWSK29 only ( * P = 0.048 and ** P = 0.003, respectively, t-test). Results in A498 cells are shown as mean and standard deviation. Data from three experiments in quadruplicate for adherence and from two experiments in triplicate for IL-8 culture medium was inhibited by neutralizing antibodies. While the number of viable curliated bacteria did not differ after 30 min ( Figure 4C, left), the number of bacteria lacking curli was significantly higher in the presence of LL-37-specific antibodies compared to the samples treated with an irrelevant isotype control antibody ( Figure 4C, right, P = 0.047).
We further investigated the influence of curli expression on bacterial sensitivity to LL-37 and mCRAMP more specifically by a broth dilution method. When bacteria were initially grown in biofilm, the concentration of LL-37 at which bacterial growth was inhibited to 50% (IC 50 ) was 12 mM for the curliated strains. However, the IC 50 for the non-curliated strains was only 6-7 mM ( Figure 4D). At 10 mM LL-37, the relative growth of the curliated strains was significantly higher than growth of the non-curliated strains (P = 0.001 and P,0.001 in the presence and absence of cellulose, respectively, Figure 4E). Similar results were obtained for the mouse cathelicidin mCRAMP ( Figure 4F+G). The IC 50 value was higher for the curliated strains than for the noncurliated strains (7 versus 4 mM, Figure 4F). At 5 mM mCRAMP, the relative growth differed significantly between the curliated and the non-curliated strains (P = 0.034 and P = 0.037 in the presence and absence of cellulose, respectively, Figure 4G). The same bacteria were then pre-grown planktonically, where curli expression is suppressed. When grown under such conditions, no significant difference in the resistance against both LL-37 and mCRAMP was observed between the strains (data not shown). These data indicate that curli is a biofilm component that counteracts the bactericidal effect of cathelicidins and may contribute to the increased resistance of E. coli growing in biofilm.
In contrast to curli, cellulose did not influence the IC 50 of cathelicidins ( Figure 4D-G).
LL-37 binds to curli fimbriae
In order to elucidate one possible mechanism that could influence the increased resistance of curliated bacteria against LL-37, the binding of LL-37 to wild-type curli and recombinant CsgA was assessed. A precipitation assay showed a pronounced decrease of LL-37 in supernatants from samples containing wildtype curli or polymerized CsgA ( Figure 5A). Further, LL-37 binding to both monomeric and polymeric CsgA was demonstrated by surface plasmon resonance ( Figure 5B). By comparing the response during loading of the peptide in a time frame of 0-180 s, the sensogram of LL-37 demonstrated a higher association with CsgA than the control peptides, i.e. scrambled LL-37 (sLL-37) and the vasoactive intestinal peptide (VIP) [27]. Furthermore, the binding curves reveal that the control peptides had faster dissociation rates than LL-37, indicating a weaker binding to CsgA. This was especially pronounced for the binding to polymeric CsgA. Determination of binding constants was precluded, since LL-37 and in particular CsgA forms oligomers and polymers, respectively, and thus generate several different assemblies.
LL-37 prevents adherence and biofilm formation in vitro
At concentrations below the IC 50 for bacterial growth, LL-37 inhibited curli-mediated biofilm formation with a reduction of more than 80% at 2.5 mM for both the wild-type (data not shown) and the cellulose-negative E. coli strain ( Figure 6). To investigate induction are presented. (E) Mice were infected with the isogenic E. coli strains for 1 h. The curliated mutant were isolated from kidneys in significantly higher numbers than the double knockout ( # P = 0.026, Mann-Whitney U test). Individual values from n = 8-10 mice/group and medians are shown. (F) Levels of MIP-2 were measured in kidney tissue of infected mice after 16 h. In the absence of cellulose, the curliated mutant strain induced higher levels of MIP-2 compared to the non-curliated strain ( ## P = 0.001, Mann-Whitney U test). Expression of cellulose reduced the induction of MIP-2 in the presence ( *** P,0.0001, Mann-Whitney U test) and absence of curli ( ** P = 0.001, Mann-Whitney U test). Individual values from n = 5-10 mice/group and medians are shown. doi:10.1371/journal.ppat.1001010.g002 were infected with bacteria for 30 min and adherent bacteria were subjected to LIVE/DEAD staining. Curli and cellulose expression enhanced bacterial resistance to antimicrobial properties ( ### P,0.001 for curliated strains versus non-curliated strains, * P = 0.023 and ** P = 0.003 for cellulose expressing strains with or without curli, t-test). Combined data from four experiments are shown. (B+C) Bacteria were exposed to conditioned medium of bladder epithelial cells T24 stimulated with phenylbutyrate to enhance LL-37 production. Curli expression enhanced bacterial survival over 30 min ( ## P = 0.006, t-test) (B). Results from three experiments in triplicate are shown. Conditioned medium was incubated with neutralizing anti-LL-37-antibodies (nAb) or isotype control antibodies (Co) prior to bacterial inoculation. Neutralizing of LL-37 had no effect on viability of the curliated strain (left) but enhanced viability of the double knockout (right, * P = 0.047, t-test) (C). Results from four experiments in triplicate are shown. (D-G) The susceptibility to LL-37 and mCRAMP of E. coli strains expressing or lacking curli or cellulose was tested by the broth dilution method. The expression of curli increased the resistance to both LL-37 (D+E) and mCRAMP (F+G). A significant difference of bacterial growth was observed at 10 mM LL-37 between curliated and non-curliated strains ( ### P,0.001, t-test). The curliated strains were also significantly more resistant to 5 mM mCRAMP than bacteria not producing the specificity for the inhibitory capacity, sLL-37 and VIP were analyzed in the same assay. Our results showed that the same concentration of these peptides reduced biofilm formation by only 10%, which is a significantly lower reduction than the effect of LL-37 (P = 0.001, Figure 6B). This indicates a sequence-specific inhibition of curli-mediated biofilm by LL-37.
LL-37 inhibits the polymerization of CsgA
To explain a possible cause for the inhibition of biofilm formation by LL-37, the effect of LL-37 on curli formation was investigated. For this purpose, we utilized monomeric recombinant CsgA, the major subunit of curli, which spontaneously polymerizes [28]. CsgA polymerization was monitored with thioflavin T (ThT), a fluorescent dye that binds to polymerized, but not to monomeric CsgA. Our results demonstrated that CsgA polymerization started immediately after incubation at 37uC and reached a stationary phase after approximately 300 min (red line in Figure 7A). After prolonged incubation, the fluorescence declined, most likely due to degradation of ThT and/or curli ( # P,0.05, t-test). An increased resistance to both cathelicidins was not observed for cellulose. Mean and standard deviation from data of two separate experiments in triplicates are shown. The IC 50 is indicated by a broken line. doi:10.1371/journal.ppat.1001010.g004 caused more than 80% reduction of biofilm production, whereas the same concentration of the control peptides gave a reduction of only ,10%. Mean and standard deviation from data of two separate experiments in triplicates are shown. The difference between LL-37 versus sLL-37 or VIP at 2.5 mM was statistically significant ( *** P = 0.001, ttest). Similar results were obtained for the wild-type strain expressing both curli and cellulose (data not shown). doi:10.1371/journal.ppat.1001010.g006 precipitation of fibers [28]. The polymerization was inhibited by LL-37 in a dose-dependent manner, and at a molar ratio of 1:1 (CsgA:LL-37) fiber formation was completely inhibited ( Figure 7A, left panel). The control peptides sLL-37 and VIP had a less pronounced effect on CsgA polymerization than LL-37 ( Figure 7A, right panel).
Similar results could be achieved using confocal microscopy. After 20 h incubation of CsgA alone, we could clearly detect fiber structures stained with ThT ( Figure 7B) or Congo red (data not shown). In line with the results described above, these fibers were not detected when LL-37 was present in a molar ratio of 1:1, giving a fluorescence of only 0.05 arbitrary units compared to CsgA alone (1 arbitrary unit, Figure 7C). The control peptides sLL-37 and VIP reduced the fluorescence intensity to approximately 0.2 arbitrary units, suggesting a lower inhibitory capacity than LL-37. Thus, inhibition of CsgA fiber formation was evidently stronger for LL-37 (P = 0.007 and P = 0.011 versus sLL-37 and VIP, respectively, Figure 7C). The structure and the levels of CsgA monomers remain stable in the presence of LL-37 To confirm the inhibition of CsgA polymerization we sought to analyse the stability of the CsgA monomer in the absence and presence of LL-37. Freshly purified, monomeric CsgA (10 mM) was incubated for 20 h at 37uC without or with different concentrations of LL-37 and was subsequently separated by SDS-PAGE. After staining with Coomassie Blue, bands corresponding to LL-37 and/or CsgA in monomeric, dimeric or tetrameric form were visualized. When CsgA was incubated alone, monomers were not visible although sometimes dimers and/or tetramers could be observed. This finding indicates spontaneous formation of CsgA oligomers and/or larger polymers that can not migrate into the gel due to their size. However, in the presence of LL-37, a band migrating at 15 kDa, the predicted size of monomeric CsgA, could be observed ( Figure 8A). This was already seen at a molar ratio of 16:1 (CsgA:LL-37). To exclude degradation of CsgA as a possible explanation for the lack of the gel band, polymerized CsgA was treated with 90% formic acid, dissolving polymeric CsgA into monomers. After this treatment, a band corresponding to monomeric CsgA was detectable in the gel ( Figure 8A).
The impact of LL-37 on the structure of curli was investigated with circular dichroism (CD) spectroscopy. In line with previous findings, polymeric CsgA exhibited a beta-sheet conformation with a minimum around 218-220 nm ( Figure 8B) [28]. Furthermore, the low signal amplitude indicates a decreased solubility due to polymerization ( Figure 8B). In contrast, CsgA together with LL-37 displayed a random coil structure as has been described for monomeric CsgA [28]. This result indicates that LL-37 is able to stabilize CsgA in an unstructured form.
Discussion
In the present study, we show that the majority of uropathogenic E. coli from uncomplicated community-acquired UTI adheres stronger and produces more biofilm compared to commensal bacteria. Two major extracellular components in E. coli biofilm are curli and cellulose. We here sought to explore their role in the course of UTI and their interaction with the human antimicrobial peptide LL-37. During early stages of UTI, curli promote colonization and immune induction. Cellulose in contrast reduces MIP-2 induction, followed by impaired bacterial eradication by neutrophils. The antimicrobial peptide LL-37 produced by uroepithelial cells and neutrophils in the urinary tract interacts with curli-mediated biofilms. Curli bind LL-37, and thus protects the bacterial cell against the bactericidal activity of LL-37. On the other hand, by binding to CsgA monomers and likely also shorter oligomers, LL-37 inhibits CsgA polymerization and curli formation.
We here for the first time provide evidence that curli are present on E. coli in fresh urine of infected patients that are not catheterized ( Figure 1B). The expression of curli or cellulose was equally common among E. coli isolates from UTI and commensal fecal isolates. However, the combined expression of curli and cellulose was the most common phenotype among uropathogenic isolates whereas most of the fecal isolates expressed only curli.
In the pathogenesis of infection, curli fimbriae have previously been implicated in the attachment and invasion of host cells, interaction with host proteins and activation of the immune system [11,29,30]. Cytokine induction by Salmonella has been associated with binding of CsgA to toll-like receptor 2 [31], which is expressed on bladder and renal epithelial cells [32]. In our clinical samples, urinary IL-8 levels did not correspond to curli expression, and did not differ between isolates expressing different biofilm morphotypes (data not shown). However, virtually all tested clinical UTI isolates expressed type 1 fimbriae. Since type 1 fimbriae and other bacterial factors such as lipopolysaccharides are potent inducers of IL-8, the impact of curli on IL-8 induction was possibly masked [32,33]. In addition, it can not be ruled out that the lag time between onset of symptoms and the first visit to the hospital, when urine samples were obtained, also influenced the results.
However, we did observe a clear correlation between curli expression and IL-8 induction in bladder and renal epithelial cells ( Figure 2B) as well as MIP-2 in mice infected with the isogenic strains ( Figure 2F). Interestingly, curli-dependent IL-8 induction was also observed in A498 kidney cells, which have been found to lack toll-like receptor 2 [32,34]. It has been reported for this cell line, that IL-8 induction is probably increased by type 1 fimbriaemediated attachment [32,35]. We speculate that adherence enhanced by curli could similarly increase the immune induction in A498 cells in our experiments. Another explanation would be that there is an alternative route not yet identified mediating the immune response.
Recruitment of neutrophils is mediated by IL-8 and MIP-2 in humans and mice, respectively [33,36]. The crucial function of MIP-2 and neutrophils in the defense of the urinary tract [25] is illustrated here by less efficient elimination of the curliated, highly immunogenic cellulose mutant strain in neutrophil-depleted mice ( Figure 3B). In contrast, clearance of the wild-type strain expressing cellulose is not significantly influenced by the lack of neutrophils ( Figure 3B), consistent with low levels of MIP-2 detected in wild-type infected mice ( Figure 2F). It is well known that in wound healing bacterial cellulose itself does not induce inflammation [37]. The role of bacterial cellulose in the pathogenesis of infections, however, has previously not been established. Our results suggest a protective role against the immune system. Cellulose might mask bacterial surface structures, hence avoiding immune recognition and cytokine induction, or alternatively, actively decrease the immune response. In the cell culture model, we see a significant reduction of IL-8 after infection with the wild-type strain compared to the mutant expressing only curli, whereas there is no reduction due to cellulose in the absence of curli. It is reasonable to believe that cellulose might be able to cover the relatively short curli fibers but not longer structures such as type 1 fimbriae, which are expressed by all four strains utilized in this study. In the mouse model, we see a reduced MIP-2 induction in the presence of cellulose also in the absence of curli ( Figure 2F). Moreover, the inhibitory effect of cellulose on MIP-2 induction appears to be stronger. Thus, the production of cellulose might be an efficient protection for bacteria not only against environmental conditions but also against immune defense mechanisms in vivo.
We have recently shown that LL-37 plays a crucial role in urinary tract innate immune defense [20]. Here we demonstrate increased resistance of curliated bacteria towards the antimicrobial properties of uroepithelial cells ( Figure 4A-C), which is at least partly based on increased resistance against LL-37 ( Figure 4C-E). We confirmed the relevance of this observation for the mouse infection model by investigating the susceptibility of wild-type and mutant E. coli against mCRAMP, the murine LL-37 ortholog. Our results revealed that the curliated strains are more resistant also against the mouse cathelicidin ( Figure 4F+G), indicating a similar interaction with curli as demonstrated for LL-37.
Curli fibers are mainly composed of polymerized CsgA. By precipitation of LL-37 in the presence of wild-type curli or recombinant polymeric CsgA, we demonstrate binding between the peptide and the protein ( Figure 5A). Antimicrobial peptides including LL-37 kill their target cells by a peptide-bacterial membrane interaction that leads to lysis of the bacterium [24]. Our finding suggests that LL-37 might be trapped in the net of curli covering the bacterial surface. This prevents LL-37 from reaching the bacterial membrane and lysing the cell. In contrast, bacteria without curli lack this protection and are more easily killed during adherence and invasion into the uroepithelium. A similar protective role against cationic antimicrobial peptides has been observed for the biofilm components alginate in Pseudomonas aeruginosa [38] and the polysaccharide intercellular adhesion (PIA) in Staphylococcus epidermidis [39]. We also observed partial protection mediated by cellulose, the polysaccharide component in E. coli biofilm ( Figure 4A+B). However, this effect was not as pronounced as the protection by curli and could not be related to LL-37, since cellulose production did not influence bacterial susceptibility to the cathelicidins LL-37 or mCRAMP in vitro ( Figure 4D-G). Interestingly, we demonstrate that LL-37 inhibits the formation of curli-promoted biofilm formation in vitro ( Figure 6). We also show that LL-37 prevents CsgA polymerization (Figures 7+8), and speculate that LL-37 inhibits biofilm establishment by direct interference with CsgA assembly. The binding of LL-37 to both monomeric and polymeric CsgA might block reactive surfaces that are crucial for the CsgA-CsgA interaction [40]. There are five segments in the CsgA amino acid sequence that are conserved and share similarity to each other. They are characterized by conserved Ser, Gln and Asn residues [11,40], but these repeats also contain acidic residues that may contribute to an electrostatic interaction with the cationic peptide LL-37. Based on the general structures of amyloids [41], each of these repeats is predicted to form a strand-loop-strand motif in a strong hydrogen bonding network [11], which might be prohibited by the binding of LL-37. Considering the function of curli during infection, the prevention of curli generation would provide an effective host defense mechanism. Our in vivo results demonstrate, despite the initial advantage of curliated bacteria, that they are eradicated more efficiently at later stages of infection. The expression of many virulence factors is highly regulated by environmental conditions, and this has also been shown for curli. Curli are maximally expressed in stationary phase and participate in the initial stage of biofilm, i.e. irreversible attachment, whereas expression might be down-regulated during biofilm maturation correlating to later stages of infection [11,42]. Reduced curli expression by bacteria colonizing the kidney makes them more vulnerable towards LL-37. In addition to increased bacterial sensitivity, incoming neutrophils release high amounts of LL-37 that contributes to the antibacterial defense. In growing bacteria, the generation of new curli fibers might be inhibited by LL-37, reducing both protection and ability to colonize the host tissue. Remarkably, biofilm-inhibitory concentrations of LL-37 were much lower than bactericidal concentrations and within a range which can be present in vivo [20,43]. In contrast, subinhibitory concentrations of exogenous antimicrobial drugs, e.g. aminoglycosides, seem to stimulate bacteria to produce biofilm [44]. Moreover, bacteria grown in biofilm are less susceptible to most exogenous antimicrobial agents [45].
Biofilm inhibition by antimicrobial polypeptides has previously been described for Pseudomonas aeruginosa. Both LL-37 [46] and lactoferrin [47] increased bacterial surface motility mediated by type IV pili. A direct interaction with biofilm components was not investigated in these studies. The effect was rather related to an influence of LL-37 on the bacterial gene expression profile [46] or an influence of lactoferrin on free iron [47]. These and our findings stress an important anti-biofilm role of antimicrobial polypeptides in host defense. It is likely that the anti-biofilm activity is a general strategy for these host defense molecules to keep potential pathogenic bacteria more vulnerable to killing in various tissues, including the urinary tract.
In conclusion, we demonstrate that uropathogenic E. coli by expressing curli are able to modulate the immune response and display increased virulence. Cellulose, on the other hand, may reduce adherence and immunogenicity by masking bacterial surface structures, thereby evading the immune system. We also show that defense mechanisms in the urinary tract interfere with these biofilm components; curli protect the bacteria from being killed by LL-37, in contrast LL-37 is inhibiting the formation of curli fibers. This inhibition might be an important host defense mechanism in the protection against UTI.
Patients
The studies have been approved by the ethics committee of the Karolinska University Hospital, and written informed consent has been obtained from the patients and parents of the children, respectively, in accordance with the ethics permission.
The clinical study included 98 patients with UTI; 36 children [20] and 62 adult women [48]. One woman suffered from two episodes of UTI with different E. coli isolates. The diagnostic criterion of acute UTI was the presence of $10 5 CFU of E. coli per ml of freshly voided urine. Except for bacteriuria, the diagnostic criteria of acute pyelonephritis were: body temperature $38uC and laboratory signs of systemic inflammation, either C-reactive protein $20 mg/liter or erythrocyte sedimentation rate $20 mm/h, respectively. In addition, fecal commensal E. coli isolates were collected from 77 adults in connection with routine outpatient health examination. None of them had a history of symptomatic UTI or recent gastrointestinal disease, and their urine did not yield E. coli on cultivation [48].
Human cell lines
Two human bladder epithelial cell lines and one human renal epithelial cell line were utilized. Virus-immortalized bladder epithelial cells UROtsa were kindly provided by Prof. Scott Garrett, Department of Pathology, University of North Dakota and cultured as described previously [20,49]. Bladder epithelial cells T24 (HTB-4; American Type Culture Collection (ATCC), Rockville, MD, USA) were cultured in McCoy's 5A medium containing glutamine (Invitrogen Life Technologies, Carlsbad, CA, USA) supplemented with 10% FBS (Invitrogen). Human renal epithelial cells A498 (HTB-44; ATCC) were cultured as described before [20].
Bacteria for in vitro and in vivo experiments
For further investigation, E. coli isolate No. 12 from a child with pyelonephritis was chosen. This was a typical isolate expressing both curli and cellulose as well as type 1 fimbriae and yielding an approximately median level of biofilm as measured on microtiter plates. One-step knockout of bcsA and csgBA was carried out according to the protocol of Datsenko and Wanner with modifications [17,50]. The following mutants were constructed using oligos listed in Table 1; WE1 bcsA::Cm, deficient in cellulose production; WE11 csgBA::Cm, deficient in curli production; and WE16 csgBA bscA::Cm, deficient in both cellulose and curli production. Expression of curli and cellulose in strain No. 12 was confirmed by these knockouts. Production of type 1 fimbriae under the culture conditions used for experiments was confirmed by yeast agglutination (see below).
For relevant control experiments, complementation equivalents for the mutants WE1 and WE11 were constructed. The complementation of strain WE11 was achieved as follows: First, the chloramphenicol cassette of the curli-deficient strain WE11 csgBA::Cm was removed by Flp-catalyzed excision as described elsewhere [51]; resulting in the Cm-sensitive, curli synthesisdeficient strain WE11_1. The removal was confirmed by PCR using oligos Csg fw and Csg rev ( Table 1). The DNA region comprising the csgBA operon was amplified from strain No. 12 using the above mentioned oligos and cloned into vector pWSK29 [52]. Finally, the obtained plasmid pWSK29-csgBA was transferred into strain WE11_1. The ability of strain WE11_1 containing pWSK29-csgBA to produce curli was demonstrated by morphotype assessment on Congo red plates.
The complementation of the cellulose production-deficient strain WE1 proved to be more complex, since plasmid-based complementation approaches failed. Therefore, strain B23 was constructed which originates from wild-type strain No. 12 and carries an inducible promoter upstream of the bcs operon. In short, the previously described kmRExTET cassette [53] which contains the anhydrotetracycline (aTc) inducible tetA promoter was amplified using oligos kmRExTET-bcs fw and kmRExTET-bcs rev (Table 1) and inserted upstream of the bcs operon using the protocol of Datsenko and Wanner with modifications [50]. Insertion of the kmRExTET cassette was confirmed by PCR using oligos bcsE control fw and yhjQ control rev (Table 1). In the resulting strain B23, cellulose production became an aTcdependent occurrence due to the insertion of the kmRExTET cassette as previously communicated [54]. In the absence of aTc, the morphotype of strain B23 is consistent to the morphotype of strain WE1 bcsA::Cm. In presence of aTc, cellulose production in strain B23 is restored to comparable levels than in the wild-type strain No. 12 as judge on Congo red plates. Thus strain B23 grown under inducing conditions can be used as a complemented equivalent.
Biofilm assays
Microtiter plate method to measure bacterial adhesion and thickness of biofilm. To measure the ability of the bacteria to adhere and to form biofilm a crystal violet assay in polystyrene microtiter plates (Costar, Corning, NY, USA) was performed [17]. Bacteria were grown in Luria-Bertani (LB) broth without salt for 24 h at 37uC without shaking. Biofilm was then stained with crystal violet (3%). The dye was solubilized with ethanol (95%) and the optical density was measured at 570 nm.
Morphotype analysis on Congo red and Calcofluor
plates. Bacteria were grown at 37uC for 24 h on Congo red and Calcofluor plates and analyzed as described previously [17]. The mutants generated from strain No. 12 served as controls for the analysis of clinical isolates.
Expression of type 1 fimbriae. Expression of type 1 fimbriae was tested by mannose-sensitive agglutination of yeast cells. Bacteria were grown in LB broth without shaking to induce expression of type 1 fimbriae, centrifuged and suspended in PBS (approximately 10 10 CFU/ml). Bacteria were mixed 1:1 with a suspension of Baker's yeast (Saccharomyces cerevisiae, 3% in PBS) and inspected for agglutination. Specificity of the reaction was tested by the inhibitory effect of mannose (5% in PBS). Bacteria were subcultured in broth up to three times before considered negative for expression of type 1 fimbriae. To quantify fimbrial expression in the isogenic strains, the bacterial suspension was subjected to serial two-fold dilution and mixed with an equal volume of yeast suspension. Agglutination was monitored and the optical density of the highest dilution giving a positive result was recorded. Specificity of agglutination was confirmed by mannose sensitivity.
Cell experiments
All assays were performed using both bladder (UROtsa, T24) and renal epithelial (A498) cells grown on 24-well plates (Costar). Experiments were performed in quadruplicates and repeated at least three times independently. Wild-type and mutant E. coli strains were cultured for 24 h at 37uC on LB agar plates without salt to promote the formation of biofilm. Medium was supplemented with ampicillin (100 mg/ml) or aTc (50 ng/ml) if appropriate. Colonies were scraped off and suspended in PBS. Bacterial cell clusters were then removed by centrifugation at 1506g for 10 min. The number of bacteria was determined spectrophotometrically at 600 nm and confirmed by viable count on blood agar plates after serial dilutions in PBS.
Cell infection. The experiments were performed in serumfree medium supplemented with gentamicin (40 mg/ml). Confluent layers of cells were infected with 10 6 CFU/ml of E. coli. Cells incubated with medium only served as controls. After incubation at 37uC in 5% CO 2 and 80% humidity for 24 h, medium was aspirated, centrifuged at 3506g for 10 min and stored at 220uC prior to ELISA analysis. The viability of cells during the experiments was confirmed using the Trypan blue method. Adhesion assay. To evaluate the ability of bacteria to adhere to epithelial cells, 10 6 CFU/ml of E. coli was added to cell culture wells in serum-and antibiotic-free medium and incubated at 37uC. After 10 or 30 min, cells were washed three times with PBS (37uC) to remove non-adherent bacteria. To collect cell-associated bacteria, ice-cold PBS with 1% Triton X-100 was added. Lysates were plated on blood agar plates after serial dilution in PBS and bacterial numbers were counted after over-night incubation at 37uC.
Epithelial cell antimicrobial assays
To access the inhibitory activity of epithelial cells on E. coli growth and viability two experimental settings were employed.
Antimicrobial activity on adherent bacteria. The viability of cell-adherent bacteria expressing or lacking curli and/or cellulose was investigated by LIVE/DEAD staining. For this purpose, E. coli were cultured as described above and 10 7 CFU/ml in serum-and antibiotic-free medium were added to confluent cell layers grown on sterile glass cover slips. After 30 min of incubation at 37uC, non-adherent bacteria were removed by washing three times with PBS. In order to include intracellular bacteria in the staining process, cells were permeabilized with saponin (0.2% in PBS, Sigma-Aldrich, St. Louis, MO, USA) for 5 min before addition of the two-component LIVE/DEAD BacLight stain (Invitrogen) diluted in 0.2% saponin. After 5 min, cells were washed with PBS and lightly fixed in freshly diluted 0.1% paraformaldehyde (PFA) for 15 min. At this concentration, PFA did not affect the fluorescence intensity of any of the dye components. Cover slips were mounted in ProLong Gold antifade mounting medium (Invitrogen) and immediately examined in a Leica TCS SP5 confocal microscope. To ensure correct discrimination between live and dead bacteria, stained green and red, respectively, microscope filter acquisition settings were adjusted using preparations of live and/or dead bacteria. Staining of the cell nucleus with both the cell permeable green dye and the cell impermeable red dye served as a positive control for cell permeabilization. Red or green bacteria were manually counted in microscope images of two separate preparations of three or four independent experiments.
Antimicrobial activity of conditioned medium. In order to investigate secreted antimicrobial components and to specify the active compound, bacteria were incubated in conditioned medium. To especially investigate the influence of LL-37 on bacterial viability in relation to curli and cellulose expression, production of LL-37 was stimulated with phenylbutyrate prior to medium collection [26]. Cells were grown in complete medium to reach ,80% confluence and medium was then exchanged to serum-free medium supplemented with 4 mM 4-phenylbutyrate (Tocris Bioscience, Bristol, UK). After additional 48 h, medium was collected and cells were removed by centrifugation at 3006g for 10 min. Bacteria grown as described above and suspended in PBS were added to conditioned medium at a final concentration of 10 4 CFU/ml. Aliquots of 150 mL were transferred to wells of a polypropylene microtiterplate (Costar) and incubated at 37uC for 30 min with shaking. Thereafter, the number of live bacteria in the conditioned medium was determined by viable count and expressed in relation to the concentration in the inoculum.
To relate the antimicrobial effect to LL-37, monoclonal mouse anti-LL-37 antibodies [55] were added to the conditioned medium at a concentration of 1 mg/ml and incubated for 30 min at 37uC prior to inoculation of bacteria. For control purposes, supernatants were equally treated with mouse IgG1k isotype control antibodies (BD Biosciences, San Diego, CA, USA). The antimicrobial activity was determined as described above.
Measurement of IL-8 and MIP-2 levels
Before ELISA analysis, urine was centrifuged at 3506g for 10 min to remove cells and larger particles. ELISA kits for human IL-8 or mouse MIP-2 were obtained from R&D systems (Abingdon, UK). IL-8 or MIP-2 levels were determined according to the manufacturer's instructions. The lower detection limit for IL-8 and MIP-2 was 31.3 pg/ml and 15.6 pg/ml, respectively. The urinary levels of creatinine were analyzed colorimetrically, and the levels of IL-8 were expressed as IL-8/creatinine ratios.
Electron microscopy of E. coli in fresh urine
Urine was collected from patients with E. coli UTI and without having a catheter. A drop of urine was incubated on carbon/ Formvar-coated 400-mesh copper grids (GilderGrids, Lincolnshire, UK) for one minute. Immunostaining was performed as described previously with minor modification [56]. Briefly, grids were blocked with 1% BSA/PBS for 5 min, then incubated with anti-CsgA [28] (1:200 in 0.1% BSA/PBS) for 60 min at 37uC, followed by incubation with anti-rabbit IgG-10-nm gold antibodies (1:15 in 0.1% BSA/PBS; Sigma-Aldrich) for 30 min at 37uC. Grids were stained with 2% tungstophosphoric acid (Merck, Darmstadt, Germany) at pH 6. Analysis were performed using a FEI Tecnai Spirit electron microscope (Eindhoven, The Netherlands) at 80 kV accelerating voltage.
Dot blot analysis
Urine was centrifuged at 3006g for 10 min to remove cells and larger particles. Bacteria from the supernatant were collected by centrifugation at 35006g for 10 min. A 2-mL aliquot of the pellet suspended in a minimal volume of PBS was transferred to a nitrocellulose membrane (Invitrogen) and air dried for 15 minutes. Immunostaining was performed as described previously [56]. Briefly, membranes were blocked with 1% milk/1% BSA/PBS for 2 h at room temperature, incubated with anti-CsgA [28] (1:5000 in 1% milk/1% BSA/PBS) for 1 h, followed by incubation with anti-rabbit horse-radish-peroxidase conjugated antibodies (1:3000 in 1% milk/1% BSA/PBS; Bio-Rad Laboratories, Hercules, CA, USA) for 1 h.
Animal experiments
Mouse experiments were approved by the Northern Stockholm Animal Ethics Committee and experiments were carried out according to FELASA guidelines and in compliance with the Committee's requirements.
Bacterial infection. Female NMRI mice of 8-10 weeks age were caged in groups of 3-5 animals in standard cages. After water deprivation for 4 h, mice were anaesthetized with isoflurane (Forene TM ; Abbott Scandinavia, Solna, Sweden) and infected transurethrally with 50 ml of a bacterial suspension of 10 9 CFU/ ml, using a soft sterile polyethylene catheter (outer diameter 0.61 mm, inner diameter 0.28 mm; Clay Adams, Becton Dickinson, Franklin Lakes, NJ, USA) with lubricant. Sterile PBS was used for control mice. After 1, 16 or 48 h, mice were sacrificed by cervical dislocation and bladders and kidneys were aseptically removed. Bladders taken 1 h p.i. were cut open and washed in icecold PBS three times to remove non-adherent bacteria. The organs were homogenized in 1 ml PBS and serial dilutions of the homogenate were plated on blood agar plates for viable count. For measurement of MIP-2, homogenized samples were centrifuged at 3506g for 10 min and the supernatants were stored at 220uC prior to ELISA analysis. A total of 126 animals including controls were infected. Mice without bacterial growth in any of the organs were regarded as non-infected and therefore excluded from further analysis; only MIP-2 values from infected organs were used for statistical analysis.
Neutrophil depletion. To induce neutropenia, 100 mg of RB6-8C5 monoclonal antibody (R&D systems) was administered intravenously 24 h before infection [20]. Control animals received an equal volume of sterile saline. Neutropenia at the time of infection and at the end of the experiment was confirmed on Giemsa-stained blood smears.
Bacterial sensitivity to LL-37 and mCRAMP
The susceptibility of wild-type and mutant E. coli strains to synthetic LL-37 and mCRAMP (Innovagen AB, Lund, Sweden) was determined using a broth microdilution method. Briefly, bacteria were grown overnight at 37uC on LB plates with or without salt, inhibiting or promoting the formation of biofilm, respectively. Then, bacteria were suspended in PBS and diluted in LB broth without salt in a concentration of 10 5 CFU/ml. The bacterial concentration was verified by viable count after serial dilutions in PBS. Bacteria (90 ml suspension) were grown in 96-well plates in the presence of 10 ml aqueous solution of synthetic LL-37 or mCRAMP in final concentrations ranging from 0.6 mM to 20 mM in 2-fold dilutions. After 20 h, bacterial viability was measured colorimetrically by reduction of Alamar blue (BioSource International, Camarillo, CA, USA) for 1 h at 37uC [57]. The IC 50 was determined as the peptide concentration that gave 50% reduction of the absorbance at 570 nm relative to bacteria grown without peptide.
Effect of LL-37 on formation of biofilm
Plates were filled with 90 mL bacterial culture of planktonic cells and 10 mL of aqueous solution of LL-37 in the same concentrations as in the susceptibility assay. As control peptides, sLL-37 (Innovagen AB) and VIP [27] with similar structural properties as LL-37 were utilized. The plates were incubated without shaking at 37uC for 20 h. After incubation, the amount of biofilm formed was determined as described above.
Purification of recombinant CsgA-His 6
CsgA-His 6 and CsgG, a lipoprotein that is required for CsgA secretion [58], were overexpressed in LSR12 (C600::csg). The purification procedure of recombinant CsgA-His 6 was performed as previously described with some modifications [59]. The filtrate was incubated with nickel-nitrilotriacetic acid (Ni-NTA) agarose (Invitrogen) for 1 h at 4uC with shaking, centrifuged at 2006g for 5 min and transferred to a polyprep chromatography column (Bio-Rad, Hercules, CA, USA). To confirm protein identity of CsgA-His 6 , the isolated protein was prepared and subjected to SDS-PAGE as described below, and thereafter the protein was transferred from the gel to a polyvinylidene fluoride (PVDF) membrane (Invitrogen) at 160 mA for 60 min. The band of 15 kDa, corresponding to the molecular weight of CsgA-His 6 , was excised from the membrane stained with Coomassie Blue and analysed with N-terminal sequence analysis as has been described [60]. Since it has previously been shown that CsgA-His 6 has the same polymerizing properties as wild-type CsgA [59], we refer to CsgA-His 6 as CsgA.
Isolation of wild-type curli
The cellulose-deficient isogenic mutant of E. coli No. 12 was used to purify wild-type curli, as has been described [17]. Colonies were harvested and suspended in 0.05 M Tris-buffer by using an omnimixer. Bacterial cell debris was discarded by centrifugation at 80006g for 15 min and curli protein was precipitated by adding 0.1 M MgCl 2 and 0.15 M NaCl. The aggregated curli were centrifuged at 160006g for 15 min, and the pellet was dissolved in 10 mM TRIS, 1 mM EDTA, pH 7.5 with 2% 3-((3-cholamidopropyl)dimethylammonium)-1-propanesulfonate (CHAPS). After incubation for 45 min at 95uC, the solution was centrifuged at 200006g for 10 min. The pellet containing curli was washed three times with water. Finally, curli were suspended in PBS and used for binding studies with LL-37.
Binding of LL-37 to CsgA
Precipitation of LL-37 with curli. LL-37 (0.1 mM in PBS) was mixed with 5 mM wild-type curli, or 5 mM recombinant polymerized CsgA. As control, the same concentration of LL-37 without curli or CsgA was utilized. The samples were incubated for 1 h at 37uC with shaking and centrifuged for 30 min at 100006g. An aliquot of the supernatants was analyzed for the presence of LL-37 with Western blot.
Surface plasmon resonance. Binding analysis of LL-37 to both monomeric and polymeric CsgA was performed with surface plasmon resonance (Biacore 3000 instrument; Biacore AB, Uppsala, Sweden). Both monomeric and polymeric CsgA was immobilized on a CM5 sensor chip (Biacore AB) surface by amine coupling. The chip surface was normalized in 70% glycerol and lanes on the chip were activated with injection of 35 mL 0.05 M Nhydroxysuccinimide (NHS)/0.2 M N-ethyl-N9- [3-dimethylamino) propyl]carbodiimide (EDC). Polymeric or monomeric CsgA (1 mM in 10 mM sodium acetate, pH 4.5) was then immobilized on the chip to 7000 response units and 2000 response units, respectively. After immobilization, the lanes were subjected to 60 mL ethanolamine (1 M, pH 8.5) to deactivate remaining activated carboxylic groups. One lane, activated and deactivated, was used as negative control (without addition of CsgA). Standard Biacore HBS-EP (Biacore AB) was utilized as running buffer, and 0.1 mM LL-37, sLL-37 or VIP were injected at 20 ml/min for 3 min. The surface was regenerated after each cycle with 100 mM HCl.
Western blot analysis
Sample preparation and SDS-PAGE was carried out as described above and Western blot was performed as previously described [20]. The antibodies used were monoclonal mouse anti-LL-37 (0.6 mg/ml in 5% fat-free milk/PBS) [55] and horse radish peroxidase-conjugated anti-mouse IgG (diluted 1:5000).
CD spectroscopy
Samples containing 40 mM CsgA in 50 mM potassium phosphate buffer (KPi) and 0.02% NaN 3 , pH 7.2, with and without 10 mM LL-37 were incubated for 60 h at 37uC. The samples were then assayed with a Jasco J-810 spectropolarimeter from 190 to 250 nm in a quartz cell with a 1-mm path length at 20uC. The spectrum of buffer alone and LL-37 in buffer was subtracted from the spectra for CsgA alone and for CsgA together with LL-37, respectively.
Thioflavin T assays
Tecan plate reader. Purified recombinant monomeric CsgA (10 mM) with or without different concentrations of the peptides LL-37, sLL-37 or VIP was mixed with 20 mM of the fiber-specific fluorescent probe ThT (Sigma-Aldrich). Fluorescence was measured with a Tecan infinite M200 reader (Tecan Nordic AB, Tä by, Sweden). The excitation and emission wavelength was 430 and 490 nm, respectively. Measurements were conducted at 37uC every 10 min after shaking the sample for five seconds. Background fluorescence of the peptides themselves was subtracted.
Confocal microscopy. Recombinant monomeric CsgA was mixed with equimolar LL-37, sLL-37 or VIP and 20 mM ThT, and incubated over night at 37uC, washed twice, and suspended in 10 mM Kpi, pH 7.2 before mounting in ProLong Gold antifade mounting medium (Invitrogen). Images were acquired on a Leica TCS SP5 confocal microscope using a 206 objective.
Fluorescence intensity was quantified using ImageJ software [61]. The difference in fluorescence intensity of CsgA with and without peptides was calculated and corrected for the contribution of fluorescence from the peptides themselves.
Data analysis
Data were compared with student's t-test, Mann-Whitney U test or Fisher's exact test as appropriate. P values of less than 0.05 were considered as statistically significant.
|
2014-10-01T00:00:00.000Z
|
2010-07-01T00:00:00.000
|
{
"year": 2010,
"sha1": "c8db46e9f2766fbc500474b49892f04338a68066",
"oa_license": "CCBY",
"oa_url": "https://journals.plos.org/plospathogens/article/file?id=10.1371/journal.ppat.1001010&type=printable",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "c8db46e9f2766fbc500474b49892f04338a68066",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
53678197
|
pes2o/s2orc
|
v3-fos-license
|
Case of Sepsis and Probable Septic Arthritis with Plesiomonas Shigelloides in a Patient with Sickle Cell Anemia
Sickle cell disease (SCD) patients are at increased risk for Gram negative and unusual bacterial infections. Plesiomonas shigelloides is a Gram negative, facultative anaerobic rod, member of the family Enterobacteriaceae that has been isolated from environmental sources and a wide range of animals including mammals, birds, fish, water-dwelling reptiles and amphibians [1]. This report is the first to document a case of probable Plesiomonas septic arthritis in a pediatric SCD patient.
ISSN 2471-4925 clavulanic acid, ceftriaxone, cefepime, levofloxacin, meropenem, piperacillin tazobactum and trimethoprim sulfamethoxazole.All blood cultures obtained after the initiation of antibiotics remained sterile.Infectious Diseases was consulted due to the unusual organism isolated in the blood and recurrence of fever after 48 hours (Figure 1).
On repeat examination on hospital day 2, she was found to have pain, swelling, warmth and limited range of motion of the right shoulder.Plain radiograph showed mild thickening and irregularity of the medial cortex of the right mid and distal humerus.Magnetic resonance imaging with contrast of the right shoulder showed moderate sized effusion with synovial enhancement and osteonecrosis of the right humeral diaphysis (Figure 2).Hence on day 3 of hospitalization, she underwent a right shoulder arthrotomy with exploration and drainage, and a drain was left in place which gave moderate sero-sanguinous discharge.Fluid from her shoulder joint was obtained, however no cell count was performed and cultures were sterile.The fever resolved after the 8th day of intravenous ceftazidime.
Upon clinical improvement and removal of the right shoulder joint drain, she was discharged after 11 days of hospitalization with instructions to continue treatment with amoxicillin-clavulanic acid to complete a total of 6 weeks of antimicrobial therapy for probable septic arthritis of the shoulder and possible osteomyelitis of the proximal humerus.At follow up after 4 weeks of therapy, she had remained afebrile and had improved range of motion of the right shoulder.A plain radiograph of her right shoulder was normal and her inflammatory markers had also normalized (CRP: 0.6 mg/dL; ESR: 17 mm/hr).
Discussion
P. shigelloides belongs to the family Enterobacteriaceae [1].These microbes are Gram-negative anaerobic rods that have been associated with diarrhea and dysentery [2].The primary natural reservoirs for P. shigelloides are water and soil surfaces as well as fish and other marine animals, especially oysters.The organism is recovered from freshwater and estuaries in temperate and tropical regions and occasionally from seawater during the summer months [3].Open Access 2 Extra-intestinal P. shigellosis disease is less common than intestinal disease and it occurs in both immunocompromised and immunocompetent hosts [4,5].Reported clinical manifestations include acute respiratory distress syndrome, disseminated intravascular coagulation, splenic abscess [6], ophthalmitis [7], neonatal meningitis [8,9], cellulitis [10,11] and orchi-epididymitis [12].Though uncommon, Plesiomonas bacteremia has been associated with biliary disease, especially in older patients [13].Invasive P. shigelloides infection has been associated with a mortality rate of 50% to 60% [14].
It is important to take notice of this organism in causing invasive disease in pediatric patients with sickle cell disease as it may lead to severe consequences.In addition to our case, three other cases of P. shigelloides sepsis have been reported in SCD patients (Table 1).Among them, two were associated with end organ disease (pneumonia and splenic abscess).There were no deaths reported.Similarly, our case illustrates the potential for Plesiomonas to cause serious infection in the context of bacteremia and sepsis in SCD patients and stresses the importance of having a high level of suspicion, and performing careful and repeated physical examinations to evaluate potential foci of infection such as a joint in patients with continuous or recurrent fever while on appropriate antimicrobial therapy.Development of joint pain in settings of high WBC count, MRI finding of moderate effusion, need for arthrotomy and drain placement are consistent with septic arthritis though synovial fluid culture remained negative.As osteonecrosis is a normal disease progression of underlying SCD, a joint infection could be missed.Moreover, the impaired reticuloendothelial clearance associated with the disease has been shown to predispose patients with sickle cell anemia to bacterial infections [15].P. shigelloides is inherently resistant to ampicillin, but amoxicillin or piperacillin in combination with beta-lactamase inhibitors (clavulanate and tazobactam) can be used as antiinfective agents.Isolates generally are susceptible to cephalosporins, quinolones and carbapenems [9,[16][17][18].Aminoglycosides are generally ineffective.Duration of therapy depends on the type of infection.No definitive source of infection was identified in our patient.
Early diagnosis with blood cultures and timely use of appropriate antibiotics were key to the management of this infection and daily evaluation to identify additional foci of infection resulted in a prompt diagnosis of probable septic arthritis, which required surgical drainage in addition to prolonged antibiotic therapy to resolve.
Conclusion
Ours is the first case of P. shigelloides bacteremia and sepsis associated with probable joint infection in a pediatric SCD patient.An increased awareness among clinicians regarding this Gram negative organism is necessary as Plesiomonas has been associated with sepsis and bacteremia, and has the potential to involve other organs like joints, spleen and lungs in vulnerable populations such as patients with SCD.
Day 2 :Figure 1 :
Figure 1: Clinical course of a sickle cell patient with Plesiomonas sepsis and probable septic arthritis, managed with intravenous (IV) ceftazidime and surgical drainage, showing clinical recovery Abbreviation: Cx: culture; R: right
|
2018-11-16T04:48:07.228Z
|
2017-01-01T00:00:00.000
|
{
"year": 2017,
"sha1": "a1525244cc560c7ce2f78c2b4310892633372d3e",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.16966/2471-4925.135",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "a1525244cc560c7ce2f78c2b4310892633372d3e",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
226271232
|
pes2o/s2orc
|
v3-fos-license
|
Comparative analysis of self-care assessment scales in patients with chronic heart failure, advantages and disadvantages
Aim To compare Russian versions of the scales for assessment of self-care ability in patients with chronic heart failure (CHF), European Heart Failure Self-Care Behavior Scale (EHFScBS_9) and The Self-Сare of Heart Failure Index (SCHFI, version 6.2). Material and methods Assessment of the self-care ability was performed with Russian versions of EHFScBS_9 and SCHFI (version 6.2) scales in 130 patients with NYHA functional class II-IV CHF primarily of ischemic origin (78.5 %). Mean age of patients was 63.2±9.6 years; most of the patients were men (n=92; 70.8 %). Patients were managed in accordance with effective guidelines ESC / HFA 2016 and Russian guidelines 2018. Results Along with an increase in SСHFI scores, a decrease in EHFScBS_9 scores was observed (r= -0.31, p<0.001). The patients participating in the study showed a low self-care ability at baseline according to results of both scales. Conclusion The presence of certain differences between the study scales does not exclude a possibility of using them alone or together for more detailed assessment of the self-care ability.
Introduction
Following current European and Russian guidelines for the diagnosis and treatment of acute and chronic heart failure (CHF), training patients in self-control and self-care is an important factor in determining the successful management of this category of patients [1,2]. The gold standard of self-control of patients in CHF can be defined as "daily activities to support the patient's clinical stability" [3]. Training patients in effective self-control and self-care will contribute to the earlier diagnosis of symptoms of decompensated heart failure, timely medical care encounters, and better commitment to treatment. Moreover, it was shown that the timely training of patients with CHF is associated with better prognosis and quality of life [4][5][6].
Thus, there was a need to develop effective tools to assess the self-control and self-care ability of patients with CHF. Two scores are mostly used for this purpose: the European Heart Failure Self-care behavior Scale 9 item version (EHFScBS_9) [7] and the 22 item Self-Care of Heart Failure Index (SCHFI) [8]. According to Jaarsma et al. [7], such scores can be used in scientific research and in real-world clinical practice. They enable evaluation of patients' self-care ability, and the making of joint decisions with the patient regarding his/her self-care, including during long-term followup. The EHFScBS_9 and SCHFI scores have been recently translated into Russian and approved for use in the Russian Federation [9,10]. However, there has been no comparative evaluation of the advantages and disadvantages of the Russian-language versions of selfcare ability scores, which are the subject of this study.
Our objective was to compare the Russian-language versions of EHFScBS_9 and SCHFI used to assess the self-care ability of patients with CHF and to identify their advantages and disadvantages.
Material and Methods
The study included 130 patients with CHF of predominantly ischemic etiology (n=102; 78,5%) who received outpatient cardiology care. The inclusion criteria were that the patients were aged between 18 and 80 years of age, CHF, and a signed informed consent. Exclusion criteria were an age of less than 18s or more than 80, a history of myocardial infarction or unstable angina, cardiac surgery, percutaneous coronary ORIGINAL ARTICLES § inter ven tion within 30 days before inclusion in the study, inability to read and understand Russian, and disorientation. The study was carried out following the Good Clinical Practice and the Declaration of Helsinki. The Regional Research Ethics Committee approved the study protocol. All patients were informed and signed a consent to participate in the study.
The self-care ability of patients with CHF was assessed using the EHFScBS_9 [9] and SCHFI (version 6.2) [10] scores. Each patient received individual oral instructions on how to complete the questionnaires. Patients answered the questionnaires in the presence of a cardiologist consistently, first EHFScBS_9, then SCHFI, within a single day. If a patient experienced difficulties in answering the questions, additional explanation was provided. Timekeeping was used to register the time required by the patient to complete the questionnaires and for the physician to calculate the results. Difficulties in the process of completing the questionnaires and analyzing the results were assessed on the basis of the patient's and physician's opinions, respectively, expressed in a free form.
EHFScBS_9 consists of nine items describing the ability for self-care. The score is based on the fivepoint Likert scale, where the minimum (1 point) corresponds to the answer "strongly agree" and the maximum (5 points) is "strongly disagree". The total score varies from 9 to 45, where the lower the score, the better self-care ability [7,9].
The SCHFI score consists of 22 items reflecting compliance with the recommendations, monitoring and recognition of HF symptoms, and patient's confidence in self-care. This score includes 3 sections: A -self-care maintenance (10 questions), B -self-care management (6 questions), C -self-care confidence (6 questions). Questions are scored from 1 to 4, except for questions 11 and 16 where 0 can be selected [8,10]. The total score can be from 20 to 89. The higher the score, the better the ability to self-care. The SCHFI score uses formulas for conversion to standardized measures. The maximum possible sum of standardized scores is 305 [8,10].
The results were statistically processed using the STATISTICA 16.0 (SPSS 16) and Microsoft Office Excel software. The total score, results for individual items and questions, mean values (M), standard deviation (SD), and correlation criteria were estimated. The Kolmogorov-Smirnov and Shapiro-Wilk tests were used to confirm the normality of the distribution of the quantitative variables. The differences were statistically significant at p<0.05.
The mean score and standard deviation for each item of EHFScBS_9 are shown in Table 1.
The mean self-care score was 21.3±8.1. The minimum EHFScBS_9 score was 9, and the maximum was 43. The best results (lowest score) were observed in items 8, 9, 5, 7, and 2.
The mean SSCHFI scores and standard deviations for each question are shown in Table 2.
The total SCHFI score was 50.8±8.8, and the standardized score was 136.8±40.3. The minimum standardized EHFScBS_9 score was 32, and the maximum was 274. The best result was obtained in Section B (Manage ment of self-care) with 46.5±20.1. Section A (Maintenance of self-care) was 45.6±14.9, and Section C (Confidence in self-care) was 44.5±18.8.
Detailed examination of items in both scores found several analogies: item 8 of EHFScBS_9 ("I take my medication as prescribed") was consistent with question 5 of SCHFI score ("Keep doctor or nurse appointments?"), item 9 of EHFScBS_9 ("I exercise regularly") with question 4 of SCHFI ("Do some physical activity?"), item 7 of EHFScBS_9 ("I eat a low-salt diet") corresponded to question 6 of SCHFI ("Eat a low salt diet?"), items 2 ("If shortness of breath increases I contact my doctor or nurse") and 3 ("If legs/ feet are more swollen, I contact my doctor or nurse") of EHFScBS_9 are consistent with question 15 of SSHFI ("If you have trouble breathing or ankle swelling, how likely are you to try one of these remedies? Call your doctor or nurse for guidance").
Correlation analysis revealed that the relationship between the total EHFScBS_9 and SSHFI scores was naturally negative (r= -0.31, p<0.001).
23 (17.7%) patients with CHF had no difficulty with completing the EHFScBS_9 and SSHFI questionnaires. According to the patients, the main difficulties in using the EHFScBS_9 score were associated with the five-point Likert scale. As for the SCHFI score, the main difficulties were associated with a large number of questions to be answered. The mean duration of completing the EHFScBS_9 and SSCHFI questionnaires was 2.6±1.1 and 5.2±1.5 minutes, respectively. The mean duration of computing the results by the physician was 0.2±0.04 minutes and 3.2±0.5 minutes, respectively.
Discussion
Given the increasing role of educating patients with cardiovascular diseases, including CHF, and creating motivation for self-control and self-care, a need was defined to develop tools that would allow for effective control of self-care skills in real-world clinical practice. The best-known tools are the European Heart Failure Self-care Behaviour score 9 item version (EHFScBS_9) [7] and the 22 item Self-Care of Heart Failure Index (SCHFI) [8]. However, no comparative assessment of the advantages and disadvantages of these scores has yet been performed. Both self-care ability assessment scores have been approved in the Russian Federation [9,10], which is why a comparison is required. Only 10 (7.7%) patients score 9, which is the EHFScBS_9 score. The results for the SHSFI score were similar. Only 3 of 130 (2.3%) patients had a total (non-standardized) score of more than 70; 13 (10%) patients scored more than 60. Interestingly, despite the small but significant differences between the results of all three sections of the SCHFI score (p<0.0001), the highest score was obtained in Section B (Self-care Management). This advantage was likely due to patients' initial basic knowledge of CHF symptoms and additional measures to improve their condition. Thus, comorbid patients with CHF included in our study demonstrated low selfcare ability. A similar situation occurred during the analysis of self-care ability in patients with CHF in other countries [4,5].
The best self-care scores (minimum EHFScBS_9 and maximum SSCHFI) were reported in the corresponding items of both scores on the control of swelling syndrome, salt intake, the administration of additional diuretics, compliance with guidelines, and the use of various devices to help remember to take medicines.
In our opinion, the SSCHFI score (22 questions) enables evaluation of the self-care ability of patients with CHF in more details than the EHFScBS_9 score (9 questions), by dividing and focusing attention on three main self-care sections: maintenance, management, and confidence. Much attention is paid not only to the patient's ability to seek advice from a doctor or nurse, but also to take certain measures to improve his/her state beyond daily self-monitoring.
At first glance, the EHFScBS_9 questions are almost the same as the SCHFI questions in Sections A and B. However, a detailed comparison of the scores revealed only four issues of similar meaning. Section C in the SCHFI score is a significant addition, allowing for the evaluation of self-care confidence in patients with CHF. Moreover, unique formulae are used to compute the SCHFI scores and calculate standardized scores for each questionnaire section. We believe this to be essential. For example, in 10-12 patients who most often had the same non-standardized score of 59, the standardized score varied from 162 to 179.
When patients were required to complete the questionnaires and doctors required to calculate the results, the SCHFI score took more time than the EHFScBS_9 score. This is entirely expected since the SCHFI score contains more questions, while unique formulae are needed to calculate the total score. On the other hand, patients with CHF were more likely to face difficulties completing the EHFScBS_9 questionnaire, requiring additional explanations from the physician.
Conclusion
Thus, a comparative analysis of the European Heart Failure Self-Care Behaviour Scale and the Self-care of Heart failure Index (version 6.2) showed that both scores are simple, and effective peer-to-peer tools for assessing self-care in patients with CHF. Although there are some differences between these scores, they may well be used both independently and in combination, complementing each other in real-world clinical practice, as well as in clinical trials.
No conflict of interest is reported.
The article was received on 04/02/2020
|
2020-11-07T14:06:53.893Z
|
2020-09-07T00:00:00.000
|
{
"year": 2020,
"sha1": "1c4edc810cb309a54bc756e93fe7d5ee211eb917",
"oa_license": "CCBY",
"oa_url": "https://lib.ossn.ru/jour/article/download/1114/781",
"oa_status": "HYBRID",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "7c527f9e4ec6f736604b8baf7ef6f823fd0f3790",
"s2fieldsofstudy": [
"Medicine",
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
219109757
|
pes2o/s2orc
|
v3-fos-license
|
It is Never too Late to Learn: The Role of Physical Exercises in the Learning Process among the Older Adult Population
Physical exercises are known as a great method for improving overall health. Recent research suggests that physical exercises are beneficial to improve memory and other cognitive processes [8], [43]. Learning is a critical aspect in the lives of individuals because it does promote not only knowledge but also fosters quality of life. As the number of older adults will continue to rise through the next years, these data are suitable to support future interventions to increase active life expectancy within financial constraints among governments, especially those that provide public health care.
Introduction
The 2019 Revision of World Population prospects shows that worldwide by 2050, one in six individuals will be people in the age range of 65 and older. This proportion is even greater in European and North American countries where projections indicate that one in four people living in these areas will be older adults. A similar trend is observed in South America, Asia and Africa. Although there is great variation related to older age classification across nations with developed countries classifying older adults as those who are 65 years old and over, and in developing countries defining older people as those who are either 55 or 60 years old and over, older age has become a critical topic in health policies. Based on the increase in the population, there are limited options to help older adults stay connected and active in society [12]. This fact is critical because as the number of this population increases, the need to offer programs that can help older adults to maintain a healthy body and mind also increases. With this concern in mind, this research focused on searching articles from five electronic databases, including Angeline, Psych Info, ERIC (EBSCO), Sage Research Methods Online, and Cochrane Central. The keywords used were ageing, older adults, physical exercises, and learning. After analyzing titles and abstracts of over 3,000 articles, 59 were selected. While many articles strongly emphasize the importance of physical exercises to foster the learning process among older adults, others also suggest that the combination of physical exercises and learning can decrease social isolation, depression and improve quality of life [34], [41].
Some argue that services that can potentially be beneficial to older adults are related to learning activities [6], [20], [57]. Congruently to this fact, studies indicate that older adults are interested in learning programs, and this type of activity is often mentioned as one that promotes successful ageing. Learning is a process that happens when people engage in studies or is taught content that results in new or modified knowledge, skills, behaviours, attitudes, values and preferences. Also, learning occurs as a result of lived experiences [2], [44]. Lower education and illiteracy are associated with a higher risk of disability and poorer health outcomes among older adults. Hence, opportunities to engage in life-long learning activities facilitate the development of self-confidence, decision-making, sense of control, and skills older adults need in order to adapt and stay independent as the ageing process develops over time [7], [58]. In turn, more adapted and healthy older adults foster the disability-free life expectancy that can lower healthcare expenses caused by disability issues that may affect older adults [57]. Besides maintaining mental and physical health, learning activities are shown to be among older adults' interests, for it can help them to stay connected to society preventing critical issues related to isolation [30]. Also, learning can support many aspects of individuals' life; for instance, maintaining cognitive functioning, health management, and social relationships.
Moreover, learning might increase the socioeconomic status and well-being of older adults [9]. Although some cognitive processes might change as one age; for instance, it becomes harder to block irrelevant information during the learning process, and reaction-time slows down, given the opportunity, encouragement, and enough time older adults can successfully learn new skills, concepts and knowledge [39], [8]. Also, studies examining ageing and cognitive processes show evidence that individual differences within and between age groups in learning ability are often associated with personality traits and life habits [29], [35], [59].
It is critical to note that previous studies on agerelated changes suggested brain volume and weight declines during the ageing process. This view was strongly based on the assumption that the brain would lose neurons resulting in reduced learning capacity among older adults. However, finds indicate that the mean number of neurons of the cerebral cortex of men and women between the ages of 20 and 110 does not change [26]. A study suggests that what happens as one age is that the nerve cell body and synaptic density in some regions of the brain might suffer some decrease. However, this fact might not be directed related to ageing, but to the quality of neural development during the young years of life [3]. One needs to be reminded that the ageing "is a fundamental biological process that can be defined, measured, described and manipulated"; therefore, as one age, many changes might not be related simply to the ageing process, but the lifestyle, and some influences of the environment [3]. The Nun Study noted that more mentally, and physically active nuns had fewer issues with declines in brain processes, and even those who had some physical alterations in the brain, known as dementia-related characteristics; for instance, the amyloid plaques, had not shown cognitive changes due to the ageing process [45].
Changes in the cognitive function associated with the ageing process might affect reaction time and attentional capacity. A study comparing reaction time between young and older adults suggest that young adults process information more rapidly than older adults; however, this variation only implies that older adults need more time to execute a learning activity that does not necessarily interfere with older adults learning ability and capacity. Also, older individuals use different strategies than younger adults for problem-solving [56]. "It might take longer to remember some things or to solve complex problems, but the power to think remains the same" [36]. The attentional capacity -the ability "to maintain a goal-oriented behaviour when there are multiple other competing distractions" -undergo modifications as people get older. Generally, it has been observed that older adults have more difficulty ignoring other perceivable information. Hence, when older adults engage in learning a new task or information, besides allowing more time, it is beneficial to provide an environment that has fewer distractions [38], [56]. Also, learning does not happen in a vacuum; contrary, it occurs within a social context that encompasses cognitive and emotional processes. Learning involves a complex interplay between biological, emotional and social factors [28]. Finally, most declines in cognitive functions are triggered by lack of practice, illnesses, behavioural and social factors, so learning activities can support healthy cognitive function preventing cognitive declines as people age [7].
Physical exercises and the brain
Physical exercises are planned, designed, and repetitive set of body movements that will cause energy expenditure and has a goal to improve health or skill-related activity [11]. It is critical to distinguish physical exercise from physical activities as the latter is defined as any movement performed by the skeletal muscles of one's body in which results in energy expenditure; for instance, walking to the supermarket or doing the dishes [11]. Physical exercises are known to be a great method of improving physical and mental health. Many studies show that individuals who engage in physical exercises have less risk of developing certain types of illnesses such as diabetes and metabolic syndrome. Also, physical exercises have been used as an additional form of therapy to decrease symptoms of depression [13], [42]. Besides the fact that physical exercises can ameliorate several functional capacities in the human body, it is also considered beneficial to improve memory and other cognitive processes [8], [25], [43].
Research suggests that lifestyle factors like the practice of physical exercises might be closely related to the improvement of learning and memory on all individuals, especially older adults [37]. A meta-analyze study conducted to verify the effects of physical exercises in the cognitive processes found that sustained aerobic physical exercises can enhance the executive process of the brain [14]. Executive processes refer to operations responsible for coordinating mental activity in order to achieve a particular goal. It includes working memory, reasoning, problem-solving, planning and execution [56]. Different routines of exercises, intensity and duration consistently show improvements in acquisition and retention of new information in older adults [15], [51].
Moreover, exercises may protect the brain against atrophy in certain areas important for the learning process [14], [51], [55]. A study shows that moderate physical exercises like walking are positively associated with larger hippocampal volume in older adults. Good memory ability, lower risk of memory decline and dementia have been liked to larger hippocampus size [52].
Influences of exercises in the hippocampus
Although a full understanding of the underlying factors between physical exercises and the brain is still under investigation, some studies suggest that higher cardiorespiratory fitness, neurogenesis and angiogenesis are involved with the benefits associated with learning. Cardiorespiratory fitness refers to the circulatory and respiratory systems' ability to supply enough oxygen to the skeletal muscle system during sustained physical exercises. Besides stimulating these systems to work more efficiently, physical exercises increase the amount of oxygen that is inhaled, enabling more blood that, in this case, is rich in oxygen to distribute blood to all body systems, including the brain. Older adults who show higher cardiorespiratory fitness also show increased grey matter volume which results in better cognitive performance [54]. The literature suggests that high cardiorespiratory fitness influences neurogenesis and angiogenesis, factors related to cognitive performance and learning improvement. Neurogenesis refers to the process by which new neuron cells are produced in the brain, and angiogenesis is the new blood vessel formation from pre-existing vessels. For many years, researchers believed that neurogenesis was not possible to occur since neurons would only be generated before birth and would never change after birth [3]. Although most of the neurons in the human brain are indeed generated before birth, new studies show that neurogenesis occurs across the lifespan [17].
Other studies also indicate the positive relationship between physical exercise and neurogenesis. Findings show that exercises may be one of the conditions that induce neurogenesis. The study reveals that the practice of aerobic exercises is associated with increased hippocampal volume in older women with mild cognitive impairment [10]. This finding is consistent with another study by Erickson and colleagues [19] showing that healthy older adults who went to the intervention program of aerobic physical exercises for 12 months had a significant increase in the hippocampal volume as a result of neurogenesis.
Chemical factors associated with physical exercises and brain function
At least two growth factor proteins BDNF (brainderived neurotrophic factor) and VEGF (vascular endothelial-derived growth factor), and one hormone IGF-1 (insulin-like growth factor) are believed to contribute to neurogenesis enhancing the learning process. BDNF promotes the growth, development, and maintenance of neurons. Also, BDNF plays an active role in brain processes by regulating synaptic plasticity that is crucial for learning and memory [4], [23], [33], [47]. VEGF is produced by cells that stimulate new blood cell formation, angiogenesis [27].
The IGF-1 is an essential anabolic hormone that the body uses for the regulation of many physiological functions; for instance, the skeletal muscle function, neuronal activity and cognitive function [3], [31]. Furthermore, BDNF and IGF-1 "are crucial mechanisms underlying improved learning response to exercise" [15]. IGF-1 and VEGF are considered exercise-induced and seem to coordinate and stimulate neurogenesis and angiogenesis, respectively [21], [48].
In addition, some studies indicate that IGF-1 might be related to important neurodegenerative diseases such as Alzheimer's disease (AD). AD is a serious illness that impairs cognition, memory, and many other brain functions (Alzheimer's Association, n.d.). IGF-1 pathway disruption is associated with AD, which shows the importance of IGF-1 and its relationship with cognitive processes [50]. Moreover, IGF-1 disturbances are connected to insulin problems. Gasparini and Xu [22] suggest that patients with Alzheimer's disease have a defective response to insulin-related to alterations in the IGF-1 and insulin level. This alteration might cause problems in the clearance of an important substance called beta-amyloid. Beta-amyloid is considered one of the key factors in the development of Alzheimer's disease, and its regulation appears to be strongly dependent on the levels of insulin and IGF-1 [3], [46]. High levels of insulin seem to disrupt the betaamyloid breakdown, while low levels of IGF-1 disturb the clearance of beta-amyloid cells [24], [40], [32]. Hence, IGF-1 plays a critical role in certain brain mechanisms that maintain a healthy brain. Due to the strong relationship between physical exercises and IGF-1 and insulin regulation might be one of the key factors that induce the production of IGF-1. It is essential to note that the relationship between physical exercises and the production of IGF-1 and VEGF has been established on experiments in which IGF-1 and VEGF were blocked. This process prevented the exercise-induced effects in the brain. Interestingly, studies show that older individuals who adhere to the practice of physical exercises have a lower risk of developing Alzheimer's disease and other cognitive problems [5]. The fact that these biological chemicals are critical to the neurological function demonstrates their role in learning. Furthermore, the evidence showing that these substances might be exercise-induced demonstrate the important role of keeping a lifestyle that involves the practice of physical exercises. In addition, these findings show that older adults maintain their learning capacity, which can be even improved due to exercises.
Limitations
The articles reviewed for this study do not show specific information about the ideal duration of exercise per session and for how long one needs to be engaged in constant physical activity to show improvements in the learning process. Also, there are other factors related to learning like the level of education and socioeconomic status among older adults, which should be examined in order to have more information about the influence of these factors among older adults who engage in physical exercises to improve the learning process.
Conclusion
The increasing number of older adults in the population, circumstances related to cognitive decline, and older adults' desire to maintain social roles raise the attention to find ways to increase and promote learning. Although changes in cognitive functioning may accompany the ageing process, they are greatly associated with lifestyle and environment that lack elements to promote brain function. Older adults maintain their learning capacity, and methods like physical exercises might be beneficial to cognitive processes that can foster not only a healthier body but also a healthier mind. Furthermore, older adults who practice physical exercises might be at lower risk for certain serious diseases such as Alzheimer's disease. Promoting a healthier ageing process through physical exercises might increase the disability-free life expectancy, which in turn can lower the costs to both individuals and the healthcare system.
|
2020-05-07T09:11:52.199Z
|
2020-03-31T00:00:00.000
|
{
"year": 2020,
"sha1": "2a0531740ad49f933afacbd14796efe81702e7ca",
"oa_license": null,
"oa_url": "https://doi.org/10.20533/licej.2040.2589.2020.0445",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "9f5ca3ac745a8d07b319cdff0492777bc4fb73cc",
"s2fieldsofstudy": [
"Education",
"Medicine"
],
"extfieldsofstudy": [
"Psychology"
]
}
|
118464834
|
pes2o/s2orc
|
v3-fos-license
|
Infrared Behavior of 3-Point Functions in Landau Gauge Yang-Mills Theory
The three-gluon and ghost-gluon vertices of Landau gauge Yang-Mills theory are investigated in the low momentum regime. Due to ghost dominance in the infrared we can use the known power law behavior for the propagators to determine analytically the complete momentum dependence of the dressing functions. Besides a uniform, i. e. all momenta going to zero, divergence, we find additional singularities, if one momentum alone goes to zero, while the other two remain constant. At these asymmetric points we can extract additional infrared exponents, which corroborate previous results and expand the known fixed point solution of Landau gauge Yang-Mills theory, where the uniform infrared exponents for all vertex functions are known. Calculations in two and three dimensions yield qualitatively similar results.
Infrared Behavior
Many non-perturbative aspects of Yang-Mills theory are encoded in the infrared (IR) behavior of its Green functions. For very small momenta below the intrinsic scale Λ QCD they can be described by power laws. This scaling solution [1] is in accordance with the scenarios for confinement of Gribov-Zwanziger and Kugo-Ojima. The general expression for the so-called IR exponent of an arbitrary Green function with m gluon and 2n ghost legs under the assumption that all momenta go to zero uniformly is [2] This formula is valid in d = 2, 3, 4 dimensions for the corresponding values of κ. As it is clear that we can have more than one independent momentum for vertex functions, the question arises what happens when only one of these goes to zero? Interestingly it turns out that additional divergences can occur that do not change the uniform solution [3].
For the calculation of the three-point vertices we used the known power laws for the propagators and the following truncations for the Dyson-Schwinger equations (DSEs), which is motivated by ghost dominance in the IR [1,4] For the dressed ghost-gluon vertex we used the bare one, justified by a simple argument of Taylor [5], which is supported by lattice simulations [6] and the DSE solution [7]. The truncation considers only the first order of the skeleton expansion. For more details see [8].
Three-point integral
The ghost-triangle diagram of the three-gluon vertex is decomposed in the IR into ten tensors τ i µνρ (p 1 , p 2 , p 3 ) and ten corresponding dressing functions E i (p 1 , p 2 , p 3 ). The applied tensor decomposition [9] reveals that only ten instead of the expected fourteen scalar functions E i and tensors τ i µνρ are necessary. The former consist of massless three-point integrals, where ν 1 , ν 2 and ν 3 are non-integer numbers. We employed the Negative Dimensions Integration Method (NDIM) which yields a full analytic solution in terms of Appell's series F 4 . Using several different analytic continuations we can calculate the momentum dependence of the diagram for arbitrary d. As the variables of the Appell's series are the momentum ratios p 2 1 /p 2 3 and p 2 2 /p 2 3 we plot the scalars E i from the tensor decomposition as functions of p 2 1 and p 2 2 with p 2 3 fixed. The plotted region in fig. 1 represents the accessible ratios for Euclidean momenta and is restricted by momentum conservation. The ghost-gluon vertex can be decomposed similarly into two parts.
Three-point vertex functions
As an example for the kinematic dependence of the dressing functions we show E 5 and E 10 in fig. 1. The behavior when one of the three momenta gets small compared to the others, i.e. the region around the asymmetric points (0, 1), (1, 0) and (∞, ∞), is of special interest. This case corresponds to the emission/absorption of a soft gluon. We extracted the exponents in d dimensions and indeed found kinematic singularities, which are in agreement with the power counting analysis in ref. [3]. In fig. 1 the results in four dimensions are given.
To verify the self-consistency of the assumption of a bare ghost-gluon vertex we calculated its momentum dependence for uniform scaling and the case of only one small momentum. The results for the former clearly support the bare version, whereas those for the latter show that the structure of the ghost-gluon vertex is richer than expected and features a divergence (1 − 2κ) for the longitudinal scalar function when the gluon momentum vanishes. However, this is no contradiction to Taylor's argument because this scalar function is multiplied with the soft momentum and the divergence is suppressed.
|
2019-04-22T13:04:25.628Z
|
2008-01-01T00:00:00.000
|
{
"year": 2008,
"sha1": "7a60a80a973cfb6204270bb56125d0887334111a",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Arxiv",
"pdf_hash": "3edabb25a34decf79131e16b5015c21525f6ba1a",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
262089558
|
pes2o/s2orc
|
v3-fos-license
|
Sintilimab treatment for chronic active Epstein–Barr virus infection and Epstein–Barr virus-associated hemophagocytic lymphohistiocytosis in children
Background Chronic active Epstein–Barr virus infection (CAEBV) and Epstein–Barr virus-associated hemophagocytic lymphohistiocytosis (EBV-HLH) are rare but life-threatening progressive diseases triggered by EBV infection. Glucocorticoid/immunosuppressants treatment is temporarily effective; however, most patients relapse and/or progress. Hematopoietic stem cell transplantation (HSCT) is a potentially curative therapy; however, there are risks of transplantation-associated complications. Currently there is no standard treatment for CAEBV and EBV-HLH. Programmed death protein 1 (PD-1) inhibitors have achieved a high response in many EBV-related diseases. Sintilimab (a recombinant human IgG4 monoclonal antibody against PD-1) disrupts the interaction between PD-1 and its ligand, leading to T cell reinvigoration. Methods A retrospective analysis was performed on three children with CAEBV or EBV-HLH in the Children’s Hospital of Soochow University between 12 December 2020 and 28 November 2022. The efficacy of sintilimab was evaluated. Results Three patients, including two males and one female, were analyzed. Among them, two children were diagnosed with CAEBV with intermittent fever for more than four years, and one child was diagnosed with EBV-HLH. After sintilimab treatment and a mean follow-up of 17.1 months (range 10.0–23.3 months), patients 1 and 3 achieved a complete clinical response and patient 2 achieved a partial clinical response. All three children showed a > 50% decrease in EBV-DNA load in both blood and plasma. EBV-DNA copies in sorted T, B, and NK cells were also markedly decreased after sintilimab treatment. Conclusion Our data supported the efficacy of PD-1 targeted therapy in certain patients with CAEBV and EBV-HLH, and suggested that sintilimab could provide a cure for these diseases, without HSCT. More prospective studies and longer follow-up are needed to confirm these conclusions.
Background
The infection rate of Epstein-Barr virus (EBV) in the population worldwide is more than 95%, and the impaired balance between the host immune response and EBV can lead to various EBV-associated lymphoproliferative disorders (LPDs) of B, T, or natural killer (NK) cells [1,2].Chronic active EBV infection (CAEBV) and EBVassociated hemophagocytic lymphohistiocytosis (EBV-HLH) are rare but life-threatening diseases.To date, a standard treatment approach for CAEBV and EBV-HLH has not been established.Conventional therapies, including antiviral drugs and immune-modulatory agents, can lead to temporary remission; unfortunately, most patients relapse and progress [3].Etoposide-based HLH-1994 and HLH-2004 regimens are widely used; however, some patients are refractory or intolerant to intensive chemotherapy [4,5].Hematopoietic stem cell transplantation (HSCT) was considered as the only potentially curative method; however, it led to numerous transplantationassociated complications [3].
Programmed cell death-1 (PD-1) is a representative immunosuppressive checkpoint and is mainly expressed in activated T cells, B cells, NK cells, macrophages, dendritic cells, monocytes, and myeloid cells, and in immune-privileged sites [6].The interaction between PD-1 and its ligands leads to inhibition of T cell proliferation, activation, cytokine production, and cytotoxic T lymphocyte killer functions [7][8][9].In chronic infections or tumors, lasting antigen-exposure leads to permanent PD-1 expression, which can limit immune-mediated clearance of pathogens or neoplastic cells [10].The overexpression of PD-1 on virus-specific T cells has been documented in EBV and other virus infections [11,12].Immune evasion via the PD-1 pathway has been confirmed to play an important role in various EBV-positive cancers [13,14].PD-1 inhibition has achieved a remarkable response in EBV-positive lymphoma and EBV-associated gastric cancer, in which it is believed to reverse EBV or cancer-mediated immunosuppression by restoring immunity and releasing T cells [15][16][17][18][19].However, there have been few reports of the treatment of CAEBV and EBV-HLH with PD-1 inhibitors [20][21][22][23].Sintilimab is a recombinant human IgG4 monoclonal antibody against PD-1 that disrupts the interaction between PD-1 and its ligand, leading to T cell reinvigoration [16].The present study discussed the use of sintilimab in CAEBV and EBV-HLH combined with clinical experience in three patients.
Methods
We retrospectively analyzed the clinical data of three children diagnosed with CAEBV or EBV-HLH who were treated with sintilimab in the Department of Nephrology and Immunology, Children's Hospital of Soochow University between 12 December 2020 and 28 November 2022.Sintilimab was provided by the Xinda Biopharmaceutical (Suzhou, China) company.Real-time fluorescent quantitative PCR and TaqMan hydrolysis probes were used to detect EBV-DNA in peripheral blood and plasma.Intracellular EBV-DNA copies in sorted peripheral blood mononuclear cells (PBMCs) were also determined using quantitative PCR.
We used previously described criteria for response assessment [20,24,25].A clinical complete response (clinical CR) was defined as the resolution of all clinical signs and symptoms, including fever, liver dysfunction, progressive skin lesions, or vasculitis, accompanied by a significant decrease in EBV-DNA.Resolution of some of the above symptoms was defined as a clinical partial response (clinical PR).A molecular complete response (molecular CR) comprised a significant decrease in EBV-DNA load in both blood and plasma (< 10 2.5 copies/mL).A 50% drop in EBV-DNA load in either blood or plasma was defined as a molecular partial response (molecular PR).
Case 1
A Chinese male aged 6 years and 2 months (bodyweight, 16 kg) was admitted to our department with intermittent fever accompanying elevated liver enzymes for more than 4 years.The child had visited other hospitals many times and relevant examinations showed increased EBV-DNA copies in his peripheral blood and plasma, accompanying hepatosplenomegaly, liver function abnormalities, and lymphadenopathy (Table 1).There were no significant improvements in his symptoms and signs after treatment with antiviral drugs, intravenous immunoglobulin (IVIG), and glucocorticoids.Upon admission to our hospital, high EBV DNA loads in his peripheral blood, plasma, and bone marrow (1.97 × 10 6 copies/mL, 3.15 × 10 4 copies/mL, and 2.25 × 10 5 copies/mL, respectively) were observed.Intracellular EBV-DNA copies in sorted PBMCs were also high (CD3+CD4+ T cells: 7.7 × 10 5 copies/mL, CD3+CD8+ T cells: 1.1 × 10 5 copies/mL, CD3-CD19+ B cells: 1.2 × 10 4 copies/mL, and CD56+ NK cells: 2.7 × 10 6 copies/mL, respectively).Whole exome sequencing (WES) revealed no clear genetic mutations.The virus capsid antigen (VCA) IgG antibody, EBV nuclear antigen (EBNA) IgG antibody, and early antigen IgG antibody tests were positive.The VCA IgM antibody test was negative (Table 2).The above clinical manifestations and examination results met the recently revised diagnostic criteria for CAEBV [26].
During a follow-up of 23.3 months, fever and relevant signs improved significantly.When the 13th and 14th cycle interval was lengthened to 6 weeks, he developed fever again; however, the child's body temperature returned to normal and stabilized when the medication interval returned to 3 weeks.EBV-DNA loads in peripheral blood and plasma decreased markedly after sintilimab treatment (Fig. 1).EBV-DNA loads in sorted PBMCs decreased markedly, especially in CD3+CD8+ T cells (Table 4).The CD3+ T cell counts were maintained at a normal level, and the levels of CD3+CD8+ T cells were higher than those of CD3+CD4+ T cells (Fig. 2).After treatment with 30 cycles of sintilimab, patient 1 achieved clinical CR and molecular PR, and methylprednisolone and cyclosporine were gradually stopped.
Case 2
A Chinese girl aged 12 years and 6 months (body weight, 26.7 kg) was admitted to our department with intermittent fever accompanying skin rash and oral ulcer lasting more than 6 years (Table 1).The child had visited other hospitals many times, and her condition had not improved.Upon admission to our hospital, her EBV-DNA copies in peripheral blood, plasma, and bone marrow were elevated (1.41 × 10 7 copies/mL, 2.19 × 10 4 copies/mL and 5.34 × 10 3 copies/mL, respectively), and intracellular EBV-DNA copies in sorted PBMCs 2).The above clinical manifestations and examination results met the recently revised diagnostic criteria for CAEBV [26].She accepted the combined therapy of sintilimab, prednisone acetate (1 mg/kg/day and then gradually reduced), valganciclovir and tacrolimus.Sintilimab was administered via intravenous infusion at 3 mg/kg every 3 weeks (Table 3).
She never developed a fever; her skin rash and oral ulcer decreased gradually without recurrence during 10 months of follow-up.We observed 80.6% and 90.1% decreases in EBV-DNA copies in peripheral blood and plasma, respectively (Fig. 1), and a > 37% decrease in intracellular EBV-DNA copies (Table 4).The CD3+ T cell counts increased, and the levels of CD3+CD4+ T cells were higher than those of CD3+CD8+ T cells (Fig. 2).After regular treatment with 13 cycles of sintilimab, patient 2 achieved clinical PR and molecular PR.Prednisone acetate was reduced to a low dose (2.5 mg/ day) and tacrolimus was stopped after the 13th cycle.
The body temperature of the child improved significantly.He had a transient fever again before the 3rd and 6th cycles.EBV-DNA in plasma fell from 2.73 × 10 3 copies/mL to an undetectable level and remained stable after six cycles of sintilimab (Fig. 1).EBV-DNA copies in CD56+ NK cells decreased from 1.8 × 10 4 copies/ mL to an undetectable level after the 7th cycle (Table 4).The CD3+ T cell counts were maintained at normal and above levels, and the levels of CD3+ CD8+ T cells were higher than those of CD3+CD4+ T cells (Fig. 2).After regular treatment with 20 cycles of sintilimab and a follow-up of 18.1 months, patient 3 achieved clinical CR and molecular PR.Currently, all drugs had been stopped except for regular IVIG every month.
Discussion
In this retrospective clinical data analysis, we reviewed the clinical and immunological characteristics of three children with CAEBV or EBV-HLH.After sintilimab treatment and a mean follow-up of 17.1 months (range 10.0-23.3months), two patients achieved a clinical CR and one achieved a clinical PR.All three children showed a > 50% decrease in EBV-DNA load in both blood and plasma, which suggested a molecular PR.EBV-DNA copies in sorted PBMCs were also significantly decreased.
After primary EBV infection, individuals develop robust EBV-specific T cell immune responses, with EBVspecific CD8+ and some CD4+ T cells functioning as cytotoxic T cells, defending against the virus [27].Strong T cell immunity plays a key role in controlling infection by decreasing the viral load and eliminating infected cells.However, the continuous viral antigen burden during the course of chronic viral infection leads to persistent stimulation of antigen-specific T cells, resulting in T cell exhaustion [10,28].Studies have found that patients with CAEBV have a large number of myeloid-derived suppressor cells (MDSCs) that decrease the function of effector T cells, resulting in persistent uncontrolled EBV infection [29].EBV-induced HLH is the most common subtype of secondary virus-associated HLH during childhood, and is characterized by uncontrolled activation of T lymphocytes and macrophages [27].Kasahara et al. reported that EBV infection was predominant in CD8+ T cells in patients with EBV-HLH, whereas the dominant EBV-infected cell populations in patients with CAEBV were non-CD8+ lymphocyte subpopulations [30].Analysis of PBMCs in some patients with EBV-HLH showed a reduction in CD4+ T cells and abnormal activation of CD8+ T cells [27,31].
T cell activation relies mainly on a two-signal model [32].The first signal confers specific recognition of cognate antigenic peptides presented by major histocompatibility complex (MHC) molecules and the T cell receptor (TCR).The second signal comprises co-stimulatory and co-inhibitory signals, which modulate TCR signaling positively or negatively to direct T cell function [32,33].A group of inhibitory or stimulatory molecules expressed on immune cells, antigen-presenting cells, tumor cells, or other types of cells are regarded as immune checkpoints [34].PD-1 (also known as PDCD1 and CD279) is a representative immunosuppressive checkpoint that plays a key role in programmed death signaling to regulate T cell-mediated responses [35].PD-1 is activated by binding to programmed cell death 1 ligand 1 (PD-L1) or programmed cell death 1 ligand 2 (PD-L2) [36].Upon ligand binding, SH2 domain-containing protein tyrosine phosphatase 2 (SHP-2) is recruited to the immunoreceptor tyrosine-based switch motif (ITSM) of PD-1, leading to SHP-2 dephosphorylation of different targets downstream of TCR [7,9].
Similarly, high PD-1 expression on virus-specific T cells has been observed in infections with lymphocyte choriomeningitis virus (LCMV), human immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus (HCV) [11,41].An increase in the frequency of PD-1 on the surface of CD8+ T cells was also found during symptomatic primary EBV infection, and was associated with elevated EBV loads [12].However, reports of CABEV and EBV-HLH treated with PD-1 inhibitors are rare.You et al. [21] reported that sintilimab helped to treat mixed chimeric and reactivated EBV in a patient with adult-onset CAEBV after allo-HSCT, in which EBV-DNA was ultimately undetectable and a stable donor chimerism was obtained.Ma et al. [22] reported that 16 patients with CAEBV were treated with PD-1 inhibitors, including pembrolizumab (2/16), sintilimab (9/16), or nivolumab (5/16), and 12 patients responded to PD-1 inhibitors.Song et al. [23] investigated the combination therapy of sintilimab and lenalidomide in patients with CAEBV and reported an overall response rate of 54.2%.Liu et al. [20] reported that seven patients with relapsed/ refractory EBV-HLH were treated with nivolumab, among which 71.4% of the patients reached clinical CR.Clinical studies support the efficacy of PD-1 targeted therapy in a subset of patients with CAEBV and EBV-HLH.
In summary, our findings suggested that the PD-1 inhibitor sintilimab could achieve remarkable results in pediatric patients with CAEBV and EBV-HLH, and might provide a cure for these disease without the use of HSCT.PD-1 inhibition might restore immunity and release T cells, providing benefits for patients with CAEBV and EBV-HLH.However, further clinical trials and mechanistic studies are needed to verify its effectiveness.
Table 2
EBV DNA loads and EBV-specific antibodies before sintilimab treatment
Table 3
Sintilimab combined with other treatments
|
2023-09-22T13:51:34.767Z
|
2023-09-22T00:00:00.000
|
{
"year": 2023,
"sha1": "cef4a6c364734255d59360da2938e7c14641633d",
"oa_license": "CCBY",
"oa_url": "https://ojrd.biomedcentral.com/counter/pdf/10.1186/s13023-023-02861-9",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "8b99a848d9a67de66be731807c324f0b87cfeb60",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
248627717
|
pes2o/s2orc
|
v3-fos-license
|
Prevalence of multidrug-resistant strains in device associated nosocomial infection and their in vitro killing by nanocomposites
Background As per WHO, global burden of healthcare-associated infections (HAIs) ranges between 7% and 12%. There is a dire need to screen Device associated nosocomial infections (DANIs) in hospitals(1). To investigate the prevalence of microbes in hospitals in DANI cases and analyse in vitro control of multi-drug resistant strains by nanotechnology intervention. Methods Patients diagnosed with DANI were enrolled and monitored. Identification and antibiotic susceptibility pattern of the etiological agent of DANIs were made by the phenotypic method and Vitek 2 automated systems according to standard protocol. In addition, biosynthesized nanocomposite was analysed for their antimicrobial activity by agar well-diffusion method, CFU count and DNA degradation analysis. Results There were a total of 324 patients diagnosed with DANIs. Total 369 microbial pathogens were isolated from DANI patients. The majority (87%) of the pathogenic microbes were gram-negative bacilli and all were multidrug-resistant. 41.5% of the gram-negative isolates were ESBL producers. Methicillin-resistant Staphylococcus aureus contributes about 7.3% of the total isolates in gram-positive bacteria. Nanocomposite showed 100% bactericidal activity at 5 mg/ml concentration within 3 h of incubation, whereas 2.5 mg/ml concentration of nanocomposites takes 6 h to inhibit complete growth. Conclusions DANI, which was found in patients of all age groups, us due to multidrug-resistant gram-negative bacteria. The most commn causative agents were Acinetobacter baumannii and Citrobacter species. Nanocomposites can provide an alternative solution to prevent the DANIs.
Introduction
Patients requiring life-saving devices are perpetually admitted to the hospital's Intensive Care Unit (ICU). They regularly go through invasive strategies together with intra-tracheal intubation for mechanical ventilation or insertion of intravascular and urinary catheters. In these instances, if the right care package isn't followed, there will be development of device-associated Nosocomial Infections (DANIs). Increased nosocomial infections leads to excessive morbidity and mortality. Incidences of infections amongst patients inside the ICU are 5-to 7-fold higher than trendy inpatient admissions of all nosocomial infections in a hospital [1]. There is a worldwide escalation in each community-and hospital obtained infections because of Antimicrobial-resistant (AMR) microorganisms compromising the capacity to deal with those sufferers effectively, thereby underscoring the need for endured surveillance, suitable prescribing of antibiotics, implementation and implementation adherence to stringent contamination control measures [2].
Studies have proven the occurrence of pathogens, along with the resistant genotypes of Methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococcus, Extended-Spectrum Beta-Lactamase (ESBL)-generating Escherichia coli and Klebsiella species& carbapenems-resistant E.coli, Klebsiella speciesProteus species, Pseudomonas aeruginosa, and Acinetobacter baumannii inflict HAIs, especially inside the ICU setting. Available healing alternatives for AMR organisms are seriously restrained as those organisms often showcase multidrugresistance. It is well known that inappropriate and irrational use of antibiotics to treat infections leads to the emergence of Multi Drug Resistant (MDR) strains among the common bacterial isolates [4]. This translates into a prolonged hospital stay, a significant increase in morbidity and mortality, and an escalating economic burden. The frequency of infections among patients admitted to the ICU may vary from one geographical region to another, from one hospital to another, and even among the ICUs within one hospital. The type of infection, the profile of pathogens causing these infections, their antimicrobial susceptibility patterns also vary according to the location. It is, therefore, imperative for the treating clinician to have adequate information of the spectrum of microorganisms and the AMR patterns prevalent in that particular setting for initiating empirical therapy with appropriate antimicrobial agents [5]. Nanomaterials inhibit bacterial growth or activity that results in infections. Nanoparticles penetrate the bacteria and biofilm leading to reactive oxygen species (ROS) generation that eliminates bacteria. Thus, nanoparticles are a novel approach to combat drug-resistant bacterial infections [6]. Additionally, the ionic activity of nanoparticles can modulate the bacterial signal transduction leading to the inhibition of bacterial growth or inactivating the enzymes by interacting with them [7]. In the present study, an attempt was made to analyse antibacterial effect of nanocomposite against MDR strains and its potential use in combat against nosocomial infections. This study was conducted in MMIMSR and associated hospital, a tertiary care teaching hospital located in the Mullana, Ambala Haryana, India, with an active infection control committee. We studied the prevalence of VAP, CLABSI and CAUTI, causative organism, antimicrobial resistance, and MRSA prevalence in S.aureus, ESBL and Metallo-β-Lactamase (MBL) producing Gram-Negative Bacteria (GNB). Additionally, we have also analysed antibacterial susceptibility of carbon quantum dots decorated dual Z scheme Manganese Indium Sulphide/Cuprous Oxide/Silver oxide against MDR strains.
Material & methods
Prospective, site-specific surveillance of DANI was carried out from October 2018 to July 2021 in various ICUs of Hospitals in Maharishi Markandeshwar (Deemed to be University), Mullana,-Ambala, India. The approval was taken from Institutional Human Research Ethics Committee (IEC no. 1147). Based on the CDC's National Nosocomial Infections Surveillance (NNIS) system criteria, samples for three DANIs: ventilator-associated pneumonia (VAP), catheter-associated urinary tract infection (CA-UTI) and central line-associated bloodstream infection (CLABSI) were taken into consideration.
Patients selection
For VAP the Patient with pneumonia and placed on Mechanical Ventilation for >2 calendar days was selected as per CDC guidelines, including onset of purulent sputum or change in the character of sputum, increased respiratory secretions, increased suctioning requirements, new onset or worsening cough, dyspnea, tachypnea, bronchial breath sounds, worsening gas exchange, increased oxygen requirements, or increased ventilator demand [8]. Similarly CDC guidelines were followed for enrolling patients for CAUTI and CLABSI [10,11]. In all the cases only those patients were considered for DANI where the symptoms appeared after two calendar days of admission in ICU [9,10].
Sample collection and identification of microbes
From patients suspected of DANI, the appropriate clinical samples were collected and microbial analysis was performed with the help of a trained Infection Control Nurse the samples were collected as per clinical and laboratory standards institute (CLSI) guidelines. All the clinically relevant details were collected from patients, including age and sex. For microbial analysis of VAP, deep tracheal aspirate from the endotracheal tube was collected. For CAUTI, urine samples were aseptically aspirated from the sampling port of the urinary catheter. For CLABSI, the peripheral blood sample, the central line removed aseptically, and the distal 5 cm of the catheter were collected and processed for analysis (11).
According to the laboratory Standard Operational Procedures, phenotypic and automated identification of microbes was made based on CLSI recommendations. Phenotypic identification consisted of Gram staining followed by a series of biochemical tests specific for each group of microorganisms. Yeasts were isolated on Sabouraud agar and identified by culturing on CHROM agar based on their colonies' colour, texture, and shape. Moreover, we have employed an automated VITEK® 2 system to identify strains. Samples exhibiting microbial growth on blood agar or MacConkey agar were inoculated into specific identification cards of the automated VITEK® 2 system [12].
Antibiotic susceptibility test (AST) of bacterial isolates
AST of the identified strains was done by Kirby Bauer disc diffusion method and VITEK® 2 automated systems using the standard protocol.
Detection of MRSA strains was done using cefoxitin (30 μg) disc by modified Kirby-Bauer disc diffusion method [13]. The phenotypic analysis of extended-spectrum β Lactamase was done by Double-Disc Diffusion as described earlier [14]. Similarly, the phenotypic confirmation of Metallo β Lactamase was done by Combined Disc Method as described earlier [15].
Analysis of antimicrobial property of nanocomposite
We further wanted to test the effect of nanocomposites on the growth of multidrug-resistant strains. In our recent study, we found that nanocomposites could control the growth of bacterial strains. We tested the antibacterial activity of carbon quantum dots decorated dual Z-scheme Manganese Indium Sulphide/Cuprous Oxide/Silver oxide Nanocomposites using the agar well diffusion assay and by enumerating the CFUs as described earlier. Different concentration of nanocomposite (1.25 mg/ml, 2.5 mg/ml & 5 mg/ml) was used [16][17][18].
Effect of nanocomposites on genomic DNA of MDR bacterial strains
MDR strains (0.1 OD 600 ) were incubated with and without nanocomposite (2.5 mg/ml) at 37 • C overnight [19]. After the incubation, cells were harvested for genomic DNA isolation as described earlier [20]. Isolated DNA was analysed and compared with the untreated sample by agarose gel electrophoresis.
Prevalence of DANI
During our study from October 2018 to July 2021, 7050 patients were admitted in different ICUs with indwelling devices. There were 2160 patients with an endotracheal tube carrying a mechanical ventilator, 1590 patients with a central venous catheter, and 3300 patients with a urinary catheter. The total number of device days of 7050 patients was 64800. Based on clinical signs and symptoms in correlation with microbial culture (discussed below), 324 patients were diagnosed with DANI (124 patients of VAP, 18 patients of CLABSI, and 180 patients with CAUTI). During the study period, CAUTI was the most commonly diagnosed DANI found in 180 patients with 3.75 cases per 1000 device days, followed by VAP in 126 patients with 11.21 cases per 1000 device days and CLABSI at 18 patients with 3.18 cases per 1000 device days. The crude infections rate of DANIs was 5 cases per 1000 device days (Table I). The age distribution as shown in Table II indicated that DANI was more prevalent in 18-40 age groups.
Identification of microbes causing DANI
The samples were processed as described in Materials and Methods to know the responsible microbial pathogens of DANI by standard tests. We found 369 microbial pathogens from 324 DANI patients. Acinetobacter baumannii was the most common microbial pathogen contributing 23.57% of the total isolates followed by Citrobacter species 17.88%, Klebsiella species 57%, and Escherichia coli 13%. Other microbial pathogens isolated from DANI patients were Staphylococcus aureus 10.84% (68.28% MRSA & 35.72% MSSA), Enterococcus species 1.62% and Proteus species 0.81%. Candida was the only fungal isolate contributing in DANI. Proteus was the most minor contributor in DANI (Table III).
Antimicrobial sensitivity pattern of gram-positive cocci isolated from DANI patients
Further, we wanted to analyse the drug sensitivity pattern of the isolated pathogens. Our results indicated that all gram-positive isolates were sensitive to linezolid and vancomycin. MRSA strains showed sensitivity to G and TE at 55% and 44%, respectively. The MRSA showed the least sensitivity to E, AZ & OF with 0%, 0%, and 11%. MSSA strains showed sensitivity to all antibiotics tested. Enterococcus species showed 50% sensitivity to COT as well (Table IV).
Distribution of methicillin resistance staphylococcus obtained from DANI patients
Methicillin resistant S. aureus was detected in 64.28% of the total S. aureus whileremaining 35.72% of the staphylococcus was identified as methicillin sensitive S. aureus (Fig. 1 A).
Antimicrobial activity of nanocomposites
The isolated strains from DANI patients were resistant to conventional drugs so, we further wanted to explore if nanocomposites could control these pathogens. Antimicrobial activity of nanocomposite was investigated against the resistant strains using the agar well diffusion assay and by enumerating CFUs in the presence of nanocomposites. Zones of inhibition (mm) around each well containing different concentration (0.25 mg/ml, 0.5 mg/ml, 1 mg/ml and 2 mg/ml) of nanocomposite and Colistin (10 μg) are represented in (Table VI). All MDR isolates tested by agar well diffusion assay showed zones of inhibition ranging 20 mm-24 mm (Table VI). Furthermore, we found complete bactericidal activity at 5 mg/ml of nanocomposite within 3 h of incubation. At a lower concentration of 2.5 mg/ml, complete growth inhibition was observed after 6 h whereas at the lowest concentration tested of 1.25 mg/ml the antibacterial activity was 85.5% in 6 h (Fig. 3).
DNA degradation by nanocomposite
One of the methods by which the nanocomposite could show bactericidal activity is by producing reactive oxygen species which could degrade DNA of MDR strains. To analyse this, bacterial strains were treated with 2.5 mg/ml nanocomposite at 37 • C for 6 h. After incubation DNA extraction was done and it was observed that bacterial strains treated with nanocomposite exhibited complete degradation of DNA (Fig. 4, lanes 2-6) as compared to the untreated bacterial strain (lane 1).
Discussion
During the study period, 7050 patients were hospitalized in the different ICUs for 64800 days. We found that 324 patients acquired DANI. The overall DANI rate was 5 per 1000 device days, whereas the overall rate of getting VAP, CLABSI & CAUTI in patients was 11.21, 3. 2017 showed overall DANI rates of 74.9 per 1000 device days, it was due to longer duration of ICU stay serves greater exposure to pathogens, frequent invasive procedures, lack of a proper hospital infection control and monitoring system. Moreover, with respect to gender male patients developed DANIs higher than female patients but there seems no relation between age and gender with respect to development of DANIs [24].
A total of 369 pathogens were isolated from DANI patients. Majority (82.11%) were gram-negative organisms, the rest, 11.38%, were grampositive and 6.51% were yeast (Candida sp.) were the etiological agent. All of the isolated microbial pathogens showed resistance to at least one antibiotic of three to four groups of antibiotics and considered as multidrug-resistant bacterial strains. According to previous study infections due to gram-positive organism are more prominent in the Western world ICUs, in contrast to this, gram-negative organism were the major contributor in causing DANI in India and Asia-Pacific region [25].
The National Nosocomial Infections Surveillance System reported a significant increase in the proportion of Acinetobacter species in causing hospital acquired infections among all gram-negative aerobes [26]. Acinetobacter baumannii was also the predominant (23.57%) bacterial pathogen isolated form DANI patients in our study. Interestingly, we found that except MSSA, all the bacterial isolates exhibited MDR phenotype. In brief Acinetobacter baumannii and Citrobacter species showed 100% sensitivity to TGC followed by IMP, MRP, GEN, PIT and AK. Similar to our findings, Ghanshani R et al., 2016 revealed in their study that ≥95% bacterial pathogens were sensitive to colistin, Klebsiella and Pseudomonas were >50% resistant to 3rd generation cephalosporin and carbapenems, while E. coli was still >50% sensitive to carbapenems and Acinetobacter> 50% sensitive to 3rd generation cephalosporin. Gram-positive organisms showed zero sensitivity to penicillin, oxacillin, and tetracycline. MSSA were 100% sensitive to vancomycin, and 50% sensitive to linezolid and gentamycin. Enterococcus was 100% sensitive to linezolid, 50% sensitive to vancomycin [27]. Study conducted by Dutta V et al., 2017 found that S. aureus and Enterococcus were 100% sensitive to linezolid and vancomycin and more than 50% resistant to gentamicin, erythromycin, and ciprofloxacin. Indiscriminate use of antibiotics for prolonged and inappropriate duration is the possible explanation of such high levels of multidrug resistance in these organisms [28].
In addition to this, 68.75% Escherichia coli, 57.89% Klebsiella species, 100% Proteus species, 21.42% Pseudomonas aeruginosa, 27.58% Acinetobacter baumannii and 31.81% Citrobacter species were ESBL producers. This suggested that cephalosporin group of antibiotics were the most favored drug of first-line treatment and use of these antibiotics leads evolution of ESBL production by microbial pathogens resulting in their lower efficacy. Also, 31.81% Citrobacter species, 27.58% A. baumannii, 26.31% Klebsiella species and 7.14% Pseudomonas aeruginosa was MBL producer. We found that Escherichia coli &Proteus sp. didn't produce MBL. Mathai et al., showed 3.6% strains of Enterobacteriaceae were MBL producer [29]while Patro S et al., 2018 observed thatMBL was positive in 17.64% non-fermenters and 17.39% in Enterobacteriaceae [30]. Due to their efficacy of broad spectra and low toxicity β-lactam antibiotics are the major bulk of prescribed antibiotics in ICUs across the globe. However, irrational use of beta lactam antibiotics had been resulted in the development and spread of drug resistant bacterial pathogens mainly by production of ESBL enzymes. For the treatment of ESBL producing pathogens carbapenems group of antibiotics were the drug of choice hence frequent use of carbapenems, pathogens rapidly adapt and modify it self to produce Carbapenemase which becomes major healthcare burden [31].
Nowadays, research in the area of nanotechnology indicates the use of nanocomposite in controlling bacterial infections. These nano composites can be used in drug deliver as well as in making nanomaterials to control infections [7]. We found that carbon quantum dots decorated dual Z-scheme Manganese Indium Sulphide/Cuprous Oxide/Silver oxide nanocomposite was able to control the MDR strains. It efficiently kills the bacteria within 3 h at 5 mg/ml concentration by degrading its DNA. Nanoparticles/nanocomposite penetrate the bacteria leading to reactive oxygen species generation that eliminates bacteria. Coating nanocomposite on surface of medical devices such as endotracheal tube, central line catheter, and urinary catheter may impede microbes responsible for the development of DANIs and provide a great tool to combat drug-resistant bacterial infections. Different nanomaterials, such as nanoparticles and nanotubes can be directly used in biomedical devices to prevent spreading infections.
Conclusions
An intensive care setting is a high-risk area for acquiring DANIs. The high rate of infections in ICUs is due to frequent use of medical devices, increased device days, length of hospital stays, and patient disease severity. Our study suggested for an alarming increase in MDR in DANI patients which indicate that there is a dire need to control these MDR.
Ethical approval
All procedure for research has been approved by the ethics committee of IEC (institutional ethical clearance) number: 1147.
Sources of funding
No funding.
Consent
None.
Author contribution
Shahbaz Aman: conceptualisation and designed the study, drafted the initial manuscript, and reviewed and revised the manuscript, Narinder Kaur: Coordinated and supervised data collection, Divya Mittal: carried out the initial analyses, and reviewed and revised the manuscript, Shalini Shriwastav: data collection, data interpretation, Shubham Chauhan: data collection, data interpretation, Pardeep Singh: data collection, data interpretation, Sheetal Sharma: resources, data collection, data interpretation, Reena V. Saini: reviewed and revised the manuscript, Hardeep Singh Tuli: data interpretation and writing manuscript, Adesh K. Saini: Coordinated and supervised data collection, and critically reviewed the manuscript for important intellectual content.
Registration of research studies
Name of the registry: Unique Identifying number or registration ID: Hyperlink to your specific registration (must be publicly accessible and will be checked):
Declaration of competing interest
None.
|
2022-05-10T16:24:44.852Z
|
2022-05-01T00:00:00.000
|
{
"year": 2022,
"sha1": "1e1911ea7017eeaa478db44ad3c3172928c84438",
"oa_license": "CCBYNCND",
"oa_url": "https://doi.org/10.1016/j.amsu.2022.103687",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "91377044b0c7f2f82567c86155cebe1de30e3dd4",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
244490039
|
pes2o/s2orc
|
v3-fos-license
|
Antioxidant and tyrosinase inhibitory activities of traditional fermented Rosa from Dali Bai communities, Northwest Yunnan, China
Traditional fermented Rosa (TFR) is a typical food and medical product among the Dali Bai people, and its popularity is growing. A few studies have looked into TFR's medicinal advantages, linked germplasm resources, traditional processing procedures, and functional food qualities. Our goal was to look into Rosa's traditional processing, examine the dominant strains in TFR, and prove how these strains affected antioxidant and tyrosinase inhibitory activities. We used a snowball selection strategy to pick 371 informants for a semi-structured interview, supplemented with direct observations and sample collection. A microbial strain was isolated and identified from a TFR sample collected in the field. We synthesized TFR in the lab using the traditional way. Both of 2, 2-diphenyl-1 picrylhydrazyl (DPPH) free radical scavenging and tyrosinase inhibitory properties of the fermented solution of Rosa 'Dianhong' have been tested in this study. Altogether 15 species belonging to the genus Rosa, which are utilized in herbal medicine and fermented foods. Rosa 'Dianhong' was the Bai community's principal species with considerable cultural value and consumption. Raw Rosa petals included 15 major flavonoids and phenols, which were identified as TFR's active components. TFR-1 was discovered to be the dominating microbial strain in TFR, increasing total phenolic and flavonoid content in the fermented solution of Rosa 'Dianhong' by 0.45 mg GAE/ml and 0.60 mg RE/ml, respectively, after 30 days. TFR-1 also exhibited promising activity in terms of DPPH free radical scavenging and tyrosinase inhibition. TFR showed potent antioxidant and free-radical scavenger properties and is beneficial in skincare and nutrition, according to the findings. TFR's medicinal and edible properties suggest that it could be used as a cosmetic or nutraceutical product.
Fermentation has a long history in human food production and is a valuable process in food industries, making food products available throughout the year 1,2 . Fermentation of edible items improves product quality and creates diverse flavor components, which boosts consumer acceptability [3][4][5] . Furthermore, by adding salt and producing acid and ethanol, the fermentation process can improve the nutritional and functional qualities of foods while also extend their shelf life 2 . Microorganisms and their enzymes cause fermentation, which is a biochemical alteration of the fundamental food matrix. Traditional knowledge of the fermentation of edibles has shown pharmacological activity that can be used to cure a variety of diseases. Fermentation, for example, allows for the manufacture of broccoli products that are safe, stable, and high in sulforaphane. These fermented foods are fantastic dietary supplements 6 . It also contains anti-cancer 7,8 , anti-diabetes 9 , and anti-obesity 10 properties, as well as helping to alleviate behavioral issues linked with autism spectrum disorder 11 . Fermentation also enhances the antioxidant activity and tyrosinase inhibitory activity of phenols 12 and flavonoid antioxidants 13 .
The adaptive nature of the fermentation process within a given region arises from centuries of human relationships with the microbial community in the environment. Microorganisms, inorganic components, and their interactions all play a role in fermentation processes and product creation 14 . In traditional fermentation processes, such as those of Brassica oleracea L., Oryza sativa L., and Glycine max (Linn.) Merr., microbial sources are either internal to plants or derived from external surfaces and the surrounding environment. Fungi and bacteria, particularly yeast and lactic acid bacteria populations among the microorganisms that increase the quality of traditional fermentation products 3,4 . Traditional fermentation is a promising method for isolating certain microorganisms' impact on biosynthesis or breakdown of bioactive substances.
Traditional fermented Rosa (TFR) is a biocultural heritage among Dali Bai communities in northwest Yunnan, China, which has evolved over a long time as a result of interactions between local cultures and their environment 15 . TFR is widely acknowledged in Dlai Bai communities as a medical food that nourishes the body and keeps the skin smooth. However, until today, knowledge of traditional TFR processing has been preserved within the community. TFR's plant resources, fermentation method, and bioactivity have never been comprehensively described. When the globe is confronting a pandemic like Covid-19, research into traditional knowledge around such a vital food source and its bioactivity is critical. Functional foods with health advantages are in high demand. To that end, the goal of this study was to determine the main microbial strains and relative bioactivities of TFR, as well as clarify the plant resources and conventional processing methods employed in TFR. Our research not only contributes to biocultural preservation, but it also gives critical information for the future development of TFR-based pharmaceuticals and prospective nutraceuticals.
Diversity of the genus of Rosa in Dali Bai communities. Our informants told the interviewers about
15 Rosa species used in Dali for traditional applications. Both wild and cultivated species are in use as food and medicine. Table 1 shows the usage information for these 15 species. The informants reported that Rosa species have anti-aging properties; cure rheumatism and dehydration; activate blood circulation; and possess detoxification, insecticidal, and diuretic properties. These plants are used as medicine, food, and fragrance, with different methods and frequency of use.
The Dali Bai communities primarily use 10 of the 15 species for medicine, 7 for food, and 2 for fragrance. Rosa damascena is one of the main species used for essential oil extraction. Of the seven edible species, Rosa 'Dianhong' scored the highest informant consensus factor (ICF) 16 and use frequency (f) 17 values, indicating its cultural value and extent of consumption (Table 1). Rosa 'Dianhong' is used in petal tea, wine, sugar, and TFR. TFR is the most popular among these uses.
The TFR in Dali Bai communities. TFR in Dali was prepared by natural fermentation. Petals picked from locally available edible Rosa plants were the main ingredient in TFR (Fig. 1). The traditional planting of edible Rosa resources including Rosa 'Dianhong' , Rosa 'Mohong' , Rosa laevigata, Rosa rugosa, Rosa roxburghii, Rosa gallica, and Rosa centifolia. Rosa 'Dianhong' was mainly used in TFR preparations, while Rosa 'Mohong' was used as an ingredient of TFR only in Heqing county. According to our informants, the thinner petals of Rosa 'Dianhong' have better flavor after fermentation compared with those of Rosa 'Mohong' .
All informants mentioned that TFR benefits the skin. The majority of informants (83.3%) told us that TFR can nourish the stomach, and a few informants mentioned that TFR is hepatoprotective and refreshing. Although none of the informants could explain the pharmacological reasons, these traditional uses hint at potential therapeutic value.
Microbial strain identification. Microbial strains were isolated from TFR prepared traditionally by the Bai people in Dali. The ITS fragment from the dominant strain, TFR-1 (Table 2), shared 99.84% similarity with that of Saccharomyces rouxii. Therefore, the strain TFR-1 was named Saccharomyces rouxii TFR-1. The strain is now preserved in the China General Microbial Culture Collection Center (CGMCC) with the sample number CGMCC19335.
Pharmacological properties of FSR. Antioxidant activity. The free radical scavenging activity of FSR after different periods of fermentation were examined. As shown in Fig. 2. At low concentration (i.e., 0.025 mg/ ml, 0.05 mg/ml, and 0.1 mg/ml), samples FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30 showed better DPPH free scavenging activities than FSR-0. At a concentration of 0.05 mg/ml, this difference was significant (P < 0.05). www.nature.com/scientificreports/ FSR-21 had the highest DPPH scavenging activity, reaching 44.60%. At concentrations greater than 0.2 mg/ ml, there was no significant difference in the DPPH scavenging activity of FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30 compared with that of FSR-0. However, with increasing concentration, the trend for greater scavenging activity with longer fermentation began to flatten. DPPH radical scavenging rates of these samples were higher than 85% after 30 days. Table 3 shows the IC 50 values for FSR and the positive control.
Tyrosinase inhibition activity. At a concentration of 0.125 mg/ml, tyrosinase inhibition activity of FSR was insignificant at all time points compared with FSR-0, except in the FSR-30 sample (5.48%; P < 0.05) (Fig. 3). However, as the concentration of FSR increased, tyrosinase inhibition activity also increased. At a concentration of 0.5 mg/ml, tyrosinase inhibition by FSR-7, FSR-14, FSR-21, and FSR-30 was significantly different from that by FSR-0 (P < 0.05). The confidence level for the FSR-30 sample was higher (P < 0.001), and FSR-30 was also significantly different from the FSR-21 sample (P < 0.05). Although only the inhibitory activity of FSR-30 showed a statistical difference compared with FSR-0 at a concentration of 2 mg/ml, the mean value of tyrosinase inhibition increased with increasing fermentation time. At a concentration of 10 mg/ml, tyrosinase inhibition by FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30 was significantly higher than that of FSR-0 (P < 0.001). The highest inhibition rate was 86.58% and was for FSR-30. Among all samples, longer FSR fermentation duration resulted in greater inhibition of tyrosinase activity. Table 4 shows the IC 50 values for FSR and the positive control.
Nutrients in FSR and phytochemical profile. The total phenolic contents of FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30 were significantly higher than that of FSR-0 (P < 0.001) ( Table 5). Longer duration of fermentation increased the total phenolic contents of FSR. However, none of the differences between adjacent fermentation intervals, FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30, were significant. (Table 5). However, different from the total phenolic contents, across the whole fermentation process, the values for total flavonoid contents of FSR-7, FSR-21, and FSR-30 were significantly different from those at the previous time point (P < 0.05). This indicated that fermentation duration plays an important role in increasing total flavonoid content in FSR.
Correlation analysis. Our results revealed that an increase in total phenolic and flavonoid contents produced by increased fermentation time enhances the DPPH free radical scavenging rate and tyrosinase inhibition activity of FSR (Fig. 4). Pearson analysis demonstrated a moderate correlation between total phenolic content and DPPH free radical scavenging activity, but the difference was not statistically significant (P > 0.5) (Fig. 4A). Total phenolic contents were highly correlated with tyrosinase inhibition activity (P < 0.5) (Fig. 4B). These results indicate an increase in phenolic content after fermentation as an important factor triggering tyrosinase inhibition activity of FSR.
Pearson analysis showed that total flavonoid contents after fermentation were highly correlated with DPPH free radical scavenging rate and tyrosinase inhibitory activity (P < 0.5) (Fig. 4C,D). Moreover, the total flavonoid contents changed significantly with increases in fermentation time (Table 5).
Main effective compounds in the genus Rosa used by Bai people in Dali. Petals from 15 plant species of the genus Rosa used in traditional applications in Dali Bai communities were collected; most were included in the previous phytochemical and pharmacological studies. The contents of the petals varied, but most of them contained flavonoids and phenols, which are known to be bioactive in vitro. Antioxidation is an important bioactivity of TFR in vitro. For example, according to Vinokur et al. 18 , the radical scavenging activity in rose petals is mostly due to the high content of phenolic compounds, especially free gallic acid (1; Fig. 5). protocatechuic acid (2), syringic acid (3), anthocyanin (4), 4-hydroxybenzoic acid (5), chlorogenic acid (6), and catechin hydrate (7) are also predominant phenols in rose petals 19 . These phenols have more hydroxyls, including o-dihydroxy, which demonstrate strong free radical scavenging and antioxidant abilities 20 . Flavonoids possess a galloyl ester in the C ring, which are important structures for chelation of metal ions, formation of complexes with metal ions, and inhibition of metal-initiated lipid oxidation. Therefore, flavonoids are able to effectively scavenge hydroxyl and peroxyl radicals [21][22][23] . The high concentration of flavonoids in Rosa resulted from the presence of a large amount of naringenin (8), with quercitrin (9), hesperidin (10), quercextin (11), luteolin (12), apigenin (13), and kaempferol (14) also comprising the main effective flavonoid compounds in rose petals; these flavonoid compounds contribute towards the antioxidant capacities of rose 19 . According to Jin 24 , the main flavonoids in rose petals include rubin (15), quercetin (11), kaempferol (14), and their derivatives; these compounds contributed more than 60% to the flavonoids in TFR rose petals, and our correlation analysis showed that they were significantly related to antioxidant activity. Flavonoids also slow tyrosinase activities by interacting with copper ions essential to the active site of tyrosinase. Tyrosinase, a key enzyme in skin pigmentation, catalyzes hydroxylation of 26,27 . Therefore, flavonoids are considered to be a natural tyrosinase inhibitor.
Discussion
Traditional edible flowers have been utilized as a therapeutic food in China for thousands of years 28 because of their emollient, antibacterial, and anti-inflammatory properties. Many Dali Bai groups use specific plants because they are considered healthy, according to our ethnobotanical survey. Specific species, biochemical ingredients, or pharmacological qualities are the subject of most studies on the beneficial properties of food plants. Many researchers, for example, are looking for potential nutritional supplements against cancer in food plants. Antioxidants play a role in cancer prevention, and many types of research have been conducted on these substances. Apium nodiflorum, Humulus lupulus, Silene vulgaris 29 , Nasturtium officinale 30 , and Leopoldia comosa bulbs 31 all have strong antioxidant capabilities. Our survey showed that Rosa 'Dianhong' petals are most commonly employed in traditional meals as a classic sauce, and their phenolic and flavonoid components have been linked to health benefits. Fermentation changes the chemical composition of food, potentially increasing its biological activity 32 . Fermentation is traditionally employed for food preservation 3,33 , but it is also an important biotransformation process for producing new products or crude materials 6,34 . Fermentation appears to increase the total phenolic and flavonoid content of FSR, as suggested by our findings. According to Pearson analysis, there is a strong link between total phenolic and flavonoid content in FSR and tyrosinase inhibition, consistent with prior research 35 . TFR can play a crucial role in conferring antioxidant and tyrosinase inhibitory action by boosting phenols and flavonoids. PDA isolated TFR-1, which was later recognized as belonging to Saccharomyces rouxii. It is a yeast commonly employed in soy sauce fermentation and safe to consume as it has been consumed for a long time in human history. This strain can withstand high osmotic pressure because it produces sugars, alcohols, acids, and other chemicals involved in sugar and energy metabolism to protect cells under hypertonic culture conditions 36 . Traditional Rosa fermentation uses a 50 percent sugar concentration to retain sweetness and taste under sugar stress to generate a high osmotic pressure environment that inhibits putrefactive, pathogenic, and Only one strain, TFR-1, could tolerate the high osmotic pressure and play an important role in TFR processing in Dali Bai populations. Because of its excellent osmotic pressure resistance, non-toxicity to humans, and other possible functions in healthcare, the strain in this study is worth further investigation.
Conclusion
In the Dali Bai villages in Northwest Yunnan, China, TFR is a traditional medical food. As evidenced by our ethnobotanical survey, local people sustain the traditional TFR process, which has been passed down from generation to generation through these activities. Our findings show that TFR prepared traditionally contains strain TFR-1 (Saccharomyces rouxii) as the most important microbial content that facilitates fermentation and impacts TFR quality. It increased the antioxidant activity and tyrosinase inhibitory activity of FSR. These critical functional activity alterations provide applicable research for TFR products, mainly cosmetic or nutraceutical products. Traditional medical food culture promotes environmental protection, protects fast vanishing traditional knowledge, and has numerous applications in other domains of human activity. To that end, this work underlines the importance of continuing research on traditional fermented food products and related traditional knowledge as a method of discovering new strains and expanding the commercial potential. Our findings also highlight the importance of traditional items in creating modern healthcare, food, and cosmetic businesses. 40 . First, the purpose of the study was explained to key informants of Bai communities. Local community committees were then visited to obtain field study permission and request assistance. The assistance included introduction to community members and heirs of the intangible cultural heritage of TFR and organization of representative workshops. All field studies were carried out with informed consent. An eth- (Table 6) using the snowball sampling method to select potential informants 41 . Herbalists, farmers, merchants, and indigenous people were among the informants. They were engaged in the collection, production, sales, and use of the genus Rosa.
Study area. This study was conducted in Dali Bai
Women are frequent users of the genus Rosa and accounted for 55% of informants. Men who are herbalists, farmers, and vendors are associated with the value chain of Rosa. All had long-term experience with the applications of the genus Rosa. Semi-structured interviews were conducted with the consent of local people. The interviews focused on the following questions: (1) Does the community use the Rosa species? If yes, how many species are in traditional use? (2) Where do you get these plants? (3) How do you use these plants? Voucher specimens of 20 Rosa species naturally occurring in the study area were collected. A taxonomist and ethnobotanist at Kunming Institute of Botany identified the voucher specimens. All specimens collected during the field survey were deposited in the Herbarium of Kunming Institute of Botany.
Isolation, purification, and identification of strains. TFR prepared traditionally by the Bai people in Dali was used for strain isolation. TFR solution was prepared in sterile water at different concentrations (10 -1 -10 -5 ). The dilution coating method was applied 42 . The diluent was inoculated onto a PDA (potato dextrose agar) medium for culture of fungi 43 and onto NA (nutrient agar) medium for culture of bacteria 44,45 . After growth, fungal and bacterial colonies were purified using the plate streak method 45 . Purified materials, which were similar to yeast, were stored at 4 °C for preservation. PDA and NA medium were both from Qingdao Rishui Biotechnology Co., Ltd. The strain was named TFR-1 and deposited at the Kunming Institute of Botany, Chinese Academy of Sciences.
To ascertain the identity of strain TFR-1, its total genomic DNA was extracted using a DNA extraction kit from Beijing Tsingke Xinye Biotechnology Co., Ltd. (Tsingke) using the CTAB/SDS method 46 . The ITS1 region from the DNA sample was amplified by PCR using an Applied Biosystems 2720 thermal cycler, while fungal sequences were amplified using universal primers (Tsingke). Amplified PCR fragments were sequenced by Tsingke and used as a query sequence in a BLASTN search against the NCBI public database, following the method described by Blanc et al. 47 .
Preparation of Rosa fermentation and sampling. TFR was prepared in the laboratory following the traditional method. Petals of Rosa 'Dianhong' that is traditionally used in TFR preparation in Dali were purchased, and 0.5 kg of petals was cut into small pieces to maximize extraction. Brown sugar (0.5 kg) was added and the mixture was kneaded manually until fully wet. The mixture was then placed into a 1-l fermentation bottle and 500 ml sterile water was added. Three bottles of this solution were prepared as replicates. The bottles were exposed to ultraviolet light for 45 min for sterilization on a super purgative working table. After 24 h, 10 ml Rosa 'Dianhong' solution from each bottle was removed and stored as sample FSR-0. Then, TFR-1 solution (15 ml) with a concentration of 4.15 × 10 6 CFU/ml was added to the fermentation bottles, which were sealed with caps and stored at 28 °C in a thermostatic incubator. Fermentation solution of Rosa 'Dianhong' (FSR) from each bottle (10 ml) was sampled on days 3, 7, 14, 21, and 30 and labeled FSR-3, FSR-7, FSR-14, FSR-21, and FSR-30, respectively. Each of the samples was centrifuged at 4000 rpm for 15 min, and the supernatant liquid was stored at − 20 °C to cease further fermentation.
Determination of total phenolic contents. Total phenolic contents of FSR were determined spectrophotometrically using Folin-Ciocalteu's method as described by Kang et al. 48 with slight modification. Diluted FSR (0.4%, v/v) was used in the analysis. A 10-μl aliquot of FSR was mixed with 0.25 ml 1 N Folin-Ciocalteu's reagent in a 5-ml test tube. The mixture was covered and kept still for 2 min in the dark before 0.5 ml 12% (w/v) aqueous solution of Na 2 CO 3 and 1.24 ml distilled water were added. The mixture was incubated for 1 h at room temperature and then ultraviolet absorbance was measured at 765 nm using a UV-5500PC (METASH) against a control. The control sample contained the same amount of chemicals with the FSR replaced by distilled water. Gallic acid (GA) was used as a standard for preparing a calibration curve. Total phenolic contents were expressed as mg GA equivalents per FSR.
Determination of total flavonoid contents. Total flavonoid contents of FSR were measured using the aluminum chloride colorimetric assay 49 with slight modification. Diluted FSR (concentration at 1.6%, v/v) was used in the analysis. A 40-μl aliquot of FSR was added into a 5-ml test tube containing 1.31 ml of distilled water and 75 μl of 5% (w/v) NaNO 2 was then added. After 5 min, 75 μl 10% (w/v) AlCl 3 was added and allowed to react for 6 min before 1 ml of 4% (w/v) NaOH dissolved in distilled water was added. The solution was mixed and kept for 12 min at room temperature before ultraviolet absorbance was measured against the control at 510 nm using a UV-5500PC (METASH). Control samples contained the same amount of chemicals except the FSR was replaced with distilled water. Rutin was used as a standard for constructing a calibration curve. Total flavonoid contents were expressed as mg rutin equivalents per of FSR.
Free radical scavenging activity. The free radical scavenging activity of FSR was examined in vitro using DPPH (2, 2-diphenyl-1 picrylhydrazyl) radical as described by El Atki et al. 49 . Different concentrations of FSR or GA were added to ethanolic solution and mixed with DPPH dissolved in ethanolic solution so that the final concentration of DPPH was 0.1 mmol. The absorbance of the mixture was measured using a UV-5500PC (METASH) at 517 nm after 30 min of incubation at room temperature in darkness. A control mixture in which FSR was replaced by the equivalent amount of ethanolic solution was prepared following the same procedures. The percentage of inhibition was calculated using the following equation: %DPPH scavenging activity = 1 − OD sample /OD control × 100 www.nature.com/scientificreports/ Here, OD control is the absorbance of the negative control and OD sample is the absorbance of the sample. GA served as positive control. IC 50 values were calculated as the concentration of causing a 50% inhibition of DPPH radical.
Tyrosinase inhibition activity. Tyrosinase inhibitory activity was examined in vitro as described by Elena et al. 50 . Different concentrations of FSR or kojic acid (KA) were mixed with tyrosinase dissolved in 0.2 M phosphate buffer (pH 6.8) and kept still for 15 min at 37 °C. Next, 12.5 mM L-Dopa dissolved in 0.2 M phosphate buffer (pH 6.8) was added, so that the final concentration of tyrosinase was 25 U/ml and that of L-Dopa was 1.25 mM in the solution. Absorbance at 475 nm was measured using a UV-5500PC (METASH) after incubation for 5 min at room temperature. FSR was replaced by the equivalent amount of phosphate buffer in the control mixture prepared following the same procedures. The percentage of inhibition was calculated using the following equation: OD control is the absorbance of the negative control and OD sample is the absorbance of the sample. KA served as a positive control. IC 50 values were calculated as the concentration of causing a 50% inhibition of tyrosinase.
Statistical analysis. SPSS 22 (Statistical Product and Service Solutions, https:// www. ibm. com/ produ cts/ spss-stati stics) was used for statistical analysis. Probit regression analysis was used to analyze the IC 50 of DPPH free radical and tyrosinase inhibition. Pearson correlation analysis was used to determine the correlation of total phenolic and flavonoid contents with DPPH free radical scavenging activity and tyrosinase inhibition activity. Differences between means were determined using the least significant difference test at P < 0.05, and figures were drawn using OriginPro 2017 (https:// www. origi nlab. com).
Ethics approval and consent to participate. The Authors confirm that no animal/human studies have been carried out in the present. This study was part of a wider project entitled "Study on traditional fermented rose in Dali Bai Nationality". We conducted this research in accordance with International Society of Ethnobiology (2006), ISE Code of Ethics (with 2008 additions), and the protocol was approved by Kunming Institute of Botany, Chinese Academy of Sciences (KIB) ethics committee (Supporting documents S1) and Center of Biodiversity and Indigenous Knowledge(CBIK) ethics committee (Supporting documents S2). Before data collection, we described the goals of this research to local informants and asked them to sign a Free andInformed Consent Term. We were authorized to collect plant specimens by Forestry and grassland Bureau of Dali Bai Autonomous Prefecture.
|
2021-11-24T06:18:13.909Z
|
2021-11-22T00:00:00.000
|
{
"year": 2021,
"sha1": "e47f1f08304c0b323fd94331cdb5b5365f33e998",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s41598-021-02160-y.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "df8530fee83a5d4b9d9daf76df20b6a508bbf83b",
"s2fieldsofstudy": [
"Agricultural and Food Sciences",
"Environmental Science"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
5971178
|
pes2o/s2orc
|
v3-fos-license
|
Recombinant Production of Horseradish Peroxidase Conjugates with Fab Antibodies in Pichia pastoris for Analytical Applications.
Recombinant immunoconjugates of marker enzymes with antigens or antibodies present considerably more advantages than those obtained by conventional methods of chemical synthesis; i.e. they are homogeneous, have a strictly determined stoichiometry, and retain the functional activity of both a marker protein and an antigen/antibody. Based on the pPICZαB shuttle vector, we first managed to obtain a recombinant conjugate of key marker enzyme horseradish peroxidase (HRP) withFabfragments of antibodies against atrazine. The resulting genetic construction allows us to switch to any other antibody sequence, via the simple re-cloning of variable parts and an additional reporter enzyme. Conjugates were successfully produced in thePichia pastorismethylotrophic yeast expression system. The target activity of the conjugates (both enzymatic and antigen-binding) has been demonstrated by ELISA method.
INTRODUCTION
Enzyme immunoassays for the detection and quantitative analysis of various substances are based on coupling of marker enzymes such as horseradish peroxidase (HRP, EC 1.11.1.7]) with antigens or antibodies. However, all major approaches used for the chemical conjugation of proteins and haptens result in the partial inactivation of the enzyme and conjugate heterogeneity, which affects the specificity and sensitivity of the ELISA. Genetic engineering can be used to obtain recombinant conjugates of proteins with antigens or antibodies. Such conjugates present a number of advantages; firstly, they have a homogenous composition, secondly, they possess 1 : 1 stoichiometry, thirdly they retain the functional activities of both the marker protein and that of the antigen/ antibody, in addition to the reproducibility and the fact that they are relatively simple to produce. Recombinant conjugates of antibodies with alkaline phosphatase [1][2][3], luciferase [4], and peroxidase Arthromyces ramosus [5] were obtained earlier.
The recombinant conjugate of HRP with the earlier obtained fatty-acid-binding protein (FABP) [6] was expressed in Escherichia coli cells and used as an immunotracer when performing immunoenzyme assay aimed at the early diagnosis of myocardial infarction.
The functional expression of the recombinant conjugate of HRP and antibody fragments in E. coli is associated with a number of difficulties, since there is no post-translational glycosylation of proteins in E. coli cells, resulting in low solubility and aggregation of the expressed/obtained protein. This problem can be solved by replacing the expression system. For instance, it has been shown that methylotrophic yeast Pichia pastoris is a more suitable organism/system for antibody expression than E. coli cells [7,8].
HRP [9] and antibody fragments [10] were successfully expressed individually in P. pastoris cells, both in the single-stranded form scFv [11,12] and in a Fab form [13]. Moreover, certain immunoconjugates have also been created using this expression system [14][15][16]. It has been demonstrated that gene expression in the P. pastoris system in the secreted form considerably simplifies the scaling of the process for biochemical applications [17].
The recent advance in the functional expression of HRP and antibodies in secreted form paves the way for the construction of recombinant HRP-antibody conjugates to be used in immunoassays. Firstly, we obtained recombinant conjugates of HRP and Fab-fragments of antibodies against atrazine, in order to study the opportunities provided by this approach. In these chimeric proteins, the peroxidase part is combined with the N-and C-terminal parts of the heavy chain of an antibody via a short linker sequence. The universal vectors for the expression of conjugates of HRP and variable chains of Fab fragments of antibodies were obtained (a simple replacement of the variable part of a heavy and light chain of any other antibody by re-cloning at the PstI/BstEII and BamHI/XhoI sites, respectively) in the secreted form in P. pastoris cells. A functionally active HRP-Fab (atrazine) conjugate was obtained, possessing antigen-binding properties that are similar to those of monoclonal antibodies, which has been attested by single-stage competitive immunoassay of atrazine (IC 50 ~ 3 ng/ml).
Reagents
The reagents were purchased from the companies Sigma, Fluka, and Difco and used without further purification. Protein electrophoresis (SDS-PAGE) was performed according to the standard procedure, using a low molecular weight protein kit (LMW, Bio-Rad) as the molecular weight standards. The preparative work with DNA was performed using a QIA prep Spin Miniprep Kit and a QIAquick Gel extraction Kit (Qiagen, Germany). Enzymes for DNA restriction and modification were purchased from New England Biolabs, Boehringer-Mannheim, GIBCO-BRL-Life technologies, and MBI. Oligonucleotides for sequencing and PCR were purchased from ARK Scientific, MWG Biotech, or Interactiva (Germany).
Data processing and presentation
The gene engineering part of the study was planned using CloneManager software (Scientific & Educational Software, Cary, United States). The spatial structures of immunoconjugates were simulated and visualized on the InsightII (BioSym Inc., United States) software package (BioSym Inc., United States) on an SGI R4400 operating station. The experimental data were prepared for publication using software from the OpenOf-fice.org (www.openoffice.org) and GIMP (GNU Image Manipulation Program) packages.
Preparing competent cells. E. coli cells were grown overnight in 50 ml of the LB medium until OD 600 was 0.4-0.6 and were isolated from the culture medium by centrifugation (3500 rpm, 4 о С) for 10 min. The cell precipitate/pellet was re-suspended in a TSS buffer (buf--suspended in a TSS buffer (buf-suspended in a TSS buffer (buffer based on a LBS medium containing 10 g of PEG-6000, 5 ml of DMSO, and 0.6 g of MgCl 2 in 100 ml; рН 6.5) and then kept on ice for 1 h, aliquoted (200 µl), and quickly frozen at -80°С.
Recombinant antibodies and their conjugates with HRP were expressed using P. pastoris X33 (Invitrogen) and shuttle vector pPICZαB (Invitrogen) for cloning.
The NotI site was removed using forward and reverse primers (Table). A three-stage PCR was used (primers listed in the Table), in order to incorporate the HRP gene behind the gene of the heavy antibody chain and to remove the restriction sites BspCI, ApaI, PstI, BstEII, BglII, XhoI, BamHI, SacI, and PvuI.
DNA modification and cell transformation
Manipulations with DNA included the standard procedures [18]. E. coli cells were transformed via the addition of plasmids or a ligation mixture to the unfrozen competent cells. P. pastoris cells were also transformed by plasmids preliminarily linearized at the PmeI site via electroporation. P. pastoris cultivation and secretion of the recombinant conjugate P. pastoris cells were cultivated in a YPD medium (1% yeast extract, 2% Peptone, 2% D-glucose). The target protein was synthesized in the glucose-free YP medium, using 0.5 vol % methanol as an inducing agent. The YPDS medium (YPD containing 1 M sorbitol) was used for transformation of P. pastoris cells. The solid medium contained 1.5% of Bacto Agar. The transformants were grown in the YPDS medium at 30°С under stirring (200 rpm) until OD 600 = 15 units was obtained. The cells were centrifuged at 3,000 g and 4°С, washed with YP medium, and OD 600 was brought to 1. The induction was performed for 96 h by adding 0.5 vol % metha-nol every 24 h. The supernatant was concentrated via membrane ultrafiltration (Amicon, 10 kDa).
Determination of activity of the recombinant conjugate by ELISA ELISA was performed overnight using plates ("NUNC" MAXI-SORP) with the preliminarily sorbed BSA-atrazine (1 : 100 dilution) conjugate or BSA (10 µg/ml) in a 10 mM carbonate buffer, pH 9.0, at 4°С. The samples of supernatant of a P. pastoris culture medium were successively diluted in PBS, added into the plate wells, and incubated at 37°С for 1 h. Then, the plate was washed thrice with PBS containing 0.1% of Tween-20 (PBS-T), and 50 µl of a TMB substrate mixture was added (0.6 mg/ml TMB and 8 mM H 2 O 2 in 0.1 М acetate buffer, pH 5). The reaction was stopped by adding 50 µl of 2 M H 2 SO 4 ; the optical density was measured at 450 nm.
Competitive ELISA for determining atrazine 150 µl of the calibration sample (0.1, 1.0, 10, 20, 50, 100, 500 ng/ml of atrazine in PBS-T) and 40 µl of a recombinant conjugate solution were placed into plate wells with the preliminarily sorbed BSA-atrazine conjugate and incubated at 37°С for 1 h. The plate was washed three times with PBS-T, and 50 µl of the TMB substrate mixture was added into each well. The reaction was stopped by adding 50 µl of 2 M H 2 SO 4 ; the optical density was measured at 450 nm.
RESULTS AND DISCUSSION
Recombinant conjugates are chimeric proteins combining the structural components of both marker enzymes and an antigen/antibody. The use of modern approaches has almost solved the problem of obtaining such recombinant enzymes as alkaline phosphatase, β-galactosidase, luciferase, and horseradish peroxidase that are used as markers in the ELISA methods. However, the production of recombinant conjugates is an appreciably complicated task, since it remains thus far impossible to reliably predict the structure of the desired conjugate; hence, loss of the functional activity of both the marker enzyme and antigen is possible, due to the incorrect folding of two components.
Recombinant conjugates comprising bacterial enzymes (β-galactosidase and alkaline phosphatase) that can be easily expressed in soluble form in E. coli cells, as well as some other enzymes, were earlier obtained. The major problem associated with the use of β-galactosidase and alkaline phosphatase within conjugates is their tetrameric and dimeric structures, respectively, which results in a considerable increase in conjugate affinity in comparison with a free antibody. This phenomenon is particularly undesirable when designing ELISA competitive schemes. Meanwhile, horseradish peroxidase, one of the marker enzymes that have been most widely used in ELISA, is expressed in E. coli cells only in the form of inclusion bodies, which until recently has impeded the obtaining of an active enzyme.
The recent advance in the heterologous expression of antibody genes in the cells of methylotrophic yeast P. pastoris offers great opportunities for using this system for the synthesis of the conjugates of horseradish conjugates with antibodies in the secreted soluble and functionally active forms.
Designing the expression vector to obtain a recombinant conjugate of HRP with a Fab fragment of antibodies in P. pastoris The expression system for obtaining recombinant conjugates of HRP and Fab fragments of antibodies was elaborated on the basis of the pPICZαB vector. The genetic construction was placed under the control of the AOX promoter containing the PmeI site for the subsequent linearization and recombination into the yeast genome. The vector also contains the signal sequence of α-factor, which is necessary for the directed secretion of recombinant protein into the culture medium. Gene sh ble provides zeocin resistance of both E. coli and P. pastoris cell types. The possibility of introducing a hexahistidine sequence at the C-terminus of the recombinant protein is provided to simplify the procedures of extraction and purification of the product.
We used the plasmids earlier obtained (pPIC-VCL and pPIC-VCH [19] containing the corresponding fragments of variable regions of the light and heavy chains of K4E7 monoclonal antibody against atrazine [20], respectively) as the starting material (Fig. 1). Both these vectors contained the NotI site behind the cloned gene.
To design the universal construction, we planned to leave only one NotI site in the vector behind the gene of the antibody heavy chain. PCR (the QuickChange mode [21]) using a special primer pair (Table) was employed to remove the NotI site from the pPIC-VCL plasmid. The vector obtained by this procedure is known as pPIC-VCL dNot. Then, the BglII/BamHI fragment of the pPIC-VCH plasmid containing the heavy-chain gene was cloned at the BglII site of plasmid pPIC-VCL dNot prior to the light chain gene. The expression vector pPIC-Fab was used. The cloning scheme is given in Fig. 1.
The universal pPIC-Fab gene obtained contains SacI/XhoI and PstI/BstEII site pairs for simple cloning of the genes of the heavy and light chains, encodes the C-terminal hexahistidine fragment for simplifying the purification of the target protein using metal chelate chromatography, and the NotI site for cloning the marker protein (such as HRP, green fluorescent protein (EGFP), luciferase, etc.) at the C-terminus of the heavy chain of the antibody.
A vector for the expression of the recombinant conjugate of peroxidase with Fab fragments of antibodies was designed simultaneously. For the simplicity of cloning, restriction sites PstI, BstEII, BglII, XhoI, SacI, PvuI, ApaI, BamHI and BspCI were removed using the primers listed earlier (Table) from the initial HRP gene [22] that was preliminarily cloned in the corresponding pPIC vector. Either before the HRP gene or behind it the fragments of antibody genes were simultaneously cloned. Thus, three-stage PCR was used to obtain two genetic constructions in which the HRP gene was linked with the sequence encoding the N-terminal region of the variable part of the heavy Fab chain or the C-terminal region of the constant part of the heavy chain via a short linker sequence (Gly 4 Ser) 3 (Fig. 2). It should be mentioned that in order to avoid the formation of nonfunctional dimers of the light chain, genetic constructions were cloned in the pPIC-Fab vector at the sites PmeI/BstEII and BstEII/NotI, respectively; the heavy chain was selected for the cloning of the marker protein gene. The mutual arrangement of genes in plasmids pPIC-Ab-HRP and pPIC-HRP-Ab was confirmed by restriction analysis and sequencing.
Expression and purification of recombinant conjugates Fab-HRP and HRP-Fab P. pastoris X33 cells (Invitrogen) were transformed via the plasmid vectors pPIC-Ab-HRP and pPIC-HRP-Ab using electroporation with an efficiency of approximately 100 clones per 10 µg of plasmid DNA. The expression of the target protein was tracked by increasing the peroxidase activity in the supernatant of the culture medium, reaching a plateau on day 5 of cultivation. Ten clones were analyzed with each construction. The HRP activity with respect to the TMB substrate was detected only in three clones out of 20: two clones (1.1, 1.2) corresponding to pPIC-Ab-HRP and one clone (8) corresponding to pPIC-HRP-Ab. These clones were selected for further consideration. As shown by SDS-PAGE electrophoresis (the data are not provided), the blurred bands located below the band at 100 kDa correspond to the recombinant conjugates HRP-VCH and VCH-HRP. This blurriness of the bands is accounted for by the microheterogeneity of conjugates, conditioned by excessive glycosylation that is typical for P. pastoris; correlating with our data and the data published on the expression of the HRP gene [9]. A considerably greater excess (by a factor of 3-4) of light-chain molecules was observed (the band at 25 kDa) in comparison with that found during the expression of the Fab fragment [19]. Unexpectedly, it occurred that the recombinant conjugated manifested no enzymatic activity toward another peroxidase substrate, ABTS, as opposed to TMB. It is well known that the site of binding with ABTS is located in the hydrophobic region on the HRP surface, in the so-called "Phe patch" zone [25]. This zone is noticeably distant from the active site of HRP, and it can be assumed that the substrate binding to it is complicated due to steric reasons -as a result of excessive glycosylation or the presence of a Fab fragment of antibody. The first hypothesis is more probable, since the same effect is observed under both positions of the heavy chain of the antibody with respect to HRP. Moreover, a similar effect was earlier observed upon expression of the HRP gene in P. pastoris (the data have not been published).
The total yield of recombinant conjugates was approximately 3-10 mg per 1 l of the P. pastoris culture supernatant. A relatively low yield of secreted conjugates correlates with the yield upon expression of the HRP gene only. We believe that one of the factors that have a negative effect on the yield of the secreted product is the excessive glycosylation of the peroxidase component of the conjugate, which is typical of P. pastoris cells. In order to verify this hypothesis, it may be reasonable to remove all N-glycosylation sites in HRP or replace HRP with another reporter protein, such as EGFP.
Characterization of recombinant conjugates by ELISA
In order to confirm the antigen-binding activity of recombinant conjugates, we selected the scheme of indirect competitive single-stage ELISA (Fig. 3) carried out on the wells with an immobilized atrazine-BSA conjugate. The binding of recombinant conjugates to atrazine was preliminarily studied (Fig. 4). The data obtained attest to the presence of both catalytic and antibody 2) have similar characteristics, and specimen 1.1 was used for further ELISA determination of atrazine. The typical calibration diagram (Fig. 5) allows one to determine the atrazine concentration over a wide range, from 0.1 to 50 ng/ml; the variation coef-ficient being no higher than 8%. IC 50 is equal to 3 ng/ml, which agrees well with the results of atrazine determination by a two-stage ELISA procedure using recombinant Fab fragments of the same antibody K411B [19] and with the data on the single-chain mini-antibody (scFv) obtained earlier in E. coli [24]. Meanwhile, in the initial monoclonal antibody, the IC 50 value was equal to 0.2 ng/ml [19]. As is evident in the majority of similar cases, the fact that the IC 50 value differs from that of recombinant antibodies is in all likelihood connected with the bivalence of the initial monoclonal antibody. Thus, the recombinant conjugates of peroxidase with Fab fragments of antibody against atrazine obtained in the present study possess functional activity and can be used to determine atrazine via ELISA.
CONCLUSIONS
The possibility of using a recombinant, functionally active HRP (as a marker enzyme) conjugated with Fab fragments of the antibody against atrazine was shown for the first time. In the present study, recombinant conjugates were obtained in which the Fab fragment of an antibody is bound both to the N-and the C-terminuses of the marker enzyme. Both these variants manifest immunological and catalytic activity.
The functional secretion of recombinant conjugates of HRP with Fab fragments of antibodies offers opportunities for broad application in ELISA. The results obtained will be used to design highly sensitive immunobiosensors of a new generation, based on the recombinant DNA technology.
The study was partially supported by the German Federal Ministry of Science and Technology (BMBF, grant № 0311574).
|
2017-03-31T13:20:03.740Z
|
2011-07-01T00:00:00.000
|
{
"year": 2011,
"sha1": "aff34198d5f4da3bd07a520ebed62a0971005a21",
"oa_license": "CCBY",
"oa_url": "http://pdfs.semanticscholar.org/44d9/3a8dfbdc5d38ef8e3ce30540bcb2c86046dc.pdf",
"oa_status": "GREEN",
"pdf_src": "PubMedCentral",
"pdf_hash": "e9932215ef278cc5d20253d57b56079c2a3921a6",
"s2fieldsofstudy": [
"Chemistry",
"Engineering"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
248725966
|
pes2o/s2orc
|
v3-fos-license
|
Comparison of different treatment strategies in the management of endogenic caesarean scar pregnancy: a multicentre retrospective study
Background The aim of this study was to evaluate the effectiveness and safety of different treatment strategies for endogenic caesarean scar pregnancy (CSP) patients. Methods According to Vial’s standard, we defined endogenic-type CSP as (1) the gestational sac growing towards the uterine cavity and (2) a greater than 0.3 cm thickness of myometrial tissue at the caesarean scar. A total of 447 endogenic CSP patients out of 527 patients from 4 medical centres in China were enrolled in this study. A total of 120 patients were treated with methotrexate (MTX) followed by surgery, 106 received ultrasound-guided curettage directly and 221 received curettage combined with hysteroscopy. The clinical information and clinical outcomes of these patients were reviewed. Successful treatment was defined as (1) no additional treatment needed, (2) no retained mass of conception and (3) serum β subunit of human chorionic gonadotropin (β-hCG) level returning to a normal level within 4 weeks. The success rate was analysed based on these factors. Result Among 447 patients, no significant difference was observed in baseline characteristics between groups except for foetal heartbeat. The success rate was significantly different (p<0.001) among the three groups. The highest success rate of 95.9% was noted in the hysteroscopy group, and the lowest success rate of 84.0% was noted in the curettage group. In addition, the MTX group reported the longest hospital stay and highest expenses, but the curettage group showed the shortest and lowest expenses, respectively. Nevertheless, no difference in blood loss was observed between the groups. Conclusion The combination of curettage and hysteroscopy represents the most effective strategy. Pretreatment with MTX did not result in better clinical outcomes. Ultrasound-guided curettage directly should not be considered a first-line treatment choice for endogenic CSP patients. Supplementary Information The online version contains supplementary material available at 10.1186/s12884-022-04633-y.
CSP is commonly accomplished by ultrasound [1]. The diagnosis is based on the presence of a gestational sac at the site of the previous uterine incision and an empty uterine cavity and cervix as well as thin myometrium adjacent to the bladder [3].
In 2000, Vial proposed the classification of CSP into 2 types based the position of the gestational sac: endogenic (type 1) and exogenic (type 2) CSP [4]. The first type is due to the implantation of the amniotic sac on a scar with progression of the pregnancy in the cervicoisthmic space and the uterine cavity, whereas the second is a deep implantation in a caesarean scar defect with progression towards rupture and bleeding. In either type, CSP could lead to severe haemorrhage, uterine rupture and hysterectomy [1,3]; therefore, pregnancy termination should be considered after diagnosis. A multitude of treatment modalities have been proposed for the management of CSP, including expectance, medicine, uterine artery embolization (UAE), surgery and combination [3]. However, the best approach in terms of patient safety and clinical effectiveness has yet to be determined due to their low prevalence.
Medical treatment is one of the most commonly used treatment methods. Among all medicines, methotrexate (MTX) is most often utilized given its history of use in ectopic pregnancy. It is a folic acid antagonist that inhibits the enzyme dihydrofolate reductase, thereby interfering with DNA synthesis in rapidly dividing cells, such as trophoblasts [5]. However, the use of MTX as the firstline approach for CSP is often associated with a low success rate (50-60%) [6][7][8][9], and additional interventions, such as surgery, are often required. As the most convenient and inexpensive method of surgery, curettage is always the first choice after MTX for many doctors. Several studies provided data demonstrating that MTX followed by curettage might represent the best treatment for CSP [10,11]. On the contrary, studies also provided evidence that MTX combined with curettage resulted in more blood loss, a long duration of hospitalization and a normal β-human chorionic gonadotropin concentration [12].
Doctors previously held the notion that direct surgery without medical pretreatment would lead to massive uncontrolled bleeding. Some studies compared several different strategies in CSP treatment and reported that surgery is also a good choice for CSP for patients with a long gestational age, a large gestational sac diameter, high β-hCG levels, or an ample blood supply [13]. Recently, hysteroscopy has been extensively developed as a minimally invasive surgery and has gradually been considered by many doctors [14,15]. Given that endogenic CSP has a lower risk of bleeding, direct hysteroscopic surgery is becoming a first-line treatment for many gynecologists.
In this study, we aimed to evaluate three different strategies in the management of endogenic CSP, including MTX followed by surgery, ultrasound-guided curettage and curettage plus hysteroscopy without MTX. For the evaluation, we introduced the success rate, blood loss during surgery, hospital time and expenses to reflect the availability and security of treatments.
Patients
This retrospective study analysed 527 endogenic CSP patients treated at Qilu Hospital, Jinan Central Hospital, Taian City Central Hospital and Liaocheng People's Hospital, Shandong, China from August 2013 to May 2020. Data were collected through medical records and short-term follow-ups. The study was approved by all the hospitals' ethics committees. Informed consent from the patients was exempt because of the retrospective nature of the study.
The diagnosis of CSP relies on clinical presentation and sonographic signs. A woman with a positive pregnancy test and prior caesarean delivery should undergo an early transvaginal sonographic (TVS) assessment. The ultrasound criteria include the following: (1) an empty uterus and cervical canal, (2) a gestational sac at the caesarean scar site, (3) a vascular area noted at the previous caesarean scar in Doppler scan, and (4) thin or absent myometrial tissue between the bladder and the gestational sac [16]. According to Vial's standard and our clinical experience, we defined the endogenic type as (1) the gestational sac growing towards the uterine cavity and (2) the thickness of myometrial tissue at the caesarean scar was no less than 0.3 cm given that a thinner myometrium means a greater possibility of deep implantation. Summarizing the above information, gynaecologists made the final diagnosis of endogenic CSP. This study only included patients in the first trimester because those in the second trimester require a more detailed assessment of placental invasion and other complications. Patients who were unsuccessfully treated and then transferred to these hospitals were also removed. We also excluded patients with other complications, including ovarian or fallopian tube cysts and uterine myomas. Finally, a total of 447 endogenic CSP patients were enrolled in this study.
Treatments
The criteria of selecting treatment strategies for CSP patients without life-threatening situations in these departments plurally combined symptoms, Ultrasonic findings, laboratory indicators and patients' wishes. Patients could have several choices including expectant management, medical treatment, uterine artery embolization (UAE) and different kinds of surgeries. In this study, we included patients treated with MTX followed by surgery, which was defined as the MTX group, and surgery directly. The surgery included ultrasound-guided curettage only (defined as the curettage group) and curettage combined with hysteroscopy (defined as the hysteroscopy group). MTX was administered intramuscularly at a dose of 50 mg/m 2 body surface area or interventionally through the uterine artery at a total dose of 50 mg. Then, the surgery was performed after a conspicuous decline in the serum β subunit of human chorionic gonadotropin (β-hCG), which typically occurred within 1 week after MTX treatment. However, surgery was performed as soon as patients were diagnosed in the curettage and hysteroscopy group. Curettage and hysteroscopy were performed by well-trained gynaecological surgeons. Before any surgery, 6 units of pituitrin diluted in 20 mL saline were injected into 3 and 9 points of the cervix. In both curettage group and hysteroscopy group, curettage was performed under ultrasound monitoring with a negative pressure of 400-500 mmHg. After suction, it was necessary to check the villi and embryonic tissue in the container. In the hysteroscopy group, hysteroscopy was performed by experienced gynecologists after curettage to examine whether residual pregnancy tissue was present in the uterine scar site. Hysteroscopy was performed under general anaesthesia with the woman in dorsolithotomy position. The main surgical procedures are similar to that of Wang et al. [17]. Uterine distension was achieved using 5% mannitol solution with a pressure <100 mmHg. The intervention began by an overview of the uterine cavity. The endometrial cavity was empty and a diverticulum could be seen at the scar of the anterior wall of the uterus. The residual pregnancy tissue was excised by wire loop electrode in an 80 W monopole current and blood vessels were coagulated in a 100 W coagulation current.
Data collection
We defined a successful treatment as (1) no additional treatment needed, (2) no retained mass of conception and (3) serum β-hCG level returning to a normal level within 4 weeks. The success rate is equal to the number of patients who are successfully treated over the total number of patients under a certain treatment. The clinical characteristics of the patients, including age (year), number of previous caesarean sections, years from last caesarean section, gestational age (day), foetal heartbeat activity, vaginal bleeding, abdominal pain, myometrial thickness of uterine scar (mm) and gestational sac mass diameter (cm), were collected. The gestational sac/mass diameter was defined as the largest length of the sac or mass. Serum β-hCG levels (mIU/ ml) before and after treatment, time of surgery (min), blood transfusion and blood loss during surgery (ml) were collected to evaluate the safety and availability of surgery. Blood loss during surgery was assessed by the surgeons according to blood volume in the container and sterile pads. Postoperative serum β-hCG value was measured on the second day after surgery. Beyond that, weekly measuring of serum β-hCG levels and TVS were required until serum β-hCG levels returned to normal levels and no mass was found in the uterine cavity. Here, β-hCG differences between treatments, rate of β-hCG decline, time of hospital stay and hospitalization expenses were also collected to evaluate the availability of different treatments. The β-hCG difference is calculated based on the serum β-hCG level before a specific treatment minus that after treatment. The rate of β-hCG decline was equal to the β-hCG difference over the initial value before treatment. Hospitalization expenses include expenditures for medicine, surgery and nurse care during hospital stays.
Statistical analysis
Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS, Chicago, USA) version 25.0 software. Categorical variables are presented as numbers and percentages, and between-group differences were assessed by the chi-squared test. Normality tests were performed for all quantitative variables, and none of the variables exhibited a normal distribution. Therefore, all quantitative data are presented as the median (interquartile range) and were assessed by nonparametric tests. A P value less than 0.05 was considered indicative of a statistically significant difference.
Baseline characteristics among patients in the three groups
A total of 447 endogenic CSP patients were enrolled in this study (see Additional file 1). Among these patients, 120 (26.85%) patients were treated with MTX followed by surgery, 106 (32.42%) were treated with ultrasoundguided curettage, and 221 (67.58%) were treated with curettage combined with hysteroscopy. The baseline characteristics of these patients in the three groups are described in Table 1. Foetal heartbeat significantly differed among the three groups. There were no significant differences with respect to age, number of previous caesarean sections, years from last caesarean section, vaginal bleeding, abdominal pain, gestational age, gestational sac/mass diameter, myometrial thickness of uterine scar, or serum β-hCG and haemoglobin (HGB) levels before surgery.
Clinical outcomes for the three groups
Characteristics reflecting the availability and security of three different strategies are compared in Table 2. According to our definition, the success rates of the three groups are 85.8, 84.0 and 95.9%. In addition, success treatment, surgery time, hospital time, hospitalization expenses, β-hCG after surgery, and rate of β-hCG decline (P < 0.05) were significantly different among these groups. No significant differences were observed with regard to blood loss during
Complications
Among the 17 patients in the MTX group who were not successfully treated, one patient underwent a curettage operation, and the remaining patients received medical treatment, including repeated MTX injection and the administration of a contraction-promoting drug (such as misoprostol), until serum β-hCG levels returned to normal. All 17 patients in the curettage group who required greater than 4 weeks to return to normal β-hCG levels received mifepristone or a contraction-promoting drug. Among the 9 patients in the hysteroscopy group who were defined as unsuccessfully treated, one patient underwent a repeated hysteroscopy operation, and the remainder received medicine. No cases of uterine rupture, bladder injury, hysterectomy and haemorrhagic shock were noted in these patients.
Discussion
CSP was first reported by Larsen and Solomon and refers to a special type of ectopic pregnancy in which embryos are implanted on the uterine scar [18]. Given the significant increase in the percentage of caesarean deliveries, CSP diagnosis has increased and represents a challenge for contemporary obstetrics [19]. The high risk of death and serious complications leads to heated discussion about the management strategies for CSP. However, no internationally recognized diagnosis and treatment program has been published thus far. In this study, the clinical information of 447 patients was collected, and a significant difference was observed between the three groups. Among the three different management options, the highest success rate of 95.9% was found in the curettage plus hysteroscopy group followed by 85.8% in the MTX group and 84.0% in the ultrasound guided curettage group. These results were similar to previous studies [13,20]. However, previous studies were limited by their small number of cases. The surgery time in the hysteroscopy group was relatively longer. A more careful operation may be considered in the surgery group, and easer surgical options were chosen after MTX pretreatment given that doctors used to believe that MTX pretreatment would reduce the amount of bleeding. However, the likelihood of MTX completely inactivating trophoblast cells is very low. Under this premise, the choice of a non-visible or uncertain surgery method after MTX treatment was not optimal. This traditional concept may be responsible for the lowest success rate in the MTX group. According to our results, the β-hCG level between different treatments showed no difference, but the rate of β-hCG decline in the MTX group was significantly greater than that in both surgery groups. This result indicates that the β-hCG value in the MTX group declined more despite the much longer hospital stay. Due to the time needed by doctors to observe the patient between MTX treatment and surgery, the MTX group showed significantly the longest hospital time and the highest expenses. Similarly, previous study even reported the long duration of hospitalization of 19.38 ± 8.75 days in MTX pretreatment group [12]. This result demonstrates that that MTX treatment was not highly efficient, although a greater reduction in β-hCG levels was observed. The curettage group demonstrated the shortest hospital stay, lowest expenses and a relatively greater reduction in β-hCG levels. However, this is not an optimal strategy given that it exhibited the lowest success rate. Interestingly, the surgery group, especially the hysteroscopy group, reported a significantly higher rate of foetal heartbeat but no difference in blood loss, indicating that foetal heartbeat presence may unwittingly affect the choice of different treatment options but actually does not influence the amount of blood loss or the safety of treatments. Considering that no study has supported foetal heartbeat as an independent risk factor so far, more clinical research was urgently need to explain the connection between foetal heartbeat and treatment options.
To the best of our knowledge, this is the first study concentrating on endogenic CSP patients and providing detailed success rates for three different strategies. According to our results, patients pretreated with MTX showed a lower success rate, longer hospital stay, and higher expenses but proportionable blood loss compared with those in the hysteroscopy group. This result demonstrates that MTX pretreatment did not result in better outcomes. Moreover, endogenic CSP patients treated with ultrasound-guided curettage only showed the lowest success rate of 84.0%, which was similar to previous studies in CSP patients [20][21][22]. Therefore, we concluded that curettage combined with hysteroscopy tends to be the most effective strategy, but curettage alone should not be considered a first-line treatment choice for endogenic CSP patients. In addition, past work on CSP was invariably limited by the small sample size. The limited number of cases is associated with a greater possibility of selection bias and a worse generalization ability of the results. In this study, a large number of patients treated at four medical centres were analysed, which strengthened our conclusion from a statistical point of view.
There were also several limitations in this study. First, given its retrospective nature, detailed information was occasionally not available. For example, the blood loss in our study was assessed mainly by surgeons, so it was simply not evaluated as accurately as expected, which might lead to our result indicating no significant difference between groups. In addition, due to the absence of a standard surgical procedure for CSP, it is possible that the skill and experience of different surgeons affect our conclusion. These findings also support the urgent need for an optimal treatment consensus and gynaecologists specializing in CSP.
Above all, our study provided an accurate success rate of different treatment strategies for endogenic CSP patients by analysing the clinical outcomes of 447 cases. Our results demonstrated that curettage followed by hysteroscopy could be a recommended safe strategy with possible advantages. However, further studies including a large series with comprehensive information, detailed follow-up data and multiple patient sources are needed to support this finding. Medical departments specializing in CSP have been established at our institutions, and detailed treatment as well as integrated follow-up data of CSP patients are being collected. Thus, more convincing studies and prospective research will be provided in the future.
Conclusions
Caesarean scar pregnancy is one of the most severe complications of caesarean delivery with no internationally recognized treatment program. Endogenic Caesarean scar pregnancy could also lead to severe haemorrhage, uterine rupture and hysterectomy. Thus, the security and availability of surgery should be fully considered for endogenic caesarean scar pregnancy patients. Analysing clinical data from 447 endogenic CSP patient, we came to the conclusion that the combination of curettage and hysteroscopy represents the most effective strategy, pretreatment with MTX did not result in better clinical outcomes.
|
2022-05-13T13:50:13.013Z
|
2022-05-12T00:00:00.000
|
{
"year": 2022,
"sha1": "8d747cb9c25be3006f0f3dd84d948656f9289441",
"oa_license": null,
"oa_url": null,
"oa_status": null,
"pdf_src": "Springer",
"pdf_hash": "8d747cb9c25be3006f0f3dd84d948656f9289441",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
269634264
|
pes2o/s2orc
|
v3-fos-license
|
Discovery of a brain penetrant small molecule antagonist targeting LPA1 receptors to reduce neuroinflammation and promote remyelination in multiple sclerosis
Multiple sclerosis (MS) is a chronic neurological disease characterized by inflammatory demyelination that disrupts neuronal transmission resulting in neurodegeneration progressive disability. While current treatments focus on immunosuppression to limit inflammation and further myelin loss, no approved therapies effectively promote remyelination to mitigate the progressive disability associated with chronic demyelination. Lysophosphatidic acid (LPA) is a pro-inflammatory lipid that is upregulated in MS patient plasma and cerebrospinal fluid (CSF). LPA activates the LPA1 receptor, resulting in elevated CNS cytokine and chemokine levels, infiltration of immune cells, and microglial/astrocyte activation. This results in a neuroinflammatory response leading to demyelination and suppressed remyelination. A medicinal chemistry effort identified PIPE-791, an oral, brain-penetrant, LPA1 antagonist. PIPE-791 was characterized in vitro and in vivo and was found to be a potent, selective LPA1 antagonist with slow receptor off-rate kinetics. In vitro, PIPE-791 induced OPC differentiation and promoted remyelination following a demyelinating insult. PIPE-791 further mitigated the macrophage-mediated inhibition of OPC differentiation and inhibited microglial and fibroblast activation. In vivo, the compound readily crossed the blood–brain barrier and blocked LPA1 in the CNS after oral dosing. Direct dosing of PIPE-791 in vivo increased oligodendrocyte number, and in the mouse experimental autoimmune encephalomyelitis (EAE) model of MS, we observed that PIPE-791 promoted myelination, reduced neuroinflammation, and restored visual evoked potential latencies (VEP). These findings support targeting LPA1 for remyelination and encourage development of PIPE-791 for treating MS patients with advantages not seen with current immunosuppressive disease modifying therapies.
www.nature.com/scientificreports/mice show significantly improved clinical scores in the myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalitis (MOG-EAE) model 6 .LPA1 blockade also alleviates inflammation and fibrosis, both of which are contributing factors to MS [7][8][9] .These data combine to suggest that an LPA1 antagonist may enable a multipronged approach towards disease treatment, and that the identification of a potent, selective, brain penetrant LPA1 antagonist could make a significant impact on the treatment of inflammatory demyelinating diseases such as MS.
We have identified and profiled PIPE-791, a potent, small molecule, brain-penetrant, LPA1 antagonist.Using a combination of in vitro and in vivo models, we show that PIPE-791 induces OPC differentiation into oligodendrocytes and demonstrates efficacy in the MOG-EAE mouse model of MS.We also show that LPA1 may impact other MS-related mechanisms beyond OPC differentiation, including inflammation and fibrosis.Careful pharmacokinetic characterization of PIPE-791 also revealed a slow binding kinetic that resulted in unusually long CNS receptor occupancy.Together, these results encourage the further development of PIPE-791 as a treatment for multiple sclerosis.
PIPE-791 is a potent, small molecule LPA1 antagonist
Though several LPA1 antagonists have been optimized for peripheral indications, none have demonstrated adequate brain penetration for CNS studies or indications.As such, we developed PIPE-791, a brain-penetrant, orally bioavailable, small molecule LPA1 antagonist.
We first tested PIPE-791 in a competitive membrane filter binding assay using membranes from cells overexpressing human LPA1.Several concentrations of PIPE-791 were co-incubated with these membranes in the presence of an LPA1 selective radioligand and then filtered 10 .We found that PIPE-791 bound human LPA1 with high affinity with a calculated K i of 0.752 nM (IC 50 2.63nM, Table 1).
Next, we examined the kinetics of PIPE-791 binding to human LPA1 receptor (hLPA1) in a recombinant membrane setting.We used a forward kinetic method where 1 µM PIPE-791 was mixed with 0.25, 0.5, or 1 nM of [ 3 H]-PIPE-791.Binding was initiated by adding hLPA1 overexpressing membranes at different time points (ranging from 1 min to 24 h) and filtered.We found that PIPE-791 exhibited slow kinetics with a calculated t 1/2 of 8.65 h (Fig. 1A, Table 2).
The binding affinity of PIPE-791 to the LPA1 receptor was also assessed in a native tissue setting by performing saturation binding experiments in mouse and human brain homogenates.Using [ 3 H]-PIPE-791, saturation binding in human brain tissue revealed a K d of 0.68 nM and a B max of 462 dpm (disintegrations per minute, Fig. 1B).In mouse brain tissue, we observed a K d of 1.7 nM and a B max of 1649 dpm (Fig. 1C).The similarity between these two species suggests that findings in the mouse brain should be translatable to human.
To assess selectivity, PIPE-791 was tested in a functional calcium mobilization assay in cells overexpressing hLPA1 and using either a 30 min or 24 h pre-incubation period prior to LPA addition (Table 1).The slow on-rate kinetics of PIPE-791 likely contribute to the potency shift observed between the 30 min and the 24 h calcium mobilization assay.PIPE-791 also showed selectivity against the two most homologous LPA receptor isoforms, LPA2 (35.4-fold) and LPA3 (35-fold) (Table 1).We also evaluated PIPE-791 in a Eurofins off-target panel (SAFETYscan E/IC50 ELECT) and observed no appreciable activities at a test concentration of 30 µM (Supplementary Fig. 1).
In vivo pharmacokinetics and binding kinetics
Using in vitro methods, we have shown that PIPE-791 exhibits unusually slow association kinetics.We wanted to see if this property could be recapitulated in vivo, particularly in the context of plasma concentration and brain LPA1 receptor occupancy across time.
In the first experiment, mice were dosed intravenously with [ 3 H]-PIPE-791 and brains harvested at different time points.In agreement with the kinetics observed in vitro, receptor binding of [ 3 H]-PIPE-791 in the brain gradually increased over time, plateauing between 24 and 72 h.Receptor bound radioligand levels then declined by 7 days, demonstrating that binding was reversible.Specific binding was determined at the 2 and 24 h timepoints by competition with unlabeled PIPE-791 that was administered orally at a dose of 3 mg/kg (Supplementary Fig. 2).
The in vivo receptor occupancy of PIPE-791 was characterized using a novel, selective, and brain-penetrant LPA1 radioligand ([ 3 H]-OPC3497) (Supplementary Fig. 3).This radioligand displays relatively fast on and off kinetics making it more suitable for in vivo occupancy studies.Following a single oral dose, PIPE-791 inhibited [ 3 H]-OPC3497 binding to mouse brain in a dose-dependent manner with an ED 50 of 0.17 mg/kg and 0.19 mg/ kg when administered 2 h or 24 h prior to radioligand, respectively (Fig. 2A).The unbound brain EC 50 (C u, brain ) was 2.2 nM (Fig. 2B).When PIPE-791 was administered once daily for 4 consecutive days to reach steady state, the LPA1 receptor occupancy ED 50 decreased to 0.03 mg/kg while the unbound brain EC 50 remained relatively unchanged at 2.4 nM.The shift in ED 50 following repeat dosing is consistent with the slow receptor binding kinetics associated with PIPE-791 (Fig. 1A).
To further explore the pharmacokinetic/pharmacodynamic (PK/PD) relationship, mice were dosed orally with 3 mg/kg of PIPE-791 and its plasma concentration and the corresponding brain receptor occupancy were evaluated out to 7 days.LPA1 occupancy was sustained for 24 h and slowly decreased out to 7 days.Plasma concentrations, on the other hand, peaked early (2 h) and decreased rapidly by 24 h and were below the limit of detection after 3 days.This disconnect in the PK/PD relationship is again consistent with the slow receptor off-rate observed with PIPE-791 (Fig. 2C).
LPA1 is expressed in oligodendrocyte precursor cells
Having identified and characterized PIPE-791 as an LPA1 selective antagonist in both in vitro and in vivo contexts, we sought to better understand LPA1 expression CNS, particularly in the context of multiple sclerosis.An earlier report had described the expression of LPA1 in OPCs 11 .We have since independently confirmed these data by examining Lpar1 expression in OPCs enriched using the OPC-specific surface marker O4.O4 + OPCs were isolated from rat cortex and LPA1-5 receptors were assayed by quantitative PCR.In addition to confirming the presence of Lpar1, we observed that Lpar2-5 mRNA expression was significantly lower in OPCs (Fig. 3A).
We then examined Lpar1 expression as OPCs differentiated into oligodendrocytes.Using platelet-derived growth factor (PDGF) withdrawal to induce OPC differentiation, we examined Lpar1 expression at several timepoints 12 .Upon PDGF removal, we observed an increase in the oligodendrocyte marker, myelin basic protein (Mbp) and a concurrent decrease in the OPC marker Pdgfra, thereby confirming differentiation.Interestingly, over the course of differentiation, we also observed an increase in Lpar1 expression, consistent with literature and databases showing that mature oligodendrocytes also express Lpar1 (Fig. 3B).
Given the expression of Lpar1 on OPCs, we wanted to know how direct, exogenous application of LPA would affect OPC differentiation.Using the same PDGF withdrawal paradigm to induce OPC differentiation, we added various concentrations of LPA to the culture media and quantified MBP + oligodendrocytes 3 days later.In doing so, we observed a dose dependent inhibition of OPC differentiation in response to LPA treatment (Fig. 3C) thereby suggesting LPA receptor activation by LPA as a negative regulator of this process.Although these results alone do not directly implicate LPA1, cultured OPCs from LPA1 knockout mice also show enhanced OPC differentiation compared to wildtype controls 5 .
Macrophage-secreted factors suppress OPC differentiation
During MS, inflammatory macrophages invade the CNS, destroying the myelin sheath and releasing factors that prevent subsequent myelin repair 13 .While macrophage-oligodendrocyte co-culture systems have been used to show that macrophages phagocytose oligodendrocytes, such cultures have not been used to study the secretion of diffusible factors [14][15][16] .
To better understand the impact of macrophage-secreted factors on OPCs, we utilized a Transwell culture system to spatially separate macrophages and OPCs but still allow for factor exchange.Rat peritoneal macrophages were plated in the bottom compartment while rat OPCs were plated on a Transwell insert above the macrophages.Upon PDGF withdrawal, we observed an overall suppression of OPC differentiation; specifically, a decrease in MBP + oligodendrocytes and an increase in PDGFRα + OPCs, when OPCs were cultured in the presence of macrophages (Fig. 3D).
Autotaxin (ATX), an enzyme responsible for LPA synthesis, is elevated in CSF samples from multiple sclerosis patients and may be an underlying source of elevated LPA levels in the CNS during disease.By www.nature.com/scientificreports/immunocytochemistry, we confirmed that the CD68 + macrophages express autotaxin (ATX), suggesting that LPA may be one of possibly several secreted, diffusible factors involved in suppressing OPC differentiation (Fig. 3E) 6,17 .
To confirm the release of LPA from these macrophages, macrophage-conditioned media was taken 30 min or 48 h after plating and various LPA species were measured by mass spectrometry 18 .We observed a significant increase in several LPA species over time.Most notably, 18:1 LPA which has been implicated in LPA-mediated disorders such as neuropathic pain, showed the most significant increase over time 19 .Importantly, the increase in LPA was also inhibited in the presence of the autotaxin inhibitor, PF-8380 (1 µM, Fig. 3E).Together, these data build on previous observations that in addition to destroying oligodendrocytes, macrophages may also limit myelin repair by suppressing OPC differentiation through the release of diffusible factors such as LPA.
Autotaxin expressing cells are found near lesions in MS patient brain tissue
To see whether LPA may be involved in MS, we obtained human MS tissue and examined autotaxin expression around the lesion border.Fresh frozen MS tissue was stained with Sudan Black to demarcate demyelinated regions 20 .The tissue was immunostained for ATX and HLA-DR (a marker for inflammatory microglia and macrophages, 21 ).At the lesion, HLA-DR + cells were abundant near the lesion border.Importantly, there was substantial autotaxin colocalization with HLA-DR + cells suggesting that cells such as microglia or macrophages may act during multiple sclerosis to increase local levels of LPA at the lesion sites.(Fig. 3F and Supplementary Fig. 4).
PIPE-791 induces OPC differentiation in dissociated culture and can overcome inhibition by macrophage-released factors
Having characterized PIPE-791 as a potent antagonist of LPA1, we determined whether the treatment of OPCs with PIPE-791 could induce their differentiation.Primary rat OPCs were isolated and cultured in the presence of PDGF.One day later, PDGF was removed and various doses of PIPE-791 were added to the cultures.These cultures were fixed 3 days post-PIPE-791 treatment and immunostained for MBP.We observed a dose dependent increase in the number of differentiated oligodendrocytes with an EC 50 of 108 nM (Fig. 4A).
We next tested whether PIPE-791 induced oligodendrocytes could functionally myelinate axons using a rat primary cortical myelination assay described previously 22 .In this assay, cells from embryonic day 15 rat cortex were plated.Over time, axons in the culture are myelinated by oligodendrocytes and myelin segments can be measured.We observed robust myelination following PIPE-791 treatment, with an EC 50 of 2.6 nM (Fig. 4B).AM152 (BMS-986020), a clinical stage, peripherally-restricted LPA1 antagonist, was used as a reference control.
We then revisited the macrophage-OPC Transwell co-culture to assess whether PIPE-791 could derepress differentiation inhibition by macrophage-secreted factors.Following PDGF withdrawal, we observed a 41% decrease in the number of MBP + oligodendrocytes when co-cultured with macrophages.In the presence of macrophages, addition of PIPE-791 resulted in a 2.1-fold increase in differentiation versus vehicle treated cultures (Fig. 4C).
PIPE-791 induces OPC differentiation in vivo
To evaluate whether PIPE-791 could increase oligodendrocyte number in vivo, mice were dosed once with 3 mg/ kg PIPE-791 and 5 days later, brains collected, and tissue sectioned.Sections were immunostained for the mature oligodendrocyte marker, CC1, and the oligodendroglial marker, OLIG2 to represent the total OPC/oligodendrocyte population.In doing so, we observed a significant 163% increase in CC1 + /OLIG2 + oligodendrocytes suggesting OPC differentiation (21).
PIPE-791 induces OPC differentiation in human slice culture
PIPE-791 potently antagonizes the human LPA1 receptor and triggers the differentiation of mouse and rat OPCs into functional oligodendrocytes.To assess whether PIPE-791 could similarly induce OPC differentiation in a human context, we evaluated oligodendrocyte markers in human cortical slice cultures after treatment.Cortical slices from regions containing both white and gray matter were generated from fresh human adult donor tissue and cultured for 10 days.PIPE-791 was then added to the culture for 9 days.Analyzing Mbp by qPCR showed a dose-dependent increase with an EC 50 of 4.2 nM (Fig. 5A).
We also used an immunohistochemical endpoint and quantified the number of CC1 + cells present after PIPE-791 treatment.At 300 nM, we observed significant differentiation following PIPE-791 treatment.The magnitude of the effect was also comparable to 1 µM AM152.In light, these data suggest that PIPE-791 has the potential to induce OPC differentiation in a human setting (Fig. 5B-D).
PIPE-791 inhibits microglial activation but does not inhibit phagocytosis
LPA1 has been implicated in the activation of microglia in several disease models, including MOG-EAE 6,24,25 .In light, we wanted to test whether treating microglia with PIPE-791 could inhibit their activation.
Hippocampal slices from postnatal day 20 mice were generated and transferred into basal media containing PIPE-791 or AM152 for 2 h.Microglial activation was then triggered by the addition of 10 µM LPA, slices incubated for 3 additional hours, then fixed.Slices were immunostained against IBA1 and microglial perimeters were measured as a mean of differentiating between ramified/resting versus rounded/activated microglia.After LPA treatment, we saw a significant increase in the number of rounded IBA1 + cells as measured by a decrease in cell perimeter length, indicative of activation 26 .This decrease in perimeter length was inhibited in slices that were treated with LPA in the presence of PIPE-791 or AM152 (Fig. 6A,B and Supplementary Fig. 6).
To see if PIPE-791 impacted the ability of microglia to phagocytose myelin debris, microglia were isolated and cultured in an activated state using complete serum media which inherently contains high levels of LPA 27 .At this point, microglia already displayed an activated, amoeboid morphology.Microglia were then treated with various concentrations of PIPE-791.Fluoromyelin-labeled myelin debris was added for 90 min and then the
PIPE-791 reduces meningeal fibroblast activation
Recent work from Dorrier et al., has implicated persistent fibrotic scarring in the CNS in the mouse EAE model and that the contributing cells were of meningeal origin.As LPA and LPA1 have been implicated in the induction and progression of fibrosis, we wanted to see whether PIPE-791 could inhibit the activation of brain meningeal fibroblasts 8,9 .We cultured primary rat or human brain meningeal fibroblasts and applied 10 µM LPA to induce activation.Upon LPA stimulation, we saw a change in morphology that could be quantified as a change in collagen I (COL1 + ) area.Preincubation with PIPE-791 inhibited the increase in COL1 + with an IC 50 of 31.8 nM in rat and 4.5 nM in human meningeal fibroblasts (Supplementary Fig. 8).www.nature.com/scientificreports/
PIPE-791 protects oligodendrocytes from cytokine insult
TNFα and IFNγ are two cytokines that, while not normally expressed in the CNS, are elevated in MS 29 .These cytokines can be secreted by both macrophages and microglia 30,31 .Receptors for these cytokines (TNFR1, TNFR2, and IFNGR) are expressed in oligodendrocytes.Of note, TNFR1 and 2 have been found on oligodendrocytes bordering MS lesions 23,32,33 .Further, although receptor expression for these cytokines on OPCs is unknown, addition of either of these cytokines impact differentiation in isolated OPC cultures 28,34 .Cultured rat OPCs were differentiated with T3 for 24 h then treated with both TNFα and IFNγ Upon cytokine addition, we observed significant cell death consistent with previous observations (Fig. 6C, left 22,[35][36][37] ).To test whether PIPE-791 could afford protection after cytokine insult, cells were treated with TNFα and IFNγ along with various concentrations of compound.In the presence of PIPE-791, we observed dose-dependent protection of MBP + oligodendrocytes with an EC 50 of 125 nM (Fig. 6D, right).Concurrent assessment of total cellular viability with Alamar blue yielded an EC 50 of 39.6 nM (data not shown).While the mechanisms underlying this observation remain unknown, the results do suggest PIPE-791 promotes oligodendrocyte survival in addition to differentiation in the context of an inflammatory microenvironment.
PIPE-791 induces oligodendrocytes in lysolecithin treated mouse organotypic slice cultures
We next used an ex vivo organotypic brain slice culture to assess remyelination following a demyelinating insult.Mouse cortical brain slices were treated with lysolecithin, which induces acute demyelination through a non-specific lipid-based mechanism 38 .Lysolecithin was removed 18 h later and replaced with media containing PIPE-791.After 3 additional days, the brain slices were processed for Mbp.We observed a decrease in Mbp following lysolecithin insult in vehicle treated slices and a dose-dependent increase in Mbp in slices treated with PIPE-791, EC 50 of 12.1 nM (Fig. 7A).Slices were also evaluated using a histological endpoint.Here, following demyelination, slices were incubated in PIPE-791 for 5 days, then fixed and processed for immunostaining against MBP.Caspr clustering was used as a marker for nodes of Ranvier formation which implicates the presence of functional axons that have responded to myelination 39,40 .We quantified MBP + area and observed a dose dependent increase following PIPE-791 treatment www.nature.com/scientificreports/with an EC 50 of 74 nM (Fig. 7B).We also observed an increase in the number of Caspr puncta with an EC 50 of 17.9 nM (Fig. 7C).
Altogether, these data suggest that following a demyelinating insult, PIPE-791 promotes OPC differentiation and induces the formation of functional, myelinating oligodendrocytes in an organotypic slice culture.
PIPE-791 efficacious in MOG-EAE
We next tested PIPE-791 in the MOG-EAE (myelin oligodendrocyte glycoprotein-induced experimental autoimmune encephalitis) mouse model.This model recapitulates several of the features thought to occur in MS and is a widely used model of inflammatory demyelination.
During the EAE model, body weights provide a simple, real-time indication of disease severity.In the vehicledosed MOG group, the mean body weight decreased from Day 11 until Day 19 at which time mean body weights were relatively stable until study termination.Although MOG-EAE mice treated with PIPE-791 at 0.3 or 3 mg/ kg experienced some weight loss, such loss was significantly mitigated relative to vehicle (Fig. 8A).The mean cumulative body weight of the vehicle group over the duration of the experiment was significantly lower than that of the non-MOG control group and this effect was mitigated in the PIPE-791-treated groups (Fig. 8A).
Clinical disability was also measured as an indicant of disease severity.In the vehicle group, onset occurred at day 13, reaching a mean peak score of approximately 3.2, then decreased to a mean score of 2.5 at the end of the study (Fig. 8B).The mean cumulative disease index (CDI) in the vehicle group was 27.4 by study end.In contrast, when dosed with PIPE-791 at 0.3 or 3 mg/kg, clinical disability onset was delayed, and scores were reduced.Specifically, in mice dosed with 3 mg/kg PIPE-791, disease onset was delayed to day 14 and peak clinical scores only reached 1.4.At study end, scores had reached 1.5.Efficacy was also seen at 0.3 mg/kg, again with delayed onset (to day 14), a peak score of 2.1, and a score of 1.5 by day 24 (Fig. 8B).When clinical scores from day 1 through 24 were summed to generate a cumulative disease index, significant improvement was seen at 3 mg/kg PIPE-791 with trending improvement at 0.3 mg/kg (p = 0.0575, Fig. 8C).
Clinical score is largely influenced by demyelination in the lumbar spinal cord; hence g-ratios (the ratio of the inner to outer diameter of a myelinated axon) were quantified in this area.In control, non-EAE mice, 90.9% of the axons were myelinated (g-ratio < 1).In vehicle treated MOG-EAE mice, only 73.4% of axons were myelinated.When MOG-EAE mice were treated with 3 mg/kg PIPE-791, we observed significant improvement with 84.6% of the axons myelinated.Significant improvement was also seen at 0.3 mg/kg with 83.3% of the axons myelinated.
Remyelinated axons canonically have thin myelin 41 .As such, we examined axons with g-ratios ≥ 0.8 but less than 1 as a surrogate measure of a remyelinating axon.Analysis of these thinly myelinated axons suggest that there was significant remyelination after 3 or 30 mg/kg PIPE-791.Specifically, where MOG-immunized mice showed only 23.6% axons with a 1 > g-ratio ≥ 0.8, with PIPE-791 treatment, this percentage was significantly higher, namely 32.3% at 0.3 mg/kg, and 31.9% at 3 mg/kg.Non-MOG control animals showed 31.2% (Fig. 8D,E).
To further assess myelin status, we measured conductance velocity in the optic nerve, a heavily myelinated tract that is sensitive to demyelination.To measure conductance, we utilized visual evoked potentials (VEPs), a measure for which clinical translatability makes it an especially salient endpoint 42,43 .VEPs were measured on day 21.Non-EAE control mice exhibited an N1 latency of 56.4 ms.In comparison, latency in MOG immunized animals had slowed to 60.9 ms.In mice treated with PIPE-791, significantly shorter latencies were observed, namely 52 ms in the 3 mg/kg, and 54.6 ms in the 0.3 mg/kg groups.Since conduction velocity is driven largely by myelination, these data suggest that more axons were myelinated in the PIPE-791 treated groups (Fig. 8C).To confirm this, g-ratios in the optic nerve were similarly quantified.Like spinal cord, a loss in myelinated axons was observed in MOG-immunized mice.In non-MOG-EAE mice, 86.4% of the axons were myelinated.MOG-EAE resulted in a reduction to only 68.7% axons being myelinated.With 3 mg/kg PIPE-791, 86.1% of the axons were myelinated.At 0.3 mg/kg, 85.5% were myelinated.When thinly myelinated axons were quantified, 30.2% in the MOG-treated vehicle group were observed, compared to 50.7% in the non-MOG control.Upon treatment with 0.3 mg/kg PIPE-791, this increased to 44.7% and with 3 mg/kg, 50.6% axons were thinly/ remyelinated (Fig. 8F,G).
Because our in vitro data suggested that PIPE-791 might inhibit microglial activation, we further performed immunostaining of the spinal cord with an IBA1 antibody.We observed a significant increase in the number of activated microglia in MOG-EAE mice compared to non-MOG-EAE.In mice treated with PIPE-791 (3 mg/kg), the level of IBA + staining was significantly reduced (Supplementary Fig. 9).
Discussion
Multiple sclerosis is an inflammatory disease that causes profound demyelination, resulting in deficits in signal transduction and subsequent neurodegeneration.Due to its complex etiology, a multifaceted approach to MS, either by way of combination therapies or by the identification of single targets that act on multiple pathways, would be advantageous.With its involvement in OPC differentiation, inflammation, and fibrosis, antagonizing LPA1 presents such a possibility for addressing several of the mechanisms that drive MS.While LPA concentrations are not high in the healthy brain, in diseased contexts such as MS, there is evidence of elevated LPA levels in the CNS, likely due to blood infiltration from the periphery or through increased production in the CNS.In support, we observed the presence of autotaxin expressing HLA-DR + cells proximal to MS lesions.As macrophage incursion is a key feature of MS, we focused our attention on these cells, specifically asking whether macrophagederived secreted factors could impact OPC differentiation.We show that macrophages secrete diffusible factors (one of which may be LPA) that inhibit OPC differentiation.Direct application of LPA also inhibited OPC differentiation into oligodendrocytes.
We have identified a small molecule antagonist against LPA1, PIPE-791.PIPE-791 is a selective, brainpenetrant, orally bioavailable, small molecule LPA1 antagonist.Due to its binding kinetics, PIPE-791 persists in the brain long after clearance from the plasma.Such extended CNS coverage poses several advantages such as reduced dosing frequency.It is worthwhile highlighting that while most small molecule ligands against GPCRs typically have half-lives on the order of minutes 44,45 , tiotropium (Spiriva) is another rare example of a small molecule GPCR antagonist with slow kinetics, having a t 1/2 of 24.5 h on the muscarinic M3 acetylcholine receptor 46 .
We have shown that PIPE-791 promotes OPC differentiation into oligodendrocytes and that these oligodendrocytes are capable of wrapping axons in dissociated culture.Mechanistically, LPA signals via LPA1 to a number of pathways including Rho kinase 47 .While further investigation into the signaling mechanisms surrounding LPA1-mediated OPC differentiation are needed, previous work from Pedraza et al. 48has shown that inhibition of the Rho pathway with fasudil, an inhibitor of ROCK (Rho-associated kinase) and a known LPA1 effector, leads to OPC differentiation and may be thus be a possible mechanism 48 .
In vivo, direct dosing of PIPE-791 in naïve mice induced an increase in CC1 + oligodendrocytes relative to vehicle.This suggests that a single dose of PIPE-791 can induce OPC differentiation.It also implies that a baseline level of LPA is present in the brain.Given the dearth of reports on LPA in the naïve brain, further investigation around this observation is warranted.We also showed that PIPE-791 increases the number of CC1 + oligodendrocytes in an adult human brain slice culture.
In acutely demyelinated mouse brain slice cultures, following PIPE-791 treatment, we observed an increase in Mbp RNA, MBP protein, as well as an increase in the node of Ranvier marker, Caspr.PIPE-791 also reduces both microglial and fibroblast activation in vitro.Next, using the MOG-EAE model of inflammatory demyelination, we showed that PIPE-791 dose-dependently improved functional recovery as measured by both body weight loss and clinical disability score.Furthermore, both 0.3 and 3 mg/kg doses of PIPE-791 resulted in significant improvement in optic nerve myelination as assessed by VEP (a highly sensitive, clinically translatable endpoint).Importantly, we saw significant restoration of myelin at both 0.3 and 3 mg/kg PIPE-791 treatment groups as measured by electron microscopic analysis of g-ratios in both the spinal cord and the optic nerve.
Several LPA1 antagonists such as BMS-986278 or BMS-986020 have demonstrated Phase 1 tolerability and have either progressed to Phase 2 (BMS-986278) or have completed a Phase 2 study (BMS-986020) 49,50 .Current small molecule antagonists are generally intended for peripheral indications such as idiopathic pulmonary fibrosis; thus little is known about LPA1 inhibition in the human CNS.Observations in rodent models may provide possible insight, and although LPA1 null and conditional knockout mice do exist, knockouts with the temporal kinetics of a small molecule, e.g., an inducible knockout, have not been generated.As such, the observations made in constitutive knockouts may not reflect antagonism with a small molecule.Contos et al., observed that LPA1 knockout mice had 50% neonatal lethality and was likely due to developmental abnormalities in the olfactory bulb or cortex 51 .Further, in addition to cells of oligodendroglial lineage, LPA1 is also expressed in neural progenitor cells, particularly in the subventricular zones of mice.Interestingly, while LPA1 labels neural progenitor cells in the mouse, LPA1 is not co-expressed with markers for human neural progenitors.Specifically, LPA1 had been considered a marker for the Gra.Neu5 cluster.Upon closer examination, however, it appeared that the cluster consisted of oligodendrocytes and granule cells and LPA1 had been unintentionally misassigned to the granule cell population 52,53 .Thus, while mouse data implicates a role for LPA1 in neurodevelopment, given the constitutive nature of a knockout and the discrepancies between mouse and human, it remains to be seen how a brain-penetrant, small molecule LPA1 antagonist will behave in human.
In conclusion, these data describe the characterization of PIPE-791, an LPA1 antagonist with favorable CNS drug-like properties.These data also confirm a role for LPA1 in OPC differentiation and remyelination both in vitro and in vivo.Further, we show that antagonizing LPA1 with PIPE-791 may address other mechanisms involved in the manifestation of MS, such as neuroinflammation.Together, these data support LPA1 as a target for remyelination and pave the way for an oral, "first-in-class" LPA1 antagonist for the treatment of MS and other diseases associated with myelin loss.
Animal care and tissue collection
Mice and rats were multi-housed in Innorack IVC mouse racks (Innovive, San Diego, CA) with access to water and standard rodent chow ad libitum.Animals were subject to a 12h light-12h dark cycle.For tissue collection, rodents were euthanized by exposure to CO 2 followed by cervical dislocation or decapitation unless otherwise specified.All experiments were conducted in accordance with procedures as approved by the Contineum
Statistics
Unless otherwise noted, statistics were performed using GraphPad Prism software (San Diego, CA) using the specific analyses described in text and legends.
OPC differentiation assay
OPC isolation: For immunopanning purification of OPCs, cortical hemispheres were dissected from postnatal day 8 Sprague-Dawley rat pups.Tissue culture dishes were incubated overnight with goat IgG secondary antibodies to mouse (Jackson Laboratories, Bar Harbor, ME).Dishes were rinsed and incubated at room temperature with primary antibodies for Ran-2, GalC and O4 (gift from Jonah Chan lab).Rodent cortical hemispheres were minced and dissociated with papain (Thermo Fisher, Waltham, MA) for 1 h at 37°.After dissociation, cells were resuspended in a panning buffer (0.2% BSA in DPBS) and incubated at room temperature sequentially on three immunopanning dishes: Ran-2 and GalC were used for negative selection before positive selection with O4.OPC Assay: OPCs were plated in poly-l-lysine (Millipore Sigma, St. Louis, MO) coated 96-well Viewplates™ (Perkin Elmer, Boston, MA) at a seeding density of 35 × 10 5 cells per well.Cells were maintained in media composed of DMEM (Invitrogen, Carlsbad, CA) supplemented with B27 (Invitrogen, Carlsbad, CA), N2 (Invitrogen, Carlsbad, CA), penicillin-streptomycin (Invitrogen, Carlsbad, CA), N-acetylcysteine (Millipore Sigma, St. Louis, MO), forskolin (Millipore Sigma, St. Louis, MO), and 25 ng/ml PDGF-AA (Peprotech, Cranbury, New Jersey) for 24 h at 37°, 5% CO 2 .Compounds were diluted 10× at a concentration range of 0.001-3 µM.After 24 h, media was gently removed from cells and the same media minus PDGF ± compound was added.Cells were returned to incubator at 37 °C, 5% CO 2 for 72 h.For immunofluorescence, OPCs were fixed with 4% paraformaldehyde and Hoechst in PBS.Well plates were blocked with 20% normal goat serum and incubated with rat monoclonal antibody to MBP (Millipore Sigma, St. Louis, MO) overnight at 4 °C, and then incubated with secondary antibodies for 1 h at room temperature.For the oligodendrocyte protection assay, 2 h after plating, a mixture of TNFα and IFNγ (both Peprotech, Cranbury, NJ) diluted to a final concentration of 1 ng/ml and 10 U/ml respectively in OPC complete media minus PDGF ± 30 ng/ml T3 or various concentrations of PIPE-791 were added.Cells were returned to incubator at 37 °C, 5% CO 2 .
OPC-macrophage Transwell cultures
OPCs were isolated as previously described.Cells were plated on the membrane layer of poly-l-lysine (Millipore Sigma, St. Louis, MO) coated Transwell culture plates (VWR, Radnor, PA) at a seeding density of 1 × 10 6 cells per well.Transwell culture plates have microporous membrane which allows for factors to flow between the top and bottom chamber.This permits us to minimize direct contact between the macrophage and OPCs while allowing for a constant, renewable source unlike conditioned media which can be depleted.Cells were maintained in media composed of DMEM (Invitrogen, Carlsbad, CA) supplemented with B27 (Invitrogen, Carlsbad, CA), N2 (Invitrogen, Carlsbad, CA), penicillin-streptomycin (Invitrogen, Carlsbad, CA), N-acetylcysteine (Millipore Sigma, St. Louis, MO), forskolin (Millipore Sigma, St. Louis, MO), and 25 ng/ml PDGF-AA (Peprotech, Cranbury, New Jersey) for 24 h at 37°, 5% CO 2 .Compounds were diluted 10X at the concentrations reported.After 24 h, media was gently removed from cells and the same media minus PDGF ± compound was added.Cells were returned to incubator at 37 °C, 5% CO 2 for 1 h.
Isolation of macrophages.Male Sprague Dawley rat macrophages were isolated from intraperitoneal lavage fluid.Using a 10 ml syringe, cold PBS −/− was injected into the intraperitoneal cavity.After Fluorescent images from cultured oligodendrocytes and macrophages were collected using a Nikon A1R confocal microscope.Images were acquired using a 20× objective, each well was read in a 5 × 5 matrix.Data were analyzed using GraphPad Prism (La Jolla, CA) and statistical significance was determined by ANOVA.
Determination of LPA concentration in culture media
Rat macrophages were cultured as described above.Media from wells at 30 m or 48 h post-plating were taken.Media from macrophages treated with 1 µM PF-8380 (autotaxin inhibitor, Sigma, St Louis, MO) for 48 h was also collected.For LPA determination, 20 µl of culture media was aliquoted into glass 12 × 75 mm culture test tubes.500 µl of 30 mM citric acid/40 mM disodium phosphate buffer (pH 4.0) with 1 ng/ml 17:0 LPA as an internal standard (IS) were also added to the test tubes.2 ml of 1-butanol were then added to the test tubes, tubes covered with parafilm, vortexed for 10 s, and centrifuged for 10 min at 1000g at 4 °C.After centrifugation, the top layers were transferred to clean test tubes.The bottom layers were extracted again with 1 ml of water-saturated butanol.
The test tubes were covered with parafilm, vortexed for 10 s, and centrifuged for 10 min at 1000g at 4 °C.After the second centrifugation step, the top layers were extracted and pooled into the glass tubes containing the first extractions.The pooled extractions were then dried under a nitrogen stream at 25 °C using a nitrogen drier.When the samples were dry, they were reconstituted with 100 µl of methanol, vortexed for 10 s, and transferred to low-volume HPLC vials with 300 µl fused inserts.After reconstitution, the samples were analyzed by LC-MS/ MS using an AB Sciex 6500 + .Chromatographic separation was conducted using a reversed-phase column (Luna Omega C18 1.6 µm, 100 × 2.1 mm).The temperatures of the autosampler and column were set to 15 and 40 °C, respectively.The injection volume was 10 µl.The LC flow rate was set to 0.2 ml/min.Gradient initial conditions were 80% mobile phase A (95:5 water:methanol, 5mM ammonium acetate, 0.1% formic acid) and 20% mobile phase B (5:95 water:methanol, 5 mM ammonium acetate, 0.1% formic acid).After one minute of initial conditions, mobile phase B increased to 85% in 1 min.Mobile phase B then increased to 100% over 7 min.Mobile phase B was held at 100% for 6 min, then decreased to 20% in 1 min.Final conditions were held for another minute before the next injection.
Liquid chromatography was coupled to an AB Sciex 6500 + mass spectrometer.The analysis was conducted under negative ionization mode with turbo ion-spray voltage at − 4500 V, turbo-ion-spray source temperature at 500 °C, and curtain gas at 30 psi.Sample analysis for the analyte, LPA 18:1, was performed using multiple reaction monitoring (MRM).The mass transition used for LPA 18:1 was 435.141 > 153.013.Declustering potential (DP), collision energy (CE), and collision cell exit potential were set at -95 V, − 30 V, and -17 V, respectively.
Recombinant membrane binding
Binding assay: Membranes were prepared from B103 cells stably expressing human LPA1 receptor.Cells were resuspended in a hypotonic buffer containing 0.25 M sucrose/0.2mM EDTA.After 30 min on ice, cells were centrifuged at 41,415 × g.Membrane pellet was resuspended in 20 mM HEPES, pH 7.4 storage buffer and centrifuged at 41,415 × g.Membranes were resuspended in storage buffer + 1 mM dithiothreitol.Protein was quantified using the DC Protein Assay (Bio-Rad, Hercules, CA), aliquoted, and stored at -80 °C until assay.A tritiated LPA1 selective antagonist 10 was used as radiolabeled ligand.3 µl/well of PIPE-791 was serially diluted in 100% DMSO at a concentration range of 0.1 nM to 10 µM; 30 µl/well of radioligand diluted in 50 mM HEPES + 100 mM NaCl, Tween-20 pH 7.4, and 267 µl/well of membrane diluted in 50 mM HEPES + 100 mM NaCl + 2 mM EDTA, pH 7.4.The plates were shaken at 300 rpm for 2 h at room temperature.GF/B plates assay filter plates (Perkin Elmer, Boston, MA, Waltham, MA) were prepared by adding 50 µl of 0.5% polyethylenimine (Millipore Sigma, St. Louis, MO) and incubated at room temperature for > 30 min.The assay was terminated by vacuum filtration (Brandel Harvester).Plates were washed 3 × with 2 ml ice cold buffer containing 50 mM HEPES + 100 mM NaCl + 2 mM EDTA, pH 7.4.Once dry, 50 µl of Betaplate Scintillation Cocktail (Perkin Elmer, Boston, MA) was added to each well.After a 20 min incubation, plates were read on the Perkin Elmer TopCount.
Kinetics assay: [ 3 H]-PIPE-791 was used as a radioligand.To a deep-well assay plate (Thermo Fisher, Waltham, MA) the following was added: 3 µl/well of either 100% DMSO or 1 µM PIPE-791 diluted in 100% DMSO and 30 µl/well of radioligand diluted in 50 mM HEPES + 100 mM NaCl, Tween-20 pH 7.4 at 0.25, 0.5 and 1 nM.At each time-point 267 µl diluted membranes were added.Fifteen total time points were taken (1 min-24 h).Each time point was assayed in quadruplicate.The plates were placed on a plate shaker at 300 rpm at room temperature.www.nature.com/scientificreports/ The assay was terminated by using vacuum filtration (Brandel Harvester) to rapidly filter plates through the PEIcoated GF/B plates.The plates were washed 3 × with 2 ml ice cold buffer containing 50 mM HEPES + 100 mM NaCl + 2 mM EDTA, pH 7.4.Once dry, 50 µl of Betaplate Scintillation Cocktail (Perkin Elmer, Boston, MA) was added to each well.After a 20-min incubation to allow the filter to saturate, plates were read on the Perkin Elmer TopCount.All data were analyzed using GraphPad Prism™ software.Dissociation binding was established by measuring the off-rate for [3H]-PIPE-791 dissociating from the receptor using the association kinetics algorithm.
Functional calcium mobilization assay
B103 cells stably expressing human LPA1 receptor were plated in 96-well black walled, clear bottom plates at 5 × 10 4 cells/well and incubated for 24 h at 37 °C/5% CO 2 .Intracellular calcium was monitored with Fluo-4 (Molecular Devices).Dye was made up in HBSS with 0.1% (w/v) BSA and 2.5 mM probenecid.Plates were read using a FlexStation 3 (Molecular Devices, Sunnyvale, CA).Calcium response was generated by the simultaneous addition of the EC 80 of LPA.24-h assay: Antagonists were added to plates prior to cell seeding at a 10× concentration in DMEM (Invitrogen, Carlsbad, CA).On the day of assay, media was removed and replaced with dye and 3× fresh compound in HBSS−/− with 0.1% BSA and incubated for 1 h.
30-min assay: Antagonists were added to plates after dye loading was complete at a 3× concentration in HBSS−/− with 0.1% BSA and incubated for 30 min.
LPA2: Evaluation of hLPA1 and 2 was conducted at Eurofins (Fremont, CA) using a 3 h preincubation in PIPE-791.Cells were challenged at EC 80 LPA.
Cortical myelination assay
Sprague-Dawley rats embryonic day 15 (Charles River) were used.All procedures were approved by the local Institutional Animal Care and Use Committee.
Generation of dissociated culture: Methods used were adapted from those previously described by 22 .Briefly, forebrains were collected from embryonic day 18 rats and finely minced with a scalpel.Tissue was digested in papain (Worthington) for 15 min, washed with 20% HBSS, then triturated with a P1000 pipet.Tissue was centrifuged (500 × g), the supernatant removed, the pellet resuspended in Neurobasal (Invitrogen, Carlsbad, CA) supplemented with N21 Max (R&D) and penicillin/streptomycin (Invitrogen, Carlsbad, CA) and plated at a density of 20,000 cells/100 µl/well onto 96-well Viewplates (Perkin Elmer, Boston, MA) coated with polyd-lysine (Millipore Sigma, St. Louis) and laminin (Millipore Sigma, St. Louis).On day 4, 100 µL of myelination media (MyM) was added.The next day, PIPE-791 was diluted to 3 × of the final concentrations in MyM and 100 µL added to each well.
Immunohistological evaluation of MBP: Cells were fixed 9 days later with 4% paraformaldehyde (EMS) for 15 min followed by three washes with DPBS (Invitrogen, Carlsbad, CA).Cells were incubated in rat MBP (Millipore) and mouse Tuj1 antibodies (BioLegend) diluted in 10% donkey serum containing 0.1% Triton X-100 (Fisher) overnight.Cells were then washed with 3xDPBS followed by 1 h incubation with secondary antibodies (anti-rat, Alexa488; anti-mouse, Alexa647, Hoechst counterstain) in blocking buffer.All secondary antibodies were produced in goat (Invitrogen, Carlsbad, CA) and used at 1:250.Images were acquired using a Nikon A1R confocal microscope and NIS-Elements software.Image analysis was performed using ImageJ.Only MBP segments that co-localized with Tuj1 axons were measured and the average myelin length per oligodendrocyte was calculated.
Mouse organotypic brain slice culture
Female CD-1 mice at postnatal day 17 (Charles River) were used for the following experiments.All procedures were approved by the local Institutional Animal Care and Use Committee.
Evaluation of Mbp transcript in mouse brain slice culture: Brains were collected from P17 mice and placed into ice cold HBSS containing 20% FBS.Brains were bisected sagittally and 250 µm coronal slices made with a manual tissue chopper.Only slices anterior to the hippocampus displaying a discrete corpus callosum were used.Approximately 12 slices are obtained per animal.Slices were laid on a 30 mm MilliCell organotypic culture insert in a 6-well culture dish containing 1.1 mL growth media (DMEM, 25% HBSS-Ca 2+ /-Mg 2+ , 25% heat inactivated horse serum, glucose (5 g/l), 25 mM ascorbic acid and penicillin/streptomycin).Slices were cultured in vitro for 72 h with a 50% media change at 24 h.After 72 h, slices were demyelinated using media containing 0.5 mg/ ml lysolecithin for 18 h.Following demyelination, lysolecithin was replaced with media containing varying concentrations of PIPE-791 and treated for 72 h.Slices were snap frozen and RNA extracted using the RNEasy Mini kit (Qiagen).Reverse transcription was performed using QScript (QuantaBio) followed by qPCR with Perfecta PCR SuperMix (QuantaBio) on a StepOne Plus Thermocycler (ABI).Primers used were as follows: 18S rRNA F: GTC TGT GAT GCC CTT AGA TG R: AGC TTA TGA CCC GCA CTT AC; Mbp: CTA TAA ATC GGC TCA CAA GG R: AGG CGG TTA TAT TAA GAA G. Cycling parameters: 95 °C 30 s, then repeat 45×: 95 °C for 5 s, 60 °C for 15 s, followed by melt curve analysis.Data was calculated using 2 −ΔΔCt , normalizing to 18S RNA and vehicle.
Human organotypic slice culture
Human donor brain was received 14 h after death on wet ice.Upon receipt, the brain was immediately placed into-ice cold DPBS without calcium or magnesium.Cortical regions containing gray-white border were isolated then sliced on a McIlwain tissue chopper set at 400 µm thickness.Slices were laid on a 30 mm MilliCell organotypic culture insert in a 6-well culture dish containing plating media (DMEM (Invitrogen, Carlsbad, CA), B-27 (Invitrogen, Carlsbad, CA), 25 mM ascorbic acid (Millipore Sigma, St. Louis), sodium pyruvate (Millipore Sigma, St. Louis), Glutamax (Invitrogen, Carlsbad, CA), and penicillin/streptomycin (Invitrogen, Carlsbad, CA), 1 mM HEPES (Millipore Sigma, St. Louis).Slices were maintained in plating media for a minimum of 1 h, then replaced with culture media.Half of the media was replaced every other day for 10 days.At day 10, culture media was replaced with culture media containing vehicle, PIPE-791 (at 0.3, 3, 30, and 300 nM) or AM152 (at 300 nM) and cultured an additional 9 days, replacing half of the media (containing compound) every other day.Slices were then processed for qPCR or immunohistochemistry.
Evaluation of Mbp transcript in human brain slice culture: After compound treatment, slices were immediately transferred to lysis buffer and RNA extracted using the RNEasy Mini kit (Qiagen).Reverse transcription was performed using QScript (QuantaBio) followed by qPCR with Perfecta PCR SuperMix (QuantaBio) on a StepOne Plus Thermocycler (ABI).Primers used were as follows: 18S rRNA F: GTC TGT GAT GCC CTT AGA TG R: AGC TTA TGA CCC GCA CTT AC; Mbp: CTA TAA ATC GGC TCA CAA GG R: AGG CGG TTA TAT TAA GAA G. Cycling parameters: 95 °C 30 s, then repeat 45×: 95 °C for 5 s, 60 °C for 15 s, followed by melt curve analysis.Data was calculated using 2 −ΔΔCt , normalizing to 18S RNA and vehicle.
Immunohistological evaluation of oligodendrocytes in human brain slice culture: After compound treatment, slices were fixed in 4% PFA overnight.Slices were gently lifted from the membrane insert and washed with at least five 15 min washes in DPBS containing 0.5% Triton-X 100 (0.5%-DPBS) followed by overnight incubation in primary antibodies in blocking solution.Antibodies used were CC-1 (mouse, 1:250, Millipore, Temecula, CA), Olig2 (rabbit 1:250 CellMarque).Slices were washed with at least five 15 min washes in 0.5%-DPBS followed by incubation with secondary antibodies (anti-mouse Alexa488; anti-rabbit, Alexa568, and Hoechst counterstain) in 0.5%-DPBS.All secondary antibodies were produced in goat (Invitrogen, Carlsbad, CA, Carlsbad, CA) and used at 1:250.Filters were excised from insert, mounted with Fluoromount (Millipore Sigma, St. Louis) on a microscope slide and coverslipped.2 images per slice were acquired using a Nikon A1R confocal microscope and NIS-Elements software.Images were thresholded and counted using ImageJ.
Microglial activation and phagocytosis assays
Brains were harvested from mice (postnatal day 20, Charles River).Hippocampi were dissected out and immediately placed into ice cold HBSS containing either 0.1% DMSO or 3 µM PIPE-791 (dissection media).400 µm thick coronal slices were cut on a McIlwain tissue chopper and returned to dissection media and separated manually.Slices were then transferred into culture media containing DMEM + 0.01% fatty acid free BSA (Gibco) with 3 µM PIPE-791 or DMSO.Slices were incubated for 2 h at 37 °C (4 slices per condition).Microglial activation was initiated by adding LPA (18:1, Avanti Polar Lipids) to a final concentration of 10 µM and incubated for 3 h.Slices were then fixed in formalin and for 2 h.
For phagocytosis, microglia were harvested from cortical cultures generated from P2 rats.Cortices were isolated, meninges removed, minced and digested in 0.05% trypsin (Invitrogen, Carlsbad, CA, Carlsbad, CA) for 15 min at 37 °C.Tissue was allowed to settle, trypsin removed and HBSS (Corning, Glendale, AZ) with 20% FBS (20% HBSS) was added to halt further digestion.Tissue was washed again in 20% HBSS and resuspended in DMEM with 10% FBS and penicillin/streptomycin (10% DMEM).Cells were plated in a T150 flask and cultured for 10 days with media changes every other day.Microglia were mechanically isolated by tapping flask against the hood surface.Microglia were transferred to a conical and centrifuged at 800 × g for 15 min and resuspended in 10% DMEM, plated in 96 wells at a density of 5000/well and cultured for 3 days.Culture media was replaced with fresh 10% DMEM containing PIPE-791 and incubated for 6 h.Fluoromyelin labeled myelin debris was added to each well and incubated for 90 min.Microglia were then fixed and immunostained for IBA1 (Invitrogen, Carlsbad, CA, Carlsbad, CA).Fluoromyelin debris labeling: Myelin was isolated from corpus callosum of adult CD-1 mice and triturated several times through a 25-gauge syringe needle until homogeneous then centrifuged at 19,000 × g for 15 min.Supernatant was removed and washed 3 × in PBS then frozen at -20 °C.Prior to applying to microglia, the pellet was resuspended in PBS, fluoromyelin (Invitrogen, Carlsbad, CA, Carlsbad, CA) was added at 1:300 and incubated for 30m.Labeled myelin was pelleted (19,000 g × 5 m), supernatant removed and resuspended in 10% DMEM.
Human meningeal fibroblasts
Meningeal fibroblasts were obtained from IxCells (San Diego, CA) and plated at 5000 cells per well according to manufacturer's protocol.After 1 day, cells were starved overnight in DMEM and treated with PIPE-791 for 6 h followed by 4 h with 10 µM LPA.Cells were stained and imaged as described with rat meningeal fibroblasts.
In vivo dose-occupancy
Occupancy using [3H]-OPC3497 as radiotracer: Female C57BL6/N mice (Envigo, Indianapolis, IN) were dosed with vehicle or PIPE-791 by oral gavage.[ 3 H]-OPC3497 was diluted to 9.8 µCi/mL in saline and administered via IV injection to the lateral tail vein at a dose volume of 5 ml/kg at the times indicated.Mice were euthanized by decapitation, trunk blood collected into K 3 EDTA tubes and stored on wet ice, each brain rapidly dissected and the forebrain isolated, weighed and placed into a 5 ml polypropylene tube.Tissues were diluted with a 10 × volume of ice-cold binding buffer (50 mM HEPES, 100 mM NaCl, 2 mM EDTA, pH 7.4).Brains were homogenized and 350 µl of the homogenate was filtered over Whatman GF/B filters (GE Life Sciences, Marlborough, MA) prewetted with 0.5% polyethyleneimine.Filters were washed with cold 50 mM TRIS-HCl, 154 mM NaCl, 0.05% Tween 20, pH 7.4 twice and dried in an oven (~ 40-50 °C). 5 ml Ultima Gold F scintillation fluid (Perkin Elmer, Boston, MA, Boston, MA) was added to each tube and samples read on a Beckman LS6500 liquid scintillation counter.After all samples were collected, dissected, and filtered, the previously collected blood in K 3 EDTA tubes was centrifuged (1450 × g, 10 min, 4 °C) to separate plasma.Plasma was aliquoted into 96-well polypropylene plates and stored at -80 °C until analysis for PIPE-791 by LC-MS/MS.Occupancy using [3H]-OPC3497 as radiotracer: [ 3 H]-PIPE-791 was diluted to 9.8 µCi/ml in saline and administered via IV injection to the lateral tail vein at a dose volume of 5 ml/kg at various times (1 min, 5 min, 30 min, 2 h, 4 h, 24 h, 72 h, 7 days, 10 days, or 14 days) prior to sample collection at Time 0.
A subset of mice from the 2 h and 24 h [ 3 H]-PIPE-791 groups were dosed with PIPE-791 (3 mg/kg for nonspecific binding) by oral gavage 2 h prior to [ 3 H]-PIPE-791 administration.At time 0, mice were euthanized, trunk blood collected into K 3 EDTA tubes and stored on wet ice (PIPE-791-dosed animals only), forebrain isolated, weighed and placed into a 5 ml polypropylene tube.Tissues were diluted with 10 × volume of ice-cold binding buffer.Brains were then rapidly homogenized and 350 µl of the resulting homogenate was filtered in duplicate over Whatman GF/B filters (GE Life Sciences, Marlborough, MA) which had been pre-wetted with 0.5% polyethyleneimine prior to loading onto the Hoefer vacuum manifold.Filters were washed twice by applying 5 ml ice-cold wash buffer to the manifold.
In vivo CC1 induction assay
Female C57Bl/6 mice (9-11 weeks old) were used in this experiment, n = 5/group.On day 0, mice were dosed with vehicle (1% HPMC, 0.1% Tween-80) or 3 mg/kg PIPE-791.On day 5, mice were sacrificed and brains submerged in formalin.Tissue was sectioned at 40 µm on a sliding microtome (ThermoFisher, Microm HM450, Waltham, MA).Four sections per mouse were immunostained.Sections were permeabilized and blocked by incubation with 20% goat serum (Millipore Sigma, St. Louis, MO) and 0.2% Triton X-100 (Millipore Sigma, St. Louis, MO) in PBS.Oligodendrocytes were detected with a rat monoclonal antibody to Anti-APC (1:500, Calbiochem, San Diego, CA), and the total pool OPC/oligodendrocyte pool were detected with a rabbit monoclonal antibody to Olig-2 (1:500, Invitrogen, Carlsbad, CA).Alexa Fluor 594 and 647 IgG secondary antibodies against rat and rabbit (1:1000, Invitrogen, Carlsbad, CA) were used for primary antibody detection.Cell nuclei were identified with Hoechst (1:1000, Invitrogen, Carlsbad, CA).Fluorescent images from 4 brain slices per mouse with 2 regions of interest (adjacent and superior to the corpus callosum) per brain slice were collected on a Nikon A1R confocal microscope.Detection and quantification of the cells were performed using NIS Elements imaging software (Nikon, Minato, Tokyo).
MOG-EAE induction and assessment
On Day 0, EAE was induced using a MOG 35-55 /CFA Emulsion PTX Hooke Kit™ (Hooke Laboratories, Lawrence, MA).Briefly, per manufacturer's instructions, each mouse was administered a total of 0.2 ml emulsion delivered via dorsal subcutaneous injections of 0.1 ml each to the mid-scapula and lower lumbar regions.Pertussis toxin was prepared per lot-specific instructions and administered via intraperitoneal injection of 0.1 ml (200 ng) on Day 0, 2 h post-MOG administration, and again on Day 1, 24 h post-MOG.
Vol:.( 1234567890 Supportive care (subcutaneous saline, Nutrical supplement, HydroGel) were provided to all animals with a clinical score greater than 3 and/or displaying a body weight loss greater than 10% from the previous day's weight.
Visual evoked potentials
VEP recordings were conducted 21-24 days post-MOG induction (for simplification, this will be referred to as a nominal Day 21 VEP throughout this document).To do this, animals were placed in a room with red light conditions and allowed to adapt for 1 h.Animals were anesthetized with an intraperitoneal injection of ketamine/ Xyalazine (75/10 mg/kg).Pupils were dilated with 1.0% tropicamide (1 drop per eye) (Akorn Pharmaceuticals, Lake Forest, IL).One minute following application tropicamide, one drop of Genteal (Alcon Laboratories, Fort Worth, TX) was applied to each eye to maintain ocular moisture during anesthesia.
VEPs were conducted using a Celeris system (Diagnosys, Lowell, MA).Anesthetized mice were placed onto a heated platform (37C) and instrumented with subcutaneous recording electrodes placed into the snout and dorsal occipital regions.VEP execution started approximately 10-12 min following injection of the anesthetic.Each examination was comprised of at least 3 runs using settings outline in Table 3. Flash VEPs were recorded from each eye independently and simultaneously.Parameters used (setting): Pulse intensity (3 cd s/m 2 ); Frequency (1 Hz); On time (4 ms); Pulse color (white-6500K); Sweeps per result (100).VEP waveforms were analyzed using Espion (Diagnosys, Lowell, MA).N1 latency for each eye was determined by averaging N1 from 3 independent VEP traces.Similarly, P1-N1 and N1-P1 amplitudes were determined.Waveform symmetry and integrity was expressed by calculating the amplitude ratio (AR): AR = (P1-N1 amplitude in µV)/(N1-P2 amplitude in µV).An AR approximately ≥ 1 suggests normal VEP integrity, while an AR < 1 and approaching 0 suggests degradation of the N1-P2 amplitude.
Electron microscopy and g-ratios
Study termination was the last day that clinical scores were determined for all study animals (Day 24).Since VEP recordings required several days (Days 21-24), oral dosing continued through perfusion, which was performed on Day 24 or Day 25.On Day 24 or 25, and following completion of VEP recordings, mice were deeply anesthetized with isoflurane anesthesia and whole body perfused with EM grade Karnovsky's fixative (3% glutaraldehyde, 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4; part # 15732-10; Electron Microscopy Sciences, Hatfield, PA).Spinal columns and optic nerves were collected and stored in fixative at 4 °C until further processing.After at 3-5 days in fixative, spinal cords were isolated and dissected from the vertebrae.Approximately 2 mm of the most caudal lumbar region was trimmed and segmented and, along with the optic nerves, shipped to Charles River Laboratories (Durham, NC) for processing for and imaging by electron microscopy.
At Charles River Laboratories, tissues (spinal cord or optic nerves) were post-fixed in osmium tetroxide, rinsed in distilled water (2×), dehydrated through an ethanol series (50%, 70%, 95%, 100% (3×)), transitioned through propylene oxide (2x), infiltrated in Epon-Araldite (1:1 EA:PO, 3:1 EA:PO, pure EA), and embedded in Epon-Araldite blocks.The blocks were polymerized overnight at ~ 85 °C, and semi-thin sections (~ 1 μm, one per tissue) prepared and stained with toluidine blue using standard protocols.Semi-thin images were evaluated, and the most caudal region was identified as the region of focus for thin sections.Thin sections of ~ 100 nm were prepared using standard protocols and the grids were imaged at 1500 × on a JEOL JEM-1400 + transmission electron microscope fitted with an AMT 16 MP digital camera system.The direct magnification used was 1500 × or 4000 × for spinal cord and optic nerve, respectively.
Digital EM images were analyzed using ImageJ (NIH, Bethesda, MD) measure the circumference of the axon and, when present, the myelin sheath, and g-ratios were calculated.For spinal cord analysis, 4 representative images were analyzed per animal and the results pooled, and the percentages calculated from the animal total.For optic nerve analysis, 1 representative image was analyzed per optic nerve and each optic nerve was treated as a separate data point.
Figure 2 . 9 Figure 3 .
Figure 2. PIPE-791 has distinct in vivo brain occupancy kinetics.(A) Dose-occupancy was assessed 2 h or 24 h after administration of PIPE-791.For the steady state condition, mice were dosed once daily for 4 days then occupancy assessed 2 h after the fourth dose (mean ± SEM, n = 6).(B) Pooled analysis of unbound brain concentration (C u , brain ) plotted versus occupancy.(C) Time course of 3 mg/kg PIPE-791 plotted against brain receptor occupancy (left y-axis) and plasma concentration (right y-axis) highlighting its extended brain receptor occupancy and disconnect with plasma concentration (mean ± SEM, n = 6).
Figure 6 .
Figure 6.PIPE-791 inhibits microglial activation.(A) Mouse hippocampal slices (postnatal day 21) were generated and treated with PIPE-791.LPA was then added to the slices to induce microglial activation.Slices were fixed and stained with an antibody against IBA1 and counterstained with Hoechst.Treatment with LPA converts microglia from a branched, ramified morphology to round and activated and is prevented in the presence of 3 µM PIPE-791.Activation was quantified using cell perimeter length of IBA1 + /Hoechst + cells (mean ± SEM, n = 4 slices, 3 animals) ANOVA with Tukey's).Scale bar: 100 µm.(B) Example image from slices treated with 10 µM LPA and LPA with PIPE-791.Scale bar: 25 µm.(C) Rat OPCs were differentiated with T3 into oligodendrocytes then treated with TNFα/IFNγ resulting a 35% decrease in viability (mean ± SEM, n = 2, 6 wells/n).(D) Addition of PIPE-791 prevented oligodendrocyte death in a dose-responsive manner (EC 50 125 nM, mean ± SEM, 6 wells/n, graphs are background subtracted to TNFα/IFNγ and plotted as % of T3/ vehicle.Values were compared by ANOVA with Dunnett's vs vehicle, * p < 0.05).
Table 1 .
Summary table of PIPE-791 in vitro radioligand binding selectivity profile in Ca +2 mobilization.
a Selectivity assessed using a 3 h incubation of PIPE-791.
30 s with gently Vol:.(1234567890)Scientific Reports | (2024) 14:10573 | https://doi.org/10.1038/s41598-024-61369-9www.nature.com/scientificreports/agitation Meningeal fibroblasts were isolated from postnatal day 8 CD-1 rats (Charles River Labs, Willington, MA).Meninges were peeled off and digested in Accutase (Thermo Fisher, Waltham, MA, Waltham, MA) for 20 min at room temperature.20% HBSS was added to stop protease activity.Digested meninges were dissociated by trituration and cells were strained through a 40 µm nylon mesh and spun down at 800 g for 15 min.The cell pellet was resuspended in DMEM + 10% FBS and plated in a T-75 flask (Corning) for 10 days.Cells were lifted with Accutase and replated into 96-well Perkin Elmer, Boston, MA Viewplates, at 15k per well.The next day, cells were starved in DMEM for 8 h.PIPE-791 at different concentrations was added and incubated for 15 h.
|
2024-05-10T06:17:47.814Z
|
2024-05-08T00:00:00.000
|
{
"year": 2024,
"sha1": "c53b18a4c0ad37d41410f8c503bdc6a9f80c5498",
"oa_license": "CCBY",
"oa_url": "https://www.nature.com/articles/s41598-024-61369-9.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "9d30904d719180c8e2c38f72c350adc9e99bc1b3",
"s2fieldsofstudy": [
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
257075158
|
pes2o/s2orc
|
v3-fos-license
|
Metallocene-Filled Single-Walled Carbon Nanotube Hybrids
In this paper, the growth mechanism, structure, growth processes, growth kinetics, and optical, vibronic and electronic properties of metallocene-filled single-walled carbon nanotubes (SWCNTs) are considered. A description of the procedures used to fill the nanotubes is provided. An investigation of doping effects on metallicity-mixed SWCNTs filled with metallocenes by Raman spectroscopy, near edge X-ray absorption fine structure spectroscopy, photoemission spectroscopy, and optical absorption spectroscopy is described. The studies of doping effects on metallicity-sorted SWCNTs filled with metallocenes are discussed. Doping effects in metallicity-mixed and sorted SWCNTs upon the chemical transformation of encapsulated molecules are analyzed. A discussion of the modification of the electronic properties of filled SWCNTs is presented. Applications of metallocene-filled SWCNTs in electrochemistry, thermoelectric power generation, chemical sensors, and magnetic recording are discussed.
SWCNTs were first filled with metallocenes in 2005: in Refs. [54,70], ferrocene (FeCp 2 ) and cobaltocene (CoCp 2 ) were incorporated, accordingly In 2009, cerocene (CeCp 3 ) molecules were filled inside SWCNTs [77]. The formation of double-walled carbon nanotubes (DWCNTs) from metallocene-filled SWCNTs was first reported for ferrocene in 2008 [60]. In 2015, nickelocene molecules were encapsulated inside SWCNTs and DWCNTs were formed [73]. It was shown that the filled SWCNTs have interesting electronic properties and that thermal treatment leads to the growth of carbon nanotubes with a modified precisely tailored electronic structure that is important for their application.
Metallocene-filled carbon nanotubes attract the attention of researchers for six reasons ( Figure 1). First is the growth mechanism of carbon nanotubes with metal carbides as catalysts of the synthesis process. The nanoparticles can be in liquid or solid state during the synthesis process [88][89][90][91][92][93][94][95]. The chemical state of the nanoparticles can be carbidic or SWCNTs filled with metallocenes is considered. Raman spectroscopy, near edge X-ray absorption fine structure spectroscopy (NEXAFS), photoemission spectroscopy (PES), and optical absorption spectroscopy (OAS) are performed . The investigation of doping effects in metallicity-sorted SWCNTs filled with metallocenes is discussed [67].
Sixthly, applications of metallocene-filled SWCNTs attract interest. Among them, applications in electrochemistry [172], thermoelectric power generation [3], chemical sensors [173], and magnetic recording [174][175][176][177] are considered to be promising. For these applications, carbon nanotubes with homogeneous properties are needed. The listed applications are most viable. Other possible applications are in solar cells and light emission devices. The aim of this paper is to review the growth mechanism, structures, growth processes, growth kinetics, and optical, vibronic, and electronic properties of metallocenefilled SWCNTs.
In Section 2, the growth mechanism of carbon nanotubes is discussed. In Section 3, the structure of SWCNTs filled with molecules is characterized. In Section 4, the growth process of carbon nanotubes and other structures is discussed. In Section 5, the growth Secondly, the structure of SWCNTs filled with metallocenes attracts interest because it is important to fill the bundles of nanotubes. Scanning electron microscopy (SEM) [116] and transmission electron microscopy (TEM) [72,74,75,116,117] are used. The chemical state of the filler is confirmed by energy dispersive analysis (EDX) [116]. The structure of DWCNTs formed as a result of the annealing of metallocene-filled SWCNTs is characterized [116].
Thirdly, the growth process of carbon nanotubes and other nanostructures is studied with microscopy. It is shown that pyrolysis of metallocenes leads to the formation of metal-filled carbon nanotubes [118]. The treatment of ferrocene at different pressures and temperatures was demonstrated to result in the synthesis of amorphous carbon, microparticles, nanotubes, microcones, and spirals [119].
Fourthly, the growth kinetics of carbon nanotubes inside metallocene-filled SWCNTs attracts huge interest [72,75,120]. The growth process of carbon nanotubes is characterized in two stages, with carbidic and metallic catalytic particles, accordingly Each stage is characterized with activation energy and growth rate [75]. They are dependent on the diameter and chiral angle of the nanotubes, and the metal type [120].
Fifthly, the characterization of the optical, vibronic, and electronic properties of SWC-NTs filled with metallocenes is considered. Raman spectroscopy, near edge X-ray absorption fine structure spectroscopy (NEXAFS), photoemission spectroscopy (PES), and optical absorption spectroscopy (OAS) are performed . The investigation of doping effects in metallicity-sorted SWCNTs filled with metallocenes is discussed [67].
Sixthly, applications of metallocene-filled SWCNTs attract interest. Among them, applications in electrochemistry [172], thermoelectric power generation [3], chemical sensors [173], and magnetic recording [174][175][176][177] are considered to be promising. For these applications, carbon nanotubes with homogeneous properties are needed. The listed applications are most viable. Other possible applications are in solar cells and light emission devices.
The aim of this paper is to review the growth mechanism, structures, growth processes, growth kinetics, and optical, vibronic, and electronic properties of metallocene-filled SWCNTs.
In Section 2, the growth mechanism of carbon nanotubes is discussed. In Section 3, the structure of SWCNTs filled with molecules is characterized. In Section 4, the growth process of carbon nanotubes and other structures is discussed. In Section 5, the growth kinetics of carbon nanotubes are considered. In Section 6, the optical, vibronic, and electronic properties of SWCNTs filled with molecules are characterized. In Section 7, the investigation of doping effects in metallicity-sorted SWCNTs filled with molecules is discussed. In Section 8, investigations of doping effects in metallicity-mixed and -sorted SWCNTs upon the chemical transformation of encapsulated molecules are considered. In Section 9, a The authors of Ref. [88] suggest that, for the majority of high-yield CVD techniques such as injection methods for growing SWCNTs at temperatures in the order of 1000 • C [89,90], the catalyst is likely to be in the liquid state. However, in situ transmission electron microscopy (TEM) observations on the growth of SWCNTs and MWCNTs on metallic and carbidic nanoparticles at temperatures up to 650 • C demonstrated that the particles remained crystalline during the growth process, although the particles changed their shape [91][92][93][94][95]. It was shown that Ni nanoparticles with a size down to~4-5 nm stayed crystalline at temperatures as high as 540 • C [91] and 615 • C [92] while they were growing carbon nanotubes. The authors of Ref. [93] observed structural fluctuations in a solid Fe 3 C nanoparticle at 600 • C that was growing an SWCNT with a diameter as small as 1.5 nm. In another instance, they observed the growth of a~15-20 nm diameter MWCNT on the surface of a crystalline Fe 3 C nanoparticle. The crystal structure of the active catalyst particle was found to be fluctuating between Fe 23 C 6 and Fe 3 C structures with random crystallographic directions, indicating that carbon atoms were migrating through the bulk ( Figure 2) [95].
kinetics of carbon nanotubes are considered. In Section 6, the optical, vibronic, and electronic properties of SWCNTs filled with molecules are characterized. In Section 7, the investigation of doping effects in metallicity-sorted SWCNTs filled with molecules is discussed. In Section 8, investigations of doping effects in metallicity-mixed and -sorted SWCNTs upon the chemical transformation of encapsulated molecules are considered. In Section 9, a discussion of the modification of the electronic properties of filled SWCNTs is presented. In Section 10, applications of metallocene-filled SWCNTs in electrochemistry, thermoelectric power generation, chemical sensors, and magnetic recording are discussed.
Physical State of Catalyst
The authors of Ref. [88] suggest that, for the majority of high-yield CVD techniques such as injection methods for growing SWCNTs at temperatures in the order of 1000 °C [89,90], the catalyst is likely to be in the liquid state. However, in situ transmission electron microscopy (TEM) observations on the growth of SWCNTs and MWCNTs on metallic and carbidic nanoparticles at temperatures up to 650 °C demonstrated that the particles remained crystalline during the growth process, although the particles changed their shape [91][92][93][94][95]. It was shown that Ni nanoparticles with a size down to ~4-5 nm stayed crystalline at temperatures as high as 540 °C [91] and 615 °C [92] while they were growing carbon nanotubes. The authors of Ref. [93] observed structural fluctuations in a solid Fe3C nanoparticle at 600 °C that was growing an SWCNT with a diameter as small as 1.5 nm. In another instance, they observed the growth of a ~15-20 nm diameter MWCNT on the surface of a crystalline Fe3C nanoparticle. The crystal structure of the active catalyst particle was found to be fluctuating between Fe23C6 and Fe3C structures with random crystallographic directions, indicating that carbon atoms were migrating through the bulk ( Figure 2) [95].
Chemical State of Catalyst
The formation of metal carbides from pure metals and their subsequent decomposition prior to nanotube growth has been confirmed several times [96][97][98][99]. This finding strongly suggests the decomposition of metal carbides as a crucial step of carbon nanotube synthesis [99]. Iron carbide was observed immediately before the start of nanotube growth [97], and the onset of growth coincided with the decomposition of the carbide to Fe and graphite [96]. Time-resolved XPS studies of catalyzed nanotube synthesis showed chemisorbed carbon and carbidic carbon on the Fe catalyst during the incubation phase before nanotube growth; once the growth commenced, the peak of the sp 2 graphitic carbon network emerged [98]. Figure 3 shows the time-resolved evolution of the C 1s XPS peak [98]. As soon as C 2 H 2 is let into the chamber, the peak at 282.6 eV signifies that carbon is chemisorbed on the Fe catalyst. After 90 s of incubation, a carbidic carbon peak at 283.2 eV persists for 30 s. The formation of an sp 2 carbon network heralds the appearance of another peak at 284.5 eV. The graphitic peak quickly dominates over the other two carbon species before it saturates after 60 more seconds, when nanotube growth stops.
Chemical State of Catalyst
The formation of metal carbides from pure metals and their subsequent decomposition prior to nanotube growth has been confirmed several times [96][97][98][99]. This finding strongly suggests the decomposition of metal carbides as a crucial step of carbon nanotube synthesis [99]. Iron carbide was observed immediately before the start of nanotube growth [97], and the onset of growth coincided with the decomposition of the carbide to Fe and graphite [96]. Time-resolved XPS studies of catalyzed nanotube synthesis showed chemisorbed carbon and carbidic carbon on the Fe catalyst during the incubation phase before nanotube growth; once the growth commenced, the peak of the sp 2 graphitic carbon network emerged [98]. Figure 3 shows the time-resolved evolution of the C 1s XPS peak [98]. As soon as C2H2 is let into the chamber, the peak at 282.6 eV signifies that carbon is chemisorbed on the Fe catalyst. After 90 s of incubation, a carbidic carbon peak at 283.2 eV persists for 30 s. The formation of an sp 2 carbon network heralds the appearance of another peak at 284.5 eV. The graphitic peak quickly dominates over the other two carbon species before it saturates after 60 more seconds, when nanotube growth stops. It should be noted that intermediate carbide phases are in many cases not directly confirmed by in situ TEM analysis of nanotube growth on a nickel catalyst [91,92,100]. The distinction of nickel and nickel carbide is very challenging because the Ni atoms have the same structure with very similar lattice constants, and thus they cannot be easily distinguished by diffraction or imaging [88]. The distinction becomes especially hard if partial carburization below the surface the catalyst particle is considered [101]. However, in situ TEM has been successfully applied to directly see the structures of iron and cobalt carbides in active catalyst particles during the growth of nanotubes [94,95,[102][103][104][105]. Figure 4 shows rows of a high-resolution TEM image overview, a closeup of the boxed region in the overview, and the diffractogram of the closeup at different times from a continuous video sequence [103]. The Fe particle was heated to 650 • C and exposed to 10 m Torr of flowing C 2 H 2 . The deposited particle had an irregular shape and a bcc structure. After 34.7 s, the particle was a crystallite of bcc iron with {110} faces. After 37.1 s, the particle had a more rounded shape, the diffractogram confirming Fe 3 C. By 37.3 s, the particle had already grown a multiwall carbon nanotube and its diffractogram proves it was still made of Fe 3 C. distinction of nickel and nickel carbide is very challenging because the Ni atoms have the same structure with very similar lattice constants, and thus they cannot be easily distinguished by diffraction or imaging [88]. The distinction becomes especially hard if partial carburization below the surface the catalyst particle is considered [101]. However, in situ TEM has been successfully applied to directly see the structures of iron and cobalt carbides in active catalyst particles during the growth of nanotubes [94,95,[102][103][104][105]. Figure 4 shows rows of a high-resolution TEM image overview, a closeup of the boxed region in the overview, and the diffractogram of the closeup at different times from a continuous video sequence [103]. The Fe particle was heated to 650 °C and exposed to 10 m Torr of flowing C2H2. The deposited particle had an irregular shape and a bcc structure. After 34.7 s, the particle was a crystallite of bcc iron with {110} faces. After 37.1 s, the particle had a more rounded shape, the diffractogram confirming Fe3C. By 37.3 s, the particle had already grown a multiwall carbon nanotube and its diffractogram proves it was still made of Fe3C. Rows of a high-resolution overview image (a,d,g,j), a closeup of the boxed region in the overview (b,e,h,k), and the diffractogram of the closeup (c,f,i,l), all extracted from the same digital video sequence. The bar was 5 nm. The deposited Fe particle was heated to 650 °C under 10 mTorr of flowing C2H2. At 20 s (a-c) the irregular Fe particle had a bcc structure. At 54.7 s (d-f) the Fe particle was a bcc crystallite with {110} faces. At 57.1 s (g-i) the rounded particle consisted of Fe3C. By 57.3 s (j,k,l) the particle had grown a multiwall carbon nanotube and still consisted of Fe3C. Rows of a high-resolution overview image (a,d,g,j), a closeup of the boxed region in the overview (b,e,h,k), and the diffractogram of the closeup (c,f,i,l), all extracted from the same digital video sequence. The bar was 5 nm. The deposited Fe particle was heated to 650 • C under 10 mTorr of flowing C 2 H 2 . At 20 s (a-c) the irregular Fe particle had a bcc structure. At 54.7 s (d-f) the Fe particle was a bcc crystallite with {110} faces. At 57.1 s (g-i) the rounded particle consisted of Fe 3 C. By 57.3 s (j,k,l) the particle had grown a multiwall carbon nanotube and still consisted of Fe 3 C. Reprinted with permission from Sharma
Tip-and Base-Growth Models
Base-growth was observed for MWCNTs in Refs. [94,106,107] and for SWCNTs in Refs. [92,94,100,108,109]. Tip-growth was reported for MWCNTs in Refs. [91][92][93]102] and for SWCNTs in Refs. [110][111][112]. Time-resolved in situ HRTEM is arguably the most direct tool for investigating the growth of carbon nanotubes and for distinguishing the tip-and base-growth mechanisms [91][92][93][94]100,110]. Figure 5a-d shows TEM images of SWCNTs and nanocages [100]. The diameters of the SCWNTs range between 0.6 and 3.5 nm. Their lengths vary between a few nanometers and one micrometer. All SWCNTs had clean straight walls and closed Ni catalyst-free tips, which is indicative of base-growth. The diameters of Ni catalyst particles at the base are correlated to the diameter of the SWCNT.
Tipand Base-Growth Models
Base-growth was observed for MWCNTs in Refs. [94,106,107] and for SWC Refs. [92,94,100,108,109]. Tip-growth was reported for MWCNTs in Refs. [91][92][93]10 for SWCNTs in Refs. [110][111][112]. Time-resolved in situ HRTEM is arguably the mos tool for investigating the growth of carbon nanotubes and for distinguishing the t base-growth mechanisms [91][92][93][94]100,110]. Figure 5a-d shows TEM images of SWCNTs and nanocages [100]. The diam the SCWNTs range between 0.6 and 3.5 nm. Their lengths vary between a few nano and one micrometer. All SWCNTs had clean straight walls and closed Ni catalyst-fr which is indicative of base-growth. The diameters of Ni catalyst particles at the b correlated to the diameter of the SWCNT.
(e) (f) Base-growth was frequently observed under typical synthesis conditions for nanotubes [92,94,100,108,109]. There are also well documented scenarios whe Base-growth was frequently observed under typical synthesis conditions for carbon nanotubes [92,94,100,108,109]. There are also well documented scenarios where tip-growth occurs in SWCNTs [110][111][112]. In Ref. [112], a fast-heating CVD process was employed to synthesize long and aligned SWCNTs. Careful inspection confirmed that both tip-and basegrowth mechanisms had occurred during the synthesis. Moreover, the long and oriented nanotubes were produced exclusively by the tip-growth mechanism. Figure 5f shows the SEM image of the short random SWCNTs, obtained in a classical CVD process [112]. Direct evidence for the tip-growth mechanism in long and aligned SWCNTs under fast-heating conditions came from AFM imaging. The tips of all long aligned SWCNTs featured a nanoparticle. One example is shown in Figure 5e [112]. The size of the particles at the tip of the nanotubes was in most cases slightly larger than the SWCNT's diameter. The authors of Ref. [112] argue that the catalyst particles likely grow due to amorphous carbon deposition during the cooling process.
Tangential and Perpendicular Growth Modes
Besides the tip-and base-growth mechanisms there is also the distinction between tangential and perpendicular growth modes [113][114][115].
TEM data was employed to conduct a statistical analysis of the correlation between the diameters of SWCNTs and the sizes of the catalytic nanoparticles on which they grow [113]. The images unanimously proved the existence of two types of nanotube nuclei for the tangential and perpendicular growth modes, respectively. In the tangential growth mode, the diameter of the nanotube is close to that of the nanoparticle. In the perpendicular growth mode, the diameter of the carbon nanotube is much smaller than that of the particle and the carbon walls have a nearly perpendicular contact angle to the surface of the nanoparticle [113]. The diameters of nanotubes nucleated in the perpendicular growth mode are not correlated with that of the nanoparticle. If the diameter of the nanotube is less than 75% of the particle diameter it can only have been formed by the perpendicular growth mode. The ratio of tangential to parallel nucleations is not affected by the diameter of the particles. It is, however, affected by the synthesis time.
Characterization of the Structure of SWCNTs Filled with Molecules
This section is dedicated to analysis of the structure of SWCNTs filled with metallocenes, and the structure obtained from them due to chemical reactions. Figure 6 presents a scanning transmission electron microscopy (SEM) image of ferrocenefilled SWCNTs [54]. The image shows bundles of filled SWCNTs, which are homogeneously located throughout the sample. The image proves no impurities of excess ferrocene on the outer side of the carbon nanotubes' walls, which confirms the successful filling of the channels of the carbon nanotubes with ferrocene. Figure 6 presents a scanning transmission electron microscopy (SEM) image of ferro cene-filled SWCNTs [54]. The image shows bundles of filled SWCNTs, which are homo geneously located throughout the sample. The image proves no impurities of excess fer rocene on the outer side of the carbon nanotubes' walls, which confirms the successfu filling of the channels of the carbon nanotubes with ferrocene. Figure 7a shows a TEM image of ferrocene-filled SWCNTs [54]. The image shows the bundle of filled carbon nanotubes. It is visible that the interior space inside the SWCNTs on the edge of the bundle is filled with molecules. The energy-dispersive analysis (EDX) (Figure 7b) shows the presence of iron peaks in the spectrum, which corresponds to the filling of the carbon nanotubes [54]. Figure 7a shows a TEM image of ferrocene-filled SWCNTs [54]. The image shows the bundle of filled carbon nanotubes. It is visible that the interior space inside the SWCNTs on the edge of the bundle is filled with molecules. The energy-dispersive analysis (EDX) (Figure 7b) shows the presence of iron peaks in the spectrum, which corresponds to the filling of the carbon nanotubes [54]. In Ref. [117], the iron/iron carbide clusters were formed inside carbon nanotubes due to the annealing of ferrocene-filled carbon nanotubes. Figure 8 shows the metal clusterfilled carbon nanotubes. Figure 8a-c shows the filled semiconducting SWCNTs, with the inset in Figure 8b showing the individual cluster. Figure 8d-f shows the filled metallic SWCNTs, which are filled with iron. In Ref. [117], the iron/iron carbide clusters were formed inside carbon nanotubes due to the annealing of ferrocene-filled carbon nanotubes. Figure 8 shows the metal clusterfilled carbon nanotubes. Figure 8a-c shows the filled semiconducting SWCNTs, with the inset in Figure 8b showing the individual cluster. In Ref. [117], the iron/iron carbide clusters were formed inside carbon nanotubes due to the annealing of ferrocene-filled carbon nanotubes. Figure 8 shows the metal clusterfilled carbon nanotubes. Figure 8a-c shows the filled semiconducting SWCNTs, with the inset in Figure 8b showing the individual cluster. Figure 8d-f shows the filled metallic SWCNTs, which are filled with iron. The thermal treatment of ferrocene-filled carbon nanotubes can lead to the formation of double-walled carbon nanotubes (DWCNTs). Figure 9 shows a TEM image of an individual DWCNT and the cross-section of a bundle of DWCNTs [54]. The presence of two walls is clearly visible in the image. Figure 9a shows the individual DWCNT and a nearby SWCNT for comparison. Figure 9b shows the cross-section of carbon nanotubes. The thermal treatment of ferrocene-filled carbon nanotubes can lead to the formation of double-walled carbon nanotubes (DWCNTs). Figure 9 shows a TEM image of an individual DWCNT and the cross-section of a bundle of DWCNTs [54]. The presence of two walls is clearly visible in the image. Figure 9a shows the individual DWCNT and a nearby SWCNT for comparison. Figure 9b shows the cross-section of carbon nanotubes.
Cobaltocene-Filled Carbon Nanotubes
The structure of cobaltocene-filled carbon nanotubes was studied by TEM imaging. The thermal treatment of filled SWCNTs was conducted to obtain cobalt/cobalt carbide clusters inside carbon nanotubes. Figure 10 shows the metal-filled carbon nanotubes obtained after the annealing of cobaltocene-filled SWCNTs at 550 °C for 2 h [72]. The images show the presence of metal clusters inside the carbon nanotubes.
Cobaltocene-Filled Carbon Nanotubes
The structure of cobaltocene-filled carbon nanotubes was studied by TEM imaging. The thermal treatment of filled SWCNTs was conducted to obtain cobalt/cobalt carbide clusters inside carbon nanotubes. Figure 10 shows the metal-filled carbon nanotubes obtained after the annealing of cobaltocene-filled SWCNTs at 550 • C for 2 h [72]. The images show the presence of metal clusters inside the carbon nanotubes.
Nickelocene-Filled Carbon Nanotubes
Nickelocene molecules were filled inside SWCNTs, and TEM was used to investigate the structure of the filled SWCNTs and annealed filled carbon nanotubes. Figure Figure 11 shows TEM images of cobaltocene-filled SWCNTs annealed at 800 • C for 2 h [72]. The DWCNTs are formed in all images. Figure 11a shows the clean DWCNT. Figure 11b shows the DWCNTs and nearby SWCNTs forming DWCNTs. Figure 11c-e shows scanning TEM images of DWCNTs. Figure 11c shows a clean individual DWCNT. Figure 11d,e shows growing DWCNTs.
Nickelocene-Filled Carbon Nanotubes
Nickelocene molecules were filled inside SWCNTs, and TEM was used to investigate the structure of the filled SWCNTs and annealed filled carbon nanotubes. Figure
Nickelocene-Filled Carbon Nanotubes
Nickelocene molecules were filled inside SWCNTs, and TEM was used to investigate the structure of the filled SWCNTs and annealed filled carbon nanotubes. Figure 12 shows TEM and scanning TEM images of nickelocene-filled SWCNTs annealed at 200 • C (Figure 12a), 500 • C (Figure 12b), 700 • C ( Figure 12c) for 2 h [75]. It is visible that the increase in annealing temperature leads to the formation of longer metallic clusters.
Growth Processes of Filled Carbon Nanotubes and Other Structures
In Ref. [119], the treatment of ferrocene at different conditions (pressure, te ture) was shown to lead to the formation of amorphous carbon, microparticles, nan microcones, and spirals. The processes occurred due to self-organization. Fig shows the phase diagram for the growth process of synthesized structures depen the pressure and temperature of synthesis [119]. At pressures above 5 MPa and te tures above 590 °C, the mirrored structure microcones and spirals were obser higher temperatures and lower pressures, the formation of nanotubes was obser low temperatures and pressures, amorphous carbon was formed. Figure 13b-h SEM images of amorphous carbon, microparticles, nanotubes, microcones, and [119].
Growth Processes of Filled Carbon Nanotubes and Other Structures
In Ref. [119], the treatment of ferrocene at different conditions (pressure, temperature) was shown to lead to the formation of amorphous carbon, microparticles, nanotubes, microcones, and spirals. The processes occurred due to self-organization. Figure 13a shows the phase diagram for the growth process of synthesized structures depending on the pressure and temperature of synthesis [119]. At pressures above 5 MPa and temperatures above 590 • C, the mirrored structure microcones and spirals were observed. At higher temperatures and lower pressures, the formation of nanotubes was observed. At low temperatures and pressures, amorphous carbon was formed. Figure 13b-h shows SEM images of amorphous carbon, microparticles, nanotubes, microcones, and spirals [119].
the pressure and temperature of synthesis [119]. At pressures above 5 MPa and temp tures above 590 °C, the mirrored structure microcones and spirals were observed higher temperatures and lower pressures, the formation of nanotubes was observed low temperatures and pressures, amorphous carbon was formed. Figure 13b-h sh SEM images of amorphous carbon, microparticles, nanotubes, microcones, and sp [119].
Characterization of the Kinetics of the Growth of Carbon Nanotubes
The growth kinetics of carbon nanotubes was characterized in Refs. [72,75,120]. It was shown that the growth process of carbon nanotubes is characterized by two activation energies, E α and E β , of growth on carbide and metallic catalytic particles, accordingly. These two stages are characterized by two growth rates, α and β. The growth process depends on the diameter of the nanotube and the type of metal. Figure 14 shows the growth curves of carbon nanotubes with different chiralities at 580 • C on a cobalt catalyst [72]. It is visible that the carbon nanotubes grow quickly with rate α at the beginning of the growth process. They continue to grow slower with rate β.
Here the annealing times from 2 min to 3000 min are considered, and saturation is achieved.
Characterization of the Kinetics of the Growth of Carbon Nanotubes
The growth kinetics of carbon nanotubes was characterized in Refs. [72,75, shown that the growth process of carbon nanotubes is characterized by two energies, Eα and Eβ, of growth on carbide and metallic catalytic particles, a These two stages are characterized by two growth rates, α and β. The grow depends on the diameter of the nanotube and the type of metal. Figure 14 shows the growth curves of carbon nanotubes with different c 580 °C on a cobalt catalyst [72]. It is visible that the carbon nanotubes grow q rate α at the beginning of the growth process. They continue to grow slower w Here the annealing times from 2 min to 3000 min are considered, and sa achieved. Logarithmic plots of growth rates α and β were prepared using the experim on the growth of carbon nanotubes with different chiralities on cobalt and nick Logarithmic plots of growth rates α and β were prepared using the experimental data on the growth of carbon nanotubes with different chiralities on cobalt and nickel catalysts ( Figure 15) [72]. They exhibit linear behavior. The temperature range from 480 • C to 640 • C is considered, and it is shown in the upper abscissa axis. The arrows for every spot, which were considered for the fitting of the linear function, are shown. Logarithmic plots of growth rates α and β were prepared using the experi on the growth of carbon nanotubes with different chiralities on cobalt and nick ( Figure 15) [72]. They exhibit linear behavior. The temperature range from 48 °C is considered, and it is shown in the upper abscissa axis. The arrows for which were considered for the fitting of the linear function, are shown. Figure 16 shows the dependence of activation energies E α (Figure 16a) and E β (Figure 16b) on the diameter of the nanotube [120]. The authors demonstrated that activation energy E α decreases with decreasing the carbon nanotube diameter for both the nickel and cobalt catalysts. Activation energy E β shows only a small dependence on the diameter. The authors skipped a discussion of the dependence of the activation energy on the chiral angle. 4 13 of 35 Figure 16 shows the dependence of activation energies Eα ( Figure 16a) and Eβ ( Figure 16b) on the diameter of the nanotube [120]. The authors demonstrated that activation energy Eα decreases with decreasing the carbon nanotube diameter for both the nickel and cobalt catalysts. Activation energy Eβ shows only a small dependence on the diameter. The authors skipped a discussion of the dependence of the activation energy on the chiral angle. The following nine points in the observed dependences should be refined. The following nine points in the observed dependences should be refined. The activation energy of growth in carbide and metal particles decreases when the tube diameter is decreased. In the case of the activation energy of growth in carbide particles, the activation energy is larger for cobalt than nickel for larger-diameter tubes, and it is larger for nickel than cobalt for smaller-diameter tubes. This is caused by differences in the bulk structures of nickel and cobalt particles. The value for a (10,4) carbon nanotube grown on a cobalt catalyst is the maximum because of the structure of the carbon tube. The value for a (9,3) carbon nanotube grown on a cobalt catalyst is the minimum because of the structure of the carbon nanotube. In the case of the activation energy of growth in metallic particles, the activation energy is larger for a nickel catalyst than a cobalt catalyst for larger-diameter tubes, and it is larger for a cobalt catalyst than for a nickel catalyst for smaller-diameter carbon nanotubes. The value for a (12,3) carbon nanotube is the minimum for a cobalt catalyst, and the value for several carbon nanotubes (12,3), and (9,3), are maximal for a nickel catalyst among other carbon nanotubes, because of their atomic structure.
An increase in activation energy for a cobalt catalyst is caused by the modification of its atomic structure from hexagonal to cubic and then monoclinic.
An increase in activation energy for a nickel catalyst is caused by the modification of its atomic structure from cubic to hexagonal and then to monoclinic. This required additional theoretical calculations for the optimization of the crystal structure of catalyst particles.
The chiral angle's dependence on the activation energy of growth in carbon nanotubes for both catalysts (nickel and cobalt) exists in a torch-like shape on the plot, which is caused by the different atomic structure and gradual change in chiral angle of the carbon nanotube when it grows on catalyst particles at a different angle relative to the walls of the carbon nanotube (see figure). Here on the figure we see that nanotubes with the largest chiral angles have the largest activation energies of growth. The nanotubes with smaller chiral angles have smaller activation energies, but they are positioned in a range from maximal to minimal. The nanotubes with the smallest chiral angles have the smallest activation energies. The nanotubes (10,4), (9,3), and (12,3) have the largest activation energies due to their structures, and this leads to a broadening of chiral angle distributions. The distributions are not shown here, because they are broadened.
The chiral angle's dependence on the activation energy is broadened, but it is visible that cobalt particles have higher activation energies for the growth of carbon nanotubes.
The growth rates of carbon nanotubes decrease when the nanotube diameter is increased, with the exception being the chiral angle's dependence on growth, which is observed to be the same for both activation energies.
The growth rate increases for a nickel catalyst as compared to a cobalt catalyst because of differences in the diffusion rates of metal and carbon.
For a fixed-time growth, the growth rate of the carbon nanotube increases with when the tube diameter is decreased; this dependence is refined with chiral angle dependence, which is, as discussed in the above-mentioned points, due to its dependence on the activation energy of grown carbon nanotubes. It is important to note that this point should be refined.
For the dependence of fixed-time growth on temperature, the growth rate of the carbon nanotube decreases for larger-dimeter armchair carbon nanotubes, which have the largest diameter in the case of nickel and cobalt catalysts. The growth rates of carbon nanotubes increase for zigzag tubes, which can have smaller diameters.
For fixed-time growth, the growth rates are larger for a nickel than a cobalt catalyst. They have different diffusion rates for different chiral angles of chosen carbon nanotubes. This should be refined by the experiment and theoretical calculations.
Here, we discuss the points of the activation energy and growth rate dependence of carbon nanotubes for chosen data for nickelocene-and cobaltocene-filled SWCNTs. The results can vary for other data, but the discussed tendencies provide a fundamental theory with respect to the growth kinetics and growth dynamics of carbon nanotubes. This theory can be formulated in five points:
•
The growth of carbon nanotubes depends on the temperature and time.
•
The growth of carbon nanotubes depends on the metal catalyst and precursor.
•
The growth rate depends on the temperature and metal catalyst type.
•
The growth rate depends on the diffusion rate of metal and carbon.
•
The activation energy depends on the tube diameter and chiral angle, because of structural differences between the catalyst and carbon nanotube. Organic and organometallic (metallocene and metal acetylacetonate) molecules decompose at high temperatures. Powders of these substances cannot be melted, but they can be sublimed in vacuum at elevated temperatures without decomposing the molecules. The gas phase method was employed to fill SWCNTs for spectroscopic investigations of their modified electronic properties. In the gas phase method, the SWCNTs and an excess amount of the organic or organometallic powder are sealed in a quartz ampoule. One half of the ampoule is heated to gradually re-sublime the heated powder from the warm side onto the cooler side. The buckypaper of the SWCNTs is positioned in the center. The heated and unheated ends of the ampoules are typically flipped twice a day and the entire gas phase filling takes several days.
Investigation of Doping Effects in Metallicity-Mixed SWCNTs Filled with Molecules
In the literature, investigations of metallocene-filled SWCNTs by Raman spectroscopy, near edge X-ray absorption fine structure spectroscopy (NEXAFS), photoemission spectroscopy (PES), and optical absorption spectroscopy (OAS) have been performed .
Photoemission Spectroscopy
Information on changes in the bonding environment and the Fermi level in moleculefilled SWCNTs was obtained by photoemission spectroscopy. In the literature, the C 1s XPS spectra of SWCNTs filled with cerocene [77,78], ferrocene [59,61,65,67], nickelocene [73], cobaltocene [3], and nickel acetylacetonate [81] molecules have been reported. Modifications in the electronic properties of the filled SWCNTs are reflected in the shifted position, altered width, and changed shape of the C 1s peak. The filling of SWCNTs with molecules usually leads to an upshift in the C 1s binding energy and the increased asymmetry and width of the C 1s peak. These changes testify to a lower work function due to the charge transfer from the incorporated molecules to the SWCNTs. Figure 17 shows the C 1s XPS spectra of the pristine and cobaltocene-filled SWCNTs [3]. Comparison of the spectra reveals an upshift in the C 1s peak, its increased asymmetry and a decrease in the intensity of the shake-up peaks at~291 eV (denoted by the arrow) due to the electronic interaction between the nanotubes and the introduced cobaltocene.
In Ref. [63], the determination of the shift value in the Fermi level of the ferrocene-filled SWCNTs was performed by UPS. In the valence band spectra of the filled SWCNTs, the peak positions of the first and the second vHs of semiconducting SWCNTs and the first vHs of metallic nanotubes were consistently upshifted by 0.05 eV, which is equivalent to increasing the Fermi level of SWCNTs by this value. This n-doping is due to the charge transfer from the encapsulated molecules to the SWCNTs.
In Ref. [63], the determination of the shift value in the Fermi level of the ferrocenefilled SWCNTs was performed by UPS. In the valence band spectra of the filled SWCNTs the peak positions of the first and the second vHs of semiconducting SWCNTs and the first vHs of metallic nanotubes were consistently upshifted by 0.05 eV, which is equivalent to increasing the Fermi level of SWCNTs by this value. This n-doping is due to the charge transfer from the encapsulated molecules to the SWCNTs.
Optical Absorption Spectroscopy
The charge transfer in molecule-filled SWCNTs was studied by optical absorption spectroscopy. Figure 18 demonstrates the OAS spectra of the pristine and cobaltocenefilled SWCNTs with a mean diameter of 1.4 nm [3]. The spectra reveal the characteristic and absorption bands of semiconducting and metallic SWCNTs positioned at wavelengths around 900-1300 nm and 600-800 nm, respectively. Both absorption bands are red-shifted by ~30 meV in the case of the filled SWCNTs, which was explained by ndoping in the SWCNTs by the encapsulated cobaltocene [3].
Optical Absorption Spectroscopy
The charge transfer in molecule-filled SWCNTs was studied by optical absorption spectroscopy. Figure 18 demonstrates the OAS spectra of the pristine and cobaltocene-filled SWCNTs with a mean diameter of 1.4 nm [3]. The spectra reveal the characteristic E S 22 and E M 11 absorption bands of semiconducting and metallic SWCNTs positioned at wavelengths around 900-1300 nm and 600-800 nm, respectively. Both absorption bands are red-shifted by~30 meV in the case of the filled SWCNTs, which was explained by n-doping in the SWCNTs by the encapsulated cobaltocene [3]. In Ref. [57], the OAS spectra of pristine, open-ended, and ferrocene-filled SWCNTs with a mean diameter of 1.0 nm were obtained. Comparison of the spectra of the pristine SWCNTs and open-ended SWCNTs showed notable differences in the S11 region. In the latter, the peaks of the largest-diameter SWCNTs in the range from 900 to 1100 nm were suppressed and red-shifted due to p-doping and the improved screening of optical excitons [57]. After filling the SWCNTs with ferrocene, the peaks in the S11 region maintained the red shift, but their intensity increased significantly. This is due to the unchanged screening of optical excitons and the charge transfer from the incorporated ferrocene molecules to the SWCNTs, i.e., n-doping, that can even overcome the ambient p-doping in the pristine SWCNTs and increase peak intensities [57].
Investigation of Doping Effects on Metallicity-Sorted SWCNTs Filled with Molecules
In the literature, there are reports on the comparison of the doping effect of molecules on metallicity-sorted metallic and semiconducting SWCNTs. In Ref. [67], a comprehensive In Ref. [57], the OAS spectra of pristine, open-ended, and ferrocene-filled SWCNTs with a mean diameter of 1.0 nm were obtained. Comparison of the spectra of the pristine SWCNTs and open-ended SWCNTs showed notable differences in the S11 region. In the latter, the peaks of the largest-diameter SWCNTs in the range from 900 to 1100 nm were suppressed and red-shifted due to p-doping and the improved screening of optical excitons [57]. After filling the SWCNTs with ferrocene, the peaks in the S11 region maintained the red shift, but their intensity increased significantly. This is due to the unchanged screening of optical excitons and the charge transfer from the incorporated ferrocene molecules to the SWCNTs, i.e., n-doping, that can even overcome the ambient p-doping in the pristine SWCNTs and increase peak intensities [57].
Investigation of Doping Effects on Metallicity-Sorted SWCNTs Filled with Molecules
In the literature, there are reports on the comparison of the doping effect of molecules on metallicity-sorted metallic and semiconducting SWCNTs. In Ref. [67], a comprehensive spectroscopic investigation of the charge transfer in the ferrocene-filled metallic and semiconducting SWCNTs was conducted by combining XAS, XPS, and UPS. Figure 19 shows the C 1s XAS spectra of the 1.4 nm-diameter pristine and ferrocenefilled metallic, semiconducting, and metallicity-mixed SWCNTs [67]. The spectra include the π*-resonance, originating from an electronic transition from the C 1s core level to the unoccupied π*-conduction band; it demonstrates fine features, which correspond to an electronic transition from the C 1s core level to individual vHs in the conduction band of SWCNTs. The C 1s XAS spectra are fitted with a broad π*-peak at photon energies ranging from 284 to 286 eV and several components of individual vHs. The doping effects in the ferrocene-filled metallic and semiconducting SWCNTs were also investigated by XAS at the Fe edge. The XAS spectrum of the ferrocene-filled nanotubes was compared to the spectra of metallic Fe and pure ferrocene. Figure 20 shows the Fe2p3/2-Fe 3d XAS edges [67]. There are two different components resolved in the spectra of pure ferrocene and filled SWCNTs. The stronger absorption peak is positioned around 708.75 eV; it is blue shifted as compared to elemental Fe, where it is found at 708 eV. The strongest peak belongs to transitions to molecular orbitals (MO) of FeCp2 from Fe 3dxz and Figure 19. C 1s XAS spectra of 1.4 nm-diameter pristine (top panels) and ferrocene-filled (bottom panels) metallic, semiconducting (SC), and metallicity-mixed SWCNTs fitted with individual components. The experimental data are shown as black circles. The fit data for the metallicity-mixed, metallic, and semiconducting SWCNTs are shown in green, red, and blue, respectively. The data are fitted with the π*-peak (grey) and the peaks of individual vHs of metallic (M * 1 -M * 4 ) (red) and semiconducting (S * 1 -S * 5 ) SWCNTs (blue). In the filled SWCNTs, two peaks in the C 1s XAS response from pure ferrocene are indicated (dark red). In the case of the ferrocene-filled SWCNTs, there is a decrease in the intensity of the components of vHs of the metallic and semiconducting SWCNTs in comparison to the π*-peak, as well as a change in their relative intensities ( Figure 19) [67]. In the case of the metallic SWCNTs, the component of the M * 1 vHs in particular decreases in intensity as compared to the other ones, whereas the relative intensities of the components of all other vHs are kept constant. In the case of the semiconducting and metallicity-mixed SWCNTs, there is a slight decrease in the relative intensity of the components of the S * 1 and S * 2 vHs as compared to the component of the S * 3 vHs in the semiconducting SWCNTs and the component of the M * 1 vHs in the metallicity-mixed SWCNTs. In all spectra of the filled SWCNTs, two peaks in the C 1s XAS response from pure ferrocene are present. The authors of Ref. [67] explained the observed trends by the charge transfer from the encapsulated ferrocene to the nanotubes and hybridization between π-orbitals of the carbon in the SWCNTs and the molecules.
The doping effects in the ferrocene-filled metallic and semiconducting SWCNTs were also investigated by XAS at the Fe edge. The XAS spectrum of the ferrocene-filled nanotubes was compared to the spectra of metallic Fe and pure ferrocene. Figure 20 shows the Fe2p 3/2 -Fe 3d XAS edges [67]. There are two different components resolved in the spectra of pure ferrocene and filled SWCNTs. The stronger absorption peak is positioned around 708.75 eV; it is blue shifted as compared to elemental Fe, where it is found at 708 eV. The strongest peak belongs to transitions to molecular orbitals (MO) of FeCp 2 from Fe 3d xz and Fe 3d yz together with a small p*(Cp)-ligand contribution. The weaker peak at around 711.1 eV belongs to transitions to the p*(Cp) MOs from Fe 3d xy and Fe 3d x 2 −y 2 , which corresponds to a metal-to-ligand transition [67]. The red-shift of this peak to lower photon energies as compared to pure ferrocene i attributed to an increase in the electron-nuclear Coulomb attraction, which correspond to a higher valency state [67]. The effective valency of iron in the ferrocene-filled SWCNT is determined by the peak positions. The effective valencies of Fe in ferrocene inside me tallic and semiconducting SWCNTs are +2.4 and +2.3, respectively. The different valencie are illustrated in the schematic in Figure 21 [67].
These valencies of Fe in ferrocene inside SWCNTs are much bigger than the +2 va lency in pure ferrocene, where one electron is transferred onto each of the cyclopentadi enyl rings. This directly quantifies the n-doping of metallic and semiconducting SWCNT by the encapsulated ferrocene. The charge transfer is more effective for the ferrocene-filled The red-shift of this peak to lower photon energies as compared to pure ferrocene is attributed to an increase in the electron-nuclear Coulomb attraction, which corresponds to a higher valency state [67]. The effective valency of iron in the ferrocene-filled SWCNTs is determined by the peak positions. The effective valencies of Fe in ferrocene inside metallic and semiconducting SWCNTs are +2.4 and +2.3, respectively. The different valencies are illustrated in the schematic in Figure 21 [67].
are illustrated in the schematic in Figure 21 [67].
These valencies of Fe in ferrocene inside SWCNTs are much bigger than the +2 va lency in pure ferrocene, where one electron is transferred onto each of the cyclopentadi enyl rings. This directly quantifies the n-doping of metallic and semiconducting SWCNTs by the encapsulated ferrocene. The charge transfer is more effective for the ferrocene-filled metallic SWCNTs [67]. These valencies of Fe in ferrocene inside SWCNTs are much bigger than the +2 valency in pure ferrocene, where one electron is transferred onto each of the cyclopentadienyl rings. This directly quantifies the n-doping of metallic and semiconducting SWCNTs by the encapsulated ferrocene. The charge transfer is more effective for the ferrocene-filled metallic SWCNTs [67].
Investigation of Doping Effects on Metallicity-Mixed and Sorted SWCNTs upon Chemical Transformation of Encapsulated Molecules
Molecules can undergo chemical transformations inside the SWCNT channels, which leads to the modification of the electronic properties of nanotubes. The types of chemical reactions involved may include a simple thermal decomposition as well as more complex chemical transformations. The authors of Ref. [80] demonstrated a variety of chemical reactions that can be conducted in the interior of SWCNTs starting from Pt (II) acetylacetonate as precursor. SWCNTs are filled with Pt (II) acetylacetonate via the gas phase approach at 150 • C in high vacuum. Then, Pt (II) acetylacetonate can be decomposed at 500 • C with the formation of pure Pt clusters inside SWCNTs. The former react with iodine, leading to the formation of platinum iodide inside the SWCNTs. Pt (II) acetylacetonate can be inserted in the interior of nanotubes simultaneously with iodine at 150 • C in high vacuum. The thermal treatment of the mixture formed inside the SWCNT channels at 500 • C leads to the formation of platinum iodide. Finally, platinum iodide can react with sulfur at 550 • C to form platinum sulfide. The SWCNTs can also be filled with trans-bis(acetylacetonato)diiodoplatinum Pt(acac) 2 I 2 and bis(acetylacetonato)di-thiocyanatoplatinum Pt(acac) 2 (SCN) 2 via the gas phase method at 130-150 • C in high vacuum. The thermal treatment of the compounds inside the SWCNTs leads to the formation of platinum iodide and platinum sulfide, respectively. The authors of Ref. [80] studied the direction of electron transfer in the filled SWCNTs by tracing the position of the G-band peak in the Raman spectra. They concluded that there is no electron transfer in SWCNTs filled with Pt (II) acetylacetonate, platinum iodide, and platinum sulfide. In the case of SWCNTs filled with pure Pt and Pt(acac) 2 (SCN) 2 , electron transfer from the compounds to the SWCNTs, i.e., n-doping, was revealed. In the case of SWCNTs filled with Pt(acac) 2 I 2 , electron transfer from the SWCNTs to the compound, i.e., p-doping, was observed.
The most popular group of chemical reactions conducted inside molecule-filled SWC-NTs that lead to the modification of their electronic properties is the thermal decomposition with the formation of inner tubes. Annealing is a feasible way to control chemical transformations in the incorporated molecules and to control the effect of the encapsulated substance on the electronic properties of nanotubes. After the precursor molecules have decomposed, there is a local supply of carbon that can be gradually molded into inner nanotubes, which again actively participate in charge transfer. By choosing the precursor molecules and the temperature and duration of its processing, the nanochemical reaction can be fine-tuned for a stable ambipolar doping level in SWCNTs.
In Ref. [76], the thermal treatment of the nickelocene-filled semiconducting SWCNTs with a mean diameter of 1.7 nm at temperatures between 360 and 1200 • C caused a change in the doping level of nanotubes and even a switching of the doping type. The nickelocenefilled semiconducting SWCNTs were obtained by the density gradient ultracentrifugation of the filled SWCNTs, and, as a result of the centrifugation process, a semiconducting fraction was obtained in the bottom part of the centrifugation tube ( Figure 22) [76]. The thermal treatment of the nickelocene-filled semiconducting SWCNTs led to the chemical transformation of the nickelocene into nickel carbides and metallic nickel at low annealing temperatures (360-600 • C), the formation of inner tubes at temperatures higher than 600 • C, and the simultaneous evaporation of nickel from the nanotubes, which caused in unison the variation in the doping level of SWCNTs [76]. Figure 23 shows the C 1s XPS spectra of the nickelocene-filled semiconducting SWCNTs and the samples annealed at temperatures between 360 and 1200 °C [76]. The spectrum of the pristine SWCNTs includes a single C 1s peak. In the case of the nickelocene-filled SWCNTs, the C 1s peak is shifted toward higher binding energies, which cor responds to the upshift in the Fermi level of SWCNTs, i.e., n-doping. The shift in the C 1s peak increases at the annealing of the filled SWCNTs at 360 °C. At an increase in annealing temperature, the shift in the C 1s peak gradually decreases, but it stays positive until 600 °C. This means that the nickel carbides and metallic nickel formed as a result of the chem ical transformation of nickelocene cause n-doping in SWCNTs. At a further increase in annealing temperature, the C 1s peak shifts toward lower binding energies as compared to the nickelocene-filled nanotubes. This means that the inner tubes cause p-doping in the SWCNTs. The evaporation of nickel leaves empty DWCNTs, for which the p-doping leve is maximal [76]. Figure 23 shows the C 1s XPS spectra of the nickelocene-filled semiconducting SWC-NTs and the samples annealed at temperatures between 360 and 1200 • C [76]. The spectrum of the pristine SWCNTs includes a single C 1s peak. In the case of the nickelocene-filled SWCNTs, the C 1s peak is shifted toward higher binding energies, which corresponds to the upshift in the Fermi level of SWCNTs, i.e., n-doping. The shift in the C 1s peak increases at the annealing of the filled SWCNTs at 360 • C. At an increase in annealing temperature, the shift in the C 1s peak gradually decreases, but it stays positive until 600 • C. This means that the nickel carbides and metallic nickel formed as a result of the chemical transformation of nickelocene cause n-doping in SWCNTs. At a further increase in annealing temperature, the C 1s peak shifts toward lower binding energies as compared to the nickelocene-filled nanotubes. This means that the inner tubes cause p-doping in the SWCNTs. The evaporation of nickel leaves empty DWCNTs, for which the p-doping level is maximal [76].
responds to the upshift in the Fermi level of SWCNTs, i.e., n-doping. The shift in the C 1 peak increases at the annealing of the filled SWCNTs at 360 °C. At an increase in annealing temperature, the shift in the C 1s peak gradually decreases, but it stays positive until 60 °C. This means that the nickel carbides and metallic nickel formed as a result of the chem ical transformation of nickelocene cause n-doping in SWCNTs. At a further increase in annealing temperature, the C 1s peak shifts toward lower binding energies as compared to the nickelocene-filled nanotubes. This means that the inner tubes cause p-doping in th SWCNTs. The evaporation of nickel leaves empty DWCNTs, for which the p-doping leve is maximal [76]. Analogous changes in the doping level and doping type were revealed for metallicitymixed SWCNTs with mean diameters of 1.4 and 1.7 nm filled with nickelocene [73], ferrocene [61], and nickel (II) acetylacetonate [81] molecules.
Another demonstration of controlled ambipolar doping in SWCNTs was presented in Refs. [40,42]. Gd@C82-filled nanotubes were subjected to confined nanochemical reactions. The encapsulated endohedral fullerenes caused p-doping in the SWCNTs. The thermal treatment of the metallofullerene-filled SWCNTs led to DWCNTs filled with Gd nanowires, which was proved to cause strong n-doping in the nanotubes [40,42].
The annealing of cerocene-filled SWCNTs leads to an increase in n-doping level [77,78]. Cerocene filling in itself undergoes charge transfer, and the hosting SWCNTs are n-doped. The authors of Ref. [77] found that the thermal decomposition of cerocene inside the SWCNT channels and the subsequent growth of inner nanotubes led to an increased density of states at the Fermi level. The transition into a metallic state in cerium-containing semiconducting nanotubes left its signature in an increased screening of the photoexcited final state.
Discussion of the Modification of the Electronic Properties of Filled SWCNTs
The encapsulation of molecules inside metallicity-sorted and metallicity-mixed SWC-NTs with diameters of 1.4 and 1.7 nm resulted in large filling ratios. The characterization of the filled SWCNTs by OAS, XPS, and UPS showed that the incorporated organometallic molecules (NiCp 2 , CoCp 2 , FeCp 2 , CeCp 3 , Ni(acac) 2 ) cause n-doping in SWCNTs accompanied by charge transfer from the molecules to the nanotubes and an upshift in the Fermi level of SWCNTs by~0.1 eV.
The thermal treatment of SWCNTs filled with the organometallic molecules (NiCp 2 , CoCp2, FeCp2, CeCp3, Ni(acac)2) leads to variation in the doping level of nanotubes and even a switching of the doping type from n to p due to three overlapping processes: (i) the chemical transformation of molecules with the formation of metal carbides or pure metals, (ii) the formation of inner tubes, and (iii) the evaporation of metals from the nanotubes. Figure 24 shows the shift in the Fermi level of SWCNTs filled with nickelocene upon their vacuum annealing at temperatures between 250 and 1200 • C. In Figure 24a, the doping levels for different annealing temperatures are presented. In Figure 24b-d, the schematic band structures of the NiCp2-filled SWCNTs and annealed samples are demonstrated. The annealing of the NiCp2-filled SWCNTs, where an upshift in the Fermi level as compared to the pristine SWCNTs by 0.07 eV is observed (Figure 24b), at temperatures between 250 and 600 • C leads to n-doping in the SWCNTs accompanied by an increase in the Fermi level of SWCNTs by~0.05-0.2 eV (Figure 24c). This corresponds to the number of transferred electrons Ntotal (e − per carbon) ranging from 0.00013 to 0.00118 e − /C and the charge transfer density per tube length CT (e − Å −1 ) ranging from 0.0027 to 0.0240 e − /Å. The annealing of the filled SWCNTs at temperatures between 800 and 1200 • C results in pdoping in the SWCNTs accompanied by a lowering of the Fermi level of the SWCNTs bỹ 0.15-0.2 eV (Figure 24d), which corresponds to Ntotal ranging from −0.00118 to −0.00078 to e − /C and CT from −0.0240 to −0.0160 to e − /Å. zation of the filled SWCNTs by OAS, XPS, and UPS showed that the incorporated organometallic molecules (NiCp2, CoCp2, FeCp2, CeCp3, Ni(acac)2) cause n-doping in SWCNTs accompanied by charge transfer from the molecules to the nanotubes and an upshift in the Fermi level of SWCNTs by ~0.1 eV.
The thermal treatment of SWCNTs filled with the organometallic molecules (NiCp2, CoCp2, FeCp2, CeCp3, Ni(acac)2) leads to variation in the doping level of nanotubes and even a switching of the doping type from n to p due to three overlapping processes: (i) the chemical transformation of molecules with the formation of metal carbides or pure metals, (ii) the formation of inner tubes, and (iii) the evaporation of metals from the nanotubes. Figure 24 shows the shift in the Fermi level of SWCNTs filled with nickelocene upon their vacuum annealing at temperatures between 250 and 1200 °C. In Figure 24a, the doping levels for different annealing temperatures are presented. In Figure 24b-d, the schematic band structures of the NiCp2-filled SWCNTs and annealed samples are demonstrated. The annealing of the NiCp2-filled SWCNTs, where an upshift in the Fermi level as compared to the pristine SWCNTs by 0.07 eV is observed (Figure 24b), at temperatures between 250 and 600 °С leads to n-doping in the SWCNTs accompanied by an increase in the Fermi level of SWCNTs by ~0.05-0.2 eV (Figure 24c). This corresponds to the number of transferred electrons Ntotal (e − per carbon) ranging from 0.00013 to 0.00118 e − /C and the charge transfer density per tube length CT (e − Å −1 ) ranging from 0.0027 to 0.0240 e − /Å. The annealing of the filled SWCNTs at temperatures between 800 and 1200 °С results in p-doping in the SWCNTs accompanied by a lowering of the Fermi level of the SWCNTs by ~0.15-0.2 eV (Figure 24d), which corresponds to Ntotal ranging from −0.00118 to −0.00078 to e − /C and CT from −0.0240 to −0.0160 to e − /Å.
Electrochemistry
The authors of Ref. [172] investigated the electrochemical properties of redox active guest-molecules (cobaltocene and methylated ferrocene derivatives) inside 1.4 nm-diameter SWCNTs. It was shown that the filling inside the SWCNTs modifies the oxidation state of the metallocenes. The authors developed the technique for quantifying the electronic doping of SWCNTs using electrochemistry.
Cyclic voltammetry measurements were performed [172]. The Fermi level of the SWC-NTs shifted in metallocene-filled carbon nanotubes. As a result, when the electric potential was applied, higher or lower energy levels were depleted in the metallocene@SWCNT hybrids. The first case is related to the n-dopant metallocene molecule. The second case is attributed to the p-dopant molecule.
Combined with density functional theory calculations, coulometry provides an accurate indication of n-/p-doping in SWCNTs [172]. The filling of redox active molecules inside carbon nanotubes leads to hybrids with complex, interesting electrochemical properties, which are different from the properties of individual carbon nanotubes and metallocenes.
Knowledge about the correlation between electron transfer, the diameter of nanotubes, and the metallicity type of carbon nanotubes leads to a better understanding of hostguest interactions. This opens new roads to tailoring the oxidation state of metallocenes and mofidications in the electronic properties of carbon nanotubes in complex intriguing hybrid nanostructures [172].
Thermoelectric Power Generation
The combination of mechanical strength, low thermal conductivity, and high electrical conductivity renders filled SWCNTs a very promising material for efficient flexible lightweight thermoelectric devices.
The authors of Ref. [3] realized a flexible p-n type thermoelectric device that consisted of films of naturally p-doped empty SWCNTs and n-doped CoCp 2 -filled SWCNTs (CoCp 2 @SWCNT). The highly efficient power generation achieved approached the theoretical calculated limit and the device performed flawlessly without any air-protective coating. Two different freestanding films were fabricated and investigated. The first film consisted of CoCp 2 @SWCNT (Figure 25a) and the second film was made from empty SWCNTs. The scanning electron microscopy (SEM) micrograph in Figure 25b shows that the typical diameters of the bundles of CoCp 2 @SWCNT are in the range between 10 and 200 nm and their lengths exceed 5 µm. As shown in Figure 25c, there was no significant change in the normalized sheet resistance [3]. SWCNTs showed a positive Seebeck coefficient of 45.3 μV K −1 at the same temperature, which corresponds to p-type nature (Figure 25e). These additional electrons were provided by charge transfer from the encapsulated molecules to the host SWCNTs [3]. The higher electrical conductivity led to a significantly increased power factor of the CoCp2@SWCNT film (75.4 μW m −1 K −2 at 320 K) as compared to the value of the empty SWCNT film (Figure 25f). In Ref. [3], the films of the CoCp2@SWCNT and empty SWCNTs as the n-type and ptype semiconducting materials were used to fabricate a p-shaped thermoelectric device. The two films were electrically connected by a thin Ni plate (Figure 25g,h). At a temperature difference of 10 K, the value amounted to 0.67 mV. This demonstrated device performance was very close to the expected value (0.70 mV) based on the Seebeck coefficients of the two films (SWCNTs: ~30 μV K −1 , CoCp2@SWCNT: ~−40 μV K −1 ) [3]. The deviation from the calculated values observed at large temperature gradients was attributed to thermallyexcited electrons and contact resistance (Figure 25i). However, good reversibility by decreasing the temperature gradient was observed, which was tested by applying a temperature difference of 9 K after operating the device at ΔT of 20 K [3]. (c) Normalized sheet resistance of CoCp 2 @SWCNT films plotted after repeated bending (the bending radius is 3.5 mm). R 0 is initial resistivity, R is resistivity after given bending cycles. The electrical conductivity (d), Seebeck coefficient (e) and power factor (f) of the empty (black circles) and CoCp 2 -filled SWCNT films (red squares) at different temperatures. (g) The schematic of the setup for measuring thermoelectric power generation that consisted of the films of the CoCp 2 @SWCNT and empty SWCNTs as the n-type and p-type semiconducting materials. (h) The image of the thermoelectric device. (i) Measured (red circle) and calculated (black square) voltages (V) generated from the thermoelectric device as a function of the temperature gradient (∆T). Reproduced from [3]. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License.
The electrical conductivity and Seebeck coefficients of the two types of films were measured at different temperatures (Figure 25d,e). Filling with CoCp 2 led to an increase in electrical conductivity by one order of magnitude (from 4450 S/m for the empty SWCNTs to 43200 S/m for the filled SWCNTs at 320 K). The electrical conductivity in the films of the empty and filled SWCNTs was also found to be constant in the temperature range between 310 K and 420 K (Figure 25d).
The Seebeck coefficient of the CoCp 2 @SWCNT film amounted to a negative value of −41.8 µV K −1 at 320 K, which was assigned to an n-type semiconductor. The empty SWCNTs showed a positive Seebeck coefficient of 45.3 µV K −1 at the same temperature, which corresponds to p-type nature (Figure 25e). These additional electrons were provided by charge transfer from the encapsulated molecules to the host SWCNTs [3]. The higher electrical conductivity led to a significantly increased power factor of the CoCp 2 @SWCNT film (75.4 µW m −1 K −2 at 320 K) as compared to the value of the empty SWCNT film (Figure 25f).
In Ref. [3], the films of the CoCp 2 @SWCNT and empty SWCNTs as the n-type and p-type semiconducting materials were used to fabricate a p-shaped thermoelectric device. The two films were electrically connected by a thin Ni plate (Figure 25g,h). At a temperature difference of 10 K, the value amounted to 0.67 mV. This demonstrated device performance was very close to the expected value (0.70 mV) based on the Seebeck coefficients of the two films (SWCNTs:~30 µV K −1 , CoCp 2 @SWCNT:~−40 µV K −1 ) [3]. The deviation from the calculated values observed at large temperature gradients was attributed to thermally-excited electrons and contact resistance (Figure 25i). However, good reversibility by decreasing the temperature gradient was observed, which was tested by applying a temperature difference of 9 K after operating the device at ∆T of 20 K [3].
Sensors
The working principle of SWCNT-based gas sensors is based on reversible changes in the electric properties of bundles or isolated SWCNTs when they are exposed to gases. The authors of Ref. [173] demonstrated the application potential of SWCNTs in a sensor for NO2, which is a well-known highly toxic air pollutant, capable of recovering at ambient temperature. Metallicity-sorted semiconducting and metallic SWCNTs filled with nickel (II) acetylacetonate molecules as well as nickel clusters obtained by heating at 500 • C were tested. In the case of Ni cluster-filled SWCNT, the system even fully recovered at room temperature [173]. Figure 26a,b shows the C 1s core level photoemission spectra from the semiconducting and metallic SWCNTs filled with nickel (II) acetylacetonate molecules (Ni-acac@SC-SWCNTs and Ni-acac@M-SWCNTs, respectively) and nickel clusters (Ni-nc@SC-SWCNTs and Ni-nc@M-SWCNTs, respectively) before and after exposure to 80 L of NO2 (1 L ≈ 1.33 · 10 −6 mbar · s) [173]. The spectra are fitted with individual components. The dominant component around 284.5 eV in all spectra corresponds to the main carbon peak. In the case of the nickel (II) acetylacetonate filling, there is a grey shaded broad peak at slightly higher energy values as compared to the main C 1s peak, which arises from the carbon atoms associated with the nickel (II) acetylacetonate molecules. In the spectrum of nickel cluster-filled SWCNTs, this component has lower binding energy and smaller intensity, which can be associated with nickel atoms bonded to carbon atoms [173]. There are two more components at higher binding energies corresponding to two types of carbon-oxygen bonds-ketene groups (C=O) and carboxylate groups (O-C=O). The carboxylate peak is directly attributed to the nickel (II) acetylacetonate filling. The ketene groups are due to the partial oxidation of the SWCNTs during the sample preparation [173]. These components have large intensity in the spectra of the molecule-filled SWCNTs after exposure to NO2. It is also visible that these components are more pronounced for the semiconducting hosts after NO2 exposure. This suggests that the interior molecule filling increases the interaction with the external surrounding NO2. In the spectra of nickel cluster-filled SWCNTs, the carboxylate group is not observable anymore for either semiconducting or metallic hosts. It should be noted that the intensity variations and shifts are more pronounced in the semiconducting hosts, which indicates their higher sensitivity [173]. Subsequently, the samples were in situ annealed at 500 °C to transform the nickel (II acetylacetonate molecules into nickel nanoclusters inside the SWCNTs. The nickel cluster filled SWCNTs were studied at 100 K without exposure to NO2. The corresponding meas urements are pictured as the starting point (stage III) in the bottom panel of Figure 26 [173]. The C 1s core level PES peaks of the Ni-filled semiconducting and metallic SWCNT have slightly different positions, which testifies to their slightly different reactivity. I stage IV, after exposure to 80 L of NO2, the C 1s binding energies both downshifted an coincided. This means that, although the Ni-filled semiconducting and metallic SWCNT did not necessarily go through the same reaction pathway, they had a similar reactivit when exposed to NO2 [173]. When comparing the effects of NO2 exposure to nickel (II acetylacetonate-filled (the stage II) and Ni cluster-filled semiconducting SWCNTs (th stage IV), one can immediately see that the molecule-filled SWCNTs were more reactiv toward NO2, or that a physisorption process must have occurred. In such a case, the shif Figure 26. C 1s core level photoemission spectra of semiconducting (a) and metallic SWCNTs (b) filled with nickel (II) acetylacetonate molecules (Ni-acac@SC-SWCNTs and Ni-acac@M-SWCNTs, respectively) and nickel clusters (Ni-nc@SC-SWCNTs and Ni-nc@M-SWCNTs, respectively) before and after exposure to 80 L of NO 2 . The spectra are fitted with individual components of sp 2 carbon, nickel (II) acetylacetonate filling (grey shaded), ketene group (C=O), carboxylate group (O-C=O), and π-π* interactions, which are indicated by arrows. Molecular models are presented as the inset of each plot. (c) The shift of the main C 1s peak of the semiconducting (SC, blue) versus metallic (M, red) hosts with different fillings at the stages I to V. (d) The C 1s binding energy shift in the Ni-filled semiconducting SWCNTs with increasing dose of NO 2 over time at room temperature. Reproduced from [173]. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence.
The authors of Ref. [173] studied selectivity and sensitivity by exposing the semiconducting and metallic SWCNTs filled with nickel (II) acetylacetonate molecules and nickel clusters to NO 2 at different temperatures. The effects of the gas dosing level, the filler states, and the temperature on the C 1s core level PES spectra were investigated in situ. The in situ series for a sample were divided into five subsequent stages, I-V. In Figure 26c, the position of the main C 1s peak is traced for the filled semiconducting and metallic SWCNTs in these stages [173]. In stage I, the initial binding energies for the filled semiconducting and metallic SWCNTs cooled down to 100 K were~284.77 and~284.79 eV, respectively. In stage II, after exposure to 80 L of NO 2 , the C 1s binding energies downshifted by 0.26 eV and 0.23 eV for the filled semiconducting and metallic SWCNTs, respectively. This was explained by the fact that the adsorbed NO 2 molecules acted as electron acceptors and caused a charge transfer. The samples were then left to recover without inducing changes externally. The slow desorption of the NO 2 molecules led to the shift in the C 1s backwards (Figure 26c, top panel). It was accelerated by heating at an intermediate temperature of 400 • C to complete the desorption procedure and to obtain an adsorbent-free material [173].
Subsequently, the samples were in situ annealed at 500 • C to transform the nickel (II) acetylacetonate molecules into nickel nanoclusters inside the SWCNTs. The nickel cluster-filled SWCNTs were studied at 100 K without exposure to NO 2 . The corresponding measurements are pictured as the starting point (stage III) in the bottom panel of Figure 26c [173]. The C 1s core level PES peaks of the Ni-filled semiconducting and metallic SWCNTs have slightly different positions, which testifies to their slightly different reactivity. In stage IV, after exposure to 80 L of NO 2 , the C 1s binding energies both downshifted and coincided. This means that, although the Ni-filled semiconducting and metallic SWCNTs did not necessarily go through the same reaction pathway, they had a similar reactivity when exposed to NO 2 [173]. When comparing the effects of NO 2 exposure to nickel (II) acetylacetonate-filled (the stage II) and Ni cluster-filled semiconducting SWCNTs (the stage IV), one can immediately see that the molecule-filled SWCNTs were more reactive toward NO 2 , or that a physisorption process must have occurred. In such a case, the shift is assigned to charge transfer between the SWCNTs and the filler [173]. Upon the recovery of the system to reach ambient temperature at stage V in relation to III, the C 1s line position is fully restored for the Ni-filled semiconducting SWCNTs as compared to the counterpart with the metallic host tubes (Figure 26c, bottom panel). Although both Ni cluster-filled SWCNT samples reached a better recovery than the nickel (II) acetylacetonate moleculefilled SWCNTs, the authors of Ref. [173] concluded that Ni-filled metallic SWCNTs are more prone to chemisorption during exposure, while NO 2 is mainly physisorbed on the Ni-filled semiconducting SWCNTs.
Full recovery at room temperature is a key challenge for sensors. In Ref. [173], timeresolved photoemission studies on the Ni-filled semiconducting SWCNTs were performed in order to trace the C 1s peak position upon exposure to NO 2 at room temperature. The experiments were carried out increasing the dose of NO 2 over 50 min. As it is shown in Figure 26d, the C 1s spectra of the Ni-filled semiconducting SWCNTs underwent a constant binding energy shift with increasing NO 2 dosage [173]. The observed trend is consistent with the effects revealed in the above-discussed stages I to V (Figure 26c). These in situ PES studies of filled SWCNTs, which revealed remarkable results of sensitivity and recovery at ambient temperature, provide motivation for testing SWCNTs filled with appropriate substances in sensors for other reactive and poisonous gases with controlled and increased sensitivity and selectivity at room temperature.
Magnetic Recording
The encapsulation of substances inside SWCNTs allows for nanomagnets that outperform their bulky counterparts and control the magnetic properties of these nanohybrids to be obtained. The SWCNTs contribute their own properties to the nanohybrid, which facilitates applications of these nanomagnets in magnetorecording devices [174][175][176][177][178][179].
The authors of Ref. [174] encapsulated iron nanowires inside SWCNTs and revealed ferromagnetic behavior in the filled SWCNTs even at room temperature. Figure 27a shows a hysteresis loop for the Fe-filled SWCNTs at room temperature [174]. The hysteresis loop clearly evidences that the encapsulated nanowires are ferromagnetic at 300 K with a coercivity of 18 mT. The carbon shell provides effective protection from ambient oxidation [174]. In Ref. [175], behavior was ferromagnetic, whereas there was superparamagnetism at low temperatures. SWCNTs is very close to that of bulk anhydrous ErCl3. This means that the magnetizatio of Er 3+ ions is independent of their coordination environment [177].
Conclusions, Perspectives
This review was dedicated to the electronic properties of metallocene-filled SWCNTs. It was confirmed that the ionization energy or electron affinity of molecules o the work function of elementary substances and inorganic compounds are the most im portant predictors of the filler's effect on the electronic properties. They decide the sig and doping type as well as the magnitude of Fermi level shifts. The doping effect als depends on the diameter and metallicity type of the hosting SWCNT. Additionally, th filling ratio (filled vs. empty sections) directly influences the electronic properties of bul filled SWCNT. [174]. (b) Left: X-ray diffraction profiles for the pristine SWCNTs (grey) and SWCNTs filled with 3 nm (blue), 7 nm (green) and 10 nm (red) nickel clusters, q is the scattering vector. The position of carbon peak and (111) fcc nickel peak are denoted by dashed vertical lines. Right: The magnetization curves for the 3 nm, 7 nm and 10 nm nickel clusters in SWCNTs and bulk nickel measured at 5 K by SQUID. (c) The temperature dependence of the magnetization measured upon zero-field (ZFC, blue curves) and field cooling (FC, green curves) by SQUID and normalized to the XMCD data at high temperatures. Superposed onto the SQUID data are the Ni 3d magnetic moments derived from XMCD data plotted versus temperature (open rectangles). The inset shows the blocking temperature (T B ) plotted versus the reciprocal of the length (L) of the nickel cluster. Reproduced from [176]. This work is licensed under a Creative Commons Attribution 4.0 International License.
Nickel nanowires with face centered cubic (fcc) structure and different sizes were formed by the thermal treatment of Ni (II) acetylacetonate in the interior of SWCNTs [176]. Its net magnetization was minimized by cooling in absence of an external magnetic field. The finite coercivity and superparamagnetic blocking temperature were found to scale with the nickel cluster size. Figure 27b shows the normalized bulk magnetization isotherms measured at 5 K by a superconducting quantum interference device (SQUID) (Figure 27b, right) for nickel clusters encapsulated in SWCNTs with mean cluster sizes of~3 nm, 7 nm, and 10 nm evaluated from X-ray diffraction (XRD) measurements (Figure 27b, left) [176]. Upon cooling in zero field, spins of small magnetic domains are frozen to form a spin glass state with a very small net magnetization. Upon cooling in a finite field, spins are aligned, resulting in a larger net magnetization [176]. The blocking temperature TB for 3 nm, 7 nm, and 10 nm clusters equals 18, 40, and 42 K, respectively. It is inversely scaled by the cluster length (L) as illustrated in the inset in Figure 27c [176]. While SQUID is a macroscopic bulk method, XMCD is more sensitive to the nickel atoms on the surface of the clusters, because of the mean free path of low energy electrons. This applies even for the clusters with a size of 1-2 nm. The fluctuation is more visible in smaller nickel clusters. This is a manifestation of reduced dimensionality (Figure 27c) [176]. Encapsulated single-domain metal clusters have a protective carbon shell and are not affected by environmental factors. As a result, the nanohybrid can function as a stable hard magnet [176].
In Ref. [177], the magnetic susceptibility of ErCl 3 -filled SWCNTs was measured by SQUID. The magnetization of the ErCl 3 -filled SWCNTs was found to be larger than that of the purified pristine SWCNTs. The magnetization behavior of the ErCl 3 -filled SWCNTs can be fitted by the Curie-Weiss law. This value is expected from the 4f 11 electronic configuration of Er 3+ ions [177]. The observed magnetization behavior of the ErCl 3 -filled SWCNTs is very close to that of bulk anhydrous ErCl 3 . This means that the magnetization of Er 3+ ions is independent of their coordination environment [177].
Conclusions, Perspectives
This review was dedicated to the electronic properties of metallocene-filled SWCNTs. It was confirmed that the ionization energy or electron affinity of molecules or the work function of elementary substances and inorganic compounds are the most important predictors of the filler's effect on the electronic properties. They decide the sign and doping type as well as the magnitude of Fermi level shifts. The doping effect also depends on the diameter and metallicity type of the hosting SWCNT. Additionally, the filling ratio (filled vs. empty sections) directly influences the electronic properties of bulk filled SWCNT.
The growth of carbon nanotubes depends on the temperature and time, as well as on the metal catalyst and precursor. The growth rate depends on the temperature and the metal catalyst type, as well as on the diffusion rate of metal and carbon. The activation energy depends on the tube diameter and chiral angle because of structural differences between the catalyst and the carbon nanotube.
The structure of metallocene-filled SWCNTs continues to attract our interest. The development of methods of microscopy allows for the precise determination of the structure of molecules and the refinement of the molecular structure of all known metallocenes. The thermal treatment of metallocene-filled SWCNTs was first shown to lead to the formation of DWCNTs in the work of Japanese scientist Hidetsugu Shiozawa [60].
Metallocenes are a promising precursor for the synthesis of metal-filled multi-walled carbon nanotubes and SWCNTs. We demonstrated brilliant works on the synthesis of metal-filled carbon nanotubes, and investigations of the growth processes of advanced nanostructures. Such works are needed to form a fundamental insight into the chemistry of carbon nanostructures with simple chemical methods.
Investigations of the growth kinetics of carbon nanotubes develop in the direction of environmental TEM studies, in order to prove the kinetics with visual TEM images at all stages of the growth of carbon nanotubes. The experiments in the microscopes are very promising because, as shown in Section 2, they allow for the growth kinetics to be studied with a resolution of less than seconds.
The development of methods of analysis of the properties of filled carbon nanotubes allows for studying in detail the samples of metallicity sorted, single-chirality SWCNTs (which are need for applications), the precise applications of properties in devices, and the construction of devices on the basis of metallocene-filled SWCNTs. The doping effects in metallicity sorted and single-chirality SWCNTs should be studied.
Among the applications of metallocene-filled SWCNTs, those applications in electrochemistry are the most promising and recent. The thermal treatment of metallocenes leads to the formation of metal clusters. The measurement of spectra of metal clusters inside SWCNTs under electrochemical charging allows for the creation of devices with precise properties based on carbon nanotubes. The applications of metallocene-filled SWCNTs in electrochemistry open the roads to the creation of devices of metallocene-filled SWCNTs with known electrochemical properties combined with thermoelectric power generation devices, sensors, and magnetic recording devices. These devices are interesting now to investigate in projects and works. We hope that our work on the electrochemistry of carbon material is estimated positively and warmly. We do the best for the development of this topic toward the best prizes in the world [180]. We hope that referees of the project within the European Union have a warm attitude to us.
|
2023-02-22T16:04:39.696Z
|
2023-02-01T00:00:00.000
|
{
"year": 2023,
"sha1": "2b8f9256a46106ba85a685361c45ef00efab3766",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "269a6253a308e9619e01e0c970b4ed0f36b0d111",
"s2fieldsofstudy": [
"Materials Science",
"Physics",
"Engineering"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
228963583
|
pes2o/s2orc
|
v3-fos-license
|
Exploring the resident gut microbiota of stranded odontocetes: high similarities between two dolphin species Tursiops truncatus and Stenella coeruleoalba
Abstract The evaluation of symbiotic microbial communities occurring in the intestinal tract of animals has received great interest in recent years. However, little is known about gut microbial communities in cetaceans, despite their relevance in the ecology of marine communities. Here, we report an investigation using 16S rRNA gene amplicon sequencing of the resident gut microbiota of the two cetacean species Stenella coeruleoalba and Tursiops truncatus by sampling intestinal mucosa from specimens retrieved stranded along the Tyrrhenian coast of Tuscany (Italy). We found an abundance of members from Clostridiaceae and Fusobacteriaceae, which in total accounted for more than 50% of reads, in agreement with gut microbiota composition of other carnivorous mammals. Probably due to the limited number of samples available, sex, preservation status and also species, did not correlate with overall differences in the microbiota. Indeed, a high similarity of the taxonomic (family-level) composition between the gut microbiota of the two species was found. However, Pedobacter spp. was found abundant in amplicon sequencing libraries from S. coeruleoalba, while clostridia were more abundant from T. truncatus samples. Our results shed some light on the gut microbiota composition of two dolphin (S. coeruleoalba and T. truncatus) species, with specimens collected in the wild. Studies with a larger number of individuals are now needed to confirm these first results and evaluate the interspecific differences in relation to sex and age.
Investigation of the intestinal microbial content of threatened or vulnerable species may constitute an effective monitoring instrument that could reveal the presence of pathogenic microorganisms and indicate environmental health status (see Godoy-Vitorino et al., 2017). Marine cetaceans are considered particularly sensitive to stressors present in their environment and several species have suffered drastic die-offs in the last decades due to human pressures. For example, both the striped dolphin Stenella coeruleoalba (Meyen, 1833) (http://www. iucnredlist.org/details/16674437/0) and the bottlenose dolphin Tursiops truncatus (Montagu, 1821) (http://www.iucnredlist.org/details/16674437/0) have been subjected to anthropogenic threats.
Studies on the gut microbiota communities of marine mammals have been performed (Nelson et al., 2013) and research on whales, some dolphin and porpoise species is in the literature (see for instance, Wan et al., 2018;Kim et al., 2019;Miller et al., 2020, Robles-Malagamba et al., 2020. These studies showed a dominance of members from the phylum Firmicutes and for marine carnivores a high representation of Fusobacteria (Nelson et al., 2013). Within the phylum Proteobacteria, Gammaproteobacteria were particularly abundant, including members of the genus Halomonas which are thought to play a role in their hosts' digestive and immune systems (Wan et al., 2018). For the dolphin species T. truncatus and S. coeruleoalba few reports are present, and mainly the enteric microbial composition of individuals maintained in captivity has been investigated (Soverini et al., 2016;Suzuki et al., 2019). Recently, analysis of swabs taken from different body parts of the same species has been reported (Robles-Malagamba et al., 2020). In a few cases the gut microbiota from wild dolphins has been investigated (Bik et al., 2016;Godoy-Vitorino et al., 2017). These reports showed that the gut microbiota of animals from natural habitats is different from that of animals kept in captivity. However, analyses on wild animals are difficult and only very recently have data on wild animals been published (Robles-Malagamba et al., 2020). Mainly stranded (in most case dead) animals are analysed, questioning the relevance of such results for inferring the gut microbiota in normal conditions (Godoy-Vitorino et al., 2017).
The aim of this work was to characterize the associated gut microbiota from wild individuals of S. coeruleoalba and T. truncatus stranded along the coast of the Tyrrhenian Sea (Tuscany, Italy), to shed, with the precaution due to sampling stranded animals, some more light on the gut microbiota in wild conditions for such dolphin species.
Sampling and sequence production
Sections of intestine (colon) of S. coeruleoalba and T. truncatus were removed from adult individuals found stranded along the Tyrrhenian coast in the Tuscany region (Italy) and collected in the years 2014-2017 in centres associated with the network of Tuscan Observatory for Biodiversity (OTC centres of the Regione Toscana). Colon sections were surgically taken in ARPAT (Agenzia Regionale per la Protezione Ambientale della Toscana, Livorno, Italy) premises, immediately after delivery of stranded animals, under sterile conditions. For each animal, three samples of ∼5 cm in length of intestine were taken and pooled in a composite sample, representing the single animal (thereafter defined as 'specimen'). A total of 12 specimens (six belonging to S. coeruleoalba and six to T truncatus) were sampled immediately after their retrieval (Table 1). For each individual the sex (M or F) and preservation status were registered in order to assess the eventual independence of the gut microbial composition of these parameters. The preservation status of individuals was evaluated in accordance with standard guidelines for stranded cetaceans (Geraci & Lounsbury, 2005) by assigning to each carcass a numeric value ranging from 2 to 5 (2: fresh carcass, <24 h after death, normal appearance, minimal external changes, no odour, minimum dehydration and wrinkling of skin, eyes, membranes and mucous membranes, normal eyes, no swelling carcass, not protruded tongue and penis; 3: carcass in moderate decomposition, intact carcass, evident bulge, protruded tongue and penis, desquamated skin, delicate odour, still wet membranes and mucous membranes, sunken eyes; 4: carcass in an advanced state of decomposition, the carcass can be intact, but more frequently collapsed; desquamated skin, smell strong and unpleasant, altered internal organs, sunken or missing eyes; 5: mummified carcass or skeleton remains, carcass often dried with dehydrated skin stretched over the bones, often missing organs).
All samples immediately after collection were stored at −20°C until the extraction of DNA (which was done within 1 month from the collection).
Raw data processing and statistical analyses
Illumina sequences were clustered into Operational Taxonomic Units (OTUs) following the classical UPARSE pipeline (Edgar, 2013) as previously described in Abdelrhman et al. (2017a). Sequences were pre-processed with StreamingTrim (Bacci et al., 2014) in order to remove low quality nucleotides which might interfere with downstream analysis. PANDAseq assembler (Masella et al., 2012) was used for merging paired-reads into fullamplicon sequences. Singletons, namely sequences found only one time in all samples, were removed before the OTU clustering step that was performed using an identity threshold of 97% in UPARSE ('cluster_otus' command). Putative chimeric sequences were removed during the clustering step by UPARSE and no additional removal was conducted. A single representative sequence has been chosen from each cluster and taxonomically annotated using the SINA standalone classifier in combination with the 'Ref NR 99' database (Pruesse et al., 2012). All steps were implemented with an in-house pipeline available at https://github.com/ GiBacci/o2tab. Rarefaction analysis was carried out with Table S1). Good's estimator (Good, 1953) was used to calculate the percentage of coverage. In order to inspect eventual differences in the OTUs distribution between T. truncatus and S. coeruleoalba, a Detrended Correspondence Analysis (DCA) was performed using the R package 'phyloseq' (McMurdie & Holmes, 2013). Moreover, further investigations were conducted to assess whether sex and preservation status affected the OTUs distribution by means of a PERMANOVA as implemented in R (Hoffman & Schadt, 2016) and LDA Effect Size (LEfSe) (Segata et al., 2011). Specific differences in community composition were determined using similarity percentage (SIMPER) analysis as done in Past 3 software (Hammer et al., 2001).
Links to deposited data
The sequences dataset was deposited in the SRA database under the BioProject PRJNA473403.
Results
The sequencing of 16S rRNA genes yielded a total of 7483-64,067 reads per sample (Supplementary Table S1). Rarefaction analyses (Supplementary Figure S1) showed that most samples were satisfactorily sampled (Good's coverage 99.99 ± 0.01). All samples were used for the following analyses.
After assigning reads to Operating Taxonomic Units (OTUs, 97% sequence similarity) and removing OTUs not assigned or assigned to Eukaryotes, a total number of 270 OTUs assigned to the bacterial taxonomy were detected (Supplementary Table S2). The OTU matrix was very sparse, with many OTUs represented in a few specimens only, indicating a high heterogeneity of OTUs abundance among specimens. The number of OTUs per specimen (Table 1) ranged from 17 to 102 and were significantly different between S. coeruleoalba and T. truncatus (Table 1) (t-test P < 0.05). The other diversity indices were not different between species. No significant differences with respect to sex or preservation status were found.
DCA and UPGMA clustering revealed a general lack of separation of OTUs representation between T. truncatus and S. coeruleoalba ( Figure 1). Indeed, either a variance partition or a PERMANOVA analysis did not find significant differences in relation to species and on the whole dataset in relation to preservation status, but a slightly significant effect of preservation status on S. coeruleoalba was found (Supplementary Table S3). An LDA Effect Size (LEfSe) (Segata et al., 2011) confirmed a lack of significant separation between species and among preservation statuses 3. Taxonomic composition at the family level of the gut microbiota of S. coeruleoalba (black bars) and T. truncatus (grey bars). The occurrence of discrete taxa is quoted as percentage abundance. Only families contributing to at least 5% of the total microbial communities in one of either species are reported. Table 2. SIMPER analysis on genera representation. The differences with respect to conservation status on gut microbiotas were inspected separately for S. coeruleoalba (a) and T. truncatus (b). In (c) the differences between S. coeruleoalba and T. truncatus considering samples in good conservation status only are reported (data not shown). This could be due to the lack of sharing of the highly abundant OTUs (Supplementary Figure S2). On the overall dataset, taxonomic compositions of the two species are very similar to each other. At the level of phylum, Firmicutes (mean relative abundance 60-62%), Proteobacteria (14-17%) and Fusobacteria (12-22%) dominated the gut microbiota ecosystem of both T. truncatus and S. coeruleoalba (Figure 2). Clostridia accounted for nearly 50% at the class level. The dominant families (Figure 3) were Clostridiaceae 1 (36.0% in S. coeruleoalba and 38.7% in T. truncatus) and Fusobacteriaceae (16.2% in S. coeruleoalba and 21.9% in T. truncatus) contributing in total to the 52.2% and the 60.6% of the gut microbial ecosystem of S. coeruleoalba and T. truncatus respectively. Other mainly representative families are Sphingobacteriaceae (9.5%) and Vibrionaceae (9.4%) in S. coeruleoalba and Moraxellaceae (7.6%) and Family XI (5.6%) in T. truncatus. When considering T. truncatus and S. coeruleoalba datasets separately, a differential abundance of taxa (meaning relative abundance in amplicon sequencing libraries) was found in relation to a comparison between good and bad preservation statuses (2 vs 3 + 4). Indeed, under SIMPER analysis samples of S. coeruleoalba with good preservation status were more abundant in members of Pedobacter, Photobacterium and Paeniclostridium, while poor preservation status samples had higher numbers of clostridia (Table 2). Similar differential abundance of clostridia associated with samples with poor preservation status was found for T. truncatus also ( Table 2). The same approach was used to inspect possible differential occurrence of taxa between T. truncatus and S. coeruleoalba individuals in good preservation status ( Table 2). Most of the differences were due to clostridia (with two groups being more abundant in T. truncatus and S. coeruleoalba, respectively), Pedobacter, Photobacterium and Paeniclostridium. However, due to the extremely limited number of samples, these results should be treated with great caution.
Discussion
Results obtained in our work pointed out high similarities in the gut microbial composition between the cetacean species T. truncatus and S. coeruleoalba. The slight (not statistically significant) differences between the two species could be due to the diversity in their prey consumption, putatively reflecting different metabolic necessities. Indeed, it has been shown that although both cetaceans exhibit a piscivorous dietary regime, they feed on prey partially different in relation to their discrete sea habitats (see Scuderi et al., 2011). However, the gut microbiota taxonomic composition revealed in our T. truncatus specimens partially differs from that identified in previous works (Bik et al., 2016;Soverini et al., 2016). Besides bacterial taxa typically occurring in the gut of carnivorous species (i.e. Clostridiaceae, Fusobacteriaceae and Peptostreptococcaceae) (Nelson et al., 2013) that were found in our investigations, previous works on animals in captivity detected a higher presence of members of Staphylococcaceae and Lactobacillaceae (Soverini et al., 2016) (here 13% vs <5%, see Supplementary Table S2). Conversely, in a recent analysis on the gut microbiota of captive T. truncatus in aquaria in Japan (Suzuki et al., 2019), abundance of Fusobacteriaceae, Peptostreptococcaceae and Vibrionaceae was found as in our analysis, reflecting the typical taxonomic composition of carnivorous species in the wild. However, the discordance between studies from animals in aquaria suggests that local dietary supplements/condition may bias the estimates on normal gut microbiota composition (as for instance in individuals analysed in Bik et al., 2016;Soverini et al., 2016). Recently, in agreement with our report, analyses of faecal samples from free T. truncatus (Robles-Malagamba et al., 2020) found a high abundance of Firmicutes and Fusobacteria, suggesting that our samples may Table S2). The average dissimilarity (Bray-Curtis) and the percentage of contribution to variance is reported for the contrasts between individuals with good preservation status (status = 2) and bad preservation status (status 3 and 4) (a and b). See Table 1 for specifications on preservation status. In (c) only individual with preservation status = 2 are considered.
give a realistic representation of the gut microbiota of free-ranging animals. Of course, we cannot a priori exclude that the preservation status of samples may have biased our results, in particular in relation to the number of clostridia. Biases among studies can in theory be due to the use of different procedures in DNA extraction and bioinformatic analyses, possibly limiting the comparison. However, the taxonomic level of analysis we have chosen (family) strongly limits any bias in terms of single species/genus representation. Unfortunately, regarding our findings on S. coeruleoalba, no comparison can be made with previous works since the sole study focused on the characterization of the gut microbiome of this species considered only one specimen (Godoy-Vitorino et al., 2017). In that work colon microbiota were rich in Firmicutes, Fusobacteria and Proteobacteria, which is in agreement with our results. However, at genera level few clostridia were found by those authors on their single animal. However, we may expect that the same impact on the gut microbiota of captivity conditions would be present in this species also. Contrarily to T. truncatus, in S. coeruleoalba some differences in relation to preservation status were found, mainly in a possibly higher abundance in clostridia for samples with poor preservation status (as would be expected due to anaerobic digestion of carcasses) and in Bacteroidetes (Pedobacter spp.). Moreover, Pedobacter spp. was found exclusively in S. coeruleoalba. The presence of members of this latter genus may deserve further attention in relation to a reservoir of antibiotic resistant strains in S. coeruleoalba gut. In fact, Pedobacter has been claimed as a superbugs genus since species in this group are intrinsically resistant to several classes of antibiotics, including colistin (Viana et al., 2018).
Overall, our results could not confirm the supposition advanced by Sanders et al. (2015) on the importance of phylogeny on the microbial gut communities of mammalians. Indeed, our findings may suggest that the dolphin gut microbiota could be similar to that of other carnivorous (phylogenetically unrelated) mammals (Bik et al., 2016), which could lead to an hypothesis on the influence of diet (carnivorous) on the taxonomic shaping of the gut microbiota. However, since there is a limited number of samples this point deserves more attention in future sampling and a careful evaluation of biases inherent to the preservation status (such as the massive presence of clostridia).
In conclusion, our work shed light on the gut microbiota composition of wild animals of T. truncatus and S. coeruleoalba, by analysing 12 stranded individuals, with different preservation status. We emphasize here the importance of a careful recording of preservation status of stranded animals, as well as the availability of specimens from a relatively high number of animals to provide reliable estimates of the gut microbiota composition of marine mammals in the wild.
Supplementary material. The supplementary material for this article can be found at https://doi.org/10.1017/S0025315420000983.
|
2020-11-05T09:10:51.623Z
|
2020-10-29T00:00:00.000
|
{
"year": 2020,
"sha1": "c84797c07ff18fcfead8bfe7faa094ec1c4d1efe",
"oa_license": "CCBY",
"oa_url": "https://www.cambridge.org/core/services/aop-cambridge-core/content/view/77EBF3C07CF3D80E992506D65BAB480C/S0025315420000983a.pdf/div-class-title-exploring-the-resident-gut-microbiota-of-stranded-odontocetes-high-similarities-between-two-dolphin-species-span-class-italic-tursiops-truncatus-span-and-span-class-italic-stenella-coeruleoalba-span-div.pdf",
"oa_status": "HYBRID",
"pdf_src": "Cambridge",
"pdf_hash": "b5006f25dbf9a09ad65f72cc7e744a265ed2b206",
"s2fieldsofstudy": [
"Environmental Science",
"Biology"
],
"extfieldsofstudy": [
"Biology"
]
}
|
218632124
|
pes2o/s2orc
|
v3-fos-license
|
ARE ALL FOUR-DAY SCHOOL WEEKS CREATED EQUAL? A NATIONAL ASSESSMENT OF FOUR-DAY SCHOOL WEEK POLICY ADOPTION AND IMPLEMENTATION
Four-day school weeks are used in over 1,600 schools across twenty-four states but little is known about adoption and implementation of these types of school calendars. Through examinations of school calendars and correspondence with school districts, we have compiled the most complete four-day school week dataset to date. We use this unique database to conduct a comprehensive analysis of four-day school week policy adoption and implementation. We find adoption of four-day school weeks is often financially motivated and has generally remained a small, rural district phenomenon. These schedules feature a day off once a week—often Friday—with increased time in school on each of the remaining four school days that, on average, is nearly an hour longer than the national average among five-day schools. Four-day school week schedules average only 148 school days per year, resulting in less time in school than the national average for five-day schools (180 days per year) despite the longer school days. Sub-stantial heterogeneity exists in the structure of these schedules across states, which may help explain differential four-day school week effects on student outcomes across institutional settings in the previous literature.
INTRODUCTION
Many school districts have faced increased financial pressures following the Great Recession (December 2007to June 2009), due in part to stagnant or decreasing local property tax revenue, slowly recovering state foundation aid in many states (Leachman, Masterson, and Figueroa 2017), and increasing expenditures (Marchitello 2018;Costrell 2020).While some states have developed financial intervention systems to help school districts navigate these financial pressures, school districts have traditionally dealt with financial pressures by laying off teachers and administrators, increasing class sizes, closing or consolidating schools, implementing student activity fees, and/or implementing alternative school schedules. 1One type of alternative school schedule 2 that is used to reduce costs and ease financial pressures is the four-day school week.This alternative to the traditional five school days per week model eliminates one school day per week with lengthened hours on the remaining four days.Although four-day school weeks have been used in the United States since the 1930s (Donis-Keller and Silvernail 2009), use of this model has significantly increased in the wake of the Great Recession.There were approximately 1,607 schools in 662 school districts across at least 24 states 3 using a four-day school week calendar as of the 2018-19 school year.
Despite the growing use of four-day school weeks across the United States, little is known about the characteristics of the schools implementing this model, 4 the rationale for these schedule changes, and how these school schedules are being structured (e.g., school day start and end times, number of yearly instructional days, which day is used as the nonschool weekday [i.e., the "off day"], and use of any off-day activities).Previous research has largely featured case studies of the adoption and implementation of fourday school weeks in individual schools or a small group of schools. 5Our study attempts to build on these by contextualizing questions of four-day school week adoption and implementation at a scale not observed in this previous literature.
Several state-level quasi-experimental analyses have assessed the impacts of fourday school weeks on student achievement (Anderson and Walker 2015;Thompson 1.Previous research generally finds mixed evidence on the impacts of these types of interventions on student achievement and overall community welfare.Thompson (2016) finds that the financial intervention system in Ohio led to a decrease in math proficiency rates in primary schools.Literature on the achievement effects of school closures and class size changes has found positive (see Carlson and Lavertu 2016 regarding school closures), null (see Hoxby 2000 andChingos 2012 regarding class size), and negative impacts (see de la Torre and Gwynne 2009; Brummet 2014; and Larsen 2020 regarding school closures; and Angrist and Lavy 1999 and Krueger 1999 regarding class size) on achievement.2. Four-day school weeks are not the only type of alternative school schedule, but this schedule is one of the few to reduce exposure to the school environment.Other types of alternative school schedules-most notably the yearround school calendar-often attempt to extend exposure to the school environment, reduce summer learning loss, and/or ease overcrowding issues in schools.Previous literature has generally found null (McMullen and Rouse 2012) or negative impacts (Graves 2010(Graves , 2011) ) of year-round schooling on student achievement, but does find established cost savings of these school calendars (Daneshvary and Clauretie 2001;Cooper et al. 2003).3. The states with schools operating on a four-day school week schedule during the 2018-19 school year were Alaska, Arizona, California, Colorado, Georgia, Idaho, Iowa, Kansas, Louisiana, Michigan, Minnesota, Missouri, Montana, Nebraska, Nevada, New Mexico, North Dakota, Oklahoma, Oregon, South Dakota, Utah, Texas, Washington, and Wyoming.A couple of schools recently adopted four-day school weeks for the 2019-20 and 2020-21 school years in Arkansas.4. A recent report by the Institute of Education Sciences (https://nces.ed.gov/pubs2020/2020011.pdf ) details some basic characteristics for schools participating in the 2017-18 National Teacher and Principal Survey that were identified as using a shortened school week. 5.For a thorough review of this literature, see Donis-Keller and Silvernail (2009).Heyward (2018) provides an overview on the current state of four-day school week research that highlights the gaps addressed in this study.
2019; Morton 2021), juvenile crime (Fischer and Argyle 2018), and maternal labor supply (Ward 2019).Anderson and Walker (2015) find positive effects of four-day school weeks on fourth-and fifth-grade math and reading proficiency rates in Colorado, whereas Thompson (2019) finds negative effects on math and reading achievement in third through eighth grades using student test score data in Oregon.Morton finds null effects on district-level achievement in Oklahoma.Fischer and Argyle find that four-day school weeks led to an almost 20 percent increase in juvenile crime.Ward finds that mothers of primary school-age children work fewer hours and are less likely to be employed as a greater percentage of local school enrollment switches to a fourday school week schedule.The findings of these studies-in particular, the differing achievement results found in Colorado, Oklahoma, and Oregon-suggest that differences in the policy environment and/or the structure and implementation of these fourday school weeks may be a relevant mechanism to consider when analyzing these effects.Yet, to date, there has been no large-scale national assessment of the adoption and implementation of this four-day school week model.
To address this significant knowledge gap regarding four-day school weeks, we conduct a comprehensive analysis of the schools and school districts across the United States adopting four-day school weeks and detail the main features of these school schedules (e.g., day off, school day length, school start time).Specifically, this paper examines the following research questions: (1) which types of districts adopt four-day weeks?( 2) what motivates districts to adopt these policies?(3) how do these patterns vary across states and over time?( 4) what are the key features of the structure of current four-day weeks?(5) how do these structures vary across states?and ( 6) what implications do they have for explaining differences in past research on four-day school week effects?From anecdotal evidence, we expect to find that districts opting for a four-day school week schedule will be primarily small, rural school districts and will be motivated by financial, student attendance, and teacher recruitment issues.We expect these factors may vary over time, with larger school districts adopting these schedules and these decisions becoming more financially motivated during the Great Recession period.Finally, we expect there to be heterogeneity across districts and states in the structures of these school schedules in terms of off-day activities and time in school, and this heterogeneity will help reconcile some of the differences in findings from the previous literature on four-day school week effects.
To answer these research questions, we have compiled, to date, the most comprehensive four-day school week dataset through examinations of current school calendars and direct correspondence with four-day school week school districts.This dataset includes panel data on four-day school week adoption/use from 1999-2019 and a crosssection of the structure of four-day school week implementation during the 2018-19 school year.Using this unique dataset, we confirm that these school schedules are often used in small, rural school districts and are often financially motivated.These school calendars feature a day off once a week, with a majority of schools using a Friday offday.Fewer than one-third of districts offer student-based academic services on the off day.Four-day school week schools average 148 school days annually and have increased time in school that, on average, is about 45 minutes longer than the national daily average for five-day schools.Despite the increase in hours, the reduction in the number of school session days results in less school time in four-day school week schools than the national average for five-day schools.There also is substantial heterogeneity in the structure of these four-day school weeks across states, which may help explain differences in achievement and other outcomes of four-day school weeks across states.We also find that differences in instructional spending and the availability of off-day academic student programming may explain differences in cost savings among financially and nonfinancially motivated four-day school week districts.
DATA
The previous research on four-day school weeks has focused on the effects for individual school districts or multiple four-day school week school districts within particular states.A key hurdle to undertaking a nationwide study of four-day school week impacts on student and district outcomes (e.g., student achievement, cost savings, juvenile crime) has been the lack of a complete national longitudinal database of four-day school week adoption.Using publicly available data from state boards of education, combined with extensive proprietary data collection, we have, to our knowledge, compiled the most expansive four-day school week dataset.This section details the specifics of the creation of this national database and the other data used in the subsequent analyses conducted in this study.
Four-Day School Week Survey Design and Implementation
To identify which school districts had at least one school operating on a four-day school week in 2018-19, we first collected lists of four-day school week districts from state departments of education Web sites. 6Once we identified the subset of school districts in each state that were flagged as using a four-day school week, we went to each school district's Web site and school calendar to confirm the school district was indeed operating on a four-day school week during the 2018-19 school year.As available, we also collected information on the school day start and end times, total number of yearly student instructional days, and whether the off day occurred on Monday or Friday.Historical data on the use of four-day school weeks from 1999-2019 were determined from state departments of education Web sites, news reports on four-day school week adoption, and e-mail and phone correspondence with these four-day school week schools, as described below. 7 For school districts where some of this information was missing from either the school district Web site or the school calendar, we conducted e-mail and phone correspondence with school districts to collect the missing data. 8Through e-mail and phone correspondence with all four-day school week districts, we also collected information 6.In addition to these state department of education lists, the lists of four-day school week districts for several states (e.g., Arizona, Kansas) came from news articles discussing the extent of four-day school week use in those states.See Appendix B for a list of sources for these state-level four-day school week lists.All appendices are available in a separate online appendix that can be accessed on Education Finance and Policy's Web site at https://doi.org/10.1162/edfp_a_00316.7.In some cases, information was collected from historical school district Web sites using the Internet Archive.8. Based on the state-level lists, we were able to verify the continued usage of four-day school weeks for the 2018-19 school year for 85 percent of these school weeks through the examination of school calendars from school district Web sites.For those school districts without 2018-19 school year calendars posted online, we verified four-day school week status, off-day used, school day start and end times, and number of yearly instructional days through direct e-mail and phone correspondence with these districts.
on the rationale for the switch to the four-day school week schedule, whether the school provides any student services on the off day, and, if so, what form those services take (e.g., remedial instruction, enrichment activities).These two survey items were asked as open-ended questions and responses were interpreted and coded based on the key themes identified in the responses (e.g., cost savings, absences).For the full text of the survey questions from the e-mail/phone survey that was conducted with these school districts, see online Appendix C. For the 2018-19 school year, we have information on the off-day used for 659 out of the 665 (99.1 percent) school districts with at least one four-day school week school; we also have school day start/end times, and yearly student instructional days for 1,543 out of the 1,608 four-day school week schools (96 percent).We obtained full historical fourday school week use for 2,006 out of the 2,081 schools (96.4 percent) that have ever had a four-day school week. 9We received responses regarding use of off-day programming from 552 out of the 796 districts (69.4 percent) that have ever had at least one school with a four-day school week, and responses regarding rationale for the switch to the four-day school week from 533 out of the 796 (67 percent) that have ever had at least one school with a four-day school week.As noted in online Appendix table A.1, there are no notable differences 10 between districts that responded to these survey items compared to those that did not respond or could not be reached.
Other Data Sources
We augment this four-day school week information with data from the National Center for Education Statistics (NCES), including school district-level information on total enrollment; the number of schools; pupil-teacher ratio; the number of students disaggregated by sex, race, free and reduced-price lunch eligibility (via the National School Lunch Program), urbanicity, and per-pupil revenues and expenditures.We also obtain disaggregated school district financial information from the NCES F-33 files from 1999-2014 to examine the expenditure impacts of different four-day school week structures and rationales.This study also uses the 2007-08 and 2011-12 restricted-use Schools and Staffing Survey data from the NCES.These data include information on the number of school days, which allows us to determine four-versus five-day school week status.This survey also includes data on instructional time by subject, which we use to compare across four-versus five-day schools to see what specific subjects may be experiencing changes to instructional time as a result of the switch to the four-day school week.Across the two years of Schools and Staffing survey data, approximately 9. We have collected partial historical data (e.g., missing a few years of data) for the remaining 75 schools.Thus, we have partial historical four-day school week information for all of the schools that have identified as ever having a four-day school week.Reasons for missing data include lack of sufficient past records and lack of response from districts regarding historical four-day school week use.10.As noted in online table A.1, school districts that chose to respond to the survey questions are quite similar to the universe of school districts that operated on a four-day school week schedule at some point between 1999 and 2019.One notable exception is the group of four-day school districts that provided 2018-19 school year information on their four-day school week structure.These districts have smaller enrollments and higher per-pupil revenue and expenditures compared with the average of all of the districts that ever used a four-day school week schedule, but is in line with the comparable sample of 2018-19 four-day school weeks presented in panel A of table 2.
THE WHO, WHEN, AND WHY OF FOUR-DAY SCHOOL WEEK ADOPTION
Before detailing implementation specifics of these policies, we first seek to establish the policy environment in which these four-day school weeks are adopted.This section examines our first set of research questions detailing what types of districts adopt fourday school weeks, when these changes have occurred across time and geographical dimensions, and the rationale behind these school schedule adoption decisions.
Trends in Four-Day School Week Adoption
Over the past two decades, there has been enormous growth in the number of schools using a four-day school week schedule.As shown in figure 1, there were 257 schools in 108 school districts that operated on a four-day school week in 1999, but that number has risen to 1,607 schools in 662 school districts by 2019.This widespread growth in four-day school week adoption largely has been facilitated by state policy changes allowing waivers of the minimum yearly instructional day mandate of a traditional school 11.On average, the four-day school week schools surveyed in the Schools and Staffing Survey have more students and teachers than the average among all four-day school weeks in our four-day school week database.Despite their larger size, the four-day school week schools surveyed in the Schools and Staffing Survey have comparable pupil-teacher ratios, similar demographic make-up (in terms of race, sex, and free-and reduced-priced lunch participation), and similar likelihood of being located in a rural area, as the full sample of four-day school week schools in our database.The table of summary statistics comparing the four-day school week schools in these two datasets is available upon request.
calendar (generally consisting of 175-180 days).These waivers give school districts the ability to use alternative school schedules, such as the four-day school week, provided that the alternative school schedule still adheres to a minimum yearly instructional hours requirement. 12Thus, states adopting these policy changes have often precipitated the expansion of four-day school weeks nationally over the past two decades.Most notably, states like Missouri (2011), Montana (2007), and Oklahoma (2009) changed their laws and immediately saw school districts begin to adopt four-day school weeks. 13 Due to these law changes and other contributing factors (e.g., financial downturns), there has been massive dispersion of four-day school week schools geographically over the past two decades (see figure 2).In 1999, four-day school week schools were found only in a handful of states-namely, Arizona, Colorado, New Mexico, and Oregon.In subsequent years, four-day school week schedules expanded to other states in the western half of the United States, with the largest growth occurring after the Great Recession.As of the 2018-19 school year, four-day school week schools operated in twenty-four states across the United States, and these schools are geographically dispersed across many of these states.In states like North Dakota, Kansas, and Nebraska, however, many four-day school week schools are geographically concentrated near the borders of neighboring states that also allow four-day school week schedules.During the 2018-19 school year (table 1), these school schedules were most prevalent in Colorado, Oklahoma, Montana, Oregon, Idaho, and South Dakota, which all had at least 125 four-day school week schools.
Characteristics of Districts Using Four-Day School Weeks
As shown in panel A of table 2, four-day school week school districts, on average, are much more geographically concentrated in rural areas and, on average, have significantly smaller enrollments than five-day school week districts.Of the four-day school week districts implementing the schedule districtwide, 14 90 percent are rural districts and have an average total school district enrollment of 454 students. 15Among five-day school week districts, only 50 percent are rural districts and have an average enrollment of 3,735.The lower enrollments in four-day school week districts also may be driving the larger average per-pupil spending in these districts.Four-day school week 12.For states that allow for minimum instructional hours requirements, all schools, with a few exceptions, are subject to the minimum instructional hours requirement (see https://www.ecs.org/50-state-comparisoninstructional-time-policies/ for more details on these state-level school instructional time policies).The only difference is that four-day schools obtain a waiver that allows them to meet this requirement in fewer school days than the minimum instructional days statute.Thus, the choice of whether to switch to a four-day school week has little bearing on the regulations faced in terms of minimum instructional hours.However, it is certainly plausible that school districts above the minimum number of instructional hours while on a five-day school week schedule reduce their hours to the yearly minimum as a way to facilitate the implementation of the four-day school week.differ in terms of the racial and socioeconomic composition of the student body.Four-day school week districts have significantly smaller proportions of black and Asian students, and a significantly higher proportion of free or reduced-price lunch-eligible students than five-day school week school districts.
There also is notable heterogeneity in the type of districts adopting these school schedules across different states and across time.As shown in panel B of table 2, the size of four-day school week districts varies greatly across states, with Oregon having an average enrollment of 781 in four-day school week districts compared to Montana, where four-day school week districts only average eighty students.While many of the states have demographic makeups that are similar to the national average, Oklahoma is a primary exception.Oklahoma four-day school week school districts have much lower white student percentages and much higher percentages of American Indian students due to the high adoption of four-day school weeks by schools with high populations of American Indians.As shown in panel C of table 2, more recent adopters of four-day school weeks are larger, have lower per-pupil expenditures and revenues, and are less rural than early adopters (pre-2000) of four-day school weeks.
Rationale for the Four-Day School Week
In addition to examining which school districts adopt four-day school weeks, it is also critical to understand why school districts are adopting these types of school schedules.Based on our phone and e-mail survey/correspondence with four-day school week school districts, we find that the choice to adopt four-day school weeks appears to be motivated by three main rationales (see table 3).The first is cost savings, as stagnant or declining property tax revenues and declining state aid necessitate unique cost-cutting measures in many school districts.Nearly two thirds of the respondent school districts
Notes:
The "Four-day" category includes all districts with a universally adopted four-day school week (e.g., all schools in the school district operate on a four-day school week).Standard deviations given in parentheses.
Panel (65.1 percent) cited financial reasons as one of the main rationales for the switch to the four-day school week.The other two reasons most often cited by school districts were attendance issues (28.9 percent) related to things such as long commutes for schoolsponsored athletic events or family appointments (e.g., medical), and issues related to being primarily rural districts (30.6 percent), such as teacher retention, long bus commutes for students, and family farming/ranching commitments.Some school districts (11.1 percent) also cite other things, such as aligning schedules with other four-day school week schools in the area and teacher professional development, as additional reasons for use of the four-day school week.
As noted in panels B and C of table 3, there is also some notable heterogeneity in the rationale for these school schedules across states and the time period of policy adoption.Districts in Oklahoma and Colorado generally cite financial and teacher recruitment/retention motivations for the four-day school week, while districts in Oregon and Montana cite primarily financial and athletics-related absence motivations for the four-day school week.Districts that first adopted the four-day school week prior to 2000 had a more evenly balanced distribution of motivations compared with the full sample, with 55.4 percent citing financial motivations, 41 percent citing attendance motivations, and 43.4 percent citing rural-related issues.For districts adopting the four-day school weeks after 2000, there has been a general shift away from absence-related motivations and toward greater financial motivations (particularly during the 2000s and the Great Recession period) and student/teacher recruitment and retention motivations (particularly since 2012).
HOW ARE CURRENT FOUR-DAY SCHOOL WEEK SCHEDULES STRUCTURED?
Having addressed our first set of research questions, we now turn to the question of how school districts are structuring these four-day school week schedules.The design of these school schedules is critically important to understand, as decisions regarding school calendar structure may have important implications for academic achievement and other student outcomes, particularly if there are alternations in exposure to instructional hours (Lavy 2015;Cattaneo, Oggenfuss, and Wolter 2017), school meals (Schwartz and Rothbart 2020), and physical activity supports (Carlson et al. 2008;Packham and Street 2019).For example, Thompson (2019) found that schools in Oregon reduced instructional time by nearly 3.5 hours per week as a result of the switch to a four-day school week, which coincided with lower student achievement in these districts.Thus, it is important to understand what decisions school districts are making regarding which day of the week to take off, whether to offer any off-day enrichment opportunities to students, how much to lengthen the remaining school days each week, whether to move daily school start times earlier, how much to change overall instructional time, and how to distribute time among instruction and other activities (e.g., art/music/recess/physical education).
Choice of Day Off and How Day Off Is Used
We first consider the discretion school districts have over both the choice of weekday off and how that day off is used.From our data collection, we find that districts exclusively choose either Monday or Friday as the off day, yielding a three-day weekend regardless of which off day is chosen. 16Of the 665 school districts with at least one school currently using a four-day school week schedule, 84.2 percent use a Friday-off day, 14.4 percent use a Monday-off day, and less than 1 percent use a combination of off days.As observed in figure 3, most states have all or nearly all four-day school week schools operating on a schedule with the same off day. 17 While many schools choose to close their school buildings completely on the off day, some school districts choose to keep buildings open to facilitate student extracurricular events, teacher professional development opportunities, and enrichment or remedial education to students who want or need these activities to complement their normal school curriculum.Of the 552 four-day school week school districts that provided information about the off day, 48.2 percent responded that buildings were fully closed or no academic services were offered to students or teachers on the off day.An additional 29.7 percent had some type of remedial or enrichment activities on the off day. 18These programs varied in scope, from teacher office hours to field trips to very structured offday programs, and in frequency, from as needed to bi-monthly to weekly.For four-day school week school districts not offering student services on the off day, some districts (23.2 percent) provide teacher professional development activities on this day.
16.Many districts that choose a Friday-off schedule will attend school on Fridays when there is a Monday holiday.
Thus, the choice of day off does not generally impact the total number of yearly instructional days, which are an average of 149 for Monday-off schools and an average of 148 for Friday-off schools.Depending on when student activities are scheduled (e.g., sporting events), however, the choice of off-day may have important implications for student learning and potential cost savings associated with the four-day school week.17.A couple of states (namely, Oklahoma and Colorado) have some schools with Friday-off days and some with Monday-off days.18.It should be noted that some school districts are able to offer these programs through the receipt of grants (e.g., 21 st Century grants).Anecdotal evidence from school districts indicates schools discontinuing use of these programs after initial use or after grants lapse due to high costs, poor attendance, and/or other issues.Although these programs are often free to students, it may incur a cost to schools or families of providing transportation, food services, etc.As teachers are generally paid for off days regardless of whether there is any service offered, teacher costs are unlikely to factor into whether these services are offered to students.Districts that offered some of these programs in the 2018-19 school year tend to be larger, less rural, and have lower percentages of students eligible for free and reduced-price lunch, on average, compared with all four-day school week districts.This suggests that larger and more affluent communities may be the ones able to continue to offer these programs over the long term or have families that are able to incur the costs (e.g., transportation) of enrolling their students in these programs.
Choices Regarding Amount and Composition of Instructional Time
To accommodate the shortened week, school districts also have discretion regarding the structure of instructional time on the remaining four school days in order to meet minimum required hours statutes.On average, four-day school week schools have earlier school day start times, longer school days, and less annual student time spent in school than the average five-day school in the United States.Nationally, the average school start time among four-day school week schools is 7:56 AM, the average school day lasts seven hours and forty-six minutes, and the average number of student instructional days is 148.Among five-day schools in the United States, according to the 2015-16 National Teacher and Principal Survey, the average school start time is 8:07 AM, the average school day lasts six hours and fifty-four minutes, and the average number of student instructional days is 179.Based on these differences, we find that four-day school week students attend school for an average of 1,150 hours per year19 compared with an average of 1,235 yearly hours for five-day school week students. 20s shown in figures 4 and 5, there is also significant variation in school start times, school day length, yearly days in session, and yearly time in school across four-day school week schools.Four-day school week school start times during the 2018-19 school year vary from 7:00 AM to 9:00 AM, with a majority of schools (81.6 percent) starting at 8:00 AM or earlier.School day lengths for four-day school week schools range from six and a half hours to nine hours, with a majority of schools (58.4 percent) holding a school day that is longer than seven and a half hours but less than or equal to eight hours.The number of yearly instructional days varies from 131 to 165 under various types of four-day school week schedules, with a majority (59.8 percent) holding 150 instructional days or fewer. 21Given the substantial heterogeneity in both school day length and the number of instructional days, we observe significant variation in the amount of yearly time students spend in school under a four-day school week schedule.Yearly time in school ranges from 949 to 1,419.5 hours, with a majority of four-day school week schools (71.4 percent) holding between 1,100 and 1,200 annual hours of time in school.
While overall differences in time in school are important, what likely matters most for student learning and test outcomes is the time spent in specific subjects.For example, do four-day school weeks maintain instructional time in the tested subjects (i.e., math and English/Language Arts) at the expense of nontested subjects?To examine this question, we compare responses from the Schools and Staffing Survey regarding time spent in specific subjects across four-and five-day school week schools.We find that time spent on tested subjects is generally less under a four-day schedule than a five-day schedule (table 4).In particular, on average, four-day schools have twenty-five fewer minutes per week of math instruction in both third and eighth grades, and fortynine fewer minutes of English instruction in third grade.There is also greater time spent on third grade activities related to physical activity (physical education and recess) and music instruction under a four-day school week schedule.In particular, on average, third graders in four-day schools have twenty-one more minutes per week of physical education instruction, fifteen more minutes per week of recess, and twentytwo more minutes per week of music instruction.Thus, it appears that, contrary, to what we expected, four-day school week schools appear to be more heavily focusing on maintaining access to instruction that develops physical and creative skills, while possibly shifting away from maintaining instruction in key tested subjects.Given the cross-sectional nature of these comparisons, however, there is the possibility that the four-day school week schools sampled may have exhibited different instructional patterns regardless of whether they offered the four-day school week.
21.Many of the school districts on the upper end of this range of yearly instructional days implement a "hybrid" four-day school week model, in which the four-day school week is not used every week.These school calendars may feature things such as a four-day school week every other week or four-day school weeks only over certain periods of the school year (e.g., winter months when likelihood of snow days is much higher).The full-year four-day school week districts generally have fewer than 156 days of school per year.
HETEROGENEITY IN FOUR-DAY SCHOOL WEEK STRUCTURE ACROSS STATES AND IMPLICATIONS FOR FOUR-DAY SCHOOL WEEK EFFECTS
As a majority of existing four-day school week research is conducted at the individual state level, it is also important to consider how the structure of these four-day school weeks varies across states.The varying structure of these school schedules across states may help explain the differing effects noted in existing four-day school week research, and inform future work on a national scale.
Student Achievement
A key consideration regarding four-day school weeks is their impact on student achievement.The current quasi-experimental empirical literature-focused on four-day school weeks in Colorado, Oklahoma, and Oregon-provides mixed evidence on the impacts of four-day school weeks on achievement.Anderson and Walker (2015) find positive effects of four-day school weeks on fourth-and fifth-grade math and reading proficiency rates in Colorado, while Thompson (2019), using student test score data, finds negative effects on third-through eighth-grades, math and reading achievement in Oregon.Morton (2021) finds null impacts of four-day school weeks using district-level achievement in Oklahoma.
So, what might be underlying this difference in the achievement effects of four-day school weeks in these states?Thompson (2019) posits that overall time in school is a critical mechanism for understanding the achievement declines in Oregon schools and thus could potentially explain differences in achievement effects of four-day school weeks across states.Most notably, during the 2018-19 school year, the average yearly time in school was 1,116 hours for four-day school weeks in Oregon compared with 1,139 hours for four-day school weeks in Oklahoma and 1,169 hours in Colorado.In addition to a lower average annual time in school, as noted in figure 6, the distribution on the lower end of yearly time in school (<1,100 hours) is much more prominent among Oregon four-day school week schools than among those in Colorado and Oklahoma.In fact, 26.6 percent of Oregon four-day schools have yearly time in school below 1,100 hours, compared with only 5.8 percent of Colorado four-day schools and 9.5 percent of Oklahoma four-day schools.It is key to note that this difference is driven exclusively by longer school days in Colorado and not by more school days per school year, Another mechanism that could drive differences in achievement outcomes are school start times.Moving school start times earlier to accommodate the lengthening of the school day could have detrimental effects on student achievement (Wahlstrom, Wrobel, and Kubow 1998;Carell, Maghakian, and West 2011;Wong 2012;Hinrichs 2011;Edwards 2012;Heissel and Norris 2018).As noted in figure 7, a much higher percentage of four-day school week schools in Colorado (68.2 percent) have start times before 8:00 AM than four-day school week schools in Oregon (40.6 percent) or Oklahoma (48.6 percent).This distribution of school start times suggests that earlier start times are unlikely to explain the differences in four-day school week achievement effects across the three states.Finally, the use of enrichment programs on the off day could help further increase instruction time and boost achievement.However, based on our survey, a greater percentage of four-day school week schools in Oregon offer students off-day activities compared to Colorado or Oklahoma.
Cost Savings
Our survey results suggest nearly two thirds of school districts adopt four-day school weeks for financial reasons.Thus, overall cost reduction is an important metric on which to base the effectiveness of these school schedules as a potential cost savings policy alternative for school districts.There is a small but growing literature on the impacts of four-day school weeks on school district costs. 22Projections from Griffith (2011) suggest the switch to a four-day school week could produce a maximum cost savings of up to 5.4 percent, but realized savings are likely to be lower-on the order of between 0.4 and 2.5 percent.Whereas Thompson (2020) finds minimal statistically significant overall cost savings of the four-day school week nationwide, Morton (2021) finds cost savings of 1.36 percent resulting from four-day school week adoption in Oklahoma.While both studies find minimal impacts on instructional spending, they do find statistically significant reductions in spending areas where services are reduced by one day per week (e.g., food service, transportation).The reductions in these spending areas are found to be larger in Oklahoma (10.5 percent for transportation; 14.1 percent for food services) than for four-day school weeks nationally (10.1 percent for transportation; 4.3 percent for food services).Thompson (2020) also notes these cost savings are larger in school districts with four-day school weeks that are financially motivated.This finding may help explain why Oklahoma-the state with the highestdegree of financially motivated four-day school weeks-had larger overall cost savings from four-day school weeks (Morton 2021) than what was found nationally (Thompson 2020).
The results of this previous literature suggest that the four-day school week may not intrinsically be a cost savings policy, and school districts that have these financial motivations may need to put in additional effort beyond just switching the school calendar to realize overall cost savings.Thompson (2020) posits that four-day school weeks often may be used as a negotiating tool to compensate teachers for lower salaries in the face of budgetary problems. 23Beyond this, most savings appear to come from reductions in noninstructional operating expenditures, with the size of these reductions likely varying by the type of school schedule structures that are used to achieve these financial motivations.
To examine which attributes may influence cost savings, we compare average perpupil costs before and after four-day school week adoption 24 for those with and without 22.The cost impacts of other alternative schedules, such as year-round schooling, also have been considered.Graves, McMullen, and Rouse (2013) note that year-round schooling schedules can reduce the need for new construction due to the easing of overcrowding issues, and may yield additional cost savings from benefits being calculated on a twelve-month basis and changes in overall staffing and transportation needs.Daneshvary and Clauretie (2001) empirically find the switch to a year-round schooling schedule reduces total expenditures per pupil by 7.5 percent, with larger percentage savings on real estate capital (31 percent) than operations (12.3 percent).These results suggest cost savings from reducing issues of overcrowding may be much greater than the cost savings realized from shifting the composition of instructional time through use of the four-day school week.23.These results suggest that teachers are paid approximately the same amount regardless of four-day school week status.For many school districts using the four-day school week as a teacher recruitment device, the four-day school week may represent a non-monetary benefit they can offer teachers when they are unable to increase salaries further.Thus, it may not be surprising to see no change in teacher salaries as a result of this change in school calendar.Despite the lack of increased monetary compensation, the addition of a three-day weekend may be desirable for many teachers.A recent study by Turner, Finch, and Ximena (2018) finds strong support among teachers and staff for the four-day school week in Missouri, citing things like improved staff morale and an enhanced curriculum.Future work that develops a large-scale survey building off of the work of Turner, Finch, and Ximena (2018) will provide a better understanding of teachers' perceptions of these school calendars across different state contexts and four-day school week structures.24.These per-pupil expenditures are adjusted for mean growth rates in nationwide per-pupil expenditures over time to draw meaningful comparisons across the pre-and post-four-day school week periods.
reasons, but we find very little change and some slight increases in per-pupil expenditures for school districts adopting these school schedules for non-financial reasons.These differences in expenditure changes between the two rationales appear to be driven primarily by stark differences in instructional expenditures and differences in the use of student academic off-day activities.Instructional expenditures fall by over $600 per pupil, on average, in districts with financial motivations for the four-day school week, while we see almost no change for those without financial motivations.These results align with those of Thompson (2020), who notes the four-day school week may be a necessary consequence of cost savings policies focused on reducing teacher salaries.Thus, school districts with financial motivations may be more willing to reduce teacher salaries-using the four-day school week as compensation for impacted teachers-than those moving to the four-day school week for other reasons.Given these findings, future work examining the role of the four-day school week in teacher labor markets is warranted.
While the structure of the school calendar (e.g., yearly time in school, instructional days, start time) is roughly the same, on average, across the two rationale groups, there is a noticeable difference in the use of the off day between these two groups.Relative to school districts without financial motivations for the four-day school week, school districts with financial motivations are slightly more likely to use a Monday-off day-18 percent compared to 15 percent-and much more likely not to offer off-day student academic activities-50 percent to 24 percent.The notable increases in per-pupil spending during the post-four-day school week period for operations and maintenance and student support services for school districts without financial motivations suggest that operating the building and providing additional student services on the off day may come at significant cost to these school districts.Thus, more work is needed to ascertain the true costs of these programs, as well as any benefits these enrichment activities provide to students.
Juvenile Crime/Parental Labor Supply
Our survey results also suggest that a majority of school districts do not provide academic student services on the off day.This situation may create substantial unsupervised time for children or force parents to make tough financial decisions about outsourcing child care or changing work schedules to provide child care on their own during the off day.A few studies have examined the ramifications of the off day on juvenile crime and parental labor supply.Fischer and Argyle (2018) find that four-day school weeks led to an almost 20 percent increase in juvenile crime in Colorado.Given that Colorado generally has a percentage of school districts without student academic services on the off day that is comparable to the national average, we may expect similar opportunities for juvenile crime to be increased in other four-day school week settings.Ward (2019) finds in a study of four states that mothers of primary school-aged children work fewer hours and are less likely to be employed as a greater percentage of local school enrollment switches to a four-day school week schedule.Given that Ward takes advantage of data that include Oklahoma (where there is a much larger percentage of schools not offering off-day services to students), it may be the case that these results represent an upper bound on the labor supply responses of parents.
CONCLUSION
This study conducted the first comprehensive descriptive assessment of four-day school week nationally and found that the justification for four-day school weeks and structure of these school schedules varies considerably across districts.Thus, the parameters surrounding the structure of the four-day school week are likely to be an important consideration when analyzing the causal effects of these school schedules on various child and family outcomes.
Our findings suggest that certain aspects (e.g., instructional time) of these school schedules may impact student achievement.We demonstrate that instructional time in tested subjects is lower in four-day school week schools, which may have direct implications for student learning and test-score performance in these subjects.Thus, school officials considering these types of alternative school schedules should be cognizant of the potential instructional time implications of such a change.We also find that the willingness to cut instructional spending and the lack of availability of off-day academic student programming among financially motivated four-day school week districts may drive larger expenditure reductions in these districts compared with those adopting these school schedules for nonfinancial reasons.
Changes in the various parameters associated with these four-day school week schedules also could have impacts on other aspects of child well-being (e.g., health outcomes, health behaviors, absences, social/emotional outcomes).For example, school start times moving earlier as the result of the switch to the four-day school week could impact student sleep, tardiness, and so forth, while districts with an unstructured nonschool off day may lead to greater opportunities for adolescents to engage in criminal behavior (see Fischer and Argyle 2018) and/or risky health behaviors.Losing one school day per week also may diminish exposure to school counselors and school-based health services, school meal programs (e.g., free or reduced-price lunch), and other supports that could negatively affect child physical health and social-emotional development.Thus, future work should attempt to use the wide variation in four-day school week structure highlighted in this paper to examine the mechanisms underlying the causal effects of these school schedules on various student outcomes.In doing so, we may gain a much greater understanding of how the policy decisions to switch to four-day school week impact students, families, and communities, and inform policy makers and school officials about how best to structure these types of alternative schedules.
Figure 1 .
Figure 1.Number of Four-Day School Week Schools and Districts, 1999-2019 13. See online Appendix figure A.1 for state-specific time trends in four-day school week use.14.A majority of these school districts (92 percent) use these school schedules districtwide.The remaining school districts, which are generally larger than the average four-day school week district, often allow their rural schools to adopt a four-day school week schedule while the rest of the district remains on a traditional fiveday schedule.15.The density of enrollment among four-day school week districts is clustered heavily below a total enrollment of 200 students-with 43 percent of the four-day school week schools falling into this enrollment bin.
Figure 2 .
Figure 2. Geographic Distribution of Four-Day School Weeks from 1999-2019 B separates out the summary statistics for the four states with the largest number of four-day school weeks and aggregates the summary statistics of the other states with four-day school week districts.PP =
Figure 3 .
Figure 3. Geographic Distribution of Four-Day School Weeks in 2019, by Off-Day
Figure 4 .
Figure 4. Distribution of Four-Day School Week Daily Start Times and School Day Length
Figure 5 .
Figure 5. Distribution of Four-Day School Week Yearly Time in School and Days in Session
Figure 6 .
Figure 6.Distribution of Four-Day School Week Yearly Time in School in Colorado, Oklahoma, and Oregon
Figure 7 .
Figure 7. Distribution of Four-Day School Week Daily School Start Time in Colorado, Oklahoma, and Oregon
Table 1 .
Number of Four-Day School Week Schools and Districts, by State Note: These numbers reflect the number of four-day school week schools and the number of districts with at least one four-day school week school during the 2018-19 school year.
Table 2 .
Characteristics of School Districts with Four-Day School Weeks
Table 3 .
Distribution of Survey Responses About Four-Day School Week Rationale and Use of the Off Day Numbers in each cell are the number of respondents citing each type of rationale or off-day category.Percentages are computed by the number of responses for a given category divided by the number of total survey responses.Panel B separates out the responses for the four states with the largest number of four-day school weeks and aggregates the responses of the other states with four-day school week districts.FDSW = four-day school week.
Table 4 .
Differences in Weekly Instructional Time Per Subject, by Four-or Five-Day Schedule * p < 0.1.
|
2020-05-15T13:12:47.807Z
|
2020-05-14T00:00:00.000
|
{
"year": 2020,
"sha1": "b5b73f33b6e4ed447aae79c09eea3f07cebd5d9a",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.1162/edfp_a_00316",
"oa_status": "GREEN",
"pdf_src": "ScienceParsePlus",
"pdf_hash": "2f7489e71603b03a7a6a894a229197747c5985bb",
"s2fieldsofstudy": [
"Education",
"Economics"
],
"extfieldsofstudy": [
"Psychology"
]
}
|
243850722
|
pes2o/s2orc
|
v3-fos-license
|
In situ microstructure analysis of Inconel 625 during laser powder bed fusion
Laser powder bed fusion is an additive manufacturing process that employs highly focused laser radiation for selective melting of a metal powder bed. This process entails a complex heat flow and thermal management that results in characteristic, often highly textured microstructures, which lead to mechanical anisotropy. In this study, high-energy X-ray diffraction experiments were carried out to illuminate the formation and evolution of microstructural features during LPBF. The nickel-base alloy Inconel 625 was used for in situ experiments using a custom LPBF system designed for these investigations. The diffraction patterns yielded results regarding texture, lattice defects, recrystallization, and chemical segregation. A combination of high laser power and scanning speed results in a strong preferred crystallographic orientation, while low laser power and scanning speed showed no clear texture. The observation of a constant gauge volume revealed solid-state texture changes without remelting. They were related to in situ recrystallization processes caused by the repeated laser scanning. After recrystallization, the formation and growth of segregations were deduced from an increasing diffraction peak asymmetry and confirmed by ex situ scanning transmission electron microscopy.
GRAPHICAL ABSTRACT Introduction
Additive manufacturing facilitates the fabrication of arbitrarily complex geometries and tailored material properties using a range of different materials. Laser powder bed fusion (LPBF) is a process especially suited for the production of metal parts. LPBF employs focused laser radiation to melt a powder bed selectively. After laser exposure, the material solidifies immediately, and a new powder layer is applied, resulting in the layer-wise fabrication of the desired geometry. The selective and both locally and temporally varying energy input leads to a complex heat flow and temperature distribution, which govern the solidification and grain growth regime. Precise knowledge of the interaction of laser parameters and microstructural response within the workpiece opens up the prospect of tailored microstructure design.
Several properties characterize the microstructure of crystalline material, e.g., phase composition, microscopic defects such as porosity, and nanoscopic crystal lattice defects. Additionally, the grains' size, morphology, and crystallographic orientation severely impact the material's mechanical properties. The entirety of all crystallographic orientations in a polycrystalline material is summarized as its texture.
The texture is often described as the deviation from a random orientation distribution. The texture of a polycrystal is created during processes that define the microstructure, such as solidification and crystallization of a molten mass, recrystallization, and plastic deformation. The texture is determined by factors with a directed impact, such as mechanical force in plastic deformation or a directed heat flow during solidification [1].
A textured material exhibits anisotropic properties, e.g., Young's modulus, ductility, strength, and hardness. In LPBF, the directed solidification initially determines the texture as the material is exposed with a focused laser beam. The grain growth and the texture in a LPBF manufactured specimen are influenced by the melt pool geometry and the thermal gradients acting in the component [2]. As the melt pool solidifies, grains nucleate at the solidification front. During the subsequent grain growth, grains with different orientations compete [3]. Several studies found that grains with an orientation parallel to the melt pool boundary normal dominate during this phase [2,[4][5][6]. This preferred orientation results from the heat flow, which has its largest magnitude in the melt pool boundary's normal direction [7]. The preferred orientation is, furthermore, dependent on the crystal lattice. For example, face-centered cubic (fcc) metals such as the nickel-base alloy Inconel 625 grow in the preferred h100i direction [1].
The melt pool geometry depends on several factors, e.g., thermal conductivity, energy absorption, and heat flow [6,8]. Gong et al. [9] investigated the influence of different laser powers and scanning speeds on the melt pool geometry. They found that high laser powers and low scanning speeds lead to the keyhole effect. Here, the melt pool shows a broad opening at the top and tapers toward the bottom. The keyhole has a high penetration depth due to the high laser intensity and energy input, resulting in the remelting of previously solidified material.
Furthermore, keyholing leads to undesirable porosity [10]. Contrarily, a combination of low laser power and high scanning speeds leads to the balling effect. If the energy input is insufficient, the layer below will be insufficiently wetted. Spherical melt pools and bead-like structures are formed, which impede the following layers and can disrupt the manufacturing process by jamming the powder recoating mechanism [11].
The energy input also impacts grain growth. Higher laser powers lead to deeper melt pools, lower laser powers to shallower melt pools. In shallower melt pools, the normal vectors on the melt pool boundaries are more parallel than in deeper melt pools. They are parallel to the building direction of the part, therefore resulting in a preferred orientation in the building direction [6]. The melt pools are deeper when using lower scanning speeds or high laser powers, leading to a more significant variation of the melt pool boundary normal directions [4]. Therefore, crystallites do not only grow in the building direction, resulting in a lower preferred orientation [6]. Sun et al. [4] concluded that texture development could be controlled via process control. They successfully built parts from 316L, which showed a h011i fiber texture instead of the commonly found h001i for fcc alloys by using an increased laser power and a multi-scan method.
Yin et al. [5] found a dependence between the orientation of grain growth and laser power and scanning speed using FEM simulations for Ti-6Al-V4 processed via LPBF. They concluded that the layer number exerts an influence on the melt pool geometry and, therefore, on the grain orientation. For small layer numbers, the substrate plate strongly influences heat dissipation. Here, the crystal orientations show substantial deviations from the building direction. With increasing layer numbers, the melt pool becomes shallower, and the grain growth changes toward the building direction after a few layers. [5] Several approaches to tailor the microstructure via process parameter manipulation have been published to date. Roehling et al. [12] adjusted the shape of the laser beam to generate preferable microstructures and mechanical properties. Dehoff et al. [13] showed that tailored, site-specific textures could be designed by varying the process parameters. The typical route is a post-process heat treatment, usually aimed at achieving a recrystallized microstructure [14]. For Inconel 625, Marchese et al. [15] investigated the microstructural evolution after various heat treatments and found a recrystallized microstructure after a solution heat treatment. Sabzi et al. [16] recently presented the first experimental proof that dynamic recrystallization (DRX) occurs during LPBF as well, using ex situ EBSD measurements and thermomechanical modeling. However, an in situ observation of DRX during LPBF has not been reported yet.
Since DRX impacts the crystallographic texture, high-energy synchrotron radiation diffraction is wellsuited for its investigation. In situ diffraction experiments so far have been focused on phase transformations [17], cooling dynamics [18,19], and internal stresses [20,21]. Recently, Wahlmann et al. [22] presented results regarding the in situ formation of precipitates, another important constituent of the microstructure, in a nickel-base superalloy using synchrotron radiation diffraction and the LPBF device from the present study's group.
To the authors' knowledge, for LPBF, in situ texture analyses using high-energy synchrotron radiation have not been reported so far. Using a custombuilt LPBF machine designed for in situ diffraction measurements, experiments using Inconel 625 were carried out to determine the development of texture and the influence of repeated laser scanning on the microstructure.
Custom LPBF machine
In previous work, a custom LPBF process chamber was developed and integrated into the modular LPBF machine AconityMINI from Aconity3D GmbH, Herzogenrath, Germany, to realize in situ X-ray diffraction experiments with synchrotron radiation [23]. The custom LPBF machine is equipped with a 400 W Ytterbium fiber laser YLR-400-AC from IPG Laser GmbH, Burbach, Germany. The laser source emits light with a wavelength of 1070 nm in continuous wave. The laser is focused and deflected on the powder bed with an Axialscan-30 from Raylase GmbH, Wessling, Germany, with a focus diameter of ca. 60 lm in 1/e 2 specification. Before processing, the process chamber was sealed and purged with argon. During the processing, a clean argon atmosphere was maintained by using a filtration unit connected to the process chamber and coupled to a circulation pump. The powder bed was sandwiched between two glassy carbon windows. Inlet and outlet windows for the synchrotron radiation in the process chamber are made from polyimide foil. An automatic powder recoating mechanism inside the purged process chamber ensured the additive manufacturing of multi-layer parts without interrupting the experiment for manual operations.
Materials
The nickel-base alloy 625, commonly referred to as Inconel 625, was used as powder material and purchased from m4p material solutions GmbH, Magdeburg, Germany. The particles were nitrogenatomized and exhibited mostly spherical shapes, Fig. 1, with particle sizes ranging from 20 lm to 63 lm, Table 1. The median circularity of the particles, calculated with Eq. 1, was 0.85 with circ = 1 equaling a perfect circle.
Structural steel S235 plates with a size of 70 9 20 9 3 mm 3 were used as substrate material.
Process parameters and measurement modes
Two different LPBF parameter sets were investigated, Table 2. They are composed of two pairs of laser power P L and scanning speed v L ranging from a low laser power P L = 55 W and low scanning speed v L = 50 mm/s to a high laser power P L = 275 W and a high scanning speed v L = 760 mm/s. The ratio of laser power P L to scanning speed v L is denoted as line energy E L . The scanning pattern featured unidirectional scanning vectors aligned longitudinally to the incident synchrotron radiation beam, Fig. 2. Cuboid samples ca. 20 9 2.5 9 5 mm 3 in size were produced out of 100 layers with a layer thickness of Dz = 50 lm. Laser scanning was set to start at the left edge and finish at the right edge of the sample geometry. Two different measurement modes were used. In measurement mode 1 (MM1), the gauge volume (GV) distance to the working plane, z GV , is kept constant throughout the measurement of all the 100 layers. In measurement mode 2 (MM2), the absolute gauge volume position in the sample is maintained such that a defined volume element of the sample, defined by its distance to the samplesubstrate interface, f GV , is tracked throughout the process.
In this study, the axis parallel to the incident synchrotron radiation beam is called longitudinal direction (LD), which spans the working plane together with the transverse direction (TD). The third axis is parallel to the part height and called the building direction (BD).
Experimental procedure
In situ diffraction experiments were carried out at PETRA III, Deutsches Elektronen-Synchrotron (DESY) in Hamburg, Germany. The High Energy Materials Science (HEMS) beamline P07 [24], which is operated by Helmholtz-Zentrum Hereon, features a heavy load 6-axis positioning system on which the process chamber of the custom LPBF machine was mounted. A PerkinElmer XRD1621 area detector was used for diffraction pattern acquisition. For the diffraction experiments, synchrotron radiation energies of 87 and 98 keV were used. The synchrotron radiation beam size was set to 750 lm in TD and 70 lm in BD, Fig. 2. The diffraction patterns were acquired with an exposure time t = 0.1 s and frequency f = 10 Hz. Patterns were collected as layer-wise image series when the laser emission was active. The series measurement started with the laser being turned on and ended at the end of laser scanning. Therefore, not the whole cooling process after the laser impact was observed.
To complement the in situ diffraction experiments, two specimens were analyzed after production using standard metallographic preparation methods. For both parameter sets, the TD-BD plane was polished and etched in a solution of 1.5 ml hydrogen peroxide and 50 ml hydrochloric acid for t = 45 s. This preparation emphasized grain structures and melt pool boundaries in the subsequent optical microscopy using a Zeiss Axioskop 50.
Additionally, one sample, manufactured using P L = 55 W, was investigated by scanning transmission electron microscopy (STEM) at Zentraleinrichtung Elektronenmikroskopie (ZELMI) of Technische Universität Berlin. The specimen was cut using a focused ion beam on a FEI Helios Nanolab 600 FIB to ensure that the subsequent STEM analysis probed the same gauge volume as the in situ diffraction experiments in MM2. The investigations were carried out using a JEOL JEM-ARM300F2 STEM with a cold field emission gun and a probe-Cs corrector operating at an acceleration voltage of 300 kV and equipped with a JEOL Dual-EDX system using two 160 mm 2 SDD detectors. High-angular annular dark-field images (HAADF) were acquired because of their high sensitivity to the atomic number of the probed material.
Data evaluation
The detector calibration and integration were performed using the Python library pyFAI [25], pole figures were generated using MTEX [26]. The detector calibration was performed using a standard LaB6 powder sample to determine the detector-sample distance and the detector tilt for accurate integration results. Subsequently, the diffraction patterns were integrated in 5°sectors, resulting in 72 line profiles, including the first five hkl reflections for each diffraction pattern, Fig. 3. For each reflection, background subtraction was performed. Then, the integrated intensity was calculated following the trapezoid rule.
Additionally, the peak position was determined by fitting the line profile with a PseudoVoigt function using the python library LMFIT [27]. This fitting function also yielded the full width at half maximum (FWHM) of the reflections. For the fitting, only the top 60% of each peak were considered due to a slight asymmetry of the profiles, further discussed in Sect. 3.3. The peak positions and integrated intensities were converted to a format readable by MTEX to calculate the orientation distribution function (ODF), which was then plotted as a pole figure. This procedure was derived from Wenk and Grigull [28].
Representative results gathered from in situ experiments illustrate the evolution of microstructural features during LPBF. The data is presented based on two visualization styles. Pole figures represent the standard method of conveying texture. They show a color-coded intensity distribution as multiples of a random distribution (MRD). Each pole figure is calculated considering both the intensity and symmetry of the first five hkl reflections. The colormap chosen for plotting the pole figures emphasizes deviations from a random distribution, which would equal no preferred orientation. In this work, pole figures are used to visualize single texture states.
Temporally resolved texture evolution is visualized by time-dependent integrated intensity distributions, which are called processing plots in this study. These distributions show the azimuthal intensity for the (111) and (200) reflections. This way, local changes in the intensity distribution over time are visible. These visualizations do not contain the same amount of information as pole figure plots, though, which is why they are used complementarily.
Impact of laser irradiation on texture
In Fig. 4, the impact of the laser scanning over the gauge volume is shown for two different energy inputs, denoted by the different laser powers P L . For both laser powers, the changes in texture are documented for a single layer. Pole figures are given for the initial state, before laser impact, and the final state, at the end of the laser scanning time. Additionally, time-resolved intensity changes are made visible through the processing plots. Here, azimuths of 0°, 180°, and 360°correspond to TD, while 90°and 270°correspond to BD. In both measurements, the gauge volume was positioned at z GV = 150 lm below the top surface of the part. The azimuthal intensities for the (111) and (200) reflections were normalized for each sample since the respective experiments were conducted with varying incident beam intensities and energies.
The 275 W sample showed an intensification of the orientation distribution after the laser has passed over the gauge volume. The same parallel lines persist, with the intensities being much higher after the laser peak. The pole figures confirm this observation. The (200) pole figure shows an intensification of texture as the maximum MRD value increases. The local intensity maxima around BD are much more pronounced in the final state than the initial state. Similar textures were found in other build heights as well, Fig. 5, showing that this parameter set leads to a homogeneous texture in the part. The 55 W sample, on the other hand, shows a different orientation distribution after the laser impact compared to before. Initially, the highest intensities were found for the (111) reflection in both TD and BD. As the laser scans over the gauge volume, a highintensity spot appears at 120°, which fades quickly. The intensity line in BD shifts toward 45°, while the one at 180°is shifted to a larger azimuth angle. The (200) reflection shows lower intensities than the (111) reflection with less distinctive features, except for intensity lines emerging between 20°and 45°after laser impact. The (200) pole figures change drastically from the initial to the final state, reflecting the (200) reflection's response to a (111) fiber texture. In the (111) pole figure, a spot at TD is visible in both the initial and final states, while a slightly inclined spot from BD appears to shift counterclockwise in the final state.
In Fig. 6, optical micrographs of the part-substrate interface regions of both samples are shown. This section was chosen because of the clarity of the melt pool boundaries. The low laser power and high line energy result in clearly deeper melt pools than the higher laser power, which features broader and shallower melt pools. The higher laser power shows columnar grains parallel to BD, which surpass the Figure 4 Comparison of the influence of different energy inputs on texture: Intensity distribution throughout a single layer in MM1 for n = 50, z GV = 150 lm. Circled area in processing plot for P L = 55 W shows high intensity that rapidly decreases after laser impact. melt pool boundaries. The grains visible in the micrographs are larger for the high laser power than for the low laser power, where the grain morphology appears less oriented.
Impact of repeated laser scans on texture
Another phenomenon appears when observing a constant gauge volume throughout the whole process. Figure 7 shows pole figures of the same gauge volume at increasing numbers of total layers. Each pole figure represents the final state after laser scanning, similar to Fig. 4. The gauge volume was located in the center of the fifth layer of the sample at f GV = 250 lm. For each pole figure set, the total number of layers is given below the estimated total part height at that moment. For n = 7, consequently, two additional layers of solidified material were above the gauge volume, five layers for n = 10 et cetera.
Initially, a bimodal fiber texture is observed at n = 7. The (200) plane shows a strong preferred At n = 10, five layers have been exposed atop the gauge volume, equaling five additional laser passages and about 250 lm of solidified material above. Here, the (111) fiber texture seems to have almost completely vanished, while the (200) fiber texture is still visible and pronounced. Still, the maximum intensity is significantly reduced compared to n = 7. At n = 14, the intensities are further reduced. The (111) pole figure has completely changed compared to its initial state at n = 7, and the fiber texture for the (200) plane is further weakened, signified by a maximum intensity of only 2.4 compared to 6.2 for the first pole figure.
In Fig. 8, the changes in the intensity distributions are temporally resolved. The azimuthal intensity distribution for the (111) and (200) reflections are shown layer by layer from n = 7 to n = 15. A logarithmic intensity scale was used to unveil all intensity changes in an extensive range for this figure.
The laser passage over the gauge volume is visible in each plot, signified by a slight intensity change at about t = 4 s. As more material is deposited above the gauge volume, this effect gets weaker. At n = 7, the initial (111) texture seen in the corresponding pole figure in Fig. 7 is created. Before the laser impact, the intensity distribution appears homogeneous. Afterward, five significant lines emerge, which persist until the laser impact at n = 9. The (200) reflection shows changes in the intensity distribution as well. Initially, there is a slight shift of a prominent intensity at 270°to about 255°. This line persists up to n = 15 and further onwards. After n = 14, no further changes in either intensity distribution were observed, which mirrors the findings from the pole figures in Fig. 7.
Repeated laser scanning distorts peak shape
Further analysis of the peak shape was carried out to evaluate possible solid-state phenomena occurring during LPBF. Initially, the FWHM was determined as described in Sect. 2.5, as shown in Fig. 9b.
The azimuthally resolved results are shown in Fig. 9a. Each horizontal line corresponds to the azimuthal distribution of the (200) reflection's FWHM at the end of the corresponding layer. Up to layer n = 10, the image appears relatively undefined with changing distributions layer by layer. At this stage, the LPBF process has likely not reached a steady-state concerning the deposited and solidified layer thickness [29]. Generally, the FWHM values appear higher for TD than for BD in this region, though.
After n = 20, two prominent lines emerge at around 90°and 270°, corresponding to increased peak widths in the building direction. Finally, between n = 30 and n = 40, these lines reach their maximum, with the rest of the distribution not changing anymore. This increase was further evaluated since the amount of material of about 1.25 to 1.5 mm between the gauge volume and the laserimpacted powder layer seemed significant, and the intensity distributions shown in the previous chapter stayed unchanged at these layer numbers.
The peak shape was determined to be one possible factor to distort the FWHM value. The measured peak profiles are slightly asymmetrical with a less steep left side. If this asymmetry were to increase over time, it would impact the FWHM value. The integrated intensity was divided into two sections to evaluate the asymmetry: left of the peak position and right of the peak position. As mentioned before, the peak position was determined via a PseudoVoigt fit of the top 60% of the peak, where the impact of asymmetry is negligible. Therefore, the left and rightside integrals were calculated separately. The ratio of left-to-right was used as an asymmetry factor. Its evolution over time is shown in Fig. 9c for the first three hkl reflections (111), (200), and (220). The remaining two collected reflections (311) and (222) showed similar behavior.
After an initial fluctuation, the ratio reaches a local minimum between n = 20 and n = 25 depending on the reflection, after which it slowly increases linearly. Therefore, the peaks become more asymmetrical over time, all while the material stays in a solid state, and the impact of the laser on the gauge volume is expected to be reduced as the amount of material between gauge volume and top layer increases. It was therefore concluded that the asymmetry was not the deciding factor behind the FWHM increase described above.
Ex situ microstructure analysis
The influence of repeated laser scanning on the microstructure was further evaluated using highresolution scanning transmission electron microscopy (STEM). The sample using P L = 55 W observed in MM2 was investigated in the region where the synchrotron gauge volume was located. A high-angular annular dark-field image detector was used due to its high sensitivity to the atomic number and the resulting Z contrast. Atoms with a higher atomic number appear brighter in the resulting image. The overview in Fig. 10a shows an inhomogeneous microstructure with a lot of contrast. Several bright features with a size of 100-300 nm in the interdendritic regions stand out. Energy-dispersive X-ray spectroscopy (EDX) images in Fig. 10c-f reveal that these segregations have high Nb and Mo as well as Si contents.
Influence of energy input
As described in the introduction, the energy input is one of the deciding factors for forming preferred orientations in the material. In this study, the parameter set with the lower laser power of P L = 55 W had the higher energy input with a line energy of E L = 1.1 J/mm compared to a line energy of E L = 0.36 J/mm for the parameter set with the higher laser power of P L = 275 W. Generally, higher energy inputs result in deeper melt pools and less textured material, which was confirmed in the present study [4].
The high laser power parameter set showed a (200) fiber texture along the building direction regardless of build height and layer number, while the low laser power did not result in a significant preferred orientation. The parameter set with a line energy of E L = 0.36 J/mm resulted in shallower melt pools, therefore, more parallel grain growth directions. As a result, crystallites grow epitaxially, signified by the intensification of existing orientations and the columnar grain growth transcending the melt pool boundaries.
The parameter set with a line energy of E L = 1.1 J/ mm leads to deeper melt pools and a reduced preferred orientation. The changes in the intensity distribution during laser scanning imply a certain depth of the heat-affected zone that enables orientation changes. The single intensity spot visible only for a short period during laser scanning, annotated in Fig. 4, indicates the consumption of grains during grain growth.
In situ recrystallization stage
The measurements in a constant gauge volume revealed significant changes in the intensity distribution in solidified material, which coincide with a reduction in texture. The texture changes from a strong bimodal (200) fiber texture in BD and (111) fiber texture in TD to a significantly reduced preferred orientation, where only a weak impression of the (200) fiber texture remains. These changes occur between n = 7 and n = 14, with 100 lm to 450 lm of material between the gauge volume and the top layer exposed by the laser. After the total number of layers has reached n = 14, no more intensity changes are observed.
The focused laser beam and its highly localized energy input, together with the extremely high heating and cooling rates, lead to lattice defects such as dislocations in LPBF [30]. The high defect density and the thermal cycling, combined with the fact that the material stayed in a solid state for the whole observation period, lead to the conclusion that DRX is the cause for the observed texture changes. While recovery, which is characterized by reducing lattice defects, results in a reduction in the FWHM, DRX leads to reformation of the microstructure, which results in the changes in the pole figures observed together with the reduction in the FWHM. These findings prove that the repeated laser scanning acts as an in situ heat treatment, which could be used to manipulate and tailor the resulting microstructure of the final part.
In situ segregation stage
In Sect. 3.3, a peak shape analysis was carried out. Anisotropic changes in the FWHM were observed.
An initial reduction in the FWHM coincided with texture changes, further supporting the DRX hypothesis discussed in the previous section. The peak width is influenced by several factors: temperature, domain size, microstrains, defect density.
Temperature can be excluded as a cause since it would impact the whole diffraction pattern, i.e., the whole azimuthal range equally. Domain size is inversely proportional to the peak width, and smaller domains lead to broader peaks. A decrease in domain size is possible during recrystallization, but after the initial microstructural changes, repeated laser scans are expected to lead to grain growth and, therefore, an increased domain size, resulting in a smaller FWHM. It is unlikely that there is a grain size reduction in BD at later stages in the process.
Defect density and microstrains are correlated. Defects such as stacking faults, dislocations, and subgrain boundaries exert a stress field, which leads to microstrains. In this case, the initial reduction in FWHM corresponds to the annihilation of lattice defects due to recovery and DRX. The increase in later stages could be attributed to an increase in microstrains. Due to the increased FWHM values in BD, higher strains and defect concentrations are expected in this direction. As mentioned in the results section, the FWHM values appear to stay constant starting from n = 40, implying that no new defects are generated from hereon.
Additionally, the peak asymmetry was evaluated. In an earlier study [20], this asymmetry was attributed to powder material irradiated simultaneously with the solidified material. However, the changes in asymmetry observed in the present study indicate that this is not actually the case. Instead, the increase in asymmetry implies a change in the phase composition of the material.
On one hand, Inconel 625 forms coherent precipitates after heat treatment [15]. Sarkar et al. [31] reported that these precipitates show diffracted peaks overlapping to the matrix. Especially the c 00 phase leads to a peak shape also found in the present study. It is difficult to extract the exact contribution of the precipitates, which is why their presence is often analyzed via changes of the matrix peak as described by Rai et al. [32]. The c 00 phase forms a small peak that contributes to the left-sided asymmetry presented in Sect. 3.3.
On the other hand, Nb and Mo segregation is a known phenomenon for Inconel 625 processed via LPBF [15], especially in the as-built state. Chemical inhomogeneities of any kind, including segregations, will impact the diffraction peak shape because they change the lattice spacing in the gauge volume. Nb and Mo-rich areas, found in the ex situ TEM investigations, have a different lattice spacing compared to regions with the nominal chemical composition, which will also induce an asymmetry in the diffraction peak.
Therefore, the increase in asymmetry that was observed indicates an increase in segregation concentration. The repeated laser scanning apparently promotes the growth of segregations in situ. This is an interesting and new finding since the impact of the laser on the gauge volume is expected to be reduced with an increasing number of layers. From these results, it appears that the energy input is high enough to promote the growth of segregations until the very end of the process, where there are more than 4.5 mm of material between the powder layer and the gauge volume. The TEM results clearly show Nb and Mo segregations, thereby supporting the hypothesis that the segregation growth is depicted by the in situ diffraction experiments.
Conclusions
This study shows that in situ diffraction experiments are a viable tool for texture analysis during laser powder bed fusion. Several factors impact the formation and evolution of texture. The energy input plays a significant role in the texture in the part. A combination of high laser power and scanning speed, but with a lower line energy E L , results in a stronger preferred orientation than low laser power and scanning speed with a higher line energy E L .
Additionally, the observation of a constant gauge volume revealed the impact of repeated laser scanning and the heat-affected zone on the microstructure. Significant texture changes were found during laser scanning up to a distance of ten layers below the working plane. The preferred orientation was strongly reduced. This texture change is attributed to recrystallization since the material stays in a solid state during this observation period. The present study is the first to experimentally observe this phenomenon in situ to the authors' knowledge. After the recrystallization stage, further microstructural changes occur. Apparently, lattice defects reappear after an initial reduction during recrystallization with a prominent preference for the building direction. This insight was gathered from an analysis of the diffraction peak FWHM. After a total number of 40 layers, no further changes of the FWHM occurred, but the peak shape still changed. Evaluating the peak asymmetry revealed an increase in an underlying contribution to the individual hkl reflections. This underlying contribution is related to the formation and growth of segregations, which was confirmed via ex situ TEM investigations. The peak asymmetry increases up to the end of the observation period at n = 100 layers, which underlines the impact of the repeated laser scanning during the segregation stage.
The results presented in this study contribute to the understanding of microstructure formation in LPBF. The in situ observations of recrystallization and segregation formation emphasize the complexity of thermal management during LPBF and highlight the relevance and capabilities of high-energy synchrotron radiation diffraction experiments to advance the fundamental process understanding. Further investigations are necessary to fully understand the influence of process parameters on the microstructure and facilitate a targeted microstructure design during the process.
Declarations
Conflict of interest The authors declare that they have no conflict of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licen ses/by/4.0/.
|
2021-11-09T14:40:33.825Z
|
2021-11-08T00:00:00.000
|
{
"year": 2021,
"sha1": "41ef64e27364a92d657af45acdd6bdc895d19851",
"oa_license": "CCBY",
"oa_url": "https://link.springer.com/content/pdf/10.1007/s10853-021-06577-8.pdf",
"oa_status": "HYBRID",
"pdf_src": "Springer",
"pdf_hash": "b158d71e15d0f345187d3cb5c479294cdc4d441c",
"s2fieldsofstudy": [
"Materials Science",
"Engineering"
],
"extfieldsofstudy": []
}
|
211096202
|
pes2o/s2orc
|
v3-fos-license
|
Rapamycin inhibits proliferation and apoptosis of retinoblastoma cells through PI3K/AKT signaling pathway
Effects of Rapamycin on the proliferation and apoptosis of retinoblastoma cells through the phosphatidylinositol 3-hydroxy kinase (PI3K)/protein kinase B (AKT) signaling pathway were studied. The retinoblastoma Y79 cells were selected and divided into negative control group (NC group), 0.2 µM Rapamycin group and 0.4 µM Rapamycin group. Then the proliferative activity of Y79 cells was detected using Cell Counting Kit-8 (CCK8) assay, the content of reactive oxygen species (ROS), malondialdehyde (MDA) and superoxide dismutase (SOD) in cells in each group was detected using enzyme-linked immunosorbent assay (ELISA), and the apoptosis of Y79 cells was detected via terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay. Moreover, the changes in Y79 cell cycle and apoptosis were determined through flow cytometry, and apoptosis and PI3K/AKT pathway were detected using reverse transcription-polymerase chain reaction (RT-PCR) and western blotting. It was found that the number of cells and the proliferative activity were significantly reduced in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group. In 0.2 µM Rapamycin group and 0.4 µM Rapamycin group, the content of ROS and MDA was significantly decreased, while that of SOD was notably increased. TUNEL assay and flow cytometry showed that in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group, the number of apoptotic cells was obviously increased, and the cell cycle was basically arrested in S phase. The expression levels of Bcl-2, PI3K and AKT declined in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group, whereas the expression of Caspase 8 increased. Similar results were also obtained in the protein assay. The above results were significantly superior in 0.4 µM Rapamycin group to those in 0.2 µM Rapamycin group. Rapamycin inhibits proliferation and promotes apoptosis of retinoblastoma cells through inhibiting the PI3K/AKT signaling pathway.
Introduction
Retinoblastoma is the most common primary intraocular malignant tumor in infants and young children, and it ranks 3rd among all tumors only following leukemia and neuroblastoma (1,2). Despite some progress made in the treatment of retinoblastoma, some major problems have not been resolved yet. Traditional external beam radiation is used to control large tumors, but there are complications, including the secondary malignant tumors, such as osteosarcoma, which leads to a higher incidence rate of hereditary retinoblastoma in patients (3,4). Although external therapies such as chemotherapy have become indispensable treatment for retinoblastoma currently, it results in noticeable complications (5). Therefore, search for new therapeutic strategies to improve the clinical efficacy of patients with retinoblastoma is urgently needed.
Rapamycin is a kind of macrolide produced by Streptomyces hygroscopicus, which was originally developed as an antifungal agent (6). However, it has been used more for other purposes when its strong immunosuppressive and antiproliferative properties were found (7). Nowadays, Rapamycin contributes to the treatment of some cancers through inhibiting the Rapamycin target pathway in mammals (8). Over the years, it has been proven that apoptosis is the primary mechanism for eliminating cancer cells, including intrinsic or mitochondrial pathway and extrinsic or death receptor pathway (9). Therefore, it is of great significance to study the proliferation and apoptosis of retinoblastoma cells.
The phosphatidylinositol 3-hydroxy kinase (PI3K)/protein kinase B (AKT) pathway is an important signaling pathway that affects the cellular energy metabolism, cell size, cell cycle, cell proliferation, survival and apoptosis, which is closely related to other important signal transduction pathways (10)(11)(12). This pathway is composed of three major driving molecules: PI3K, AKT and the mammalian target of rapamycin (mTOR) (13)(14)(15). However, the role of the PI3K/AKT signaling pathway in retinoblastoma still needs further research. The aim of the present study was to enrich and improve the theoretical basis of the effects of Rapamycin on proliferation and apoptosis of retinoblastoma cells through the PI3K/AKT pathway.
Cell culture and grouping. The retinoblastoma Y79 cells purchased from ATCC were quickly taken out from the liquid nitrogen container, and rapidly thawed in 65˚C sterile water prepared in advance, followed by centrifugation at 2,500 x g at 20˚C for 10 min. The supernatant was discarded. After the above operation was repeated several times, the cells were resuspended with medium and inoculated into a 6-well plate at an appropriate density, followed by incubation in a thermostatic incubator. The medium was replaced every other day. Then the cells in good growth status were divided into negative control group (NC group), 0.2 µM Rapamycin group and 0.4 µM Rapamycin group (the dose of Rapamycin was determined by preliminary experiments (not shown). After stimulation for 24 h, the cell samples were collected.
Cell proliferation assay using Cell Counting Kit-8 (CCK-8).
The cells in the logarithmic growth phase in each group were inoculated into a 96-well plate and cultured in the thermostatic incubator with 5% CO 2 at 37˚C for 0, 24, 48, 72 and 96 h. Then the medium was discarded, and 110 µl of developing solution was added into each well, followed by incubation for 1 h at 37˚C in the incubator. The absorbance in each group was detected at 450 nm using an ultraviolet spectrophotometer, and plotted into the line chart to reflect the cell proliferative activity.
Detection of cytokines in each group. After stimulation, the cells in good growth status in the three groups were selected from the incubator, and the medium was discarded. The cells were collected using a cell scraper and lysed using the RIPA lysis buffer (strong), followed by centrifugal separation at 2,500 x g at 20˚C for 10 min. Then the supernatant was collected to detect the levels of ROS, MDA and SOD using the ELISA kits (HM10870, Bio-Swamp; HM10250, Bio-Swamp; CSB-E17064h, Cusabio Biotech Co., Ltd.) according to the actual situations and instructions. Finally, the absorbance in each group was measured using a microplate reader.
Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) apoptosis assay. Apoptosis of Y79 cells was detected using the apoptosis assay kit (Roche), as follows: The cells in each group were collected, and the supernatant was discarded. The cells were washed with phosphate buffered saline (PBS), added with proteinase K working solution, immersed in blocking buffer, fixed, rinsed and infiltrated with 0.1% Triton X-100, followed by fluorescein isothiocyanate (FITC) end labeling of apoptotic DNA fragment using the TUNEL assay kit. The FITC-labeled TUNEL-positive cells were observed under a fluorescence microscope, and the TUNEL-positive cells were counted in 10 fields of view.
Detection of cell cycle using flow cytometry. The cells were treated for 24 h and washed twice with PBS, and the cell sediment at the bottom of tube was collected. Then the cell sediment was added with 1 ml of pre-cooled 70% ethanol, transferred into the 1.5 ml Eppendorf (EP) tube and fixed at 4˚C overnight, followed by centrifugation at 1,750 x g at 20˚C for 5 min. The supernatant was collected. With 3 replicates in each group, the cell sediment was suspended with 500 µl of binding buffer according to the instructions, added with 5 µl of propidium iodide (PI) dye and placed at room temperature for 30 min in the dark. Finally, the cell cycle was detected using the flow cytometer according to the programmed operation. Flow cytometer (FACSCalibur; BD Biosciences) was used for analysis. Data were obtained and analyzed using the CellQuest professional software (version 3.3; BD Biosciences).
Detection of apoptosis via flow cytometry. The cells were treated for 24 h and washed twice with PBS, and the cell sediment at the bottom of tube was collected. With 3 replicates in each group, the cell sediment was suspended with 500 µl of binding buffer according to the instructions, added with 5 µl of Annexin V-binding buffer and 5 µl of PI dye and placed at room temperature for 10 min. Finally, the apoptosis rate was measured using the flow cytometer according to the programmed operation.
Detection of gene expression using RT-PCR. The RNA was extracted from cells and synthesized into DNA using the kit (Takara) in accordance with the instructions. The primer amplification system (20 µl) was constructed using 2 µl of complementary deoxyribose nucleic acid (cDNA), 10 µl of qPCR mix, 2 µl of primers and 6 µl of ddH 2 O. Then, PCR amplification was performed: pre-denaturation at 95˚C for 2 min, 94˚C for 20 sec, 60˚C for 20 sec and 72˚C for 30 sec, for a total of 40 cycles. The primer sequences of target genes and the internal reference GAPDH were designed according to those in the GenBank (Table I). The expression levels of target genes were detected via RT-PCR.
Western blotting. The cells at an appropriate density in each group were collected, from which the protein was extracted according to the instructions, and the protein concentration was calculated. After that, the protein was subjected to water bath and centrifugation at 10,500 x g at 4˚C for 10 min. Then western blotting was performed: 12% separation gel and 5% spacer gel were prepared for protein loading and electrophoresis, and the protein was transferred onto a membrane using the semi-dry method, sealed, incubated with the primary antibodies overnight and incubated again with the secondary antibodies. The protein band was scanned and quantified using the Odyssey scanner, and the level of protein to be detected was corrected using GAPDH. Finally, the protein expression was calculated through gray scan.
Statistical analysis. All raw data obtained in the experiments were statistically analyzed using Statistical Product and Service Solutions (SPSS) 21.0 software (IBM Corp.), the validity of raw data was retained, and multiple comparisons were performed. The experimental results were expressed as mean ± standard deviation (mean ± SD). Analysis of variance followed by post hoc test (least significant difference) was applied for the comparison among groups. P<0.05 suggested that the difference was statistically significant. The bar graph was plotted using GraphPad Prism 8.0.
Results
Results of cell proliferation assay using CCK8. The absorbance in each group at different time points was measured through proliferation assay using CCK8. As shown in Fig. 1, the proliferation ability of Y79 cells was significantly stronger in NC group than that in other groups at 24, 48, 72 and 96 h (P<0.05), and it was the weakest in 0.4 µM Rapamycin group (P<0.05).
Detection results of oxidative stress cytokines in each group. As shown in Table II TUNEL apoptosis assay results. The level of apoptosis in each group was determined using TUNEL staining. As shown in Fig. 2, there were fewer TUNEL-positive cells in NC group, and they could hardly be observed. The number of TUNELpositive cells in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group was obviously higher than that in NC group, and it was the highest in 0.4 µM Rapamycin group (P<0.05). The above results suggest that Rapamycin can promote apoptosis of Y79 cells.
Cell cycle detection via flow cytometry. The cell cycle in each group was detected using flow cytometry. The results manifested that there were more cells in S phase and fewer Table III). The above findings demonstrate that 0.4 µM of Rapamycin can inhibit the proliferation of retinoblastoma cells, and arrest the cell cycle in S phase.
Apoptosis detection using flow cytometry. Apoptosis level in each group was determined through flow cytometry. It was found that the apoptosis rate in NC group was lower, and (Fig. 3), indicating that Rapamycin can promote apoptosis of Y79 cells.
RT-PCR results. The results of RT-PCR revealed that 0.2 µM Rapamycin group and 0.4 µM Rapamycin group had evidently lower expression levels of Bcl-2, PI3K and AKT, and evidently higher expression of Caspase 8 (P<0.05), while the above expression levels were the opposite in NC group (Fig. 4), which suggests that Rapamycin suppresses cell proliferation and promotes apoptosis, further inhibiting the occurrence of retinal diseases.
Discussion
Retinoblastoma is the most common primary intraocular malignant tumor in infants and young children. Despite some progress made in the treatment of retinoblastoma, some major problems have not been resolved yet, including the secondary malignant tumors, such as osteosarcoma. Therefore, it is urgent to search for new therapeutic strategies to improve the clinical efficacy of patients with retinoblastoma. Rapamycin is a kind of macrolide produced by Streptomyces, which can contribute to the treatment of some cancers through inhibiting the Rapamycin target pathway in mammals (16). In the present study, the effects of Rapamycin on the proliferation and apoptosis of retinoblastoma cells through the PI3K/AKT pathway were explored. It was found through CCK-8 assay that the proliferation ability of Y79 cells was significantly stronger in NC group than that in other groups at 24, 48, 72 and 96 h, and it was the weakest in 0.4 µM Rapamycin group.
In addition, the cell cycle was detected using flow cytometry. The results manifested that there were more cells in S phase and fewer cells in G2 phase in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group. The cell cycle in 0.4 µM Rapamycin group was obviously superior to that in 0.2 µM Rapamycin group. The above findings demonstrate that 0.4 µM of Rapamycin can inhibit the proliferation of retinoblastoma cells, and arrest the cell cycle in S phase, consistent with previous studies (17). The level of ROS in cancer cells is higher than that in normal cells due to oncogene stimulation, increased metabolic activity and mitochondrial dysfunction. Under the condition of continuous oxidative stress, cancer cells are adapted through a series of mechanisms, which not only activate ROS scavenging system, but also inhibit cell apoptosis. Therefore, understanding the ROS adaptation mechanism is very important for killing cancer cells and solving the problem of drug resistance. SOD is ubiquitous and prevents the optic nerve conduction abnormality. MDA can resist the effect of SOD, with cytotoxicity (18). In this study, we chose the dose of Rapamycin (0.2 and 0.4 µM) based on our preliminary experimental results, which was the basis of this study. Results showed that in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group, the content of ROS and MDA was significantly decreased, while that of SOD was significantly increased, all of which were markedly superior in 0.4 µM Rapamycin group to those in 0.2 µM Rapamycin group, indicating that 0.4 µM of Rapamycin can inhibit the oxidative stress, further suppressing the occurrence of retinoblastoma. The mitochondrial function should be further examined to more accurately account for this effects. Apoptosis can be induced by internal and external signals, and the release of cytochrome c due to the loss of mitochondrial membrane potential is the key to the activation of Caspase 9, causing intracellular damage. The apoptosis activator transmits signals to the cytoplasm, leading to the activation of Caspase 8, and results in apoptosis through the in vitro pathway (19,20). The cleaved Caspase 8 and 9 accompanied by the cascade activation of Caspase have been observed in studies, and Rapamycin can trigger the intrinsic and extrinsic PI3K/AKT apoptotic pathways of human retinoblastoma Y79 cells, thus facilitating apoptosis (21). In this study, the apoptosis level in each group was determined using TUNEL staining. It was found that there were fewer TUNELpositive cells in NC group, and they could hardly be seen. The number of TUNEL-positive cells in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group was obviously larger than that in NC group, and it was the largest in 0.4 µM Rapamycin group. Moreover, the results of flow cytometry showed that the apoptosis rate in NC group was lower, and the apoptotic cells could hardly be observed. The number of apoptotic cells was obviously increased in 0.2 µM Rapamycin group and 0.4 µM Rapamycin group compared with that in NC group, and it was the largest in 0.4 µM Rapamycin group, indicating that Rapamycin can promote apoptosis of Y79 cells. The PI3K/ AKT pathway is a central regulator for cancer proliferation, tumorigenesis and metastasis. PI3K is a lipid kinase family that phosphorylates the phosphate-3-hydroxyl. Both AKT and mTOR are downstream targets of PI3K, and they can stimulate protein synthesis, cell growth and proliferation. mTOR is an important component of the network, as well as a PI3K-associated serine-threonine kinase that can regulate antiapoptosis and survival mechanisms through phosphorylating AKT (22,23). In this study, the results of RT-PCR revealed that 0.2 µM Rapamycin group and 0.4 µM Rapamycin group had evidently lower expression levels of Bcl-2, PI3K and AKT, and evidently higher expression of Caspase 8, while the above expression levels were the opposite in NC group. According to the results of western blotting, 0.2 µM Rapamycin group and 0.4 µM Rapamycin group had remarkably lower protein levels of PI3K and AKT and a remarkably higher protein level of Caspase 8, while the above levels were the opposite in NC group, which suggests that Rapamycin suppresses proliferation and promotes apoptosis of retinoblastoma cells through inhibiting the PI3K/AKT signaling pathway, thereby further inhibiting the occurrence of retinal diseases. Differently, Wang et al (21) found that Rapamycin disturbed mitochondrial membrane potential and subsequently helped cytochrome c release from mitochondria to cytosol and activated Caspase 8, inducing apoptosis in human retinoblastoma Y79 cells. In subsequent research, animal experiments need to be introduced to further explore the deeper regulatory mechanism of PI3K/AKT signaling pathway from in vivo and in vitro levels.
In conclusion, it was found that Rapamycin may regulate the proliferation and apoptosis of retinoblastoma cells through inhibiting the PI3K/AKT signaling pathway, so Rapamycin may be used as a therapeutic drug for patients with retinoblastoma.
|
2020-02-06T09:03:07.515Z
|
2020-01-30T00:00:00.000
|
{
"year": 2020,
"sha1": "7e5def47d0a9b07aff328e24ac26907940b339ce",
"oa_license": "CCBYNCND",
"oa_url": "https://www.spandidos-publications.com/10.3892/ol.2020.11363/download",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "1bd32f693f61881098625026563816333e386637",
"s2fieldsofstudy": [
"Biology",
"Chemistry"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
90861384
|
pes2o/s2orc
|
v3-fos-license
|
Effect of Antimicrobial Triclosan on Reproductive System of Male Rat Pharmaceutica Analytica Acta
Triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol: TCS) is a synthetic, broad-spectrum antibacterial agent used in broad range of household and personal care products including hand soap, toothpaste, and deodorants. Recently, concerns have been raised over TCS’s potential for endocrine and reproductive disruption. This review contains the information about deleterious toxic effects of TCS on reproductive system of male rat and the possible mechanism. The literature findings showed that TCS deadly affects the reproductive profile of male rats. According to literature TCS depress the testicular function of male rat including spermatogenesis and steroidogenesis by decreasing the androgen production. 3-hydroxysteroid dehydrogenase (3β-HSD) and 17β-hydroxysteroid dehydrogenase (17β- HSD) are two critical enzymes in the steroidogenesis pathway, while according to findings TCS treated rats had lowered concentration of androgen. TCS treated rats also showed significant lowered testicular weight, number of germ cells, Sertoli cells, Leydig Cells, primary spermatogonia, secondary spermatogonia, and spermatocytes. These observations suggest that TCS have degenerative and retrogressive effects on rat testes. Overall the literature findings showed that TCS decrease the production of androgen like testosterone, Luteinizing Hormone (LH) FSH resulting decreased sperm production and histopathological
Introduction
Triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol: TCS) is an effective antifungal and antibacterial agent that is frequently used throughout the world in pharmaceuticals, personal care products, plastics, and fabrics [1]. TCS have effective response against many types of bacteria and fungi, it permeates the bacterial cell wall and targets multiple cytoplasmic and membrane sites, including RNA synthesis and the production of macromolecules, and it also blocks the synthesis of fatty acids [2]. TCS may be characterized as a halogenated aromatic hydrocarbon, containing phenol, diphenyl ether and polychlorinated biphenyl substructure [3]. The chemical structure of TCS is halogenated biphenyl ether, which confers its chemical properties to many toxic compound including polychlorinated biphenyls (PCBs), bispenol A and dioxin [4].
TCS was invented over 40 years ago, but it has been used gradually more over the past few years and in 1998, worldwide annual production of TCS is about 1500 tons [5]. At present, TCS is one of the more usually detected contaminants in aquatic and global environments [6]. The popularity of antibacterial consumer products has led to increased consumer use of TCS [7] and personal care product are the main source of environment contamination of TCS, these product contain about around 0.1% to 0.3% (W/W) of TCS [8]. In China, TCS ranged from 11-478 ng/L in surface water, while in Pearl River system it ranged from 50-1330 ng/kg [9]. TCS has been detected in many human matrices, including human milk, urine and blood [10], while the National Health and Nutrition Examination Survey of 2003-2004, more than 70% of U.S. residents had detectible TCS in their urine [11]. Thus, it shows the prevalent exposure of the general population to TCS.
Endocrine-Disrupting Chemicals (EDC) includes a variety of environmental contaminants that deadly affect the endocrine system [12]. The effects of anti-androgenic chemicals are dependent on the timing of exposure, but the periods of gestational and pubertal development are particularly susceptible. Despite its high occurrence across the ecosystem, the health effects of TCS have not been well studied. Recently, TCS is suspected to be a potential male reproductive toxicant. Many studies have demonstrated that TCS might have endocrine disrupting effects in animals and humans [13]. According to different studies, the mode of action of TCS as an EDC can be viz. estrogenic or weak androgenic or anti-androgenic. TCS has also been shown to function as an anti-androgen since it inhibits testosteroneinduced transcriptional activity [14]. Fourteen days TCS exposure in Japanese medaka fry (Oryzias latipes) suggested that TCS is potentially androgenic [15]. Another study reported that TCS has high toxicity on the early life stages of medaka, and the metabolite of TCS may be a weak estrogenic compound with the potential to induce vitellogenin in male medaka, as well as delaying the hatching in females [16]. Normal reproduction depends on the androgenic activity of animal [17], while, TCS treated rats showed depressed level of hormones production, which are most crucial components for testicular function. The aim of this review was to investigate the endocrine disrupting effects of TCS on male rats and to evaluate the mechanism.
Toxicity of TCS
Triclosan is a broad-spectrum antimicrobial drug and has been classified as a Class III drug by FDA [18]. TCS was declared as Priority Existing Chemical for full assessment under the Industrial Chemicals Act [6]. Contradictory findings have been found about the link between TCS and adverse health impacts in humans and animals.
Absorption, distribution, metabolism and excretion are rapid in the case of TCS in human body and are primarily excreted via urine. It has been reported that TCS exposure in human caused skin irritation [19]. Contact with TCS with further exposed to sunlight can trigger a Photo-Allergic Dermatitis (PACD) reaction, that can result in symptoms, like as irritation on the body parts [20], that might be due to long term work of active ingredient, consequently long-standing contact to TCS in turn increases the risk of PACD. Another area of concern is related to the hypothesis that TCS augments the production of chloroform. It was reported that TCS may engage in the production of chloroform, under specific conditions can almost double the chloroform formation in the drinking water treated with chlorine [21].
Toxicity of TCS also have been widely studied in various animal models, like in mice it was found that, TCS showed negative effect on the metabolism of thyroidal hormones, resulting hypothermia and generally depression of the central nervous system [22]. Similarly, the study reported that TCS treated rats showed significant decrease in sperm production [23]. Decrease in sperm production could be due to TCS interference with the metabolism of thyroid hormone. Another study in sheep pointed out that TCS can hinder the estrogen sulfotransferase activity and raises concern about its possible effects on the ability of the placenta to supply estrogen to the foetus, which would cause negative effects in the foetus development [24]. It was also reported that long term exposure to TCS in mice induced liver carcinoma [25]. Due to rising level of TCS in the environment, bacterial strains are more probable to acclimatize by developing resistance [26]. TCS has different vital medical applications; therefore the future objective must be to maintain these important applications while eliminating the gratuitous ones for its safe use.
Effect on Male Reproductive Profile
Data from wildlife and laboratory animal studies has increased apprehension that toxic chemicals might alter reproductive development in the human population. Numbers of factors are behind compromised male reproductive health including nutritional, lifestyle and environmental factors. In recent years, a large number of data has accumulated that suggests that the trend of decreasing male fertility might be due to misuse of medicine and due to exposure to environmental toxicants.
Effect on reproductive organ
The testicles are the male sex glands; produce sperm and testosterone. The seminiferous tubules and contents are the primary contributor to testicular weight. Testis size is highly correlated with fertility, poor fertility often being associated with small testis, as because process of spermatogenesis occurs in the seminiferous tubules of the testis and lower weight of testis is sign of smaller number or length of seminiferous tubules [17]. It was reported by Kumar et al. [23] that TCS treated group had decreased weights of testis, epididymis, ventral prostate, vas deferens and seminal vesicles between 20-50%, including regressive histological changes in the seminiferous tubules resulting in the suppression of spermatogenesis. Disturbances in the synthesis of androgens have been reported due to use of TCS and it is well documented that androgenic disturbance causes negative changes in reproductive organ weights, as because constant androgenic stimulation is necessary for normal growth and functions of testis, epididymis and accessory sex organs [27]. Same result was also observed with other toxic compound like Prochloraz [28] and 5fluorouracil, including some environmental contaminant like PCBs [29], bispenol A [30] and dioxin [31].
On other hand, several other studies also suggested that TCS exhibited no significant negative effect on the weight of reproductive organ [32], the difference between results can be due to use of different method, environment and dose rate. In addition it is also have been reported that TCS treated rats group body weight was not significantly altered which is sign that TCS is not toxic to the animals as well as non-androgenic in nature, since androgens are known to possess anabolic activities like stimulating the development and growth of the skeleton and skeletal muscles.
Effect on spermatogenesis
Spermatogenesis is a complex process started at the early stage of fetal development including the division and differentiation of spermatogonial stem cells into mature spermatozoa. The spermatogenesis progression include the several phases, specifically the spermatocytes productions by mitotic proliferation of spermatogonial stem cells, following the division of spermatocyte to produce haploid round-spermatids, and in the last stage linking the early stages of round spermatids to be mature elongated-spermatids. During spermatogenesis together with germ and Sertoli cells, Leydig cells also play an important role; they produce testosterone, which is important for the maintenance of both secondary sexual functions and spermatogenesis [33]. It has been reported that TCS treated male rat showed significant decrease in the total number of sperm including decreased sperm motility, effectiveness and increased percentages of dead sperm [23]. Its shows that chemical that are toxic to the testis can alter the quantity and quality of the sperm produced via spermatogenesis resulting infertility problems in males. The decrease in number of sperm production can be due decreased testosterone production in TCS treated rat, which is important for sperm production [34].
Physiologically, testosterone is synthesized in Leydig cells under stimulation by Luteinizing Hormones (LH) [35], following LH binds to its receptor, and it stimulates adenylyl cyclase, resulting increased concentration of intracellular cAMP [36]. This increased level of cAMP, results the activation of various agents of steroidogenic cascade causing an amplified production of testosterone [37]. As Leydig cells are the primary site for testosterone, so the production of testosterone can be connected with testicular weight as because seminiferous tubules and contents are the primary contributor to testis weight. Decreased testicular weight in TCS treated rat means decreased seminiferous tubules and number of Leydig cells, resulted reduced testosterone production. Testosterone is the androgen that is important to regulate spermatogenesis. In fetal male rats, 3-5 days before birth serum testosterone levels are elevated and remains high until 8 days after birth. Testosterone level decrease gradually at reached up-to 0.2 ng/mL after 8-24 days of birth [38]. In adults at (30-55 days) testosterone levels rise to 1-2 ng/mL, stable adult level [39]. While, the physiological advantages of elevated testosterone levels is not clear, but the higher testicular concentration of testosterone is vital as because spermatogenesis process requires 70 ng/mL and spermatogenesis is considerably compromised at testosterone concentrations below 20 ng/mL [40]. Thus, decreased testosterone level can be linked with lower sperm production, which can lead to infertility because fertility depends on the number of normal sperm produced. This depressed spermatogenesis could also be due to the reduced level of serum Follicle Stimulating Hormones (FSH), a hormone directly involved in maintaining spermatogenesis in conjunction with testosterone [34]. Depressed levels of androgens also reduce the sperm dynamics thereby affecting physiological maturation of the sperm resulting in reduced sperm count, motility, and density [41]. A study conducted by Raut and Angus [42] also reported the same effect of TCS in fish, according to their founding 35 days exposure of TCS to fish caused significant decreased sperm production.
With significant decrease in number of sperm due to improper spermatogenesis process, TCS treated rats also showed, decreased sperm motility, effectiveness and increase in the percentages of dead sperm, compared to control group [43]. Another study conducted by Wang et al. [32] reported that TCS Rats treated with the high dose (200 mg/kg) of TCS showed a significant decrease in daily sperm production, changes in sperm morphology and epididymal histopathology. Considering the histopathological change in the epididymis, TCS may induce the epididymal damage due to the epididymal accumulation of that. Toxic compound can damage the sperm cell by physiological, cytotoxic and genetic mechanisms and morphological abnormalities in sperm might have been caused by alterations in testicular DNA that in turn disrupts the process of differentiation of spermatozoa.
Effect on steroidogenesis
The synthesis of steroid hormones is one of the critical processes in the endocrine regulation. Most toxicants impair steroidogenesis and decrease Leydig cell function by inducing ROS and/or by decreasing the levels of steroidogenic enzymes. Steroidogenesis in Leydig cells involve numbers of steps and most of which are dependent on apposite concentrations of cAMP [44]. While, with the help of adenylyl cyclise, cAMP is produced, thus, an appropriate functioning of the enzyme adenylyl cyclase is vital for balanced steroidogenesis. The rate of steroidogenic pathway mainly depends on the transportation of cholesterol from outer to inner mitochondrial membrane with the help of transmembrane protein, while, steroidogenic acute regulatory protein (StAR) and suitable cAMP is crucial for this step [45].
Kumar et al. [23], reported that TCS treated rats showed, decreased cAMP concentration and a decreased expression of StAR protein, thus it shows that TCS have negative effects on production of steroid hormones, which are vital for normal reproductive performance in males and females. This depressed accessibility of cAMP could be due to decreased availability of ATP to the adenylyl cyclase enzyme or by a decreased performance of the enzyme itself. Another study carried out by Kumar et al. [46] to check TCS-induced anti-androgenecity in rat Leydig cells, also showed that TCS decrease the activity of adenylyl cyclase enzyme and in turn leads to the interference of intermediate steroidogenic flow resulting decreased testosterone production. After successful transportation of cholesterol to the inner mitochondrial membrane with help of StAR protein, for regulating the steroidogenesis, the expression and availability of cytochrome P450 side chain cleavage (P450SCC) enzyme is required [47], while TCS treated mouse showed depressed level [46]. The decreased P450SCC could be the outcome of a decreased availability of cholesterol in the inner mitochondrial membrane. 3-hydroxysteroid dehydrogenase (3β-HSD), 17β-hydroxysteroid dehydrogenase (17β-HSD) are two critical enzymes in the steroidogenesis pathway, While in TCS treated rats these were also found in decreased level [23]. Decreased testosterone production in treated rat might be also due decreased expression and activity of P450SCC, 3β-HSD and 17β-HSD enzymes, as a constant and stable status of all of them is required for an optimal testosterone synthesis.
In conclusion these findings showed that the P450SCC, 3β-HSD and 17β-HSD are important targets for the actions of EDC, while the initiation of this androgen inhibition starts, when EDC disrupts the activity of adenylyl cyclase enzyme. This leads to an abridged cAMP accessibility, and disturbed the whole cAMP-dependent steroidogenic pathway, resulting depressed StAR expression and down regulation of all key steroidogenic enzymes.
Effect on histopathology
The testis consists of two types of tissues: Seminiferous tubules, including Sertoli cells, and the interstitial compartment, including Leydig cells [48]. Efficiency of spermatogenesis depends on the integrity of the seminiferous tubules, Sertoli cells and endocrine regulation, while this process is regulated by testosterone and FSH. In response to LH, Leydig cells produce androgen, like testosterone, which along with FSH bind to their respective Sertoli cell receptors to regulate spermatogenesis.
The tendency of chemicals, drugs and or compound to harm reproductive processes in animals and humans is of enormous distress to toxicologists and the public. Normal testicular functions (spermatogenesis, steroidogenesis) depend on normal histology of testis. It has been reported TCS treated rat showed a number of histopathological malformations, compared to control group [49], which possibly affect the production and maturation of the sperms. A study carried by Mahmoud and Solaiman [49] reported that TCS treated rat showed cytological and nuclear degenerative changes in seminiferous tubules, resulting significantly lowered number of germ cell, Sertoli cells, Leydig Cells, primary spermatogonia, secondary spermatogonia, and spermatocytes. These observations suggest that TCS have degenerative and retrogressive effects of TCS on rat testis. Kumar et al. [23] reported that several malformations were observed in the Lumina of epididymal tubule, vas deference, and prostate tissues, including reduced sperm density in epididymal tube. The decreased testosterone and androgenic receptor level in treated rats might have led to the degenerative changes and atrophy in the sex accessory tissue resulting decreased weight. It also shows that decreased sperm count in TCS treated rat could be due decrease spermatogenesis in testis. The reduction in rate of spermatogenesis is due to decreased FSH and testosterone production in testis, as production of these hormones depend on the normal integrity of Leydig and Sertoli cells. Degenerative changes observed in the cauda, ductus deferens and prostate glands might be attributed to the decrease in the level of androgen and androgen receptors which are known to support the functioning and continuous persistence of these organs.
Conclusion
TCS is commonly used chemical as an antimicrobial agent in various cosmetics and other applications. The frequent use of TCS and its subsequent entry into the environment is an alarm due to its deadly effects, so some regulations should be made to prevent its accumulation during the next decades. Infertility problems are increasing due to environmental contaminations and frequent use of therapeutic drugs. Present review literature showed that TCS act as an endocrine disruptor in male rats, it inhibits the production of androgen by reducing the LH and cholesterol production; depressed StAR expression and finally down-regulation of several key steroidogenic enzymes. It confirmations that TCS pose a hazard to human and animal's health, so, it is recommended that caution should be exercised in the use of it.
|
2019-04-02T13:03:06.602Z
|
2016-11-28T00:00:00.000
|
{
"year": 2016,
"sha1": "4a97d377bcdb6a5f429bb4a44ee105d295988d10",
"oa_license": "CCBY",
"oa_url": "https://www.omicsonline.org/open-access/effect-of-antimicrobial-triclosan-on-reproductive-system-of-male-rat-2153-2435-1000516.pdf",
"oa_status": "HYBRID",
"pdf_src": "Adhoc",
"pdf_hash": "4290048b5525cd51697553db344fdc06fbf00c5d",
"s2fieldsofstudy": [
"Medicine",
"Environmental Science",
"Biology"
],
"extfieldsofstudy": [
"Biology"
]
}
|
119268728
|
pes2o/s2orc
|
v3-fos-license
|
Multifractal nature of the surface local density of states in three-dimensional topological insulators with magnetic and nonmagnetic disorder
We compute the multifractal spectra associated to local density of states (LDOS) fluctuations due to weak quenched disorder, for a single Dirac fermion in two spatial dimensions. Our results are relevant to the surfaces of Z_2 topological insulators such as Bi_2Se_3 and Bi_2Te_3, where LDOS modulations can be directly probed via scanning tunneling microscopy. We find a qualitative difference in spectra obtained for magnetic versus non-magnetic disorder. Randomly polarized magnetic impurities induce quadratic multifractality at first order in the impurity density; by contrast, no operator exhibits multifractal scaling at this order for a non-magnetic impurity profile. For the time-reversal invariant case, we compute the first non-trivial multifractal correction, which appears at two loops (impurity density squared). We discuss spectral enhancement approaching the Dirac point due to renormalization, and we survey known results for the opposite limit of strong disorder.
I. INTRODUCTION
The defining attribute of a 3D Z 2 topological insulator 1 (TI) is the presence of an odd number of 2D massless Dirac bands at the material surface. 2,3 Unlike the Dirac electrons that can appear in a purely 2D system (notably in graphene), the surface states of a (strong) 3D TI are robustly protected from the opening of gap, so long as time-reversal symmetry is preserved. The protection can be viewed as a consequence of the parity anomaly, [3][4][5][6] which "holographically" links surface states separated by a topologically non-trivial bulk, and gives rise to the signature properties of the Z 2 TI state: the half-integer quantum Hall effect, quantized magnetoelectric coupling, "axion" electrodynamics, etc. 2,3 As stressed by Schnyder et al. in Ref. 7, the robust character of the surface states in the presence of quenched disorder can also be taken as a principal characteristic of a topological insulator. In particular, these states are protected from Anderson localization, 8 even in the presence of a "strong" impurity potential, so long as time-reversal invariance is preserved. 9,10 With its 2D Dirac band pinned to an exposed surface, a 3D TI is ideally suited to local probes such as scanning tunneling microscopy (STM). In spectroscopic mode, an STM captures an areal map of the local density of states (LDOS). There are several ways of analyzing such data. One is to look for quasiparticle interference (QPI) [11][12][13][14][15] in the LDOS Fourier transform. This method is useful for determining short-distance details, and contains similar information as an analysis of LDOS Friedel oscillations in the presence of a single impurity. 16 It has been applied in TIs to experimental data and analyzed theoretically in Refs. 12,13 and 14,15, respectively. In QPI, the disorder is employed primarily as a facilitator to gleam information about the clean system. 11 Multifractal analysis [17][18][19] provides a complementary method better suited to extracting large-distance, disorder-dominated features in the same LDOS data field. It is a standard tool for assaying quantum interference phenomena, and is employed in the analysis of wavefunctions near a metal-insulator transition [18][19][20][21] as well as Sketch of disorder "flavors" on the surface of a Z2 topological insulator. In the time-reversal invariant case, the impurities are neutral adatoms or charged dopant ions, depicted as spheres in (a). The effects of these on the surface Dirac theory [Eq. (3.1)] are encoded in the scalar potential V (r). In the case of magnetic disorder, the impurity spins are indicated by the arrows in (b) and (c). In the limit that the spins reside in the plane of the surface, (b), the disorder appears as a vector potential A(r). The opposite case of outof-plane polarization, (c), gives the random mass M (r). The case of generic time-reversal breaking disorder has all three potentials present. mesoscopic fluctuations in diffusive metallic systems. 22,23 In this paper, we derive new results for LDOS multifractal spectra associated to disordered topological insulator surface states. In particular, we extend the pioneering results of Ref. 24 to the generic cases of time-reversal (T ) preserving and breaking impurities. Our calculations are performed in the near-ballistic limit, 25 wherein weak disorder enters as a perturbation to the clean Dirac band structure. A key characteristic of 2D Dirac fermions is that this weak disorder regime is continuously connected to more conventional domains of multifractal analysis, i.e. the diffusive (symplectic) metal 20,23 and the integer quantum Hall plateau transition. 18,[26][27][28] These appear at strong coupling (many impurities) for dirty Dirac fermions. 9,10,24,29 We consider the case of a single flavor Dirac surface band, relevant to (e.g.) the TIs Bi 2 Se 3 and Bi 2 Te 3 (Refs. 2,3,30). The different kinds of T -preserving and T -breaking disorder are sketched in Fig. 1. We demonstrate that the LDOS multifractal spectra observed in the absence of time-reversal symmetry breaking (i.e., for non-magnetic disorder) is qualitatively weaker than that induced by magnetic impurities. In particular, the first multifractal correction obtains at first order in the impurity density for the case of broken T , while the first non-trivial amplitude appears at second order in the Tinvariant case. We compute the leading terms via oneand two-loop calculations, respectively. We also compute unnormalized spectra for the spin LDOS 31 in the case of magnetic impurities. We show that renormalization effects can enhance multifractality near the Dirac point. Finally, we summarize prior results on various strongcoupling regimes. Our goal is to sketch the full portrait of quantum interference physics on the surface of a TI, valid when interparticle interactions can be neglected.
Our results indicate that the long-distance, disorderdominated features captured by the multifractal analysis behave in many cases opposite to the short-distance characteristics that appear in quasiparticle interference. [12][13][14][15] In Ref. 14, the authors observed that QPI is strongest for the spin LDOS response to magnetic impurities, while the unpolarized LDOS pattern vanishes for magnetic disorder (in the first Born approximation). The QPI response of the LDOS to non-magnetic disorder is weak but non-zero. 14 By contrast, in this work we find that the LDOS multifractality is strongest for magnetic impurities, while the spin LDOS spectrum comparatively exhibits the same or weaker strength fluctuations, depending upon the polarization direction.
The weak influence of non-magnetic disorder is tied to the intrinsic spin-orbit coupling that defines the massless Dirac kinetic term. Multifractality is suppressed at one loop due to interference mediated by the Dirac pseudospin, which is proportional to the physical spin on a Z 2 insulator surface. The spin is also responsible for the suppression of backscattering from a single non-magnetic impurity. 32 On the TI surface, magnetic disorder Zeeman couples directly to the Dirac spin, enabling backscatter-ing in near-ballistic transport, and inducing multifractal LDOS fluctuations at the lowest order in the impurity density.
A notable problem in experiments probing topological insulator surface states has been the unintentional doping of carriers into the bulk bands, which then dominate transport measurements in large samples. 33 Even if the chemical potential is moved into the gap, it may reside far from the Dirac point, making it difficult to observe surface state carrier dynamics at low densities. In this respect, STM offers several advantages over transport experiments. First, the position of the chemical potential is no barrier to probing states at the Dirac point, since the latter can always be reached by tuning the bias voltage (although the Dirac point is not guaranteed to reside in the bulk gap). 3,30 Assuming that the Dirac point or the low density regime can be accessed by tuning the tunneling bias, the advent of a finite, even large doping of the surface and/or bulk states may actually play a beneficial role in facilitating the observation of disorder-induced quantum interference effects. This is because a finite carrier density screens the long-range Coulomb potential introduced by charged defects. The potential landscape formed by screened impurities is short-range correlated on scales larger than the screening length. Good screening eliminates the problem of electron and hole puddle formation, 34,35 which has until recently 36 occluded transport and other properties of Dirac carriers in graphene near the Dirac point. On the other hand, a low density of poorly-screened bulk dopants induces a long-range correlated potential and puddle formation, as in graphene. 13 LDOS fluctuations in the puddle regime are an important topic for future work.
Three-dimensional topological insulators provide us with an interesting paradigm flip for quantum interference phenomena. Isolating the surface state contribution in transport measurements is problematic. By comparison, direct LDOS imaging is easier than in conventional semiconductor systems, wherein the 2D electron gas is typically buried in a layered material stack. Moreover, the amount of surface disorder can to some extent be controlled; for example, magnetic impurities can be deposited across the surface of an otherwise high-quality bulk 3D sample. These can be charge-neutral adatoms or charged dopants; an example of the former (latter) is provided by iron (manganese) 37 in Bi 2 Se 3 . This paper is organized as follows. We begin in Sec. II with a lightning review of multifractal composite and spin LDOS measures. In Sec. III, we present new results for multifractal LDOS fluctuations in TI surface states, in the presence of weak disorder. We also show how renormalization can enhance multifractality close to the Dirac point. Finally, in Sec. IV, we review previous results on various strong disorder regimes relevant to the Z 2 TI surface states and LDOS statistics. In particular, we discuss the symplectic metal, the integer quantum Hall plateau transition, and the Anderson insulator. Various technical details are relegated to appendices. In Appendix A, we review the symmetry classes of Anderson (de)localization that appear in the disordered Dirac surface theory. In Appendix B, we supply some details of our perturbative calculations.
A. Definitions
We suppose that the tunneling local density of states (LDOS) ν(ε, r) is imaged at a fixed energy ε over an L×L field of view. The field is finely partitioned into a grid of boxes. The box edge length a ≪ L must be chosen larger than any "microscopic" scale l m , such as the correlation length of the random potential. 18 One introduces the box probability where A n denotes the n th box. LDOS multifractality is defined through the inverse of the participation ratio (IPR), 18,20 . (2. 2) The right-hand side (scaling limit) obtains when l m ≪ a ≪ L; corrections are down by higher powers of a/L. The exponent τ (q, ε) is the multifractal moment spectrum 18,38 for LDOS fluctuations at energy ε. The construction in Eqs. (2.1) and (2.2) is useful for characterizing a system with extended states, or for an Anderson localized system in which L ≪ ξ loc (ε); ξ loc denotes the localization length. In what follows, we assume experiments are performed at sufficiently low temperatures so that inelastic cutoffs to quantum interference can be ignored. 8,39 A clean system with plane wave states at energy ε has τ (q, ε) = 2(q − 1). Multifractality refers to the incorporation of corrections non-linear in q. Physically, these arise due to quantum interference via multiple scattering of electron waves in a dirty environment, processes that serve as the precursor to Anderson localization. 18,20,22 For weak disorder, the spectrum is typically dominated by the quadratic correction 18,20 τ (q, ε) = 2(q − 1) − θ(ε) q(q − 1), (2.3) where θ ≥ 0 gives a measure of the disorder strength. As an example, in a weakly disordered 2D metal (with L ≪ ξ loc for the orthogonal or unitary classes), one finds 20,21,23 4) where N (ε) denotes the average density of states, D is the classical (Drude) diffusion constant, and β ∈ {1, 2, 4}, depending upon the presence or absence of time-reversal symmetry and spin-orbit scattering. 23,40 At stronger disorder, higher order corrections in θq must be retained; for the diffusive metals, results are known to four loops. 20 An alternative characterization of LDOS multifractality is provided by the singularity spectrum 18,38 f (α): Over a subset of the sample grid area that scales as (L/a) f (α) , the box probability µ ∼ (a/L) α . The singularity spectrum is the Legendre transform of τ (q), For the quadratic spectrum in Eq. (2.3), one obtains In this "parabolic approximation," the strength of the multifractality is encoded in the peak position α 0 [f (α 0 ) = 2], and the width α W of the spectrum such that f (α 0 ± α W /2) = 0, Part of the power of multifractal analysis for disordered quantum systems derives from the fact that the spectra [τ (q) or f (α)] typically depend only upon a few gross measures of the impurity potential. In the case of dirty metals, the entire spectrum can be computed as an expansion in one parameter, the inverse conductance (consistent with scaling theory). 8,20,22 At a non-interacting Anderson localization transition, τ (q) and f (α) become universal functions, so that the critical point is characterized by an infinite set of critical exponents [e.g., the expansion coefficients for τ (q)].
The spectra above have been defined for data collected in a single fixed realization of the disorder. Strictly speaking, Eq. (2.3) then applies only for |q| ≤ q c , where q c = 2/θ. Outside of this range, the τ (q) associated to a fixed disorder realization is linear, a phenomenon known as spectral termination. [41][42][43] [This assumes that higher order corrections can be ignored for q ≥ q c . Regardless, the τ (q) spectrum is always linear for sufficiently large q]. Termination can be viewed as a consequence of the restriction to positive sample measures f (α) ≥ 0. 19,38 In the localized regime, the states contributing to the LDOS at a given position in the sample have a discrete energy spectrum, quantized by the typical localization volume ξ 2 loc . As a result, all non-unity LDOS moments diverge in the absence of level smearing. In a tunneling experiment, smearing can appear due to inelastic scattering (temperature), open sample boundary conditions, or due to the finite energy resolution of the instrument. To characterize an Anderson insulating state over an L × L field of view with L ≫ ξ loc , the full LDOS distribution should be examined; 44,45 sensitive dependence of the distribution shape to smearing can serve as a telltale sign of the localized regime. LDOS fluctuations in the Anderson insulator are reviewed in more detail in Sec. IV B.
B. Spin LDOS spectra
By restricting the character of the tunneling species, it may be possible to measure individual LDOS components separately. For example, in the case of a spin-polarized (ferromagnetic) STM tip, 31 the spin-projected components ν ↑,↓ can be separately resolved. The use of an unpolarized tip recovers the composite LDOS ν = ν ↑ + ν ↓ . We define the spin LDOS along the spin space direction ι, For a time-reversal invariant system (with or without spin-orbit scattering and/or disorder), one has νι(ε, r) = 0. In a system with broken time-reversal (e.g., magnetic impurities), but zero average spin polarization, the integral of νι(ε, r) over a sufficiently large region becomes arbitrarily small; we cannot use the normalized construction in Eqs. (2.1) and (2.2) to characterize spin LDOS multifractals. Instead, we employ the un-normalized inverse spin participation ratio (ISPR) Pι q (ε) ≡ n µι n q , µι n ≡ An d 2 r νι(ε, r). (2.8) In the scaling limit, where the exponent xι q is the scaling dimension for the corresponding moment operator in the disorder-averaged field theory description, and c q = 0 for even q.
III. WEAK DISORDER MULTIFRACTALITY
A. Model. Short-and long-range correlated potential landscapes The Dirac surface states of a Z 2 topological insulator (TI) are guaranteed to appear in an odd number of flavors. 2,3 In this paper, we consider the simplest case of a single flavor, relevant to (e.g.) Bi 2 Se 3 and Bi 2 Te 3 . The Hamiltonian is (in units such that = 1) where µ ∈ {1, 2}, and repeated indices are summed. The coordinates r = {x, y} chart the TI surface, while the topological bulk resides in the perpendicular z direction. In Eq. (3.1), v F denotes the Fermi velocity, and the Dirac pseudospin Pauli matricesσ are related to the physical spin 1/2 operatorsŜ via {σ µ ,σ 3 } = 2{ǫ µν S ν , S 3 }. The vector, scalar, and mass potentials {A, V, M } describe the effects of external electromagnetic fields and/or surface impurities. In the absence of time-reversal (T ) symmetry breaking, A = M = 0. (See Appendix A for an enumeration of discrete symmetry operations.) Thus, a mass gap is explicitly forbidden so long as T remains a good symmetry, a consequence of the protection afforded by the topologically nontrivial bulk. When T is broken by an external magnetic field B, the vector and mass potentials are where γ (γ ⊥ ) denotes the Zeeman coupling to the inplane field B (out-of-plane field B z ), and the orbital Non-magnetic adatoms or charge traps are encoded in the scalar potential V (r). In-plane (out-of-plane) polarized magnetic impurities additionally induce point exchange coupling to the vector A(r) [mass M (r)] fields. 16 The different types of disorder leading to V , A, and M are sketched in Fig. 1. Assuming a random surface distribution of impurities and spatial rotational invariance on average, the disorder potentials can be taken as Gaussian white noise distributed variables, The dimensionless variances ∆ V,A,M quantify the disorder strength. In the first Born approximation, these are of the form where n imp is the impurity density, andũ(q) denotes the Fourier transform of the single impurity potential. We note that a net in-plane magnetization of the surface impurities A µ = 0 can be removed by a gauge transformation, while the average scalar potential V is absorbed into the chemical potential. We will assume that there is no net magnetization perpendicular to the surface, M = 0, or that we only probe LDOS fluctuations on energy scales much larger than the induced gap 2v F M . In 2D, the single impurity potential u(r) [Eq. (3.4)] must decay faster than 1/r 2 (or oscillate rapidly enough) so that the limitũ(q → 0) exists; otherwise, the white noise assumption in Eq. (3.3) is invalidated by long range impurity potential correlations. 46 This causes a problem for charged impurities, which can become poorly screened for a small surface doping relative to the Dirac point. In graphene, the long-range correlated potential undulations induced by poorly-screened substrate impurities leads to a smearing of the Dirac point over an energy scale k B T rms ∝ v F √ n imp , and to the breakup of the sample into electron and hole puddles. 34,35 The advent of electron-hole puddles has until recently prevented the observation of various "intrinsic" phenomena associated to the Dirac carriers in graphene experiments such as velocity renormalization 36 and hydrodynamic transport near the Dirac point. In this respect, a large surface or bulk doping actually improves the situation for STM measurement of disorder-induced quantum interference, since these carriers screen the potential of surface charges. The disorder potential can be considered short-range correlated for scales larger than the screening length.
If we consider only surface doping, with an insulating bulk, then the Thomas-Fermi wavelength due to a finite surface carrier density n is given by where α ≡ e 2 /ǫv F is the effective "fine structure constant." The permittivity ǫ = (1 + ǫ TI )/2, the average of the bulk TI below and vacuum above the surface. For Bi 2 Se 3 with a surface density of n = 7×10 12 cm −2 (corresponding to a doping level of 0.3 eV relative to the Dirac point), 30 v F = 5 × 10 5 m/s (Ref. 30), and permittivity 47 ǫ TI = 113, one obtains λ TF ∼ 90 nm. This is very large, and indicates that the surface state carrier density is inadequate to screen charged impurities. A smaller screening length is possible for bulk doping, 13 or by performing experiments on thin film samples exfoliated over a metallic gate. Alternatively, one can restrict the deposition of surface impurities to non-doping adatoms, e.g. iron in Bi 2 Se 3 . 37 The disorder variance associated to Thomas-Fermi screened charged impurities is Finally, we note that the appearance in isolation of any of the three disorder potentials in Eq. (3.1) realizes three different symmetry classes of Anderson (de)localization, 7,48,49 see Appendix A for a review. The T -invariant case with ∆ A,M = 0 belongs to the spin-orbit class AII, which is also the class of the Z 2 topological bulk [ Fig. 1(a)]. In the case of broken T , ∆ V,M = 0 realizes the random vector potential model in class AIII [ Fig. 1(b)], while ∆ V,A = 0 gives the random mass model in class D [ Fig. 1(c)]. All three classes exhibit delocalized states in 2D, although this occurs only at the Dirac point for class AIII. 24 In the T -invariant symplectic case, the unpaired single Dirac flavor avoids the usual spinorbit metal-insulator transition, 9 remaining delocalized even for strong disorder due to a topological term. 10 The generic case of broken-T with all three disorder potentials non-zero realizes the unitary class A, and is believed to flow under renormalization to the plateau transition in the integer quantum Hall effect. 24,29 (See Sec. IV A 2 for a review).
Because in-plane (out-of-plane) Zeeman coupling appears in the vector (mass) potential [Eq. (3.2)], one is tempted to identify class AIII (class D) with the limit of an otherwise clean surface, dusted with charge neutral magnetic impurities randomly polarized in-plane (perpendicular to the TI surface). However, a magnetic adatom is expected to also induce a local scalar potential deformation V (r). For example, it can dope the surface or bulk, as occurs for a manganese impurity in Bi 2 Se 3 (Ref. 37)]. As discussed in Appendix A, the advent of any two flavors of disorder destroys the additional discrete symmetries enjoyed by the special class D and AIII Hamiltonians. The asymptotic long-distance LDOS scaling is then governed by the unitary class A, discussed above. Nevertheless, depending upon the relative microscopic strength of the magnetic versus potential perturbations induced by polarized magnetic impurities, the class AIII or D model may provide an adequate approximation for broken-T LDOS fluctuations on intermediate scales.
B. Results
To compute the scaling of LDOS moments in a quantum theory with quenched disorder, one employs a path integral Z to express products of fermion Green's functions as functionals of the disorder configuration. Using a trick (replicas, 8,20,22 supersymmetry, 50 or Keldysh 51 ) to normalize Z = 1, the Green's functions are formally averaged over disorder configurations (typically with a Gaussian weight). The result is a translationallyinvariant, but "interacting" field theory, where the disorder strength ∆ appears as a coupling constant. 8,50 Perturbative calculations are controlled via loop expansion for small ∆.
To determine the scaling, one decomposes the q th LDOS moment into projections upon the renormalization group (RG) eigenoperators of the disorder-averaged theory. [20][21][22] The multifractal spectrum τ (q) is determined by the most relevant (negative) 52 scaling dimension x q exhibited by an eigenoperator in this decomposition, and is given by 19,43 (3.7) 1. Broken T : random vector potential disorder (Class AIII) The properties of the model in Eq. (3.1) with shortrange correlated disorder [Eq. (3.3)] were originally studied in Ref. 24. In this work, the exact multifractal spectrum τ (q) was calculated for the broken-T , random vector potential (∼ in-plane polarized magnetic impurity) 53 class AIII model, to all orders in ∆ A . Technically, this result obtains because the disorder-averaged AIII model is conformally invariant at the Dirac point, and the exact LDOS moment spectra can be extracted using an Abelian bosonization treatment. The exact spectrum 24 is quadratic in q, and takes the form of Eq. (2.3), with Subsequent work 41,42 on the random vector potential model elucidated the mechanisms of termination and freezing, transitions that occur in the spectral statistics for large moments q > q c (∆ A ) or strong disorder ∆ A ≥ 2π. For this broken-T class, we can also examine the spin LDOS fluctuations, utilizing the same nonperturbative bosonization treatment employed in Ref. 24. The spin LDOS νι(ε, r) taken along an axisι in spin space was defined by Eq. (2.7). Moment fluctuations are characterized by the inverse spin participation ratio (ISPR) in Eq. (2.8), the scaling limit of which is controlled by the dimension xι q that appears in Eq. (2.9). The out-of-plane ISPR P3 q (ε) is associated to the "mass" fermion bilinear ν3 = ψ †σ 3 ψ. For the random vector potential model, the most relevant contribution to P3 q (ε) carries the same scaling dimension that gives the composite LDOS scaling in Eqs. (2.3) and (3.8), The chiral components of the in-plane spin LDOS are the energy-resolved U (1) Dirac current operators Moments of these are RG eigenoperators that receive no corrections. The scaling of the associated ISPR is governed by the disorder-independent (tree level) exponent Eqs. (3.8), (3.9), and (3.11) are exact results that hold to all orders in ∆ A .
Broken T : random mass disorder (Class D)
In the rest of this section, we provide new results for the broken-T , random mass (∼ out-of-plane polarized magnetic impurity) 53 class D model, the T -invariant class AII model, and the generic broken-T unitary class A model. For weak disorder, none of these are conformally invariant, and we resort to perturbation theory. In this section we summarize results; some technical aspects are sketched in Appendix B. The results obtained below hold only for small ∆ V,M ≪ 1, wherein the disorder appears as a weak marginal perturbation (at tree level) to the clean Dirac surface band structure.
For the broken-T case of random mass disorder (with ∆ V = ∆ A = 0), one obtains quadratic multifractality at one loop, again governed by Eq. (2.3), with Moments of the out-of-plane spin LDOS operator ν3 = ψ †σ 3 ψ, as well as of the chiral in-plane [U (1) current] operators ν ± = ψ †σ ± ψ constitute RG eigenoperators at one loop, with scaling dimensions 13) (3.14) Note that the first correction in Eq. (3.13) is positive (and linear in q); this should be contrasted with the AIII case, Eq. (3.9) above. On general grounds, the anomalous scaling dimension associated to the q th ≥ 1 moment of the composite LDOS, or any projected component thereof, must appear with a negative sign. The reason is that this quantity is associated to a moment of a normalized probability distribution 18
Non-magnetic disorder (Class AII)
In the T -invariant case of scalar potential disorder, it turns out that no local operator (without derivatives) exhibits multifractal scaling to first order in ∆ V . For Dirac fermions, this applies to both LDOS and energy-resolved current moments. Physically, the weak influence of nonmagnetic disorder is due to interference mediated by the Dirac pseudospin (equivalent to physical spin 1/2 on the TI surface). The Dirac pseudospin is also responsible for the suppression of backscattering from a single nonmagnetic impurity. 32 Technically, this result is derived by mapping the one-loop renormalization process of local operators to the action of a certain spin-1/2 Hamiltonian H (eff) V , and identifying renormalization group eigenoperators with states that diagonalize H The first non-trivial correction to the LDOS τ (q) appears at two loops. To this order, the spectrum is again quadratic as in Eq. (2.3). A straight-forward but laborious calculation gives the coefficient in this equation, Since ∆ V ∝ n imp [Eqs. (3.4) or (3.6)], we find that the non-trivial multifractal scaling begins at second order in the impurity density. This is qualitatively weaker than any of the broken-T regimes, where the quadratic multifractality appears already at first order, Eqs. (3.8) and (3.12). This distinction between T -invariant and T -broken surfaces is our primary result, and can be tested directly in STM experiments by varying the concentration of deposited surface disorder. Although the T -invariant case is not conformally invariant (for a discussion of renormalization effects, see Sec. III C, below), the multifractal τ (q) and f (α) spectra depend only upon a single parameter, the variance ∆ V . Eq. (3.15) can be extended to higher loops, allowing ever more precise tests against numerics or experimental data within the perturbatively accessible regime. The multifractal spectrum therefore provides a unique fingerprint for the time-reversal invariant Dirac surface state of the Z 2 topological insulator, in the presence of weak but otherwise generic non-magnetic disorder. The opposite limit of strong disorder for the T -invariant case is discussed below in Sec. IV A 1.
Broken T : generic disorder (Class A)
When T is broken and any two disorder flavors appear, the system resides in the unitary class A. The third disorder flavor is always generated under renormalization-see Sec. III C, below. The results of Eqs. (3.8) and (3.12) for the LDOS τ (q) spectrum in the random vector and mass potential models suggest that the unitary case also exhibits multifractality to first order in the impurity density n imp , since ∆ A,M,V ∝ n imp .
With multiple flavors of the disorder, solving the operator mixing problem for the q th LDOS moment re-quires the diagonalization of an effective spin Hamiltonian H (eff) , transcribed in Eq. (B10) of Appendix B. In Figs. 3 and 4, we present the results obtained by numerically diagonalizing this matrix for various combinations of {∆ V , ∆ M , ∆ A }. In these figures we plot the Renyi dimension 17 d q , defined for q = 1 via Figs. 3 and 4 show that the generic broken-T case is multifractal at one loop, and easily distinguished from the two-loop T -invariant result, in the limit of weak disorder. [Note that Fig. 4 indicates that the τ (q) spectrum is not purely quadratic in this general case.] It should therefore be possible to precisely distinguish the broken-T and T -invariant spectra experimentally, by observing the dependence of the deviation 2 − d q on n imp . The single-disorder flavor results for comparable strengths are plotted in Fig. 2 for reference.
For the multidisorder unitary model, the same RG eigenoperators dominate the scaling of composite ν and out-of-plane spin ν3 LDOS moments. The dimension x3 q that determines the spin LDOS scaling via Eq. (2.9) also enters into the LDOS spectrum in Eq. (3.7), leading to Figs. 3 and 4. By contrast, moments of the chiral spin LDOS components ν ± [Eq. (3.10)] remain eigenoperators that acquire no corrections at one loop, As reviewed in the subsequent Sec. IV A 2, for M = 0, the generic broken-T model is believed to flow to the critical state at the integer quantum Hall plateau transition. 24,29 This state exhibits strong multifractality that has been extensively studied in numerics. 18,19,[26][27][28] Fig. 3, but for different unitary class disorder strength combinations. In the data presented here, ∆V > ∆M,A. Regardless, the one-loop spectrum obtained for either ∆M,A > 0 is stronger than the two loop T -invariant case with ∆M = ∆A = 0. The latter is also shown for reference.
C. Renormalization effects
As discussed at the beginning of the previous section, the disorder-averaged Dirac surface state theory used to compute LDOS multifractal spectra is an "interacting" field theory, wherein the disorder strengths ∆ V,A,M appear as coupling constants (c.f. Appendix B). Because these parameters are dimensionless, at weak coupling the disorder constitutes a marginal perturbation of the clean Dirac band structure. The one-loop RG equations for these parameters are given by 24,54 where l = log L denotes the log of the RG length scale (e.g., the system size). Energy ε scales as where the (scale-dependent) dynamic critical exponent is In this section, we use Eqs.
A is the "microscopic" value derived from the randomly polarized in-plane magnetic impurity distribution. 53 This result in fact holds to all orders in ∆ A ; 24 in this case, the theory describing LDOS fluctuations at the Dirac point is conformally invariant. Multifractality is neither enhanced nor suppressed as one moves away from the Dirac point, defined as ε = 0. However, for non-zero energies ε = 0, in an infinite size sample all states are in fact localized. 24 The localization length diverges upon approaching the band center as ξ loc (ε) ∼ ε −1/z , with z = 1 + ∆ A /π [Eq. (3.20)]. Eqs. (2.3) and (3.8) for τ (q) hold on scales smaller than ξ loc (ε).
Broken T : random mass disorder (Class D)
For the random mass model with ∆ so that the disorder is marginally irrelevant at weak coupling. 24 Integrating this equation and using Eqs.
Non-magnetic disorder (Class AII)
Now we consider the T -invariant model. The flow equation for ∆ V is In contrast to the random mass, the random scalar potential is a marginally relevant perturbation to the clean band structure. 24 Examining lower and lower energy scales approaching the Dirac point, one observes stronger effects of the disorder. In the asymptotic scaling limit wherein the impurity potential strength becomes "large" (∆ V 1), numerical 9 results and analytical 10 arguments imply that the disordered T -invariant Dirac theory renormalizes into the "conventional" symplectic metal. The metal is distinguished from the Dirac theory by its nonzero (and non-critical) density of states at zero energy, 8 and by its τ (q) spectrum. 20, 23 We discuss the strong coupling LDOS multifractality below in Sec. IV A 1. If at energy Υ, ∆ V (Υ) ≡ ∆ (•) V ≪ 1, then for a somewhat smaller energy ε we obtain the logarithmic enhancement Eq. (3.24) implies that renormalization strengthens multifractality approaching the Dirac point ε = 0, for the T -invariant case. We emphasize that this has nothing to do with weak (anti-)localization. The latter occurs in the diffusive metallic regime with k F l mfp ≫ 1, where l mfp denotes the elastic mean free path. The diffusive regime obtains at strong coupling 9 near the Dirac point k F → 0 [Sec. IV A 1, below]. The impurity strength renormalization in Eq. (3.24) is a quantum effect deriving from the clean band structure, in the "near-ballistic" regime. 25
Broken T : generic disorder (Class A)
In the generic case of broken T , with multiple disorder coupling strengths non-zero, the system flows toward strong coupling ∆ V,M,A → ∞. As a result, multifractality is enhanced approaching the Dirac point. The RG flow ultimately terminates at a strong coupling critical point, or in the Anderson insulator, discussed in the next section.
IV. STRONG COUPLING REGIMES
In this section, we review prior results on strong coupling regimes relevant to the disordered Dirac Z 2 topological insulator surface states, LDOS fluctuations and associated multifractal spectra. These are not new, but provide complimentary information to the new results derived in the previous section.
In both generic cases of T -invariant, and T -breaking impurities, the disordered Dirac description used in Sec. III fails on the largest length and lowest energy scales (approaching charge neutrality). For a sufficiently dilute concentration impurities, the results obtained in the previous section characterize the start of the scaling regime, over energy and length scales such that the disorder strengths remain weak ∆ V,A,M (L, ε) ≪ 1. When these parameters become order one (due to renormalization down to lower energies and longer lengths), the system crosses over to one of the strong coupling regimes discussed below.
A. Delocalized states at strong disorder
T -invariant case: diffusive metal via strong disorder
In a random scalar potential field, the Dirac point vacillates in energy with spatial location; as a result, the density of states near charge neutrality is enhanced by the disorder. Due to the suppression of pure backscattering for Dirac fermions, 32 the state density enhancement more than compensates for the increased scattering introduced by the additional impurities. As a result, scalar potential disorder actually increases the (zero temperature, Landauer) conductance at charge neutrality beyond the clean ballistic result, e 2 /πh (Refs. 9,25). As in Sec. III, here we assume short-range correlated disorder, due either to charge neutral impurities or efficient screening by bulk and/or surface carriers. We do not discuss the puddle regime 34,35 in the present paper.
The effective disorder strength ∆ V is enhanced by renormalization, as indicated by the runaway flow implied by Eq. (3.23). The concomitant density of states and conductance growth suggests that the disordered Dirac theory ultimately crosses over to the ordinary diffusive symplectic metal, a result born out by numerics. 9 In the absence of time-reversal symmetry breaking, Anderson localization is prohibited on the surface of a topological insulator. 7,10 The symplectic metal possesses a finite (non-critical) average density of states at charge neutrality, and a distinct τ (q) spectrum. For a large effective diffusion constant D (induced for a Dirac fermion subject to sufficiently strong disorder, 9 or for a weakly disordered system examined on large length scales), the lowest order result for the multifractal spectrum appears in Eqs. (2.3) and Eq. (2.4), above. In the latter equation, β = 4 for the symplectic class. 23,40 For the T -invariant case, the strongest multifractality is expected at intermediate coupling.
Broken T : IQHP transition
For generic T -breaking disorder, i.e. all three ∆ V,M,A non-zero, the disordered Dirac theory is also unstable under renormalization. When the average mass is zero M = 0 (see below), the flow in Eq. (3.18) is believed to terminate at the critical point of the integer quantum Hall plateau transition. 24,29 This is the delocalized state separating adjacent Hall plateaux; it exhibits strong multifractality that has been extensively studied in numerics. 18,[26][27][28] The spectrum is believed to be universal, 18 and is approximately 28 parabolic as in Eqs. (2.3) and (2.5), with θ ∼ 0.26 (Refs. 27,28).
B. Anderson insulator
At zero chemical potential relative to the Dirac point, an average out-of-plane spin magnetization at the surface of a Z 2 TI corresponds to the presence of a nonzero Dirac mass M for the surface carriers. This insulating state resides in a quantum Hall plateau [with σ xy = sgn(M ) e 2 /2h]. 2,3,6,24 In the presence of surface disorder, the plateau state will assume the character of a localized Anderson insulator. In this section we review LDOS fluctuations in the Anderson insulator. The discussion is relevant not only to the magnetized surface of a 3D TI, but also to localized states populating the bulk gap of a disordered TI. Proposals exploiting localization to realize so-called "topological Anderson insulators" by adding impurities to clean hosts include those in Ref. 55.
To understand local density of states fluctuations in an Anderson insulator, it is useful to first study a toy problem. Consider a tight-binding model on a d−dimensional lattice, subject to nearest-neighbor hopping t, and a random on-site potential V i , distributed uniformly over the region −W/2 ≤ V i ≤ W/2. We assume the absence of spatial correlations in the disorder potential. The inverse relative strength of the disorder is measured by the ratio t/W . We consider first the extreme limit of zero hopping, t/W → 0. In that case, the LDOS is the on-site operator where η denotes an energy-smearing parameter. Physically, smearing is determined by inelastic scattering, open sample boundary conditions, or due to the finite energy resolution of the probing instrument. At the "band center" ε = 0, the distribution function for disorder-averaged LDOS moments evaluates to In this equation, the LDOS is constrained to the interval ν min ≤ ν ≤ ν max , where Using Eq. (4.1), one can compute the disorder-averaged moments of the LDOS, The average LDOS is ν = 1/W ; all higher moments are proportional to η 1−q , and thus diverge in the limit of zero energy smearing η → 0. This not surprising, because the energy spectrum in our trivial toy model is discrete, so that the LDOS operator becomes a delta function with ill-defined moments as η → 0. The moments are dominated by the power-law (Pareto) tail of the distribution, accumulating at the upper limit ν = ν max . By contrast, the typical LDOS, defined as ν typ = exp(log ν) is dominated by the infrared This vanishes in the limit η → 0. We see that observables exhibit broad statistics in the single site model, governed by the p(ν) ∼ ν −3/2 powerlaw distribution in Eq. (4.1). The moments are rendered finite only by the non-zero energy smearing parameter η. This should be compared to the LDOS statistics in a system with extended states and weak multifractality, e.g. that characterized by the quadratic τ (q) spectrum in Eq. (2.3), with 0 < θ ≪ 1. It is known 22 that the corresponding LDOS distribution has a Gaussian bulk, with a small amplitude log-normal tail responsible for the weak multifractality. For the metallic system, the result is independent of energy smearing, provided that the thermodynamic limit is taken before the smearing is set to zero. Returning to the toy insulator model, we observe that the global density ν G of states (GDOS) is self-averaging in the same limit. The GDOS is defined via where N denotes the number of sites. In the large Nlimit, the cumulant expansion can be evaluated via the saddle-point. The cumulants of the GDOS then take the form where [· · · ] q c denotes the q th cumulant, and the omitted terms are smaller by positive powers of η. Taking the infinite system size limit N → ∞ before sending the energy smearing to zero η → 0 leads to the vanishing of all q > 1 GDOS cumulants.
The calculations above can be extended to non-zero hopping via a locator expansion in small t/W , as performed by Anderson in his original 1958 paper. 56 This expansion can be formally summed to all orders in 1D and on the Bethe lattice, 57 but an explicit solution for the LDOS statistics is difficult to obtain this way; see Ref. 50 for an alternative approach.
Altshuler and Prigodin 44 succeeded in deriving the distribution generating disorder-averaged LDOS moments in a 1D system, which is exponentially localized for arbitrarily weak disorder. 8 In the thermodynamic limit for a closed sample, they obtain the "inverse Gaussian" distribution whereν ≡ ν/ν, and ǫ is the typical energy level spacing in a localization volume; ǫ −1 is also the elastic scattering lifetime. 44 In the limit of small smearing η ≪ ǫ, this distribution has moments The exact result in Eq. The take away is that the LDOS distribution in an Anderson insulator becomes very broad, with a powerlaw tail yielding divergent moments, in the limit of vanishingly small energy smearing. In an STM experiment performed at ultra-low temperature, on a large, isolated Anderson localized surface, the collected LDOS statistics should be very sensitive to the smearing induced by the energy resolution of the measurement itself.
The locally discrete energy spectrum of the LDOS in the Anderson insulator invalidates the use of Eq. (2.2) as a tool to compute the multifractal τ (q) spectrum. As advocated above, the shape of the LDOS distribution function and its sensitivity to smearing can best reveal the insulating phase. If one insists upon computing mo-ments, one must employ 20 (4.6) Since the levels are discrete, Thus, In this equation, we replace averages-of-the-logs with logs-of-the-average, a procedure that is legitimate here because spatial and disorder-averaging are expected to yield the same results on the insulating side. Noting that the LDOS moments are L-independent in the insulator for L ≫ ξ loc , we obtain the expected result 18 for localized states computed in a well-defined η → 0 limit.
V. ACKNOWLEDGMENTS
The 10 symmetry classes of disordered Hamiltonians (Hermitian random matrices) can be efficiently distinguished by the presence or absence of time-reversal T , particle-hole P, and chiral/"sublattice" symmetry C. 7,48,49 For the two-component Dirac Hamiltonian in Eq. (3.1), the definitions of these symmetries are essentially unique. In terms of the two-component Dirac spinor ψ, these appear as In the second quantized language, T and C are antiunitary transformations; the unitary P can be taken as the product of these.
The imposition of any one of the discrete symmetries upon the Hamiltonian in Eq. (3.1) in every disorder realization restricts its form, and selects a particular random matrix symmetry class. 7,24,48,49 (1) T -invariance: A = M = 0, only potential disorder ∆ V ≥ 0 is allowed. Since T 2 = −1, this is the symplectic (spinorbit) class AII, which is also the symmetry class of the (presumed T -invariant) topological Z 2 bulk. (2) Pinvariance: V = A = 0, only random mass disorder ∆ M ≥ 0 is allowed. Because P 2 = +1, this is the broken time-reversal class D.
(3) C-invariance: V = M = 0, only random vector potential disorder ∆ A ≥ 0 is allowed. This is the broken time-reversal class AIII. (Technically, it is the "topological"/WZW class AIII 1 in the language of Refs. 7,49.) Class AII is generically realized whenever time-reversal is unbroken. Magnetic impurities randomly polarized parallel (perpendicular) to the TI surface manifest as point exchange sources in the vector (mass) potentials of Eq. (3.1); we are thus tempted to identify symmetry classes D and AIII with these two limits. However, a magnetic impurity will typically induce a local potential fluctuation V (r) as well. As a consequence, the generic case of broken-time reversal symmetry corresponds to the absence of T , P, and C, which gives the unitary class A. 7,48,49 In fact, for a vanishing average mass M = 0, the surface of a topological insulator with generic time-reversal breaking disorder is expected to flow under renormalization to the critical point of the integer quantum Hall plateau transition. 24,29 On a different note, the class AIII 1 and class D versions of H in Eq. (3.1) can be realized on the surface of a bulk T -invariant 3D topological superconductor, where time-reversal is respectively preserved or broken at the surface. 7 Appendix B: Perturbation theory
Chiral Decomposition and one-loop renormalization
We write a 2+0-D fermion path integral to represent correlation functions in the disordered Dirac Hamiltonian transcribed in Eq. (3.1). The fermion operators are replaced with the Grassmann fields {ψ, ψ † } → {ψ i ,ψ i }; here i ∈ {1, . . . , n} denotes a replica index, and we are to send n → 0 at the end of the calculation. 8, 20 We employ a "chiral decomposition" of the two-component spinors, Then the action of the replicated theory is where we have introduced complex coordinates {z,z} = x ± iy, {∂,∂} = (∂ x ∓ i∂ y ), and disorder potentials The energy ε is a fixed parameter. In Eq. (B2), repeated replica indices are summed. Assuming the Gaussian white-noise variances for the disorder potentials enumerated in Eq. (3.3), the replicated theory can be averaged over disorder configurations. The post-ensemble averaged action is where S 0 is the clean Dirac action, and Different replicas become coupled through the disorder. 8,20 The disorder-averaged composite LDOS ν(ε, r) corresponds to the fermion bilinear expectation The spin LDOS νι(ε, r) was defined by Eq. (2.7). For the out-of-plane and in-plane (chiral) components, one has The overlines appearing in the left-hand sides of Eqs. (B5) and (B6) denote disorder-averaging, whereas the angle brackets on the right-hand sides represent integration in the fermion path integral, using the action S in Eq. (B3). A generic local operator corresponding to the q th moment of some fermion bilinear can be viewed as sum of "strings," where each string consists of 2q total right (R) and left (L) mover labels, arranged in some order. For example, the disorder-averaged q th moment of the LDOS is represented by the the composite operator expectation value In this equation, a product is taken over operators carrying indices in the first q ≤ n replicas. The q th LDOS moment is computed by placing one copy of the LDOS operator into each of q different replicas; before averaging, this gives ν q in a fixed realization of the disorder. (Placing instead the q copies into the same replica would give the disorder-averaged first moment of a 2q-point Green's function.) The operator in Eq. (B7) is an even weight sum of 2 q "strings", all of length 2q. I.e., The semicolons separate fermion bilinears in different replicas. Each bilinear has two entries, corresponding to the chiral identity of the barred and unbarred operators. The set of all length 2q strings forms a complete basis for q th moment local operators (without derivatives). These basis strings mix under renormalization due to the disorder. 58 In general, the composite operator (≡ weighted string sum) corresponding to the q th moment of a bilinear as in Eq. (B7) does not constitute an eigenoperator of the renormalization group. The main task is to (1) identify RG eigenoperators for each disorder type and compute the spectrum of scaling dimensions, and (2) compute the projection of the LDOS and (for broken T ) spin LDOS moment operators onto this eigenbasis, and determine the most relevant contributions.
a. Effective Hamiltonian for 1-loop renormalization
It is useful to view each string as a configuration of 2q spin 1/2 moments. We associate {R, R} → 1/2 (spin up) and {L, L} → −1/2 (spin down). Operators invariant under spatial rotations have equal numbers of up and down spins, and therefore reside in the zero total magnetization sector with J z = 0. We picture each string as a basis state for a length q chain, with two spins per site. Sites are labeled by the replica index i ∈ {1, . . . , q}. The two spins at each site are distinguished by labels "A" and "B," corresponding to barred and unbarred operators, respectively.
Renormalization occurs via the action of the disorder vertices appearing in Eq. (B4), employing the clean Dirac propagator in a standard loop expansion. Operator mixing at one-loop is encoded in the effective "Hamiltonian" In this equation, S a A/Bi denotes a spin-1/2 operator acting on the barred (A) or unbarred (B) spin in replica i. The prefactor obtains from evaluating the loop integrals using a hard momentum cutoff Λ. The first, second, and third lines in the heavy brackets arise through the action of the disorder vertices in S A , S 1 , and S 2 , respectively. The ∆ A renormalization is diagonal in the ↑/↓ (R/L) basis. By contrast, S 1 , and S 2 perform single exchanges of right and left labels. S 1 (S 2 ) mediates interflavor A ↔ B (intraflavor A ↔ A, B ↔ B) exchanges. Summing the angular momenta, where J A,B ≡ i S A,Bi .
In the general case of broken T discussed in Sec. III B 4, all three disorder parameters are present. The most relevant eigenvalue of Eq. (B10) determines the scaling of the q th LDOS moment. 59 The first few multifractal moments for various disorder configurations were obtained through numerical diagonalization; results appear in Figs. 3 and 4.
Even moments of the out-of-plane spin LDOS ν3 [Eq. (B6a)] are invariant under spatial rotations and parity. 59 In the multidisorder unitary case, even moments of the composite ν and out-of-plane spin ν3 LDOS are dominated by the same RG eigenoperator. The dimension x3 q that determines the spin LDOS scaling via Eq. (2.9) is the same that enters into the LDOS spectrum in Eq. (3.7), Figs. 3 and 4. Moments of the chiral spin LDOS components in Eq. (B6b) correspond to the highest weight states |j = q; m = ±q ; here, j(j + 1) denotes the eigenvalue of (J A + J B ) 2 , with 0 ≤ j ≤ q and −j ≤ m ≤ j. These highest weight states are annihilated by H (eff) in Eq. (B10), leading to Eq. (3.17).
Below we discuss the special cases of isolated disorder flavors.
c. Broken T : random mass disorder (Class D) For the random mass case, Eq. (B10) becomes where J ≡ J A + J B . On the second line, we have evaluated the "Hamiltonian" on a total angular momentum eigenstate |jm . For 2q spins, we have max(j) = q. The maximum eigenvalue is associated to the non-degenerate j = q, m = 0 state, which is invariant under spatial rotations. The scaling dimension is The corresponding eigenoperator |j = q, m = 0 is an equal weight symmetric sum of all permutations of q "R" and q "L" labels, and has non-zero overlap with the naive LDOS moment (a q-fold triplet product) in Eq. (B7). Using Eq. (B14) in Eq. (3.7), one obtains the result for the quadratic τ (q) LDOS spectrum in Eqs. (2.3) and (3.12). By contrast, the out-of-plane spin LDOS (mass operator) in Eq. (B6a) is a singlet; the disorderaveraged q th moment thus corresponds to the eigenoperator |j = 0, m = 0 [leading to Eq. (3.13)].
d. Non-magnetic disorder (Class AII)
We rotate the "B" composite spin by π around thê z-axis, The Hamiltonian in Eq. (B10) with ∆ M = ∆ A = 0 becomes whereJ ≡ J A +J B . Eq. (B15) has the same form as Eq. (B13), with ∆ V → −∆ M . This is consistent with a mapping between the random mass and vector potential models identified in Ref. 24. In the case of the scalar potential, the maximum eigenvalue is associated to the highly degenerate singlet sector j = 0,m = 0 , leading to the scaling dimension Using this result in Eq. (3.7) gives τ (q) = 2(q − 1). We conclude that no moment operator (without derivatives) acquires multifractal scaling at one loop for the T -invariant model.
Two loop renormalization, T -invariant class AII
In the T -invariant class AII model, the first correction to the LDOS τ (q) spectrum appears at second order in ∆ V . We have carried out a two-loop calculation and found that the naive LDOS moment in Eq. (B7) remains an eigenoperator. To this order, we find the scaling dimension
|
2012-02-22T16:09:01.000Z
|
2011-11-28T00:00:00.000
|
{
"year": 2011,
"sha1": "cf06695f9f49f17ff294a79f9b5ebb504dcbd162",
"oa_license": "publisher-specific, author manuscript",
"oa_url": "https://link.aps.org/accepted/10.1103/PhysRevB.85.085122",
"oa_status": "HYBRID",
"pdf_src": "Arxiv",
"pdf_hash": "19485b256bccb79ae3b9618011291b5ab45866a6",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
271228518
|
pes2o/s2orc
|
v3-fos-license
|
Prevalence of parenthood among hospitalized adult patients with severe mental illness: a quantitative data analysis
Introduction In the Western world, more than one-third of the patients of productive age hospitalized for severe mental illness (SMI) are parents. Each of their offspring is exposed to several stressors related to their parent’s illness and hospitalization, which puts them at an increased risk of developing mental health problems. In the Czech Republic, no statistics are currently available about the families of patients with SMI, inpatients who are parents, or data about their children (ages ≤18 years). Therefore, our research aim was to describe the prevalence of parenthood among hospitalized patients with SMI, assess the number of children and determine the extent to which offspring information was present in medical records. Methods Quantitative data from medical records (2,768 patients, aged 18–63 years, hospitalized for SMI between 2017 and 2020) from two large inpatient psychiatric facilities were examined. Parental information, demographic characteristics, number of children, and other available data were collected. Results The prevalence of parenthood among inpatients with SMI was 34.6%. Parenthood was most prevalent among female patients and patients with recurrent depressive and bipolar disorders. The total number of offspring in 957 patient-parents was 1781 (41.7% minors under the age of 18). Information on parenthood was available in 99.7% of cases; information on the age of offspring, custody, and sociodemographic situation varies, being included in 73% to 89.7% of the medical records (some details were more frequently recorded than others). Discussion The data obtained may help to better understand and address the specifics of these families and thus serve as a basis for the development of prevention programs.
Introduction
Parents comprise a significant proportion of patients with severe mental illness (SMI) (1).In this paper, consistently with other studies (2), the term SMI includes these diagnoses: schizophrenia spectrum disorders, recurrent depressive disorder and bipolar affective disorder.Children of parents with mental illness are at high risk for also developing a mental disorder, yet they receive little attention (2).Meta-analyses show that children of parents with SMI have a high risk of transgenerational transmission of psychopathology, with up to a 50% risk of developing mental illness and a 32% risk of developing severe mental illness (3).These risks are due to not only the hereditary burden (a specific risk factor) but also several non-specific risk factors (poor socioeconomic family situations, relatively high parental unemployment rate, parental stress, stigmatization of families with mentally ill members, more frequent placement in foster care, etc.) (4).
Regarding the mechanisms of risk in children, it appears that risk is determined by the severity and chronicity of psychopathology and differences in parental personality, genetic characteristics, coping style, and social circumstances, rather than the parent's diagnosis itself (5,6).During parental hospitalization, children may have difficulty in coping with their circumstances, separation from their parent, the parent's condition, and their painful and negative emotions, including grief (7).
In terms of sociodemographic characteristics, compared to other children, those in families in which a parent has a diagnosed mental illness are two times more likely than others to live in a lower income family household (e.g., 8, 9), be unemployed (8,10) and have lower education levels (8,11).Children of parents with mental health disorders are also more likely than those in the general population to live with only one parent (10,11).For example, only 10% (5 of 50) of the mothers in the study by were married.In order to provide appropriate supportive interventions, we need to look at the socio-demographic circumstances as well.
Prevention programs for the children of parents with mental illness have been implemented in some other countries (Australia, Norway, Finland, Denmark, The Netherlands, USA, etc.), and the related research shows that these programs reduce children's risk of developing mental illnesses (13,14), sometimes by up to 40% (3).Other studies have demonstrated that preventative intervention programs protect both parents with SMI and their offspring (15).However, appropriately targeting support for families with a parent with SMI requires knowing the prevalence of parenting among patients and having access to detailed information about children living in families with a parent with mental illness and the factors that influence the mental health of all family members.In the Czech Republic, no demographic statistics are currently available about the families of patients with SMI and inpatients who are parents.
We decided to use a quantitative study of medical records to investigate parental rate among hospitalized patients with SMI as used in similar studies of clinical populations (e.g., 12).However, there are several ways to examine the prevalence of parenting in the literature.For an overview of different methodological approaches and reported prevalence rates, see our recent review (16).Studies of clinical populations can take the form of census studies in which either professionals collect data by completing questionnaires about patients (17,18) or the patients complete questionnaires about themselves and their children (19,20).In studies similar to the present one, the reported prevalence of parenthood ranged from 25-36% among inpatients with mental illness (12,17,20) and up to approximately 50% among outpatients (21,22).
Therefore, the aims of this study were a) to determine the prevalence rate of parenthood among SMI patients and to obtain detailed information about their offspring and b) to determine the extent to which family characteristics are included in the medical records of patients with SMI, and c) identify relevant family demographics and characteristics data that should be part of psychiatric patients medical files to design more effective and individualized interventions.Therefore, we formulated the following research questions:
•
What was the prevalence of parenthood in patients hospitalized for SMI?
• What were the sociodemographic characteristics of the families of these patients?• Was information on the age and mental health of the children of parents with SMI included in the medical records?• How many parents with SMI had children in their custodial care?
Design
This cross-sectional study was conducted using data related to inpatients in two collaborating general psychiatric hospitals with 500-and 1,000-bed capacities in districts and towns in the Czech Republic.Data were obtained by analyzing the medical records of patients with SMI hospitalized between 2017 and 2020.
Procedure
Two collaborating psychiatric inpatient facilities were contacted for data collection.Both facilities are urban psychiatric hospitals with acute care and rehabilitation programs and a capacity of approximately 500 to 1000 patients.Data were collected in March 2022 by one authorized employee at each facility (two individuals in total).From an anonymized patient list, the authorized staff member identified all patients who, at the time of hospitalization, were between the ages of 18 and 65 years and had been admitted with SMI diagnoses at any time between January 1, 2017, and December 31, 2020.In cases of patients who had repeated hospital admissions within those years, we only included the data that corresponded to one of those admissions.In this case, the number of admissions was recorded and data from the last hospitalization were obtained.From this pool of patients, those with indications of parentage in their medical records were selected.For these parent patients, additional data were obtained from medical records, specifically most often from the part of social and family histories documented by psychiatrists at admission.
Ethics
The data were collected after obtaining approval from the ethics committees of Masaryk University and the cooperating medical facilities.Data collection and anonymization were performed by an authorized employee at a medical facility.Since only anonymous statistical data were provided in the study, informed consent from individual patients was not required according to Czech legislation.The researchers worked only on statistical data and could not identify individual patients.
Data collection
Data were obtained from the electronic medical records system, specifically from medical examinations and psychosocial histories documented by psychiatrists at admission.The data consisted of the sex, age, main diagnosis, and parental status of all patients with relevant diagnoses who were hospitalized over a 3-year-period.For patients who were parents, additional data were subsequently collected from the medical records: the number and length of the last hospitalization, suicide attempts, marital status, family situation, education, disability pension, number of children and their ages, children's health (psychological/psychiatric problems only), their relationship to people with whom they lived, and who took care of the minor children during hospitalization.
Data analyses
Quantitative data were analyzed using SPSS 27 software with descriptive statistics, and the chi-square test was used to detect differences in the prevalence of parenthood between the groups by sex and diagnosis.Information on the offspring (health status, relationship with parents, and information about caregivers) varied in the medical records.Content analysis of that information was conducted by two independent researchers (A.H. and D.H.).The occurrence of certain words was quantified into categorical variables according to established coding rules (e.g., the item 'caregivers' was coded into categories of mother/father/ grandparents/institutional care, etc.).Inter-rater agreement was tested on a selected sample.As a result of good parameters (Cohen Kappa varied from 0.91 to 0.96) 1 and the nature of the data, which showed clear and unambiguous categories, inter-rater reliability was not determined for the full sample (due to sample size and staffing).The resulting univariate categorical variables were used in chi-square analyses to determine the differences between the parents' main diagnosis and the presence of a child in the household.
Prevalence of parenthood
Of the 2,768 patients, 957 were parents (34.6%); 63.1% of whom were women and 36.9% were men.Information on parenthood was missing in 9 cases (0.3%).In the group of patient-parents, 46.2% of them had a minor child under the age of 18 at the time of hospitalization.The parents of minor children (n = 446) had 666 minor children.
The prevalence of parenthood was higher among hospitalized women (50.8%) than among men (22.5%) (x2 = 239.44;df = 1; p< 0.001).Significant differences between patients with and without children were also observed across diagnoses.Patients with depressive and bipolar disorders were more likely to be parents than patients diagnosed with schizophrenia.Further details are presented in Table 1.
Sociodemographic characteristics of patient parents
Nearly half of the hospitalized parents had at least a secondary education level of high school completion (they earned a diploma) or higher.Most parents were either employed (44.0%) or received disability pension benefits due to psychiatric indications (43.3%), and 6.6% of parents were unemployed.Almost half (47.9%) of the hospitalized parents lived in a partnership or marriage, a quarter (25.4%) lived alone, while the remainder lived with other relatives (most often their parents, siblings, or adult children) or, less often, in shelters, hostels, and nursing homes (6.7%) or homeless (2.6%).Legal capacity (ability to take independent legal action) had been restricted for 8.1% of patients who were parents.Information on education was present in the medical records of 81% of the patientparent cases, on employment in 73%, on legal capacity in 82%, and information on housing in 81% of cases.See Table 2 for full details of the percentages of available information.
Age of inpatients' offspring
The hospitalized patients had a combined total of 1,781 offspring.The mean number of children in the family was 1.9 (SD = 0.86) and ranged from 1 to 6.The age (or, at least, information about adulthood) of the offspring was reported in 89.7% of cases.Almost 60% of the offspring were adults and onethird were school-aged.Table 3 provides the detailed information.
Mental health of inpatients' offspring
Explicit information on the mental health of the inpatients' offspring ("healthy" or a specific illness) was given in only 16.4% of cases.The frequency of individual mental disorders is presented in Table 4; somatic diagnoses were omitted due to the scope of the article.Seven offspring were deceased (three had died by suicide).
Care for children of hospitalized parents
Of the 1,596 offspring for whom we knew their age, there were 666 minors under the age of 18.The mean age was 9.1 years (SD = 4.84), ranging from 0 to 17 years.Half of these children were under the care of both parents.Forty-two children (6.2%) lived with their hospitalized parents alone (without the presence of the other parent, spouse, or other relatives).All single, hospitalized parents were mothers.We know from medical records that in the case of two mothers, their child was placed in transitional foster care for the duration of their hospitalization.However, no information was available about the other 40 children.Children who were not living with any parent (i.e., living with another relative, in foster care, or with adoptive parents) accounted for 9.2% of our sample.For 13.7% of the children, we had no data.Table 5 provides additional information on this topic.
Living in a shared household: differences between sex and parental diagnosis
We divided the sample of children into those living with a hospitalized parent, which included children living with both parents, with a hospitalized parent only, and in shared parenting (63.6% of minor children), and those not living with a hospitalized parent (36.4%).If a child lived with a hospitalized parent, the child was significantly more likely to have a hospitalized mother (72.8% of cases) than a hospitalized father (27.2% of cases).If the mother was hospitalized, she was 71.3% likely to live with her child, whereas the likelihood for the father was only 49.1%.These differences are statistically significant (x2 = 32.14;df = 1; p < 0.001; Phi = -0.220).
Children of parents hospitalized with schizophrenia were cohabiting with their parents in 47.7% of cases, children of parents with bipolar disorder in 53.8% of cases, and children of parents with other psychotic disorders (72.0%) and recurrent depressive disorder (74.5%) were even more likely to live together with the hospitalized parent.The differences in the parent's main diagnosis and the presence of the child in the household were statistically significant (x2 = 38.99;df = 3; p < 0.001; Cramer V = 0.239).
Number and duration of parental hospitalizations
The parents of children were hospitalized at an average of 1.7 times (SD = 1.58, mode = 1) during the 3-year follow-up period, and the number of hospitalizations varied from 1 to 16.The mean length of the most recent hospitalization was 61.8 days (SD = 66.79), ranging from 1 to 901 days.
Suicidality of hospitalized parents of minor children
Among the hospitalized parents with children, the data indicates that there was an 8% prevalence of suicide-attempt history, with 2.4% of the patients having a history of repeated suicide attempts.
Relationship between hospitalized parent and child
Information about the relationship between the hospitalized parent and child was purposely searched in medical records.Some references to a relationship were provided in 39.5% of the 666 cases.These statements were grouped into 4 categories using content analysis: "good relationship" (195 times), "occasional contact" (22 times), "not in contact" (43 times) and "conflict relationship" (3 times).
Discussion
The prevalence of parenthood among hospitalized patients with SMI in our study was 34.6%.This result is similar to the 38.5% prevalence rate reported in a previous US study (12).That study's patient participant sample also included minor and adult offspring; however, all of the patients were women, and women generally tend to have a higher parenthood prevalence.For example, in Östman and Hansson's (20) study, 75% of parents with SMI were women, and an Australian national study (8) found 56.2% of the patients with children were women, and only 25.9% of the men were parents.The prevalence of parenthood in our study was also In comparison to other methodologically similar studies (19, 20), the prevalence identified in our sample was slightly higher (28% vs. 34.6%).This difference may be explained by the moderately different distribution of hospitalized patients; a significant proportion of parents in our study were hospitalized for depressive disorder, whereas in other studies, a larger proportion of the sample consisted of patients with schizophrenia, who were parents less often (again confirmed by our results).However, it cannot be said that there are no parents among patients with psychotic disorders or that they do not live in the same household as their children.Half of the minor children whose parents were hospitalized for schizophrenia lived in the same household as the parent.The percentage of parents with other psychotic disorders co-habiting with their children was 72.0%.This is an even higher percentage than that reported by Howe et al. (17) (34.0% to 56.4% of minors living with a parent with a psychotic disorder from 2008 to 2011).
In addition to increased psychological demands, the vulnerability of the offspring of parents who are inpatients is magnified by a worsening of the family's socioeconomic status.SMI parents are more likely to be unemployed, and SMI is a risk factor for living below the poverty line (9).The patients in our sample had lower education levels (52.5% of parents had lower education than secondary with high school diplomas versus 44.1% in the general population (23) and their unemployment rate was twice that of the general population (6.6% vs. 2-3%, see 24).In addition, 47.5% of the parents received a disability pension (43.3% because of psychiatric indications) meaning that their ability to work was limited or completely inhibited in almost every second case.This finding illustrates the significant effect of SMI on family life.Support aimed at parental employability could potentially lead to a more stable family situation and other family benefits.
In our study sample, the majority of the offspring were over 18 years old, which may be explained by the wide age range of the patient sample, because we aimed to cover the widest possible spectrum of potential parents of minor children.The analyses also revealed that medical records of 99.7% of SMI patients had at least basic information about parental status in their records, and in a significant proportion of cases, additional information about their offspring (e.g., ages or information about the adulthood of offspring in 89.7%).These data are in contrast with the UK study conducted by Gatsou et al. (25) in which parental status was presented in 62% of the cases only.Our data indicate a well-established and routinized medical assessment of parenthood during the admission process to collaborating facilities.
From the perspective of mental health problem prevention, data on the mental health of the offspring were only sporadic in the medical records (only 16.4% of the offspring).In Manderson and McCune's (26) study, an inquiry into children's welfare was carried out in 24.2% of cases.These data may indicate an insufficient focus on investigating the mental health of offspring, which is an important first preventative step in identifying families in need of additional support.However, it is important to consider that patients are often admitted in the acute phase of the illness when it is not possible to collect an adequate amount of objective information.
Minors in our study shared a household with the hospitalized parent in 63.6% of cases, which is a smaller percentage compared to a recent Norwegian study (27), in which 76.2% of minor children cohabitated with a parent with mental illness.This finding can be explained by two factors: Norwegian researchers more frequently included parents with affective disorders than we did in this study, and the topic has received more focus for a longer period in Norway.There are established intervention programs and dedicated legislation-The Norwegian Health Personnel Act of 2010-postulating, among other things, a legal obligation for health professionals to investigate parental status and offer appropriate support to offspring when needed.
Children were more likely to live with hospitalized mothers than with fathers.This finding may be related to mothers perceiving parenthood as a part of their identity (12,28,29).This is another reason why we should address the topic of motherhood in the treatment of hospitalized women, because it can serve as a protective factor against risky behavior (e.g., suicidality and remaining in harmful relationships) (29,30).The impact of fatherhood on male patients' identity is a less researched topic, but it appears that men also perceive fatherhood as part of their identity, and they feel proud of and concerned about their offspring (31)(32)(33)(34).Hospitalization also affects other parents who take over most of the childcare responsibilities.Half of the children of the patients in our sample lived in a family with both parents (only 38.9% in the Norwegian study, in which more children were raised through shared parenting -see 27).Forty-two minors (6.3%) lived in a family with a hospitalized single parent (mother) without any other adults.Information on who cared for the children during the mother's hospitalization was available for only two children (a facility for children in need of immediate support).Within this framework, there is room for increased inquiry into the family situation of hospitalized parents without an extended family background, and collaboration between clinicians and other organizations is essential.Children of single parents can be negatively affected not only by the absence of a parent but also by the difficult situation they are in during the hospitalization of their parents.Moreover, mothers may postpone their hospitalization out of concern for their child's safety, as indicated by Diaz-Caneja and Johnson (35).
We also aimed to map the frequency of risk factors for the development of mental health problems in children whose parents are hospitalized.Risk factors for transgenerational transmission of psychopathology include younger child age (5).In our sample, 12.9% of children were in the "under-5 years old" category.Additionally, the parent's diagnosis per se may not play such a large role; the severity of the illness (suicidality and psychotic symptoms) and the frequency and chronicity of the illness are more impactful (36).The average parent of a minor child in our sample was hospitalized approximately twice over three years for approximately two months on each occasion, which is almost an entire school term when the child is without one parent and the other parent or relative must take over care.Suicide attempts occurred in 8% of cases in our sample in the history of parents of minor children, but this information may be misleading because it is not clear whether the attempt occurred before or after the birth of the child.However, this is an important factor, because the likelihood of repeated suicidal behavior increases with a history of suicide attempts (37), placing children at an increased risk of developing mental health problems.On the other hand, according to research, being a parent can be a protective factor against suicide (38); therefore, addressing the topic of being a parent and parenting in clinical work may have great protective benefits not only for hospitalized parents but also for their offspring.
Research has shown that a relatively large percentage of families with parents with SMI present a range of risk factors for the development of mental health problems in children (2,3,9).Increased attention and support should be given to these families, because preventing the development of mental disorders in the next generation is an important part of the agenda of mental health professionals.If we, as professionals, realize that the patient is part of a family system that is influenced by him/her and reciprocally influences him/her, we can dramatically improve both the patient's treatment and prevent the development of mental health problems in his/her offspring.
Study strengths and limitations
In our research, we attempted to address a question that has received only minimal attention in the Czech Republic thus far and to provide up-to-date data mapping of the current situation.Prevalence research is the first step, preceding further studies aimed at mapping the needs of the families of patients with SMI.Based on this information, it is possible to integrate parenthood and parenting into the clinical assessment, treatment, and psychosocial rehabilitation of patients and their families.Similar to other countries, this study can be used as a basis for the implementation of prevention programs.
However, this study had several limitations.The first is a certain degree of simplification in the definition of SMI to diagnostic categories only because the design of the data collection did not allow for a more detailed specification of SMI, for example, according to functional criteria (Global assessment of functioning score -GAF, presence of psychosis, and duration of illness).Since medical records do not use a system of functional criteria, diagnoses are central.Data collection was conducted in two similar types of inpatient psychiatric facilities; thus, we cannot comment on the prevalence of parents among patients with SMI throughout the Czech Republic.Future studies would be enriched by information from outpatient and rural forms of care.The current data were only cross-sectional; therefore, it is not possible to generalize the results over time or, for example, to look for causal links between parental characteristics and their impact on the child.The collected data did not allow us to compare the sociodemographic characteristics of parents and non-parents.
The data were also from admission assessments in medical records only and, thus, might be influenced, for example, by the admission of a patient in an acute state of illness or by the recording patterns of individual clinicians.It would be interesting to complement future research with subjective views and attitudes of patients with SMI.Future (qualitative) research could reach out directly to patients for data collection and to explore their experiences and support needs, so that prevention and support programs could be tailored better for patients.
Recommendations for practice
This study provides several recommendations for clinical practice.The first general recommendation is a greater emphasis on parenthood and parenting in the medical assessment and treatment of patients with SMI, because it may increase motivation and compliance during the recovery process.During family history mapping, clinicans should explore in detail the relationship between the patient and their child and the care of children of hospitalized patients and offer support services to children as needed.Similarly, the mapping of the health status of offspring can serve as a safety net and, with appropriate supportive interventions, prevent the development of mental health problems in the next generation.In addition, stronger collaboration and liaison between adult and child psychiatry and clinical psychology practices can facilitate and speed up cooperation and prevent the development of serious problems in children.
Studies and materials evaluating the impact and benefits of working with families and parents should be incorporated into the curriculum of professional mental health education.Inspired by Norway (The Norwegian Health Personnel Act of 2010), we recommend consideration of building legislative endorsement of supportive interventions to be offered to families and family members who are at risk for developing mental health problems.
Conclusion
More than one-third of patients hospitalized for SMI were parents.These patients' children (often minors) are exposed to several psychosocial stressors associated with their parents' illnesses and hospitalization, in addition to the hereditary burden.Parental status is well registered in medical records, unlike other more detailed information, such as the child care during a parent's hospitalization.This study aimed to highlight this topic and provide up-to-date data on the prevalence of parenthood in patients with SMI.This study may provide a background for further enhancing clinical assessments and implementing preventative interventions in families with risk factors.As more details about patients' family characteristics-and their individual roles in their families and communities-become available, rehabilitation and recovery will become both more inclusive and more personalized.
TABLE 1
Differences between patients-parents and patients without children by sex and main diagnosis.
TABLE 3
Age of offspring of hospitalized patients.
TABLE 5
Care of minor children (n = 666) of hospitalized parents.
|
2024-07-17T15:23:45.594Z
|
2024-07-15T00:00:00.000
|
{
"year": 2024,
"sha1": "9c74ef0f2e53e9537c2ac6c0f843f85b8c372558",
"oa_license": "CCBY",
"oa_url": null,
"oa_status": null,
"pdf_src": "PubMedCentral",
"pdf_hash": "4d4442b08507fd0b9797ccfefa20311bd303a213",
"s2fieldsofstudy": [
"Psychology",
"Medicine",
"Sociology"
],
"extfieldsofstudy": []
}
|
92146427
|
pes2o/s2orc
|
v3-fos-license
|
In silico Prediction and Cytotoxicity Evaluation of siRNAs Targeting Conserved Regions of MERS-CoV in Cell Culture
The Middle East Respiratory Syndrome Coronavirus (MERS-CoV) was isolated in 2012 and is well known to cause the respiratory syndrome. The orf1ab gene is known to mediate MERS-CoV replication. In this study, we have discussed the in silico prediction of potential siRNAs targeting MERS-CoV-orf1ab gene for antiviral therapeutics. To identify the potential siRNAs, various factors were considered. We have excluded the siRNAs with off-target effects and potential binding with human mRNAs. By using available softwares, total twenty-one functional, off-target reduced potential siRNA were selected from four hundred and sixty-two siRNAs based on greater potency and specificity. We have tested only seven siRNAs initially to evaluate their performance by reverse transfection approach by lipofectamine mediated delivery in Vero cells. The evaluation results showed no cytotoxicity at various concentrations of siRNAs used. The results obtained in this study provided preliminary information about the cytotoxicity which will help us to further evaluate siRNAs in other cell cultures to find out the replication inhibition efficiency of MERS-CoV. Finally, it is concluded that the in silico prediction and designing resulted in filtration and selection of potential siRNAs with high accuracy, efficiency, and strength which can be further utilized for the development of oligonucleotide-based therapeutics.
Introduction
MERS-CoV presents a major health concern for both human and animal.The first case was identified in June 2012 from Jeddah, Saudi Arabia [1].Currently, a total of 2,279 confirmed MERS-CoV cases with 806 related deaths have been reported and spread in 27 countries (WHO-last accessed on February 24, 2019).The virus is mainly transmitted to human in family clusters, healthcare workers, through contact with camels as well as community settings [2][3][4][5][6][7].MERS-CoV causes lower respiratory infections with fever and cough followed by shortness of breath and organ failure in cases with comorbidities [8].Coronaviruses genome varies from 25.0 to 32.0 kb and have high sequence variation which favors the possible recombination and emergence of new virus strains with novel characters in the extended hosts [9].
The RNA interference (RNAi) is an important biological process found in many eukaryotes in which RNA molecules inhibit gene expression.Two molecules (miRNA and siRNAs) are known as central to RNAi.The main function of siRNA is to downregulate and silence the specific gene by degrading the mRNA after transcription [10].The regulatory role of RNAi in gene expression has been shown in many in vitro and in vivo studies [11].Short interfering RNAs (siRNA) are short sequences of RNA that varies from 21 to 23 base pairs with 5' phosphate group and a 3' hydroxyl group.They function by binding to a specific protein known as RNA induced silencing complex (RISC), RISC then forms a new complex with specific RNA sequences and degrades the RNA resulting into silencing of mRNA expression [12].As all siRNAs are not equally potent, simple rules have been suggested for the computational selection of efficient siRNAs.The designing softwares combining the structural features of the targeted RNAs as well as the sequence features of the siRNAs are the good choice.
The in vitro efficacy of siRNA was reported for the inhibition of other coronaviruses.Cytopathic effect was specifically inhibited by siRNA targeting viral RNA polymerases of SARS-CoV in Vero cells [27].The titer and levels of viral proteins were reduced to indicate blocking of viral replication.S gene of SARS-CoV was selectively silenced by DNA vector-driven siRNA in SARS-infected 293T cells.Another study showed that SARS-CoV infection and replication in fetal rhesus kidney cells (FRhK-4) was inhibited by 3 siRNA duplexes targeting viral RNA polymerases, and one targeting the S gene [28].The siRNA duplexes had a prolonged prophylactic effect, with ≤ 90% inhibition of transcription, lasting for ≥ 72 h.Combinations of siRNA duplexes against multiple targets in the viral genome showed ≤ 80% inhibition.Therapeutic siRNAs and miRNAs are found to be the most promising biopharmaceuticals in commercial space as oligonucleotide-based next-generation medicines.Currently, miRNA-and siRNA-based candidate drugs are being evaluated in ~20 clinical trials [29].
For MERS-CoV, two studies have reported in silico design of siRNA and miRNA against orf1ab replicase polyprotein using computational methods for possible use as antiviral therapeutics against MERS-CoV [30,31].Recently, we have also reviewed the design and delivery of potential siRNAs for MERS-CoV [26].In the present study, we report the in silico design, synthesis and cytotoxic effect of seven siRNAs targeting the orf1ab region of the MERS-CoV viral genome.
Sequence isolation and analysis
Total seventeen Orf 1ab gene sequences of MERS-CoV isolated from human and camels were downloaded from GenBank.Sequences were multiple aligned using BioEdit sequence alignment tool and ClustalW software, the accession numbers and their sequence identity matrix are presented in Table 1.
In silico prediction and Scoring of siRNAs
The MERS-CoV genome sequences were retrieved from NCBI database submitted from various countries and orf1ab gene was extracted for multiple sequence alignment and siRNA prediction.The overall flow for prediction and selection of potential siRNAs has been presented in Figure 1.The characteristics and thermodynamic properties of predicted siRNAs are presented in Table 2.
Screening of Off-Targets
We used two different phases of selection to avoid any off-target binding of the designed siRNAs.We performed BLAST and Smith-Waterman algorithms as implemented in ParAlign to find out the near complimentary and similarity match with human mRNA sequences in NCBI database and increase the validity percentage of the siRNAs and filtered out the siRNAs with reduced off-targets effect [32][33][34].In the next phase, siRNAs were filtered through the exclusion of seed-region like human mRNA [35].
Thermodynamics, Target Accessibility, and secondary structure prediction
The thermodynamics and target accessibility were performed by using RNAxs tool [36,37] and secondary structure of siRNA was designed utilizing RNAfold server (http://rna.tbi.univie.ac.at).This program helps in the calculation of minimum free energy and partition function of RNAs by reading the RNA sequences.
Final selection and chemical synthesis of siRNAs
The siRNAs were finally selected based on the best parameters following the rule of Ui-Tei, Reynolds and Amarzguioui [24,[38][39][40].The predicted siRNAs against the targeted Orf 1ab gene of MERS-CoV were chemically synthesized by Integrated DNA Technologies (USA) (Table 2).
3 Cytotoxicity evaluation of siRNAs in cell culture
Cell Culture and siRNAs transfection
The Vero cells (ATCCCCL-81) were grown and maintained in complete DMEM medium at 37°C and 5% CO2 and used for siRNA transfection by reverse transfection.The Vero cells (1 x10 4 ) were plated into 96well plates (in triplicates) with 60-80% confluency and transfected with siRNAs using Lipofectamine 2000 (Invitrogen, USA) as transfection reagent following the manufacturer's instructions.Briefly, the complex was prepared by diluting 50μM siRNAs stocks at various concentration (0.01 to 50 nM) in 100μl Opti-MEM medium by adding Lipofectamine and incubated for 30 minutes at room temperature.The siRNA-lipid complex (1μl) was added to the Vero cells at various siRNA concentration (0.01-50nM) and mixed gently.Vero cells without siRNA, cells with Opti-MEM and cells with Lipofectamine were used as negative controls.The transfected cells were incubated for three days at 37°C.After 72 hours post-transfection cytotoxicity assay was performed on the transfected cells.
Cytotoxicity Assay
The cytotoxicity of siRNA-Lipofectamine 2000 complex on Vero cells was evaluated by MTT [3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay.After 72 hours post-transfection, media containing transfection reagents were removed and 100 l of fresh media was added to the Vero cells.The MTT assay was performed by using Vybrant TM MTT Cell Proliferation Assay kit (Invitrogen) as per manufacturer's instructions.A total of 10l MTT stock solution (12mM) was added and incubated at 37°C for four hours.The formazan crystals were dissolved in 100L SDS-HCL solution by further incubation at 37°C for four hours.The samples were mixed thoroughly, and absorbance was measured at 570nm using SpectraMax i3x imaging cytometer.All the experiments were performed in triplicates and the mean value of OD was used for calculating cytotoxicity using the standard formula.
MERS-CoV genome analysis, in silico prediction, and selection of potential siRNAs
After performing the whole genome sequence analysis of MERS-CoV, we selected the orf1ab gene as a target region for further sequence analysis.The multiple sequence alignment results showed high conservation among all the sequences analyzed.Multiple sequence alignment of the MERS-CoV strains showed that Orf 1ab gene to be highly conserved.In silico analysis and scoring of potential siRNAs targeting the targeted Orf 1ab gene of the MERS-CoV was performed using online software siDirect 2.0 [24,39,40].Potential siRNAs with no off-target matches with any human mRNA sequences were filtered and selected.The outline for in silico prediction, filtration and final selection of potential siRNA with high scores without off-target effects has been given in figure 1.During in silico prediction, many siRNAs were found to fulfill the less favorable criteria targeting a region in the orf1ab genome of MERS-CoV.By using this strategy, we selected total twenty-one functional, off-target reduced potential siRNA from four hundred and sixty-two siRNAs based on their predicted high specificity and potency and the resulting candidates were further narrowed down to only seven siRNA as per the guidelines and rule of Ui-Tei, Reynolds and Amarzguioui [24,39,40].The predicted siRNAs were expected to be highly specific and potent against MERS-CoV.
Target accessibility and secondary structure prediction
The thermodynamic properties and target accessibility were performed using RNAxs tool and results are presented in figure 2, table 2. The target accessibility analysis provided the binding position of each siRNAs at starting point along the Orf1ab gene of the viral genome such as siRNA-1:110, siRNA-2:32, siRNA-3:1184, siRNA-4:180, siRNA-5:1230, siRNA-6:291, siRNA-7:5167 of the viral genome.The secondary structure prediction was performed using RNAfold with only seven siRNAs sequences to screen initially about the performance of selected siRNAs in the cell culture and the structures are presented in figure 3. Based on results obtained using online software the results of thermodynamic ensemble prediction, the minimum free energy was observed to be -2508.09kcal/mol and the ensemble diversity was 1611.05 for partial Orf 1ab genome of MERS-CoV.
siRNA transfection and Cytotoxicity Assay
The reverse transfection using Lipofectamine 2000 was performed to mediate the delivery of siRNAs to Vero cells.The cytotoxicity results are presented in figure 4 and table 3. The cytotoxicity was evaluated for seven siRNAs to screen the performance of these siRNAs in Vero cells and cytotoxic effects of the selected siRNA were found to be concentration dependent.The CC50 was observed to be variable for each siRNAs: siRNA-1: 220.2, siRNA-2:408.3,siRNA-3:345.2,siRNA-4:322.9,siRNA-5:270.7,siRNA-6:491.6,siRNA-7:187.3.None of the tested siRNAs showed cytotoxicity to Vero cells up to 50nm concentration.
Discussion
Since the discovery of the MERS-CoV in 2012, tremendous research has been made globally to fill in gaps in the existing knowledge and have provided detailed and advanced information about MERS-CoV (PromedMail, 2019; WHO 2019).Still, more research is needed especially in the area of vaccines and drug discovery.The treatment of various diseases by using RNA-based therapeutics has shown promising results and continuous development with improved technologies are evolving.Several studies have investigated the important role of RNAi technology in the fight against many viral diseases [41][42][43].The technology provides a very specific route to silence the function of the desired gene and is being used to develop therapeutics against many diseases [29].Earlier studies were faced by obstacles like off-target binding, delivery, and stability and stimulation of immune responses which significantly contributed to hampering the use of RNAi technology as therapeutics.But, continuous research have successfully overcome these hindrances and the technology is now in clinical trials for use as therapeutics against several infectious diseases [44][45][46][47][48][49][50][51].The specificity and potency of predicted siRNAs can be improved using updated softwares [24,39].
Recently, multiple siRNAs against HCV 5'-NTR have been designed and tested.The results showed that HCV321, HCV353, HCV258 siRNA were the most efficient siRNAs against HCV replication [50,52].The in silico predicted siRNAs were expected to be more efficacious to inhibit the MERS-CoV replication.
The MERS-CoV replication process initiates by the binding of the viral particle with cellular receptors through the S-protein.ORF1ab includes two-thirds of the coronaviruses genome, is responsible for encoding non-structural proteins [53].The RNA-dependent RNA polymerase and helicase proteins that are encoded by ORF1ab are involved in the transcription and replication of the virus by forming replication-transcription complexes [54].The replication-transcription complexes assemble at the perinuclear regions and associate with double-membrane vesicles derived from the endoplasmic reticulum [55].ORF1ab gene was selected as a target for this study for several reasons, first it is long enough to give a chance for design and selection of multiple candidate siRNAs, second because of its role in viral transcription and replication making it a possible target to control the virus replication and last because it showed high similarities (>99% based on multiple sequence alignment) among MERS-CoV strains retrieved from GenBank.Earlier studies have previously reported the design of siRNAs and miRNAs against MERS-CoV-Orf1ab gene using bioinformatics tools [30,31] but their cytotoxicity and antiviral activities have not been evaluated.The potential of RNAi technology for the treatment of viral infections has encouraged us to investigate their potential as antiviral therapy candidates against MERS-CoV.
As a preliminary test for the proof of concept, our primary objective was to in silico predict and design a potential siRNA.Then the second objective was to evaluate the cytotoxicity study of predicted siRNAs into Vero cells.In this study, we have predicted, designed potential siRNAs targeting the MERS-CoV-Orf1ab gene by using automated online software by filtering and excluding the off-target effects [24,39].Based on the extensive bioinformatics analysis, the outputs resulted into multiple potential siRNAs, but we selected only seven siRNAs against the conserved target MERS-CoV ORF1ab gene with improved target accessibility and high expected antiviral potency and no off-target effect.The designed siRNAs were chemically synthesized and we evaluated the cytotoxicity of seven synthetic potential siRNAs in Vero cells using Lipofectamine 2000 mediated delivery and found them to be non-toxic with a CC50>50nM.The rationale behind in silico designing is briefly discussed.The selection of potential siRNA starts with the target sequence input that is then reverse complemented to generate antisense siRNA 1-nt shifted overlapping 21-nt siRNAs representing putative guide sequences, excluding 21-nt siRNAs containing unwanted motifs or their complements.Amongst rest, active siRNA molecules are calculated, screened and predicted based on various sets of parameters (like base-preference rules, duplex-related issues) related to the sequences and secondary structures of the guide strands and to the secondary structure of the target mRNA.Sequences with potential identity to off-targets are further excluded using BLASTn or S-W analyses.Otherwise, if few promising or nil candidates get predicted, more antisense siRNAs can be generated and selected by using the option of sequence space expansion.Space of guide RNAs can be expanded by performing A to G and C to U base exchanges within guide sequences, which induce wobble pairing with the target but preserve target complementarity and thus silencing activity.This escalates the complementary guide siRNAs number by more than a factor of thousand for target sequences.However, this computational method may be biased towards the inception of active guide sequences and/or structures.Based on the empirical scoring functions, the selected candidate sequences are finally ranked [56].Thermodynamically during RNAs interaction, the total binding energy is the sum of hybridization energy and breaking energy.Binding can occur only at positions free from prior intramolecular base pairs.The base pairs within the target site must be opened and made accessible, and so the energy used is named as disruption or breaking energy [57].After the binding site is free of any secondary structure, intermolecular helices are formed, yielding stabilizing interaction energy.siRNA design tools (namely OligoWalk, Sirna, RNAxs) perform siRNA design aided by target accessibility criteria.
Conclusions
In conclusion, the in silico predicted and designed potential siRNAs against a specific target of MERS-CoV-orf1ab gene can be utilized to develop oligonucleotide-based therapeutics without off-target effects, high efficiency, and improved specificity.The results generated in this study has provided preliminary information to evaluate the virus inhibitory effect of siRNAs in multiple cells culture systems against MERS-CoV.This novel technology for the prediction of potential siRNA can be utilized against viruses for new anti-viral therapeutics after cell-based validation.The recent developments with improved technology to use therapeutic miRNA and siRNA shows the most valuable and significant breakthroughs not only in therapeutic molecules development but also in intellectual property rights and therapeutic business for pharmaceuticals industries against multiple diseases.Author Contributions: SSS, SAE designed and executed the experiments, ZM, SSS performed bioinformatics study.SSS, SAE, ZM wrote the manuscript.EIA: Contributed in designing and execution of experiments and critically reviewed the manuscript.All authors provided critical feedback and helped shape the research, analysis and manuscript.Funding: This study was financially supported by King Abdullah City for Science and Technology (KACST), Riyadh, Saudi Arabia for providing the grant on MERS-CoV bearing project number 39-2.Authors are grateful for financial support from the KACST, Riyadh, Saudi Arabia.
Table 1 .
Sequence Identity Matrix of Selected MERS-CoV based on full genome.
|
2019-04-03T13:11:50.463Z
|
2019-02-28T00:00:00.000
|
{
"year": 2019,
"sha1": "094932f4eb2ef998dbea56bc2e741397c77e263d",
"oa_license": "CCBY",
"oa_url": "https://www.preprints.org/manuscript/201902.0268/v1/download",
"oa_status": "GREEN",
"pdf_src": "Anansi",
"pdf_hash": "094932f4eb2ef998dbea56bc2e741397c77e263d",
"s2fieldsofstudy": [
"Biology"
],
"extfieldsofstudy": [
"Chemistry"
]
}
|
255893095
|
pes2o/s2orc
|
v3-fos-license
|
SOLAR-POWERED AERIAL FUNICULAR: A CONCEPTUAL FRAMEWORK FOR ACCESSIBILITY AND CONNECTIVITY BETWEEN THE METROPOLITAN CITIES AND THEIR SATELLITE CITIES IN PERIPHERY
The urban accessibility and connectivity are enormously important factors to measure the convenience of life and the territorial liveability for residents, and they are also the most essential prerequisites for planning satellite cities in periphery near the metropolitan cities and for promoting satellite cities’ inclusive progression. Previously relative contributions demonstrated that although the metropolitan cities and their satellite cities are located quite close, but still existing the defeat of accessibility and connectivity. The article identified the need for improvements in innovating the way of eco-friendly mobility to realize the new generation green transportation under the context of all societies facing the dual pressure of environmental pollution and energy crisis. Therefore, the article presented a conceptual framework for connecting the boundary of the metropolitan cities and their satellite cities in the periphery. Through a qualitative analysis of the structure and composition, operation mechanism and the comparative advantages of the solar-powered aerial funicular and the case of Milan-Cusago, results from the research showed that it had potential as a new initiative for clean and green mode of urban mobility to further achieve accessibility and connectivity between the metropolitan cities and their satellite cities that contributes to transition for Smart Sustainable Cities.
Introduction
Generally, the present the metropolitan cities already have relatively completely public transportation infrastructures which are still the main reason why many people choose where to live. As far as Italy is concerned, among the existing mobility modes issues, there are still imperfections of inadequately friend environment of public transportation infrastructures for effectively realizing accessibility and connectivity between the boundary of the metropolitan cities and their satellite cities in periphery. Put briefly, the overwhelming majority of residents from satellite cities in periphery cannot imagine how to travel this distance without private motor vehicles. The practices for bicycle and pedestrian travel are selected by a small number of people. However, many researchers in their research gave their suggestions for further improving bicycle lanes or changing land use for non-motorized facilities and found a few concerns among local business owners that non-motorized facilities could negatively impact the carcentric economy (Hankey et al., 2012;Volker & Handy, 2021). Thus, the practical issues of accessibility and connectivity still prevail between the metropolitan cites and their satellite cities. In fact, these two factors also constitute serious barriers to incentivizing urban dwellers to divert them to settle down in satellite cities from the metropolitan cities. Transport is one of the main contributors to air pollution in Italy, so the promotion of sustainable travel could contribute significantly to reducing CO 2 and pollutant emissions. Eco-friendly mobility is based on smart cities, electric vehicles, connected cars, and a sharing economy, but above all on a widespread and participatory cultural change that leads to alterations in the daily habits of each citizen. On 25th September 2007, the European Commission has passed the Green Paper on urban mobility as a new European agenda for urban mobility with the intention of facilitating the search for solutions (Green Paper on urban mobility, 2007). While in 2015, "UNECE and ITU developed jointly a definition of smart sustainable cities" to efficiently meet more serious challenges existing in all societies with reference to economic, social, environmental and cultural aspects through ICTs as well as other means (Sustainable Smart Cities, 2015). Obviously, as a comprehensive guide for the future development trend of the cities, the EU strategy and the SSC concept will inevitably have a profound impact on the existing patterns of public transportation infrastructures and their future constructions for the metropolitan cities and their satellite cities. Especially, the revived interest in SSC intensifies after the long-term lockdown caused by global outbreak of COVID-19 pandemic. This reflects a strong consensus related to substantial shift in awareness of the urgent need for sustainable lifestyles and there is the immediate necessity for rethinking the green mode of next generation sustainable urban transportation. Namely, taking advantage of cleaner modes develops the existing modes of urban transportation to a whole new stage. Solutions for transport by rope have been around for thousands of years (Abelle, 2022). In 1908, the aerial funicular became popular for civilian transportation in Swiss (Freire-Medeiros, 2015). As a world-famous ski resort, the passenger aerial funicular connecting the mountains is naturally the most ideal alpine transportation. According to Težak et al. (2016), "Today's built cable cars have capacities up to 2,000 persons/h for aerial tramways (or jig-back ropeways) and up to 4,000 persons/h for gondolas" (p. 1). Therefore, the development of the system and technical technologies of the aerial funicular is becoming more and more mature.
Within the broader research perspective and the frame of an innovative city, this article aims to provide a solar-powered aerial funicular -conceptual framework for the solution-focused accessibility and connectivity between the boundary of the metropolitan cites and their satellite cities in the periphery.
Since the late 1980s, many scholars have conducted various studies on car dependence (Newman & Kenworthy, 1989;Buehler, 2011;Metz, 2013;Dovey et al., 2015;Buehler et al., 2017;Yin et al., 2018;Heinonen et al., 2021;Saeidizand et al., 2022). And these studies have greatly incentiv ized the numerous studies on calling for green travel as the means to facilitate the transition to SSC. The extensive literature review shows that, at present, the green travel tools used as alternatives to private motor vehicles mainly include trains, subways, light rail, trams, buses, trolleybuses, electric buses, car sharing, bike-sharing, taxi, ride-hailing, electric vehicles, push scooter, e-scooters, rides trolleys, bicycles, e-bikes, segways, mono-wheels, sk ateboa rd s, wa l k i ng (Young, 2015;Clewlow, 2016;Tirachini, 2020;Ahmed & Bulut, 2021;Alaoui et al., 2021;Fields & Renne, 2021;Kjaerup, 2021;Schwedes & Otto-Zimmermann, 2021;Koplowitz, 2022;Newton et al., 2022). However, although these both public and private alternatives to private motor vehicles have generated significant impact on car dependency in the process of urbanization upgrading and transformation in the past few decades, they are still not sufficient to tackling tough challenges for meeting the urban residents' daily travel challenges of accessibility and connectivity between the metropolitan cites and surrounding satellite cities (Vitale Brovarone & Cotella, 2020;Lin et al., 2022). On a global scale, this issue is not only widespread but more challenging than the reality. Furthermore, "there is a lack of appropriate planning models for urban regeneration" (Newton et al., 2022, p. V). Undoubtedly, there is emerging evidence of the needing immediate attention on the accessibility and connectivity for satisfying the need of residents' daily travel by sustainable mode of transportation, particularly in the era of post-pandemic. Without forgetting that on a social level, sustainable mobility improves the quality of life of residents who are both from the metropolitan cities and especially their satellite cities in periphery. Besides, more other advantages are like such as more less traffic, cleaner air and fewer traffic accidents.
Furthermore, solar energy is a renewable energ y that has been w idely used in production and life (Li et al., 2007;Chikaire et al., 2010;Samimi et al., 2012;Lodi et al., 2018;Sansaniwal et al., 2018;Chandra Mouli et al., 2020;Robisson et al., 2022). The utilization prospect of solar energy is still very broad and continues to be deeply developed (Tyagi et al., 2020). Swiss designer Fredrik Hylten created the world's first conceptual solar-powered cable vehicle -Taiyou Ropeway which is in an energy selfsufficiency way for transporting solution in remote areas where is short of electrical power supply (Hyltén-Cavallius, 2009). While people's traditional understanding of the cable car is limited to the fact that it is the best auxiliary tool for travelers when they are tired of climbing, among which Switzerland is the most famous. In line with the "rural-urban interaction" theme, during the 2010 Shanghai World Expo, the Swiss Pavilion launched the innovatively core program of taking the cable car to experience the "city to the countryside", which made the visitors from all over the world truly feel the importance of the urban regeneration and sustainable development (Wang, 2010;Stylepark, 2010). Many countries have already applicated the technical-technologies and systems of aerial funicular as the solution for the issues related to public transportation (Brand & Davila, 2011;Bocarejo et al., 2014;Heinrichs & Bernet, 2014;Težak et al., 2016;Garsous et al., 2019;Matsuyuki et al., 2020), such as Medellin and the U.S. Portland. More than a decade ago, scholars have addressed the issue of the aerial cable transport for the urban mobility (Fistola, 2011). Thus, the existing technical-technologies and relative transportation systems of aerial funicular have demonstrated the capacities and huge potential compared to other modes of transportation.
In fact, urban cableways can play a very important role in public transport systems by integrating with existing modes of transport to fill their gaps: • they need little space and can be the optimal solution to connect specific areas or particularly sensitive zones; • they allow considerable architectural freedom of expression, particularly in the design and placement of stations and line supports; • they are able to overcome any obstacle or slope by hovering in the air, while at the same time giving travelers the opportunity to enjoy unique views; • are very safe as they do not share the route with anyone else, do not cross roads, do not risk collisions with other vehicles; • they g uarantee shor ter and more certain journey times, with reduced waiting times at stations and freedom of timetable thanks to continuous operation; • they offer spacious cabins that can be freely used by people with different motor sk ills; it is possible to load bic yc le s, pu shc h a i r s, s u itc a se s; the cabins can be equipped w ith every comfort, from heating to air conditioning, from sound diffusion to Wi-Fi networks; • they offer high transport capacity with excellent energy efficiency (a modern urban system can transport up to 5,000 passengers per hour and direction, to do the same overland would require 100 buses or 2,000 cars); • they are relatively inexpensive in relation to both construction and operating costs, they do not require travelling personnel, they have low fuel consumption and can be amortized over many users, energy consumption per user is significantly lower than any other means of transport, and electricity can be generated from renewable sources; • t h e y h a v e c o n s i d e r a b l y f a s t e r construction times than any other system.
In virtue of these characteristics, in some cities this transport system is taking on the true form of a network with connections that intersect, change direction and branch off in different directions.
In fact, there are also some limitations, in particular related to the need to arrange the line supports along an obligatory rectilinear route and the overf light limits that the presence of these systems poses, particularly when they are very high above the ground.
With the help of an extensive literature review, we have understood the current state of the research topic. And this article will explore, through qualitative analysis and problem analysis method, combining the solar-powered approach with aerial funicular mode of transportation to address accessibility and connectivity between the boundary of the metropolitan cites and their satellite cities in periphery. To be specific, the research proposal will begin with a qualitative analysis of the structure and composition, operation mechanism and the comparative advantages of the solar-powered aerial funicular then will be integrated the case of Milan-Cusago for giving explanations on the conceptual framework.
2. Why do We Need to Propose a New Alternative?
Fi rst ly, t he c u r rent major mea ns of transportation still have limitations that cannot be ignored. Compared to other shared transportation modes, private motor vehicles embody a kind of complete property ownership. This is also the main psychological factor why the vast majority of people would rather buy a private car than share it with others. However, the exhaust emission of private cars has become one of the most important sources of urban traffic pollution. Although the concept of car-sharing is widely recognized, in fact, scholars have found that the availability of car-sharing has a limited impact on vehicle ownership (Zhou et al., 2020). Due to in the context of global inflation, consumers' purchasing power and willingness to buy began to shrink, the cost of electric vehicles or foldable bicycles for ordinary residents still cannot be underestimated.
Secondly, at present, even in many the metropolitan cities, there is still a lack of a friendly convenient environment for green travel, which seriously hinders residents' active response to changing their travel methods.
Thirdly, the energy crisis is a common problem faced by all countries in the world today and it has brought a considerable impact on the existing modes of transportation.
In the long run, the shortage of energy will cause the main trend of gasoline price increase, and the retail price of refined oil will gradually increase accordingly. Thus, the importance of using urban public transportation is highlighted (Fenta, 2014).
To sum up, in the context of the disadvantages of the existing travel modes, the unfriendly travel environment and the further deepened energy crisis, in order to solve the problems of accessibility and connectivity between the boundary of the metropolitan cities and their satellite cities in periphery, the innovation solution for contributing to sustainability is urgently needed.
Why can the Solar-Powered Aerial
Funicular Proposal Overcome the Limitations of Existing Modes of Transportation?
Structure and Composition
Obviously, the aerial funicular is a road suspended in the air. The modern aerial funicular for passengers is the use of wire rope traction appeared in the 19 th century (Abelle, 2022). The load-bearing ropeway and traction ropeway are respectively used to bear the weight of the cabins and run the vehicle. In 1834, the steel cable invented by Albert was greatly improved security and speed of aerial funicular (Hoffmann & Zrnić, 2012). The towers of the aerial funicular ropeway are mainly used to support the weight of the ropeway, cabins with passengers (Abelle, 2022). On the one hand, the selection of the route should avoid excessive fluctuation of the ramp to save the cost of capital construction. On the other hand, for safety reasons, relatively open green space is one of the important considerations for route selection. The vast space of the surrounding satellite cities has the proper advantage. In addition, the distance from the boundary of the metropolitan cites to their satellite cities in the periphery is relatively short, which greatly reduces the number and corresponding investment in the funicular ropeway tower. Regardless of the length of the distance and the number of towers of the aerial funicular ropeway, the entirely aerial funicular system must have at least two stations (starting stations and final stations) which are used for cabins parking and passengers' getting on and off (Abelle, 2022). In addition, although Swiss designer Fredrik Hylten has designed a solar cable car, its fully open C-shaped design has a 360-degree view and the ropeway passenger traffic is highly adaptable to the environment, the factors of strong wind, rain, snow and especially safety are still taken into account. Therefore, the structure of the aerial funicular we propose is still the first choice for traditional enclosed cabins and transparent to facilitate viewing the surrounding scenery (Fig. 1).
The long-distance transportation of aerial ropeway is usually used as the solution of critical transportation; therefore, its application and development are relatively matured. While our proposal is to suggest applying it to the short distance between the metropolitan cites and their satellite cities in periphery, usually a distance of 10-20 kilometers, thus, we have even more reason to believe that mature technical-technologies and systems of the aerial funicular used for long distances are more feasible for short distance applications. More important, it is very beneficial to stimulate the inefficient land use and to improve poor dimensional planning by promotion of aerial funicular.
Operation Mechanism
The aerial funicular is to suspend the cabins or seat in the air through the ropeway and the towers supporting the ropeway and transport passengers through the traction of the ropeway (Fig. 2).
In general, the operation mechanism of the aerial funicular is to use a pole less wire rope, which is set on the driving wheel and roundabout wheel at both ends of the cableway and maintain a certain tension through the tensioning device. In general, the driving wheel drives the wire rope at a speed of 6.0 m/s (Težak et al., 2016). The operation of the aerial funicular is a huge technical project in which every mechanical part plays an important role. According to the different operation and construction, it can usually be divided into reciprocating and circulating ropeways. Thus, according to the local specific travel data of residents, the design should consider assuming a two-lane round-trip aerial funicular for passengers or increase the passenger capacity of the cabins. Besides, compared with the vast majority of traditional cabins, we recommend that a mature solar power storage system be loaded into the cabin body to be driven by more natural and clean energy. Solar cabins are nothing new, but it is the first time this concept has been used to address accessibility and connectivity between the boundary of the metropolitan cites and their satellite cities in the periphery as another form of green mobility. Težak et al. (2016) have argued the advantages of the aerial funicular in public transport within urban areas. Firstly, generally speaking, the technical-technologies and system of the aerial funicular are very mature (Abelle, 2022). Its structural principle and operating mechanism are very clear, and scholars are still working on improving its operating efficiency (Težak et al., 2016). Therefore, mature technical-technologies and continuous improvement of capacities of the devices and operating efficiency have laid a solid foundation for the application of the aerial funicular. Secondly, motorized and non-motorized vehicles on the ground are a major contributor to road congestion (Težak et al., 2016). Thus, one of the most obvious features of the aerial funicular is that as a public transportation mode above the ground, it does not compete with other green travel modes for the limited resources of the surface road. This advantage can only be matched by light rail, but the cost of light rail is significantly higher, and the construction work is also more extensive. Therefore, this point has a significant comparative advantage. Thirdly, usually, the traditionally green modes of transportation are electrically powered which has shown its limitations under the threat of energy crisis, and our conceptual framework of solar-powered aerial funicular is proposed as the solution which is a favorable alternative to alleviating the energy crisis. Fourthly, there is no CO 2 emissions generated by the solar-powered aerial funicular respected to other modes of transport. As a form of public transport, it will certainly make a significant contribution to the goal of zero carbon dioxide emissions. Fifthly, from the perspective of cost, the application of solarpowered aerial funicular equipment and technical-technologies in many countries has gradually developed and matured, and there is no need for complicated public transportation infrastructure construction. Therefore, in addition to having evident cost efficiency (Abelle, 2022), the corresponding time-consuming is also relatively less. Sixthly, according to the research of Težak et al. (2016), the speed of aerial funicular transport is limited to "12 m/s or 43.2 km/h" (p. 4); while Tirachini (2013) researched the average speed of city buses is at 38.9 km/h. Taking Milan as an example, the city currently has various public transportation facilities on the ground and underground. At the same time, the municipal government is also preparing to further increase the investment in urban public transportation i n f ra s t r uc t u re. However, t here a re still serious issues of accessibility and connectivity between the boundary of the metropolitan cites and their satellite cities in the periphery during the development of the metropolitan area.
The Comparative Advantages
Particularly in the municipalities of the homogeneous zone South-West of Milan where a number of specific criticalities due to a lack of careful planning of Local Public Transport (LPT) services, which create difficulties in connecting with some important public services, are noted. For this reason, a mobility development project for peripheral cities in the homogeneous area of South-West Milan is being studied.
The instrument, which takes the form of a Sustainable Urban Mobility Plan (PUMS), aims to reorganise the LPT and offer solutions to improve connections to the main existing polarities (first and foremost health and school poles), working on intermodality, in particular between road, rail and soft mobility, in a vision that can integrate the 'classic' model with new forms of 'intelligent mobility', evaluating the possibility of promoting the use of more ecological means, of the electric type, and exploiting new technologies. Our proposal for a solar funicular railway to connect the city of Cusago, which is part of the municipalities of the homogeneous area south-west of Milan, is part of this framework. Taking the city Cusago around Milan to the boundary of Milan as an example, the distance between them is only 7.5 kilometers (Fig.3).
According to Google Maps analysis, this distance takes an average of 13 minutes if a private car is used (not counting the waiting time for traffic lights caused by traffic congestion in the morning and evening peaks); if a pullman operated by the private company takes an average of 21 minutes (not counting the waiting time caused by low-frequency, the low-volume operation of vehicles due to damage, the greatly shortened operating time on Saturdays and Sundays with the waiting time of more than one and a half hours); if the use of bicycles and other non-motor vehicles takes an average of 25 minutes (the unfriendly environment factor of the bicycle path has not been considered); if the walking method is adopted, it will take an average of 1 hour and 25 minutes (but the corresponding friendly environment has yet to be established). While at 43.2 km/h, the proposed solar-powered aerial funicular will take only 10 minutes.
In the face of ground traffic congestion and inadequate construction of public transport facilities, the solar-powered aerial funicular transport mode shows a favorable comparative advantage. Finally, as one of the urban public transportation solutions, in addition to eliminating the psychological factors of property rights for private transportation, it is more important for the government to systematically coordinate the planning and development of sustainable public transportation, so as to further promote SSC in an all-round way.
Discussion and Future Research Directions
The research on the alternative solution -sola r powered aer ia l f u n ic u la r for accessibility and connectivity between the metropolitan cites and their satellite cities in periphery demonstrated the great potential of facilitating the transition of SSC.
Currently, all green travel tools used as alternatives to private motor vehicles follow the same value concept, which is to provide a more sustainable green and efficient urban transportation solution. And it is much crucial to bring up even more effective solutions for realizing accessibility and connectivity between the metropolitan cites and their satellite cities in periphery. On the one hand, the conceptual framework of solar powered funicular for resolving accessibility and connectivity between the boundary of the metropolitan cites and their satellite cities in periphery is beneficial to renewing the interest in the benefits of mass transit for public good in order to advance the sustainably urban transportation transition; on the other hand, the co-evolution of multi-level transit modes is an inevitable development trend of green transportation in line with SSC. Specifically, it is imperative to promote the establishment of governmentcentered susta i nable t ra nspor tat ion system including but not limited to public transportation policy, fund raising and investment, and the technologic operation and infrastructure maintenance.
Obviously, this article focused on the conceptual framework thus it has limit in terms of insufficient sample size for realizing a more comprehensive perspective, and there are still many supporting solutions to be solved for the practical implementation.
To be spec i f ic, t he i mplement at ion a nd rea l i z at ion w i l l be a f fec ted by many factors, such as policy, funding, supporting infrastructure, information and communication technologies, psychology: • Supporting by policy and funding. The development of public transportation is inseparable from the guarantee of policies. The government's strong credibility is more conducive to raising funds for public transport investments than to individuals or enterprises. In addition, the government can use the power of the people to actively raise funds to support the development of the public infrastructure project by vigorously seeking the issuance of special bonds. • Su p p or t i n g s olut ion s f or s ol a r energy storage facilities. Sustainable development is more and more urgent, solar energy will become the main energy source in the future, and the demand for solar panels in all walks of life will continue to grow rapidly. However, the storage of solar energy is critical (Hou et al., 2011), as the collection of solar energy is affected by the weather, therefore, a sufficient solar energy storage facilities and reserve system should be equipped to ensure that the solar panels provide sufficient power. • Alternative energy supply solutions for solar energy. We still need to consider the insufficient or problematic supply of the solar energy storage system during continuous and severe thunderstorms and grid outages. • The maintenance of solar-powered aerial funicular. The safety of transportation facilities is extremely important, and the safety of air public transport is especially life-threatening. From solar panels to aerial funicular operating systems, the updating of technical-technologies and the maintenance of facilities are especially important. • Improvement of level of traffic safety.
For passenger ropeways operating in the air, safety is the top priority. Compared with traditional closed cabins, we can consider adding the elements rubber boat-like to the bottom of such lowaltitude traditional closed cabins to further increase the safety. • Application of technology. In view of the efficient and user-friendly cabin selection system and method of aircraft and ships, we can also use online or dedicated apps to optimize occupancy rate and management of passengers' waiting time.
The application of AI technology in solarpowered aerial funicular access control and security systems also needs to be further developed and utilized.
• Psychological guidance supporting. There is an urgent need to gradually guide the public to experience the transportation mode in exposed heights under the premise of safety.
Conclusion
As an emerging new mode of transportation for addressing accessibility and connectivity between the boundary of the metropolitan cites and their satel lite cities in the periphery, solar-powered aerial funicular will comprehensively demonstrate the advantages of high efficiency, intelligence and sustainability, leading the green travel in the future. In sum, solar-powered aerial funicular's solution is not only support for SSC, but also a response to sustainability initiatives.
|
2023-01-17T19:07:10.403Z
|
2022-12-01T00:00:00.000
|
{
"year": 2022,
"sha1": "914be58afdcea00a3dc7bc80fe41da59e20db61f",
"oa_license": null,
"oa_url": "https://doi.org/10.7708/ijtte2022.12(4).01",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "188c57f37bed1d5c454786e57e9faa5fff903a7d",
"s2fieldsofstudy": [
"Environmental Science",
"Engineering"
],
"extfieldsofstudy": []
}
|
3547453
|
pes2o/s2orc
|
v3-fos-license
|
Vortices in vibrated granular rods
We report the experimental observation of novel vortex patterns in vertically vibrated granular rods. Above a critical packing fraction, moving ordered domains of nearly vertical rods spontaneously form and coexist with horizontal rods. The domains of vertical rods coarsen in time to form large vortices. We investigate the conditions under which the vortices occur by varying the number of rods, vibration amplitude and frequency. The size of the vortices increases with the number of rods. We characterize the growth of the ordered domains by measuring the area fraction of the ordered regions as a function of time. A {\em void filling} model is presented to describe the nucleation and growth of the vertical domains. We track the ends of the vertical rods and obtain the velocity fields of the vortices. The rotation speed of the rods is observed to depend on the vibration velocity of the container and on the packing. To investigate the impact of the direction of driving on the observed phenomena, we performed experiments with the container vibrated horizontally. Although vertical domains form, vortices are not observed. We therefore argue that the motion is generated due to the interaction of the inclination of the rods with the bottom of a vertically vibrated container. We also perform simple experiments with a single row of rods in an annulus. These experiments directly demonstrate that the rod motion is generated when the rods are inclined from the vertical, and is always in the direction of the inclination.
I. INTRODUCTION
Granular materials are well known examples of dissipative nonequilibrium systems that show a rich variety of collective phenomenon such as convection, wave patterns, and segregation [1][2][3][4]. Most studies utilize spherical particles to investigate these bulk properties. However, in most natural or industrial settings one can find an abundance of anisotropic granular materials viz. rice, medicine capsules, and even logs. Therefore it is surprising that very few studies of prolate granular materials have been carried out. Mounfield and Edwards [5] applied the concepts of configurational statistical mechanics to study the nature of the isotropic to nematic phase transition in a granular system of elongated particles. In recent experiments utilizing a tall narrow cylinder, Villarruel et al. [6] studied the effects of anisotropy on granular packing. They observed the appearance of smectic states with the direction given by the container walls.
In thermal systems, particle anisotropy is known to produce ordered states. Examples are rod and plate shaped colloids and liquid crystals, which show orientational order and form nematic and smectic phases [7][8][9][10][11]. The ordering mechanism was found to be entropically driven, i.e. as thermodynamic equilibrium is reached, the process of entropy maximization leads to long range order. It is not obvious that such mechanisms carry over to granular systems because the thermal energy scale is very much smaller than the potential energy required for particle rearrangement, and energy has to be supplied actively to produce sustained motion. Therefore, an interesting question arises -does shape anisotropy lead to self-organization and pattern formation in granular materials?
In this paper, we report the observation of novel vortex patterns exhibited by granular rods that are vibrated vertically inside a container. We obtain the phase diagram for the observed patterns as a function of the acceleration of the driving and the packing fraction of the rods. We find that for sufficiently large packing fractions, the rods tend to align vertically and undergo vortex motion. Using high frame rate imaging and particle tracking, we have measured the velocity fields of the vortices as a function of packing fraction and driving frequency. Based on our experimental observations, we argue that the inclination of the rods causes motion due to collisions with the bottom boundary. To bolster our claim, we conduct experiments with single row of rods in an annulus to demonstrate that motion is always in the direction of inclination. Thus we show that shape anisotropy leads to translational motion in such systems.
II. EXPERIMENTAL APPARATUS
The experiments were performed in a circular anodized aluminum container with a diameter D = 6.0 cm and depth H = 1.5 cm. (Limited experiments were also carried out in a container with D = 8.5 cm.) The container is leveled to within 0.002 cm and is attached to an electro-mechanical shaker through a rigid linear bearing that allows motion only in the vertical direction. The cell is driven by a sinusoidal signal at a frequency f , and is monitored by an accelerometer via a lock-in amplifier. The experiments were conducted using copper cylinders with uniform length l = 6.2 mm and diameter d = 0.5 mm. The cylindrical surface of the rods is coated with a grey tin-oxide layer to diminish light reflection and has a coefficient of friction µ = 0.32. The patterns that form are imaged from above using a high-frame rate digital camera (Kodak SR-1000). Since the flat ends of the rods reflect light better than the sides, they appear as bright spots when imaged from above. If the rods are inclined greater than 35 degrees, they reflect far less light compared to nearly vertical and horizontal rods.
One control parameter for our experiments is Γ = A(2πf ) 2 /g, where A is the driving amplitude and g is the acceleration due to gravity. A second control parameter is the non-dimensional number fraction n φ = N/N max , where N is the total number of rods in the container and N max = π √ 12 D d 2 is the maximum number of rods required to obtain a vertically aligned monolayer. Therefore, n φ = 1 corresponds to having one layer of triangularly packed vertical rods.
III. OBSERVED PATTERNS
The system is initialized by pouring the rods into the container and then increasing Γ, which establishes a random state. Figure 1 shows the phase diagram of the observed patterns as a function of n φ (f = 50 Hz). The rods are observed to remain static in the initial configuration until Γ ≥ 1.47. As Γ is increased above this value, the rods heap towards one side of the container similar to previous observations with spherical particles [12]. As the acceleration is increased further above Γ = 2.6, the rods are observed to spread-out evenly inside the container, and form horizontal layers with random orientations. An example of such a nematic phase observed at low n φ is shown in Fig. 2(a). Above a critical n φ , a second transition is observed. Chaotically moving domains of almost vertical rods are observed to spontaneously form and coexist with horizontal rods [13]. An example is shown in Fig. 2 As n φ is increased further, the domains of near-vertical rods that are formed coalesce and undergo vortex motion. The cylindrical surface of the rods is coated with a grey tinoxide layer to diminish light reflection and has a coefficient of friction µ = 0.32. The patterns that form are imaged from above using a high-frame rate digital camera (Kodak SR-1000). Since the tips of the rods reflect light better than the sides, they appear as bright spots when imaged from above. If the rods are inclined greater than 35 degrees, they reflect far less light compared to nearly vertical and horizontal rods.
large stable vortex made entirely of near-vertical rods in the center and inclined rods at its' boundary, is observed [ Fig. 2 If the acceleration is increased further, a gas like state is reached where the rods vibrate vigorously inside the container and the nematic and vortex states are destroyed. For the highest n φ , a pure gaseous state is not reached due to the limits of the apparatus. Instead, domains of ordered vertical rods are observed to coexist with the gas of rods.
A. Domain Growth
We now discuss how vortices nucleate and grow as a function of time. An example of the growth process is shown in Fig. 3, and a movie can be viewed at Ref. [14]. Starting from a random state, pockets of near-vertical domains are observed to nucleate uniformly inside the container. This is in contrast with previous observations where the ordered domains grow inward from the boundary [6]. The pockets of vertical domains grow in time by merging to form larger domains of almost vertical rods. The domains are then observed to collectively move as the the system becomes more ordered and then finally show vorticity.
We measure the growth of the domains by taking timelapse images of the process described above. The growth of near-vertical domains is then measured from the ratio of the domain area to that of the total container area as a function of time. The fraction of rods that are nearvertical, n v is plotted in Fig. 4 for a range of n φ at a fixed driving acceleration and frequency, Γ = 3.4 and f = 50 Hz.
We find that there exist two growth regimes [a model is discussed in Sec. V]. For short times, the growth rate depends on n φ in the following way. If n φ is below a critical value, domains appear but never organize into large vortices. However, as n φ is increased the short time growth is slowed by the high packing. That is, more time is required for small domains to form due to the lack of voids present. A void filling model for domain growth will be discussed in Sec. V. The saturation value of n v increases with n φ . The difference in the values arises due to the definitions of n v and n φ .
The variation of the relative number of near-vertical rods as a function of n φ is plotted in Fig. 5. We show separate data for the relative number of inclined rods, vertical rods and horizontal rods. Rods that are tilted between 30 and 80 degrees are considered inclined, and the rest are considered as either horizontal or vertical (the inclination of the rod was obtained by measuring the distance x between adjacent rods and calculating At short times, the growth is increasing exponentially and then gives way to an exponential decrease in growth to a saturation value determined by n φ . For n φ = 0.38, the domain growth saturates very slowly to a value less than n φ because the near vertical domains cannot be supported by the horizontal rods. For n φ ≥ 0.46 the two distinct growth regimes are apparent and for all n φ vortex motion occurs after t = 300 s. We note that for the highest value of n φ the initial growth displays a slow down, which is directly related to the void filling mechanism discussed in the text. The dashed line is a fit to the integrated growth equation discussed in Sec. V. We find that the simplified model gives an accurate interpretation of the results. sin −1 [d/x]). For n φ < 0.22, the container has only horizontal rods. As n φ increases, more rods tend to the vertical direction. For n φ > 0.71 most rods are either vertical or horizontal.
B. Spatial structure of the vortex and velocity fields
We next discuss the spatial structure of the vortex. In Fig. 6, a close-up image of a vortex pattern is displayed which shows the progressive inclination of the rods from the center of the vortex. We note that the direction of motion corresponds to the direction of inclination (for Fig. 6 the motion is counterclockwise). In this case the inclination of the rods varies from approximately 85 to 40 degrees near the edges. The change in inclination of the rods within the vortex is less at higher n φ due to the increased packing.
By acquiring images at 250 (frames s −1 ), we were able to track the ends of the vertical rods and obtain the velocity fields of the vortices. In addition to the rotational motion, rods also vibrated due to the collisonal interactions with their neighbors. We first track [15] the ends of the near-vertical rods and then perform spatial and temporal averaging to obtain the vortex fields. An example of the velocity field of a vortex is presented in Fig. 7 (here the temporal averages include 100 consecu- tive frames, and spatial averages were taken over an area of 2d × 2d). The averaged azimuthal velocity v(r), as a function of the distance r from the center of the vortex, is shown in Fig. 8(a) for a range of n φ . The data was obtained at a constant acceleration Γ = 3.00 for a frequency f = 50 Hz. We note that for small distances relative to the center of the vortices, the averaged velocity v(r) increases linearly with the distance r indicating solid body rotation. At intermediate ranges, this linear relationship is not observed, and therefore shearing occurs inside the vortices. As the boundary of the vortex is approached, velocity decreases due to friction with the horizontal rods at the edge. It is interesting to note that the horizontal rods at the edge of the vortex are often aligned tangentially with the boundary of the vortex [see Fig. 2(c)]. We also observe that the slope of the averaged velocity v(r) at small r systematically decreases as the packing fraction n φ is increased. Thus the angular velocity of the inner core of the vortex decreases with its size.
C. Frequency dependence
We also explored the rotation rate dependence of the vortices as a function of f . Figure 8(b) shows v(r) for n φ = 0.53 and Γ = 3.00 for vortices of equivalent size. As stated, Γ is held constant while frequency is increased, which decreases vibration velocity. Therefore, we demonstrate that the vortex speed systematically increases with vibration velocity.
IV. HORIZONTALLY VIBRATED CONTAINER
To test if the transition to a vertical state is dependent on the way the container is vibrated, experiments were performed with horizontal driving. There do in fact exist similarities: at low n φ heaping and nematic like domains of horizontal rods are observed; at high n φ the rods tend to align vertically and form ordered domains. An image is shown in Fig. 9. In contrast to the vertical shaking the vertically aligned domains do not migrate and coarsen into vortices. We also find that in this case the convection is stronger and destroys the motion of the vertically ordered domains.
V. DISCUSSION
We first discuss possible mechanisms responsible for the rod to align vertically at high packing fractions. The tendency of the rods to align vertically at high packing fractions may be understood in terms of a void filling mechanism. If voids (i.e. space between rods), are small then they can only be filled with a near-vertical rod. On the other hand, larger voids will accommodate a horizontal rod. Assuming that the distribution of voids decreases with void size, the most probable configuration will be regions of vertical rods. Furthermore, the decrease in the number of large voids is enhanced by the coalescence of regions of vertically aligned rods. This also drives the system to pack more closely and produces a decreased overall center of mass. Thus ordering at high packing fractions can be explained by a process that is analogous to configurational entropy driven ordering in thermal systems [5].
To describe the nucleation and growth of the vertical domains at high n φ shown in Fig. 3 and 4 we present a simple analysis. The rods are assumed to be either vertical or horizontal. We then assume that the growth rate is initially linear and asymptotically must decrease to zero as the number of horizontal rods is diminished and a steady state is reached. Then the evolution of n v is described by, ∂n where, α and β are constants that depend on n φ . Eq. 1 can be integrated and is fit to the measured n v (t) in Fig. 4. This simplified interpretation seems to capture the nucleation, growth, and saturation of the nearvertical domains in our experiments. In previous work, Villarruel et al. [6], the experiments were carried out in a system where the diameter of the container and the length of the rods were comparable. The container was tapped and thus the rods experienced considerable shearing with the side boundary. They observed that vertical domains nucleated at the boundaries and subsequently propagate to the center of the cell. These observations are consistent with our findings. We have shown that vertical domains can nucleate away from the sidewalls and form independent of the direction of the driving.
Next, we elucidate the physical mechanism that is responsible for the vortex motion. We performed additional experiments with a row of cylindrical rods with l = 5.1 cm and d = 0.6 cm in a 1-D annulus with a mean radius of 5.5 cm [see Fig. 10]. When subjected to vertical motion, the rods were always observed to move in the direction of the inclination. No translation motion is observed when the rods were vertical. A movie showing this property can be found at Ref. [14]. We also used ball point pens and pencils to check that the detailed shape of the rod and the tip is not important to this translation mechanism. We find that the translation speed depends on the driving frequency and the inclination, qualitatively similar to that observed for the vortices.
The physical mechanism for the motion of the inclined rods, based on our observation, appears to be as follows. When the inclined rods are vibrated vertically, they hit the bottom plate at a point away from their center of mass. Because the rotation of the rod about its center of mass is constrained due to the neighboring rods, it gets launched in the direction of the inclination. Thus, the greater the inclination of the rods, the further they get launched and land (viz. projectile motion), giving rise to translational motion.
From these direct observations we conclude that the inclined rods form the engine that drives the vortex. The vertical rods in the center of the vortex are simply pulled around by shear induced by the inclined rods. As the number of rods is increased and the size of the vortex increases, the inclination of the rods also decreases because of greater packing. Therefore, the speed of the very large vortices is expected to decrease with the size of the vortex, consistent with our observation [see Fig. 8(a)]. We also note that the frequency dependence of v(r) is compatible with the mechanism described above. By decreasing the frequency of the driving signal at fixed Γ the velocity of the driving plate is increased. Therefore, one expects v(r) to increase with decrease in frequency, consistent with our observations [see Fig. 8(b)].
VI. CONCLUSION
In conclusion, granular rods are observed to form vertically aligned domains at high packing fractions, independent of the direction of vibration. Novel vortex patterns are observed when the rods are vibrated vertically. We have measured the growth of near-vertical domains and have developed a simple void filling model that well describes our results. We have also shown a new translation mechanism which occurs due to anisotropy. Based on these observations, Aranson and Tsimring [16] have developed a phenomenological model that describes the formation and coarsening of the vortices.
|
2016-02-02T08:36:57.578Z
|
2002-03-11T00:00:00.000
|
{
"year": 2002,
"sha1": "5a0423760bbc2864f656530600659a641980c9be",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/cond-mat/0203236",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "5a0423760bbc2864f656530600659a641980c9be",
"s2fieldsofstudy": [
"Materials Science",
"Physics"
],
"extfieldsofstudy": [
"Materials Science",
"Physics",
"Medicine"
]
}
|
268486861
|
pes2o/s2orc
|
v3-fos-license
|
Prevalence of Lameness in Dairy Cows and Associated Risk Factors at Hawassa Town Dairy Farms, Ethiopia
Lameness is one of the greatest constraints on the productivity, health, and welfare of dairy cattle. A cross-sectional study was carried out from March 2021 to September 2021 in Hawassa town with the aim of assessing the prevalence and identifying the associated risk factors of lameness in dairy farms. The study was conducted on 440 animals belonging to 19 randomly selected intensive dairy farms. Data regarding lameness and its possible risk factors were collected both at animal and farm level using a questionnaire. The results showed that the overall prevalence of lameness was 10.2% (n = 45/440). The association of lameness prevalence with various risk factors including milking status, exercise, age, parity, milk yield, and lactation stage was statistically tested using logistic regression model. There was a significant variation in the prevalence of lameness (P < 0.05) between cattle with different milking status, age, parity, milk yield, and stage of lactation by the univariable analysis result. According to the multivariable analysis, only milk yield and lactation stage were statistically associated with the occurrence of lameness. Milking animals (8%) had higher prevalence of lameness than nonmilking (2.2%). The occurrence of lameness increased with milk yield. The highest prevalence of lameness was recorded in the early stage of lactation. Lameness was more frequent in hind limbs (6.6%) than in forelimbs (3.6%). The main causes of lameness observed in this study were both claw overgrowth 10 (2.3%), unequal claw size 10 (2.3%), solar ulcer 8 (1.8%), interdigital necrobacillosis 2 (0.5%), interdigital hyperplasia 2 (0.5%), and digital dermatitis 1 (0.2%). There was no means of early lameness diagnosis in 94.7% of farms. Lameness was found to be an important disease in dairy cows at Hawassa town. Prevention and early diagnosis leading to prompt treatment of lameness in cows should be part of dairy farm management practice.
Introduction
Dairy production plays a crucial role in Ethiopian livestock farming.Te country possesses abundant and diverse livestock genetic resources, coupled with a variety of agroecologies suitable for dairy farming.Te rising consumer demand for milk and its derivatives, favorable market conditions, and close proximity to international markets underscore the signifcant potential and opportunities for the development of the dairy industry in Ethiopia [1].
Despite this substantial potential, the dairy sector has not reached the anticipated level of development, and the overall productivity of dairy animals remains low [2,3].Various challenges, including issues like lameness [4][5][6][7], contribute to this situation.Consequently, there is a shortage in the supply of dairy products, necessitating the country to expend foreign currency on importing them from abroad [2].
Lameness, characterized by a departure from the typical walking pattern due to lesions, defects, injuries, diseases, or other factors afecting the limb or other parts of the body, is typically accompanied by pain or a certain level of discomfort [8].Within the dairy industry, lameness stands out as a critical and pressing issue [9].It is acknowledged as the foremost challenge impacting productivity, health, and welfare in dairy cattle.
Lameness represents a crucial production disease in dairy cows with notable economic implications [10].In general, clinical incidences of lameness have a considerable negative impact on milk production, resulting in a reduction of 357 kg over a 305-day lactation period [11].Te onset of lameness is notably prevalent in the initial two months of a cow's frst lactation, and the alarming statistic that 50% of animals experience chronic lameness during their life underscores its severity [12,13].Lameness emerges as a signifcant risk factor for culling throughout the lactation period.Cows treated for foot and leg problems at the onset and in the second month of lactation face a culling risk six to twelve times higher than that of healthy counterparts [14].Beyond economic concerns, lameness in dairy cows raises serious welfare issues by causing pain and impeding the movement of the animals [15].
Bovine lameness, ranking third globally in modern intensive dairy production following reproductive failure and mastitis, leads to reduced milk production, enhanced treatment expenses, increased culling rates, and prolonged calving intervals [25].In Ethiopia, several studies highlight the signifcant impact of lameness in dairy cattle.An economic loss study conducted in Wolaita Sodo revealed that lameness resulted in a loss of 7.33 USD (125.30ETB) per cow, attributed to decreased milk output and treatment costs as documented by Kife [26].Additionally, Sulayeman and Fromsa [4] determined a mean reduction of 1.63 litres in daily milk yield per cow in Hawassa dairy cattle.Among the lameness-positive animals, eight out of 15 were milking cows, while the remainder were nonmilking.Te average daily milk production per cow decreased from 6.36 litres to 4.75 litres after the onset of lameness, indicating a mean loss of 1.63 litres in milk yield per cow per day.
Despite lameness posing signifcant economic losses and impacting the health and well-being of dairy cattle in Ethiopia, there is a scarcity of studies addressing the prevalence of lameness and its associated risk factors, particularly in Hawassa town.Furthermore, although lameness is infuenced by various modifable management practices, the last examination of lameness prevalence and associated risk factors for dairy cattle in Hawassa occurred in 2012 [4].Terefore, the present study aims to update the prevalence and associated risk factors of lameness in dairy farms located in Hawassa town.
Study Area.
Te study was carried out in Hawassa town, which is located in Southern Ethiopia situated 270 km south of Addis Ababa.Te area has a latitude of 7 °3′ 0″ N and a longitude 38 °28′ 0″ E on the escarpment of the Great Rift Valley.Te altitude ranges from 1650 to 1700 meters above sea level.Te average annual rainfall of the study area ranges from 800-1000 mm, and the mean temperature ranges from 11.14 °C-29.1 °C.Te soil type of Hawassa town is lacustrine that is medium to fne textured and alluvial that includes clay, sand, and gravel.Te area is mainly covered by dry savanna and bush type of vegetation including mainly short grasses and shrubs and to some extent eucalyptus, oak, and other indigenous and exotic plants [27].Te total livestock population of the study area constituted 1,721,341 cattle, 228,941 goats, 457,465 sheep, 57,643 horses, 54066 donkeys, 725, 5540 poultry, and 44,492 beehives [28].
Study Animals.
Te study was conducted on 440 Holstein Friesian dairy cattle belonging to 19 farms kept under intensive management system in Hawassa town.Each animal was identifed by site of farm, age, parity, amount of milk per day, stage of lactation, and herd size using data fles from the dairy personnel.Te ages of the animals were determined primarily based on the information obtained from the animal owners and secondly by looking at the dentition pattern of the animals [29].Te foor system was concrete, and both roughage and concentrated feed were provided to the animals.All visited farms did not use bedding for their animals.All farms included in the study area have a practice of dung removal two or more times per day.
Study Design, Sampling Method, and Sample Size Determination.
A cross-sectional study was carried out from March 2021 to September 2021 in Hawassa town.According to the information obtained from the agricultural ofce of Hawassa town, the town has 157 dairy farms.A list of all 157 farms was prepared, and 19 farms were selected using a simple random sampling, lottery technique.All animals of each selected farm were included in the study.
Te sample size for the study was determined based on the description of Trusfeld [30] and considering as expected the prevalence of 50% as there was no previous study about prevalence of lameness in Hawassa town before this study, with the confdence interval of 95% and 5% required absolute precision.Ten, the minimum required sample size was calculated using the following formula: where N � sample size, P exp � expected prevalence, and d � required precision.By substituting the values in the formula and taking d � 0.05, Even though the calculated sample size was 384, the study was conducted on a total of 440 animals by adding 15% to increase the precision.
Data Collection.
A semistructured questionnaire which contained both animal and farm level questions was developed to collect data.Data regarding foor type, frequency of dung removal, and production status, animal's age, sex, lactation stage, type of feed, and site of lesion were collected.Te questionnaire was developed based on previous studies [4,31].In order to assure the quality of the data, a pretest of 2 Veterinary Medicine International data collection instrument was carried out on 5% of the total sample size outside the study area.Furthermore, all the animals in the selected farms were carefully observed and clinically examined for lameness.During the examination, the study animals were allowed to move and observed for any symptom of abnormal gait as described by Shearer et al. [32].
2.5.Data Analysis.Te data collected in the paper format was transferred to and stored in Microsoft Excel database.Stata/MP software version 16 was used for the analysis of the data.Te prevalence of lameness was presented using descriptive statistics.Logistic regression model was used to check the association of the abovementioned potential risk factors with the occurrence of lameness.Pearson's chisquare test was used to evaluate the association of diferent variables with the prevalence of lameness with treatment practice.In all statistical analysis executed, 95% confdence level and 5% precision were used and P value of less than 0.05 was considered as statistically signifcant.2).
Prevalence of Lameness with Associated Risk Factors.
Table 3 displays the outcomes of a univariable logistic regression examination regarding the occurrence of lameness in dairy cows, considering diferent risk factors.With the exception of the permission for dairy animals to engage in exercise (P > 0.25), all the factors explored in the research, milking status, age, parity, milk yield, and lactation stage were determined to be statistically signifcant (P < 0.25).All the independent variables that demonstrated signifcance in the initial univariable analysis underwent an assessment for colinearity using Kruskal gamma statistics.Variables with gamma values falling between −0.6 and +0.6 were deemed suitable for inclusion in the multivariable logistic regression model.Consequently, milk yield and lactation stage were selected for the multivariable analysis.Both variables incorporated into the multivariable model, namely, milk yield and lactation stage, exhibited signifcance (P < 0.05) (Table 4).Te Hosmer-Lemeshow goodness-of-ft test indicated that the model adequately fts the dataset (χ 2 � 0.732; P � 0.866).
Prevalence of Lameness and Limbs
Afected.We also described the proportion of lameness associated with the type of limbs.Lameness was observed due to problems from both limbs.However, hind limbs were more prone to lameness than forelimbs (Table 6).
Practice of Early Lameness Detection and Treatment.
Out of the 19 farms observed, only one farm had the practice to detect early signs of lameness.Te incidence of lameness in dairy cows is notably higher (10.59%) in farms lacking means for early recognition compared to those with such capabilities, where lameness is recorded at 7.7% (Table 7).Most of the treatments were carried out by veterinarians (16/ 19).Te prevalence of lameness in dairy farms where farmers are responsible for treatment (11.11%) surpasses that in farms where veterinarians handle treatment (10.03%) (Table 7).Te treatment success in the study area was 100%.Ten cows were culled in the last two years due to lameness.
Discussion
Te current investigation revealed a lameness prevalence of 10.2% in dairy farms located in Hawassa town.Tis study highlights a signifcant and widespread occurrence of lameness in the study area, emphasizing the need for appropriate preventive and therapeutic measures.Notably, this prevalence is higher than the fndings reported by Lobago et al. [33] who documented a lameness rate of 7.7% in clinically examined dairy cattle under urban and peri-urban production systems in the Addis Ababa milk shed.Similarly, Kife [26] reported a lower lameness prevalence of 4.0% in Wolaita Sodo.In contrast, our study's prevalence is lower than that reported by Abunna et al. [31] in Bishoftu, where lameness was recorded at 13.9%.Moreover, the prevalence of lameness found in the present study is lower than that reported in other countries, such as 36.8% in England and Wales [34] and 28.5% in Canada [35].Te variations in lameness prevalence among our study and those conducted in diferent countries could be attributed to diferences in the management system, climate, study duration, cow productivity, and the methods employed for lameness detection and prevention.Geographical disparities and seasonal fuctuations in the Veterinary Medicine International incidence and prevalence of lameness are also evident, as indicated by Espejo et al. [36].
In the current study, the lameness prevalence varied among the farms, ranging from 0% to 33.3%, and there was a statistically signifcant correlation between the prevalence of lameness and the examined farms (p < 0.05).Te difference in the prevalence of lameness between the farms might be due to the diferences in management system and awareness of the negative impact of lameness.
Tis study examined the risk factors associated with lameness, considering both milking and nonmilking cows.Te prevalence of lameness in milking cows was 35 (8%), while, in nonmilking cows without pregnancy, it was 10 (2.2%).Te higher prevalence in milking cows is likely linked to the mobilization of fat from various tissues to support milk production, as suggested by Green et al. [37].Consequently, the hypothesis was formulated that elevated milk yield may result in thinner digital cushions, exposing cows to conditions such as sole ulcers and white line disease [38].Te present study strengthened this hypothesis by evidencing that milk yield is signifcantly associated with the occurrence of lameness.
A higher prevalence of lameness, specifcally 10.2%, was recorded in animals aged greater than two years compared to animals aged less than two years, where the prevalence was 0%.Te fndings of our study align with those of Manske et al. [39] who observed an increase in lameness with advancing age.
Te occurrence of lameness in this study was higher in animals that yield more than 16 litters of milk per day (44.4%)compared to those produce less than 16 litters, and these might also be due to the loss of minerals such as calcium that have a great role in the strength of bone in animals.
In this study, the occurrence of lameness and the limbs afected was signifcantly associated, indicating that lameness was most common in hind limbs than in forelimbs possibly due to the fact that the hind legs were often contaminated with manure and kept wet.Hedges [40] also reported that on average, approximately 80% of lame cows are lame in the hind limbs.Singh et al. [41] also reached to similar fndings from Punjab where the distribution of lameness in cattle was 28.9% in forefoot, 54.7 in hind feet, and 16.3% in both fore and hind feet.Te same authors have 4 Veterinary Medicine International also recorded more frequent foot abnormalities in the hind (80%) than in the fore (20%) feet in bufaloes.Sadiq et al. [42] suggested that there was no signifcant link between the prevalence of lameness and parity.In contrast, our study demonstrated a noteworthy connection between lameness occurrence and parity.Te higher prevalence in animals with more than two parities is believed to be attributed to their prolonged exposure to uncomfortable barn conditions and the absence of early lameness detection methods on the farm.
Van Amstel and Shearer [43] asserted that confnement on hard surfaces alone is sufcient to induce a mechanical form of laminitis, leading to subsequent claw overloading.Similarly, Barker et al. [34] found that housing dairy cows for 61 days or more were a signifcant risk factor associated with lameness prevalence in dairy herds in England and Wales.In contrast, Bicalho et al. [38] highlighted that lameness in dairy cows can manifest at any point during lactation, similar to many other diseases.
According to the present study, lesions that were found causing lameness were sole ulceration 8 (1.8%), digital dermatitis 1 (0.2%), claw overgrowth 10 (2.3%), unequal size claw 10 (2.3%), and interdigital hyperplasia and interdigital necrobacillosis 2 (0.5%).Highly prevalent lesions causing lameness in this study were unequal claw size and claw growth, and this might be due to lack of exercise and poor practice of hoof trimming.
Conclusion and Recommendations
Te present study indicated a high and wide distribution of lameness that varied among the farms.High occurrence of lameness was recorded in cows with milking status, early lactation period, increased parity, and high milk yield.Te present study showed that hind limbs of dairy cattle are more prone to foot lesions than the forelimbs.Poor means of recognizing early cases of lameness in the farms were another fnding of the study.Terefore, providing awareness to farmers on the risk factors of lameness and management systems of dairy cattle is crucial to minimize lameness which is a serious welfare and economically important disease of dairy cows.
Table 1 :
Gender and educational level of farm workers.
Table 2 :
Te prevalence of lameness in the individual examined farms.
Table 3 :
Univariable logistic regression analysis of risk factors for the occurrence of lameness in Hawassa city dairy farms.
Table 4 :
Multivariable logistic regression analysis of potential risk factors for the occurrence of lameness in Hawassa city dairy farms.
Table 5 :
Lesions that caused lameness in dairy cows in Hawassa town.
Table 6 :
Te prevalence of lameness and limbs afected in dairy cows in Hawassa town.
|
2024-03-17T17:22:45.948Z
|
2024-03-11T00:00:00.000
|
{
"year": 2024,
"sha1": "625ce4cbf68c0e27028c552bd786a56e55891a51",
"oa_license": "CCBY",
"oa_url": "https://downloads.hindawi.com/journals/vmi/2024/2732333.pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "74abd2aefd235839b0546c0a3cbbcedd99e3d336",
"s2fieldsofstudy": [
"Agricultural and Food Sciences"
],
"extfieldsofstudy": []
}
|
247308457
|
pes2o/s2orc
|
v3-fos-license
|
Aptamer fluorescence anisotropy assays for detection of aflatoxin B1 and adenosine triphosphate using antibody to amplify signal change
Fluorescence polarization/anisotropy (FP/FA) is an attractive technology for determining small molecules in homogeneous solution based on rotation changes of a fluorescent reporter. Binding induced conformation change is a specific property of aptamers. This property has been integrated into aptamer based FA assays for small molecules. In this work, we reported aptamer FA assays for aflatoxin B1 (AFB1) and adenosine triphosphate (ATP) by using antibody conjugated complementary DNA at the 3′ end and a fluorescein (FAM)-labeled aptamer at the 5′ end. The hybridization of aptamer and cDNA induced a FAM label close to the large-sized antibody, which restricts the local rotation of FAM and gives high FA signal. With the addition of target, the aptamer probe binds with the target, and the aptamer–cDNA duplex is inhibited, causing FA signal decreases. This method achieved detection of 25 pM AFB1 and 1 μM ATP, respectively. The assay is promising for application.
Introduction
Fluorescence polarization/anisotropy (FP/FA) assays for small molecules detection have wide applications in drug discovery, clinical diagnosis, food safety, environmental sensing, and molecular interaction study. [1][2][3][4] FA analysis only requires mixing a probe and target with other reagents in homogeneous solution. As a ratiometric technique, FP/FA technique is insensitive to small uctuations in instrument variance and uorescence intensity, having high reproducibility. FP/FA assay is suitable for highthroughput analysis. Conventional immune-FP assays use antibody and dye-labeled small molecules, and they show some limitations in preparing the uorophore-labeled small molecules. 3 Aptamers show advantages in facile synthesis, easy labeling, good thermo-stability and low cost. 5,6 Aptamers can be selected for various targets, including cells, proteins, small molecules and metals. 7 Taking advantages of aptamer and FP/FA technology, a variety of aptamer-based FP/FA assays have been developed for small molecules detection. [8][9][10] As nucleic acid ligands, DNA aptamers can hybridize with its complementary DNA (cDNA). Aptamer structure change can occur due to the competition between aptamer-target binding and aptamer-cDNA hybridization. [11][12][13] This attractive property of aptamer can be applied to develop FA assays by using dye labeled aptamer or cDNA. [8][9][10]14,15 To improve FA signal changes, several signal amplication strategies for small molecule detection have been reported, such as conjugating nanoparticles or biomolecules on DNA. [16][17][18][19][20] In this work, we described an aptamer FA assay for detection of small molecules by using uorescein (FAM)-labeled aptamer at 5 0 end as uorescent probe and antibody conjugated cDNA at 3 0 terminal as the signal amplier. We took aatoxin B1, a most toxic mycotoxin causing great health risk to health, [21][22][23] and ATP, an important small molecule with many functions in organisms, as two examples of small molecules. FAM-labeled aptamer binds with antibody conjugated cDNA, and a duplex is formed, drawing FAM label into proximity with the large-sized antibody (about 150 000 g mol À1 in molecular weight) and giving a high FA signal due to the restriction of local rotation. In the presence of small molecule, aptamer binds with target, and the aptamer-cDNA duplex assembly is inhibited. FA signal signicantly decreases with addition of small molecules as the FAM-labeled aptamer shows low FA in aptamer-target complex due to the small size of aptamer and rapid local rotation of FAM. This assay enabled detection of 25 pM AFB1 and 1 mM ATP, respectively.
Aptamer FA detection of targets
First, the digoxin labeled complementary DNA and antibody specic for digoxin were incubated in the binding buffer with a ratio of 1 : 1 at 4 C for 15 min to achieve antibody-conjugated cDNA by the immunoreaction between antibody and digoxin. In the assay for AFB1 detection, 1 nM Apt AFB -5 0 FAM and 20 nM antibody conjugated C14 AFB for AFB1 were incubated in buffer A for 30 min at 4 C. In the assay for ATP, 5 nM Apt ATP -5 0 FAM and 20 nM antibody conjugated C11 ATP for ATP were incubated in buffer B for 10 min at 25 C. Aer that, 100 mL of sample solution was added to the wells of the black 96-well microplates (USA, Thermosher). Duplicate samples were tested. For each sample, FA signals were measured three times by a plate reader Synergy H1 (U.S.A., Biotek), with an excitation at 485 nm and an emission at 528 nm. The average FA values were used. For the three measurements of the same sample, the relative standard deviation (RSD) ranged from 0.4% to 5%. For measurement of duplicate samples, the RSD was less than 3%.
Detection of targets in complex sample matrix
For AFB1 detection, diluted grape juice, white wine and tap water samples were used as complex sample matrix. Grapes without the peel and nucleus were prepared for grape juice through juice extractor. Aer that, grape juice sample was centrifuged at 12 000 rpm for 10 min. Then the collected supernatant was subsequently ltered with 0.22 mm membrane and ultraltration membrane (Millipore, 3 kDa cut-off) at 12 000 rpm for 15 min. White wine samples was ltered with 0.22 mm membrane. At last, 20-fold diluted grape juice, 10-fold diluted white wine and 10-fold diluted tap water was obtained by diluting with the binding buffer, used as complex sample matrix. For ATP detection, 50-fold diluted human serum and 20fold diluted urine samples were used as complex sample matrix. Before dilution, human serum samples were centrifuged at 12 000 rpm for 15 min, and the supernatant was ltered with 0.22 mm membrane. Urine samples were from volunteer in the lab, and ltered with 0.22 mm membrane. The sample preparation and determination were the same to the FA assay procedure described above. Informed consents were obtained from human participants of this study.
Results and discussion
3.1. Principle of aptamer FA assay for small molecule detection using adjacent antibody to amplify signal change The principle of the proposed aptamer FA assay using adjacent large-sized antibody as a signal amplier is shown in Scheme 1. In this design, FAM-labeled aptamer is used as uorescent probe. Antibody conjugated complementary sequence is employed as the signal amplier. The antibody conjugated cDNA is obtained by incubating digoxin (Dig) labeled cDNA with anti-digoxin antibody. The hybridization of aptamer and cDNA causes FAM label close to the antibody. The proximity between the large-sized antibody and the FAM label signicantly restricts the local rotation of FAM and produces a much higher FA value. The addition of small molecule target in sample solution results in the formation of aptamer and target complex, preventing the assembly of duplex of aptamer probe and cDNA. The FAM label in the aptamer probe shows a lower FA signal. Thus, we can achieve the detection of small molecule by measuring the remarkable FA change upon addition of target.
Aptamer FA assay for AFB1
Aatoxins (AFs) are natural mycotoxins that are produced by Aspergillus parasiticus and Aspergillus avus. [21][22][23] Many food sources (such as rice, wheat, corn, peanuts and fruits) are easily polluted by AFs. In different types of AFs (including AFB1, AFB2, AFM1, AFM2, AFG1 and AFG2), AFB1 shows most toxic and has been classied as a primary carcinogenic compound by International Agency for Research in Cancer (IARC). 22 Developing sensitive and simple method for AFB1 and its analogues is important to ensure food safety and human health.
Firstly, we tested the feasibility of our design. FA value of 1 nM 5 0 FAM-labeled aptamer was 0.116 (Fig. 1a). The antibody-conjugated cDNA hybridized with aptamer, resulting in the proximity of antibody and FAM and the restriction of local rotation of FAM. Thus, the FA response increased to 0.253 (Fig. 1b). When antibody was distant from the FAM label (Apt AFB -3 0 FAM + antibody-conjugated cDNA, Fig. 1c). The DNA hybridization only caused slight FA increase (less than 0.020) due to the fast local rotation of FAM though antibody (about 150 000 g mol À1 in molecular weight) is much larger than the aptamer probe in size. 20 With the addition of AFB1, AFB1 bound to aptamer probe, and the cDNA did not hybridize with the 5 0 FAM-labeled aptamer probe. Thus, FAM label rotated fast and showed low FA value (0.133, Fig. 2) because the bound AFB1 was small in size. These results demonstrated that our design is feasible for detecting AFB1.
Then, we investigated some important factors for AFB1 detection, such as the concentrations of antibody-conjugated C14 AFB and incubation time. As shown in Fig. 2, the FA value of FAM label in formed duplex increased with increasing of antibody-conjugated cDNA. In the presence of 10 nM AFB1, FA value showed slight increase with increasing of antibodyconjugated cDNA. The maximum reduction in FA caused by 10 nM AFB1 increased from 0.021 to 0.120. Further increase of antibody-conjugated cDNA required more antibodies. Thus, we chose 20 nM antibody-conjugated cDNA for further experiments.
Incubation time shows great effect on aptamer and cDNA hybridization. As shown in Fig. 3, FA value of FAM label in duplex increased with the increase of incubation time from 10 min to 30 min. FA value did not increase with further increase of incubation time. It shows the incubation for 30 min is enough for hybridization between aptamer probe and the antibody conjugated cDNA. In the presence of 10 nM AFB1, similar FA signals were observed in different incubation time. Thus, FA measurement was conducted aer 30 min incubation.
Under optimal conditions, we detected AFB1 by using 1 nM aptamer and 20 nM antibody-labeled cDNA. As shown in Fig. 4, the FA value of FAM label gradually decreased with the increase of AFB1. This method allowed to detect as low as 25 pM AFB1 ($0.008 mg kg À1 ) (determined by S/N > 3), with dynamic range from 25 pM to 100 nM. As comparison, the FP (uorescence polarization) signals were also recorded (Fig. 4B). We compared our method with some FP/FA assays for AFB1 detection in terms of affinity ligand, maximum signal change and detection limit (Table S2 †). 20,[24][25][26][27] As shown in Table S2 in ESI, † our assay not only shows larger FA signal change than that in traditional immune-FA assay, but also has lower detection limit than some reported immune-FA assays and aptamer-based FP/FA assays. [8][9][10]20,24,26,27 Then, we tested the selectivity of our proposed assay for AFB1. FA decreased when AFB1 existed in the sample buffer. The FA of duplex in the presence of 200 nM other mycotoxins such as ochratoxin A (OTA), fumonisin B1 (FB1), fumonisin B2 (FB2) and zearalenone (ZAE), was similar to that of blank sample (Fig. S1 in ESI †). Simultaneously existing of other mycotoxins (each of 200 nM) showed little effect on the FA value of FAM label in the presence of AFB1 (10 nM). These results suggest that our proposed assay is highly selective for AFB1 detection.
Our method shows different FA decreases upon other aatoxin analogues (such as AFB2, AFM1, AFM2, AFG1 and AFG2) by using 1 nM FAM-labeled aptamer and 20 nM antibody conjugated C14 AFB (Fig. S2 †), which are in agreement with the previous report due to the different affinities between aptamer and these aatoxins. 24 Then, we used the concentrations of analytes corresponding to 50% FA change (EC50) to evaluate the cross-reactivity (CR) of AFB1 and other AFB1 analogues with this method. The respective EC50 values for AFB1, AFB2, AFM1, AFM2, AFG1 and AFG2 were 1.3 nM, 1.6 nM, 4.2 nM, 4.1 nM, 16 nM and 22 nM, respectively (Table S3 †). The percentage of CR was obtained by dividing the EC50 of AFB1 with the EC50 of respective aatoxin species. The cross-reactivity of AFB1, AFB2, AFM1, AFM2, AFG1 and AFG2 were estimated to be 100%, 81%, 32%, 32%, 8% and 6%, respectively (Table S3 †). Our method enabled to detect 25 pM AFB1, 25 pM AFB2, 0.39 nM AFM1, 0.39 nM AFM2, 1.6 nM AFG1 and 1.6 nM AFG2, respectively. It shows the aptamer FA method can be used to detect other aatoxin species. This method shows potential for the detection of total quantity of these aatoxins (AFB1, AFB2, AFM1, AFM2, AFG1 and AFG2) in some assays.
To evaluate the performance of the aptamer FA assay in complex sample matrix, we tested AFB1 in diluted grape juice, white wine and tap water samples (Fig. S3 †). In the tested sample matrix, FA value toward to different concentrations of AFB1 was similar to that in binding buffer. Our method enabled detection of 50 pM (0.016 mg kg À1 ) AFB1 in complex sample matrix. Our method meets the requirement of the maximum regulation levels of AFB1 setting by European Union (2 mg kg À1 ) and World Health Organization (WHO, 5 mg kg À1 ) allowed in food products. 21 These results demonstrate that our proposed assay is feasible for AFB1 detection in complex matrix, and it has potential for real sample analysis.
Aptamer FA detection of ATP
Adenosine triphosphate (ATP), an important small molecule participating in many activities in living organism was chosen for the universal test of our design. 28,29 A FAM-labeled aptamer at the 5 0 end, Apt ATP -5 0 FAM (5 nM; 5 0 -CCT GGG GGA GTA TTG CGG AGG AAG G-3 0 ), an antibody-conjugated cDNA at the 3 0 terminal C11 ATP (20 nM; 5 0 -ACT CCC CCA GG-3 0 ) (Fig. S4 †), and 10 min incubation of sample solution at 25 C (Fig. S5 †) were chosen for ATP detection to obtain relatively better FA responses. Under optimal conditions, we successfully achieved detection of ATP ( Fig. 5 and S6 †). The maximum FA change was 0.06 (the maximum FP change was 0.08). The dynamic detection range was from 1 mM to 1 mM. The detection limit of ATP was 1 mM, which is comparable to that in some previous aptamer-based FP/FA assays for ATP (Table S2 †). [8][9][10]18,[30][31][32][33] The higher detection limits of our FA assays for ATP (1 mM) than that of our FA assay for AFB1 (25 pM) can be attributed to the lower affinity of aptamer against ATP 31 and high affinity of aptamer against AFB1. 24 This method was selective for ATP detection (Fig. S7 †). This strategy also enabled detection of ATP spiked in diluted human serum and urine with a detection limit at 1 mM (Fig. S8 †).
Conclusions
In this study, a simple and sensitive aptamer FA assay was developed by using large-sized antibody to amplify signal change. We used FAM-labeled aptamer and antibodyconjugated cDNA, which induced the proximity effect between FAM and antibody in duplex. The FA assays allowed for detection of 25 pM AFB1 and 1 mM ATP, showing the method is suitable for aptamers with either high affinity or weak affinity. Moreover, the FA assays were successfully applied to rapid detection of AFB1 and ATP in complex sample matrixes, showing potential in applications.
Conflicts of interest
There are no conicts to declare.
|
2022-03-09T16:28:58.672Z
|
2022-03-01T00:00:00.000
|
{
"year": 2022,
"sha1": "a71eb2e166dab3fa964981bc4030c1d45a47bebc",
"oa_license": "CCBYNC",
"oa_url": "https://pubs.rsc.org/en/content/articlepdf/2022/ra/d2ra00843b",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "06a9bb3f28021055aef49bc01dc14c3faadecb02",
"s2fieldsofstudy": [
"Chemistry"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
258370970
|
pes2o/s2orc
|
v3-fos-license
|
Genetic Variability of HUPRA Syndrome—A Case Report
: HUPRA syndrome is a rare autosomal recessive mitochondrial disorder caused by a mutation in the SARS2 gene encoding mitochondrial seryl-tRNA synthetase (mtSerRS). It includes hyperuricemia, pulmonary hypertension, renal failure, and alkalosis. We present a case report of a boy aged 1 year 2 months with premature anemia, hyperuricemia, pulmonary hypertension, renal failure, and alkalosis and diagnosed with HUPRA syndrome. This disease is known to be progressive and fatal. A genetic test revealed a new previously undescribed heterozygous nucleotide variant in exons 14 and 1 of the SARS2 gene. The nucleotide substitution c.1295G > A (p.Arg432His) was detected in exon 14; according to the criteria of the American College of Medical Genetics (ACMG), this missense mutation is probably pathogenic. The nucleotide substitution c.227T > C (p.Leu76Pro) was detected in exon 1; according to the ACMG criteria, this missense mutation is a variant of unclear significance. We suggest that previously undescribed nucleotide substitutions in the SARS2 gene revealed in a patient with typical clinical presentation of the HUPRA syndrome should be considered as a pathogenic mutation.
Introduction
Hyperuricemia, pulmonary hypertension, renal failure, and alkalosis are symptoms of HUPRA syndrome (OMIM#613845), which is an orphan disease with an autosomal recessive inheritance, with a progressive course and leading to death at an early age. The disease was found to be caused by a mutation in the SARS2 nuclear gene encoding mitochondrial seryl-tRNA synthetase (mtSerRS) and has been known since 2011 [1]. In this article, we present a case report of a child with HUPRA syndrome with a new heterozygous mutation of the SARS2 gene.
A Case Report
A baby-boy patient (1 year 2 months old of non-consanguineous parents) was born at 36 weeks of gestation from the fifth pregnancy (the 1st and 2nd children in the family are healthy, the 3rd pregnancy ended with miscarriage at 7 weeks of gestation, and the 4th pregnancy miscarried at 22 weeks). The mother was diagnosed with thrombophilia. The boy was breast-fed until 5 months, when he presented with enterocolitis leading to poor weight gain and maldigestion and turned to bottle-feeding with a hydrolytic formula. Despite adequate food intake from the age of 7 months, the patient stopped gaining weight. At 7 months, he was also diagnosed with anemia (Hb 88 g/L [N, 110-140 g/L]), regarded as anemia of prematurity. Lab tests showed a normal serum iron level. Due to anemia, the child received erythropoietin, but this had an insufficient effect. Blood tests revealed a progressive increase in urea and creatinine, aspartate aminotransferase As the cause of kidney disease remained unknown, the boy was admitted to the Nephrology Department of the Russian Children's Clinical Hospital. Upon admission, the child had a severe condition due to a low body weight (7.5 kg, <3 percentile), while height was 74 cm (25 percentile). Arterial blood pressure remained within the normal range of 97/63 mm Hg. The child showed polyuria 800 to 900 mL per day. Table 1 shows lab tests results. Initially, the correction of electrolyte disturbances was performed. We conducted a differential diagnosis between the primary Bartter syndrome (hyponatremia, hypokalemia, and hypochloremic metabolic alkalosis), Gitelman syndrome (the presence of hypomagnesaemia), and pseudo-Bartter syndrome. We excluded the primary forms of Bartter and Gitelman syndrome due to the presence of cytopenic syndrome as well as severe hyperuricemia and early azotemia, while there were no signs of nephrocalcinosis in ultrasound examination.
According to echocardiography, the severe dilatation of the right side of the heart and signs of pulmonary hypertension were found. Cytopenic syndrome, hyperuricemia, and severe weight loss were considered as the course of an oncohematological disorder with a secondary kidney damage. To verify the diagnosis the child underwent a bone marrow puncture, which revealed evidence of dyserythropoiesis and dysmegakaryocytopoiesis. A renal biopsy was performed to determine the cause of kidney damage. After the biopsy, the child developed renal bleeding, which could not be controlled with hemostatic plasma therapy and required left-side nephrectomy. As a result of severe blood loss and secondary disseminated intravascular coagulation (DIC), the child developed multiple organ failure syndrome and died.
Renal tissue morphological examination revealed signs of focal segmental glomerulosclerosis; immature glomeruli, with large closely spaced podocytes; and dysmorphic glomeruli, including glomeruli with multiplication. Tubule changes such as foci of fibrosis and atrophy occupied about 20% of the microslide. There were signs of nephrocalcinosis (phosphate and calcium oxalate crystals) ( Figure 1). Kidney Dial. 2023, 3, FOR PEER REVIEW 3 Ultrasound examination revealed hyperechogeneous kidney parenchyma. Initially, the correction of electrolyte disturbances was performed. We conducted a differential diagnosis between the primary Bartter syndrome (hyponatremia, hypokalemia, and hypochloremic metabolic alkalosis), Gitelman syndrome (the presence of hypomagnesaemia), and pseudo-Bartter syndrome. We excluded the primary forms of Bartter and Gitelman syndrome due to the presence of cytopenic syndrome as well as severe hyperuricemia and early azotemia, while there were no signs of nephrocalcinosis in ultrasound examination.
According to echocardiography, the severe dilatation of the right side of the heart and signs of pulmonary hypertension were found. Cytopenic syndrome, hyperuricemia, and severe weight loss were considered as the course of an oncohematological disorder with a secondary kidney damage. To verify the diagnosis the child underwent a bone marrow puncture, which revealed evidence of dyserythropoiesis and dysmegakaryocytopoiesis. A renal biopsy was performed to determine the cause of kidney damage. After the biopsy, the child developed renal bleeding, which could not be controlled with hemostatic plasma therapy and required left-side nephrectomy. As a result of severe blood loss and secondary disseminated intravascular coagulation (DIC), the child developed multiple organ failure syndrome and died.
Renal tissue morphological examination revealed signs of focal segmental glomerulosclerosis; immature glomeruli, with large closely spaced podocytes; and dysmorphic glomeruli, including glomeruli with multiplication. Tubule changes such as foci of fibrosis and atrophy occupied about 20% of the microslide. There were signs of nephrocalcinosis (phosphate and calcium oxalate crystals) ( Figure 1). No arteriole changes were found. The immunofluorescent assay was negative. Electronic microscopic examination revealed flattening (fusion) of small processes of podocytes and accumulations of malformed mitochondria in tubulocytes. Mitochondria were found with an enlightened matrix and absent or single cristae ( Figure 2).
Post mortem histological examination detected fibrosis, thickening of the walls of large arteries, and interstitial focal myocardial fibrosis.
Given that the child presented with such symptoms as hyperuricemia, metabolic hypochloremic alkalosis, early onset of renal failure, and signs of pulmonary hypertension, HUPRA syndrome was suspected.
To confirm the diagnosis, a genetic examination was carried out. Genomic DNA was extracted from blood samples with the use of QiaAMP DNA-mini kit (Qiagen, Valencia, CA, USA), following the manufacturer's protocol.
Whole Post mortem histological examination detected fibrosis, thickening of the walls of large arteries, and interstitial focal myocardial fibrosis.
Given that the child presented with such symptoms as hyperuricemia, metabolic hypochloremic alkalosis, early onset of renal failure, and signs of pulmonary hypertension, HUPRA syndrome was suspected.
To confirm the diagnosis, a genetic examination was carried out. Genomic DNA was extracted from blood samples with the use of QiaAMP DNAmini kit (Qiagen, Valencia, United States of America), following the manufacturer's protocol.
Whole-exome sequencing of the patient's DNA was performed with the TruSeq DNA PCR-Free sample preparation kit on NovaSeq 6000 (Illumina, San Diego, CA, USA).
The SARS2 variants, which were revealed by massive parallel sequencing, were validated by the Sanger sequencing in both parents, confirming the biallelic state.
Whole-exome sequencing analysis revealed two heterozygous variants in the The SARS2 variants, which were revealed by massive parallel sequencing, were validated by the Sanger sequencing in both parents, confirming the biallelic state.
According to ACMG criteria, variant c.227T > C (p.Leu76Pro) is supposed to be a variant of uncertain significance. The variant is absent in the gnomAD database. ReMM (Regulatory Mendelian Mutation resource [2,3]) predicts a high pathogenicity score for this variant. The second mutant allele contains c.1295G > A (p.Arg432His), which has very low minor allele frequency in gnomAD (2 heterozygous carriers among 64,553 samples) and is likely pathogenic according to ACMG. exon 14 (Figure 3). Both variants were not described earlier in the literature or in the HGMD, ClinVar, or LOVD databases. According to ACMG criteria, variant c.227T > C (p.Leu76Pro) is supposed to be a variant of uncertain significance. The variant is absent in the gnomAD database. ReMM (Regulatory Mendelian Mutation resource [2,3]) predicts a high pathogenicity score for this variant. The second mutant allele contains c.1295G > A (p.Arg432His), which has very low minor allele frequency in gnomAD (2 heterozygous carriers among 64,553 samples) and is likely pathogenic according to ACMG.
Discussion
Mitochondrial diseases are rare conditions caused by defects in mitochondria, leading to impaired cell respiration. Nervous system damage is present in most cases of diseases, and kidney damage is also diagnosed in some forms of this condition [4]. Both kidney parenchyma and neural tissue require high energy for adequate function. The proximal tubule, the ascending part of the Henle loop, is at risk of damage due to being rich in mitochondria requiring high adenosine triphosphate (ATP) levels for proper tubule transport system function [5]. This explains the fact that tubular dysfunctions of different varieties are quite often observed in mitochondrial disorders. Fanconi syndrome is the most common tubular dysfunction in children with mitochondrial diseases such as Kearns-Sayre syndrome, Pearson syndrome, Leigh encephalopathy, and CoQ10 deficiency [6][7][8][9][10][11][12][13]. Some patients with mitochondrial disorders were reported to have Barterlike phenotypes [14,15] or isolated hypermagnesuria [6].
The organelle function is controlled by nuclear and mitochondrial DNA, and mutations in nuclear genes account for 75-95% of all mitochondrial diseases [16]. A large number of nuclear-encoded factors, s.a. aminoacyl-tRNA synthetases (ARS), are required for the tRNA function to provide mitochondrial protein synthesis [17]. Each ARS is highly specific for amino acid matching and must match with the appropriate tRNA [18]. Belostotsky et al. (2011) revealed a mutation in the nuclear SARS2 gene encoding seryl-tRNA synthetase, which is charging tRNA with aminoacylated serine [1,19,20] in HUPRA syndrome. In translation, incoming tRNAs provide the serine required for mitochondrial protein synthesis. Mitochondrial proteins contribute to the formation of the mitochondrial respiratory chain complex [1]. Thus, due to the SARS2 gene mutation, this process is impaired. However, in most tissues, residual SARS2 activity is able to maintain a sufficient
Discussion
Mitochondrial diseases are rare conditions caused by defects in mitochondria, leading to impaired cell respiration. Nervous system damage is present in most cases of diseases, and kidney damage is also diagnosed in some forms of this condition [4]. Both kidney parenchyma and neural tissue require high energy for adequate function. The proximal tubule, the ascending part of the Henle loop, is at risk of damage due to being rich in mitochondria requiring high adenosine triphosphate (ATP) levels for proper tubule transport system function [5]. This explains the fact that tubular dysfunctions of different varieties are quite often observed in mitochondrial disorders. Fanconi syndrome is the most common tubular dysfunction in children with mitochondrial diseases such as Kearns-Sayre syndrome, Pearson syndrome, Leigh encephalopathy, and CoQ10 deficiency [6][7][8][9][10][11][12][13]. Some patients with mitochondrial disorders were reported to have Barter-like phenotypes [14,15] or isolated hypermagnesuria [6].
The organelle function is controlled by nuclear and mitochondrial DNA, and mutations in nuclear genes account for 75-95% of all mitochondrial diseases [16]. A large number of nuclear-encoded factors, s.a. aminoacyl-tRNA synthetases (ARS), are required for the tRNA function to provide mitochondrial protein synthesis [17]. Each ARS is highly specific for amino acid matching and must match with the appropriate tRNA [18]. Belostotsky et al.
(2011) revealed a mutation in the nuclear SARS2 gene encoding seryl-tRNA synthetase, which is charging tRNA with aminoacylated serine [1,19,20] in HUPRA syndrome. In translation, incoming tRNAs provide the serine required for mitochondrial protein synthesis. Mitochondrial proteins contribute to the formation of the mitochondrial respiratory chain complex [1]. Thus, due to the SARS2 gene mutation, this process is impaired. However, in most tissues, residual SARS2 activity is able to maintain a sufficient energy level, and the symptoms of the disease are observed only in the most "sensitive to lack of energy" tissues [1,21]. Various diseases are associated with mutations in different types of ARS genes, and most patients suffer from neurological disorders [22]. In contrast, initially missense mutations of the SARS2 gene leading to the HUPRA syndrome were described and the kidneys were affected, resulting in metabolic alkalosis and early onset of renal failure [1,23]. However, a new SARS2 gene homozygous mutation of the splicing leading to progressive spastic tetraparesis without kidney damage was described later [24]. This fact once again indicates that the clinical aspects of mitochondrial diseases can be quite diverse. HUPRA syndrome is an extremely rare disease described in only 7 children [1,23,25,26]. The additional signs of the disease are premature birth (from 27 up to 37 weeks of gestation), developmental delay, anemia with insufficient response to erythropoietin treatment, hyponatremia, hypochloremia, hypomagnesemia, and a minimal increase in blood lactic acid. Despite progressive renal failure, metabolic alkalosis was detected. An uncommon imbalance between creatinine and urea levels was also present: urea and uric acid concentrations were significantly higher with a lower creatinine level. In this case, we observed similar metabolic disturbances. However, the noticeable difference was a persistent increase in LDH and AST serum levels that were not described in other papers. Thrombocytopenia and leukocytopenia in this clinical case were previously described in only two patients [1,26]. However, our patient did not develop diabetes mellitus, which was present in all three children in the Belostotsky et al. article (2011) [1].
An important difference between this particular clinical case and three patients described previously (underwent kidney biopsy) is the absence of the glomeruli changes [1,23] as this patient appeared to have signs of focal segmental glomerulosclerosis, as well as immature and dysmorphic glomeruli. However, tubular changes such as fibrosis and atrophy and the presence of enlarged mitochondria in tubule cells were observed both in this case and in Belostotsky's paper. Additionally, in this case the patient showed no arteriole changes or positive immunofluorescent assay [1,23].
It should be noted that several studies reported the development of focal segmental glomerulosclerosis in patients with Bartter and Gitelman syndrome [27][28][29][30]. This allows us to assume that the glomerular disorders that we have identified may be the outcome of a general pathomorphological process, which is common to these syndromes. At the post mortem histological examination, fibrosis and thickening of the walls of large arteries and myocardial fibrosis were confirmed. There were also signs of pulmonary hypertension, apparently developed after renal bleeding resulting in DIC. Prior to that, there were no signs of pulmonary hypertension requiring any medications.
The cause of prolonged kidney hemorrhage after biopsy remained unclear. Despite procoagulant drug administration, plasma transfusion, nephrectomy, and surgical hemostasis bleeding were not controlled for 72 h. The kidney hemorrhage resulted in DIC. In comparison with the previously described patient histories, only Rivera et al. (2013) reported the pulmonary hemorrhage as the cause of death in one patient [23]. In other cases, there were no coagulation disorders.
Most children with HUPRA syndrome died; life expectancy ranged from 10 to 70 months [25]. Belostotsky ). Thus, this phenotype was regarded as less severe: the girl did not show pulmonary hypertension and metabolic alkalosis. The main symptom and leading cause of death was renal failure, and she died at the age of 70 months due to uremic intoxication [25]. Göknar et al. (2022) revealed the homozygous mutation c.515A > G [p.Asn172Ser] in a girl who was 4 years 4 months old and was alive at presentation. She showed all symptoms of HUPRA syndrome, except for metabolic alkalosis. Furthermore, the patient showed severe immunodeficiency, Chiari type 1 anomaly, and Graves' disease. It is still unclear whether these conditions were associated with observed homozygous mutation or were random [26].
In this case, the child died at the age of 14 months, but the main cause of death was DIC due to kidney hemorrhage. At this age, he did not show severe pulmonary hypertension but reached end-stage kidney disease. In this child, heterozygous missense mutations c.1295G > A (p.Arg432His)/c.227T > C(p.Leu76Pro) still played an important role in the severity of the disease.
In HUPRA syndrome, the many causes of organ damage remain unclear. Pulmonary hypertension appears to be associated with the development of cardiomyopathy. Hyperuricemia is explained by the impaired fractional excretion of uric acid by the affected tubule cells or as a result of chronic kidney disease (CKD) [1], as well as increased uric acid production in ischemic cardiac tissue [31]. Our patient did not have severe cardiomyopathy, but there was a persistent increase in LDH and AST serum levels from infancy.
The main hypothesis is that the residual activity of the mutated SARS2 gene is the main cause of the symptoms. It is sufficient to maintain the function of the electron transport chain in most tissues, but the thick ascending part of the Henle loop tubule cells requires much more ATP to maintain active transport system functioning [1]. So, this tissue is affected more often, and the seryl-tRNA synthetase may have additional yet unexplored functions that affect certain types of cells at different stages of their differentiation [23].
In general, the clinical presentation of HUPRA syndrome is highly variable. It is debatable whether the severity of the disease depends on the homo-or heterozygous type of mutation. It is necessary to accumulate as much information about HUPRA syndrome as possible since this condition can be easily overlooked. Due to very few cases of the disease, each clinical case presentation may play an important role in diagnostics and treatment.
Conclusions
We presented a case report on an extremely rare HUPRA syndrome with new genetic nucleotide substitutions in the SARS2 gene, which demonstrates its genetic variability in different populations. We believe that most nephrologists and pediatricians should be familiar with HUPRA syndrome and consider it as one of the possible causes of renal failure in children.
Author Contributions: E.P. collected and analyzed the patient's clinical data and compiled and edited the manuscript. M.M. reviewed the literature and wrote the manuscript. B.K. and P.P. conducted a morphological study. P.T. and E.Z. conducted a molecular genetic study and helped to write the manuscript. M.P. edited the article and carried out the translation. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement:
The study was conducted in accordance with the Declaration of Helsinki and approved by the University Review Board, protocol code NO. 81 of 17/10/2022.
|
2023-04-28T15:10:32.118Z
|
2023-04-26T00:00:00.000
|
{
"year": 2023,
"sha1": "839c0c8d6382aaf18b2376ea2d6d790871bb0450",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2673-8236/3/2/18/pdf?version=1682471265",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "77ab3889ec0a117a7bb40df1fb1abd19de622c58",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": []
}
|
126069858
|
pes2o/s2orc
|
v3-fos-license
|
Accuracy analysis of automodel solutions for Lévy flight-based transport: from resonance radiative transfer to a simple general model
The results of accuracy analysis of automodel solutions for Lévy flight-based transport on a uniform background are presented. These approximate solutions have been obtained for Green’s function of the following equations: the non-stationary Biberman-Holstein equation for three-dimensional (3D) radiative transfer in plasma and gases, for various (Doppler, Lorentz, Voigt and Holtsmark) spectral line shapes, and the 1D transport equation with a simple longtailed step-length probability distribution function with various power-law exponents. The results suggest the possibility of substantial extension of the developed method of automodel solution to other fields far beyond physics.
Introduction
The analysis of the Green's function of the non-stationary Biberman-Holstein equation for radiative transfer in plasma and gases has shown [1] that there is an approximate automodel solution based on three scaling laws: for the propagation front (i.e. relevant-to-superdiffusion average displacement of perturbation's carrier from an instant point source) and asymptotic solutions far beyond and far in advance of the propagation front. All these scaling laws are determined essentially by the long-free-path carriers (named, by Mandelbrot [2], Lévy flights, cf. page IX in Ref. [3]). The validity of the suggested automodel solution was proved by its comparison with analytical solutions by Veklenko [4] in the 3D case of the Biberman-Holstein equation of the resonance radiation transfer for various (Doppler, Lorentz, Voigt and Holtsmark) spectral line shapes with complete redistribution over frequency (within spectral line width) in the elementary act of the resonance scattering (i.e. absorption and subsequent emission) of the photon by an atom/ion. The approach [1] was extended in [5] on the wide class of nonstationary superdiffusive transport on a uniform background with a simple long-tailed step-length probability distribution function (PDF) with various power-law exponents.
In this paper, we present the results of accuracy analysis of automodel solutions for Lévy flight-based transport, including the resonance radiative transfer (section 2) and a simple general model (section 3). Possible applications of the method [1,5] to other problems, where the role of the processes with dominating role of Lévy flights is identified and widely used, is discussed in section 4.
Self-similarity accuracy estimations for the Biberman-Holstein equation
The Biberman-Holstein equation for radiative transfer in a uniform medium of two-level atoms/ions is obtained from a system of equations for spatial density of excited atoms, F(r, t), and spectral intensity of resonance radiation. This system is reduced to a single equation for F(r, t), which appears to be an integral equation, non-reducible to a differential diffusion-type equation: where τ is the lifetime of the excited atomic state with respect to spontaneous radiative decay; σ is the rate of the collisional quenching of excitation; q is the source of excited atoms, different from populating by the absorption of resonant photons (e.g. collisional excitation). Hereafter we use the dimensionless time coordinate, assuming the normalization of time by τ. The kernel G is determined by the (normalized) emission spectral line shape εω and the absorption coefficient kω. In homogeneous media, G depends on the distance between the points of emission and the absorption of the photon: The automodel solution of equation (1) with a point instant source q(r, t) = δ(r − r0)δ(t -t0) (i.e. Green's function) was suggested in [5], which in the 3D case and for arbitrary space-time coordinates of the instant point source takes the form: where g is a function of a single variable, and its asymptotic behavior is known: g(s) = 1, s << 1; g(s) ∝ s, s >> 1. The relation between g and the exact solution of equation (1), fexact, is described by the following equations: where is the function reciprocal to the G function, ρ ≡ k0|r -r0|, k0 is the absorption coefficient for photons with frequency ω0, corresponding to the line shape center, where the functions t(ρ, s) and ρ(t, s) are determined by the relation To prove the automodel solution one has to show weak dependence (independence) of QW1 and QW2 functions on, respectively, space coordinate and time. The results of the validation of the automodel solution and the reconstruction of function g from comparison of function (3) with computations of the Green's function [4] for the Lorentz, Doppler, Voigt, and Holtsmark line shapes are shown in figure 1. shapes if we, for ρ > 30, restrict the time to t > 950, 280, 700, respectively. The largest deviations take place at s ~ 1.
Self-similarity accuracy estimations for general model (simple PDF, 1D case)
We consider the 1D transport on a uniform background, described by the equation for spatial density f(x, t) of an excitation of the background medium, which may evolve due to the exchange of excitation between various points of the medium via emission and absorption of the carriers (here the retardation caused by the finite velocity of carriers is neglected; the derivation of this equation for a possible mechanism of interaction between the medium and the carriers of medium's perturbation is given in Appendix in [5]): where W(x) is a step-length PDF (i.e. the probability that the carrier, emitted at some point, is absorbed at a distance x from that point), 1/τ is the absolute value of the emission rate (i.e. τ is the average waiting time between the absorption and reemission of the carrier), q is the source function, which is the rate of production of excitation by an external source (i.e. a source which differs from the excitation of the medium due to absorption described by the W function), and σ is the rate of the quenching of excitation. The uniformity of the background assumes that, first, the W is a function of only one variable -the distance between the points of emission and absorption -and, second, τ and σ are the constants. The latter makes the role of quenching simply described by the time exponent exp(−στ), therefore in what follows we omit this process. Hereafter we use the dimensionless time and space coordinate, assuming the normalization of time by τ and using a dimensionless PDF. We will seek for the Green's function, taking, respectively, the source function as a point instant source, q(x, t) = δ(x)δ(t).
We take the PDF in the following simple form which possesses a long tail and the infinite value of the mean square displacement: It was shown in [5] that one can derive automodel Green's function in the form or, for the PDF (9), where ρfr(t) = (t + 1) 1/γ -1. The procedure of the reconstruction of function g is quite similar to that of equations (4)- (7): where is the function reciprocal to W function, where the functions t(ρ, s) and ρ(t, s) are determined by the relation The QW2 function for various values of time coordinates is shown in figure 2 for three values of the exponent γ. It is seen from figure 2 that for given values of γ the function (11) is indeed an automodel solution of equation (8) with the PDF of equation (9) The accuracy for a weaker tail appears to be worse: for γ = 1.5 and ρ > 30 the deviation amounts to 12%, 7% and1% for, respectively, t > 30, t > 100, and t > 2400. For a stronger tail, γ = 0.5, the propagation front moves substantially faster, and high accuracy is achieved at larger distances: 5% for ρ > 100, t > 30 and 1% for ρ > 100, t > 100. The largest deviations take place at s ~ 1.
Conclusions
The results of Sec. 2 and 3 shows that the identification of main features of nonlocal (superdiffusive) mechanism of transport of resonance radiation may be fruitful for extending the method of approximate automodel solution to other problems. There is an example of physics model which gives precisely the power-law PDF of Eq. (9): see Eq.
(1) in [6] for photon-assisted transport of minority carriers in semiconductors (photo-excited holes in n-type InP [7]). The simple long-tailed step-length probability distribution function (9) with various power-law exponents may serve a bridge between physics and other fields. The general features of superdiffusion based on Lévy flights are recognized and applied in many fields (see, e.g., [1], [8]). Derivation of scaling laws, and especially of approximate automodel solutions, may be of practical interest.
|
2019-04-22T13:08:53.728Z
|
2017-12-01T00:00:00.000
|
{
"year": 2017,
"sha1": "72e2cf8770237495f2c0f034ee2ee52e6caa6257",
"oa_license": null,
"oa_url": "https://doi.org/10.1088/1742-6596/941/1/012050",
"oa_status": "GOLD",
"pdf_src": "IOP",
"pdf_hash": "552eaa0b1b3ef4c96695079fa50c352409050bea",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
222344839
|
pes2o/s2orc
|
v3-fos-license
|
Effect of Hyaluronic Acid and Poly-L-Lactic Acid Dermal Fillers on Collagen Synthesis: An in vitro and in vivo Study
Purpose Skin ageing is marked by structural and functional changes in epidermis and dermis, which result clinically in wrinkles, loss of elasticity, and rough-textured appearance. In this context, different dermal fillers have been used to overcome these negative effects associated with skin ageing, such as hyaluronic acid (HA) and poly-L-lactic acid (PLLA). Despite their low immunogenicity, these materials can cause an inflammatory reaction after application. Materials and Methods Considering high demand of HA and PLLA as filler material, this study aimed to evaluate their in vitro and in vivo effects. For the in vitro study, human dermal fibroblast cell cultures were supplemented with HA or PLLA for 24, 48, and 72 h. The following parameters were assayed: 1) cell proliferation, 2) cell viability, and 3) quantification of type I collagen by ELISA. For the in vivo study, HA or PLLA was injected in the dermis of Wistar rats and the tissues were collected after 15, 30, and 60 days for histologic evaluation and for quantification of type I collagen by Western blotting. The quantitative data were statistically analyzed using an ANOVA two-way. The significance level was set at 5%. Results At 72 h, high cell proliferation was observed for HA compared to control (p<0.05). Cultures exposed to PLLA exhibited a reduction in both cell proliferation and viability compared to control in all time points (p<0.05). Type I collagen expression was greater in cultures exposed to HA or PLLA compared to control (p<0.05). Histologic analysis showed the presence of multinucleated cells only in the PLLA group in all experimental time points. Western blotting analysis revealed high content of type I collagen in HA compared to PLLA (p<0.05). Conclusion The present study addresses a potentially unfavorable effect of dermal PLLA filler on the fibroblast phenotype, with possible clinical complications, unlike HA.
Introduction
Skin aging is a dynamic process that involves clinical and biological phenomena. The first comprises an intrinsic aging, which is similar to all internal organs. The second is extrinsic aging (photoaging), which occurs due to as consequence of external factors, chiefly ultraviolet (UV) irradiation. 1 The structural integrity of all the dermis is reduced by a range of intrinsic and extrinsic factors resulting into enzymatic degradation of the collagen by metalloproteinases and, consequently, the fibroblasts' death. 2,3 This process leads to results that can be observed clinically, such as wrinkles, atrophy, elastosis, dyschromic manifestations and volume loss. 3,4 Therefore, in terms of the aging process, it would be inadvisable to correct a defect permanently at a precise time; instead, the best approach is to apply fillers as needed, to deal with the aging signs as they appear. 5 The demand for injectable fillers continues to increase, 6 and new fillers have been developed just recently, whose safety and efficacy are supported by research. 7 There are many types of ideal fillers currently used by cosmetic and medical indication, in routine clinical practice. They can be classified as temporary, semi-permanent or permanent, and according to product composition. Primary ingredients include collagen, hyaluronic acid (HA), poly-L-lactic acid (PLLA), calcium hydroxyapatite and polymethyl methacrylate. 5 HA is a glycosaminoglycan and a natural compound of the extracellular matrix. 8 It can affect cell behavior and metabolism, 9 and increases skin hydration and fibroblast proliferation by activating fibroblasts to express type I (Col-1) collagen and matrix metalloproteinase-1 (MMP-1). In addition, HA modifies the organization of the actin cytoskeleton, influencing fibroblast shape and orientation. 9 PLLA is a biocompatible, biodegradable, immunologically inert synthetic polymer that stimulates collagen synthesis, leading to gradual volume reestablishment. As the PLLA microparticles degrade, the inflammatory reaction responsible for the degradation of the particles promotes the formation of fibrous connective tissue and neocollagenesis, leading to gradual volume replacement. 3,4,7,10 In addition, PLLA increases fibroblast activity and stimulates collagen synthesis. The result is not the effect of the product, but of the host's response to this process. 6 However, the molecular and cellular processes underpinning these procedures are still elusive. 11 This study aimed to evaluate in vitro cell viability, proliferation and type I collagen expression, using human dermal fibroblasts after HA (Juvederm Volbella ® , Irvine, CA, USA) and PLLA (Sculptra ® , Berwyn, PA, USA) supplementation. In addition, the histological aspects and type I collagen expression were investigated in vivo in an animal model.
Materials and Methods
This study was divided into two parts. The first one referred to the in vitro study, which is the main objective was to evaluate the effects of HA and PLLA on human dermal fibroblast cell cultures in terms of cell proliferation, cell viability, and type I collagen secretion. The second one referred to the in vivo study, which aimed to assay qualitatively the histologic aspects of dermis after HA and PLLA application in terms of biomaterials distribution among the collagen fibers and the presence of inflammatory cells as well as the quantification of type I collagen.
In vitro Study Cell Culture
Human dermal fibroblast cells were obtained from explants of mammoplasty and abdominoplasty provided by three separate female donors (age range 42-57 years old), according to the methods previously described by Huschtscha et al. 12 The samples were used after previous inform consent by the volunteers and approval by the Ethics Committee of the São Leopoldo Mandic Research Institute, Campinas, Brazil (#2011/0143).
The fibroblast cell phenotype was confirmed by immunofluorescence for the anti-α smooth muscle actin, antitype I collagen, and anti-vimentin antibodies, which were positive, and by negativity for the pan-cytokeratin cocktail (AE1/AE3 antibody) ( Figure 1).
Cell Proliferation and Viability Assays
The influence of dermal fillers on human dermal fibroblasts was evaluated according to proliferation and viability assays, and type I collagen expression. The cells were grown in 24 and 96 wells plates (Corning, NY, USA) at an initial density of 110 cells/mm 2 per well. After 24 h, the medium was changed, and left either unsupplemented or supplemented (C, control) with 0.5, 1.0, 5.0 and 10.0 mg/ mL 14 of HA (Juvederm Volbella ® , Irvine, CA, USA), or 0.05, 0.1, 0.5 and 1 mg/mL 15 of PLLA (Sculptra ® , Berwyn, PA, USA). After 24, 48, and 72 h, the cells were detached using 0.05% trypsin and counted using a hemocytometer to calculate proliferation indices. Using a different set of plates, with the same conditions as previously described, cell viability was determined by MTT assay, as described previously, at a wavelength of 590 nm. The experiments were performed twice under the same conditions to ensure accuracy.
Type I Collagen Quantification by ELISA
After 24, 48 and 72 h, the supernatants from the cell cultures were harvested and centrifuged at 5000 g for 15 min at 4°C. Aliquots of each sample were assayed by means of ELISA to determine the type I collagen levels, according to the manufacturer's recommendations (R&D Systems, USA). Total type I collagen was quantitated in picograms per mL (pg/mL). The results were calculated using the standard curves created in each assay. The ELISA assays were performed in a blind fashion in triplicate.
In vivo Study Specimen Preparation
Male Wistar rats (250-350 g) were obtained from CEMIB UNICAMP (Centro de Bioterismo, State University of Campinas, SP) and were given chow and water ad libitum throughout the study. The experiments were approved by the Institutional Committee for Ethics in Animal Research of UNICAMP (# 2015/005), and followed the recommendations of the Guide for the Care and Use of Laboratory Animals (AAALAC). Sculptra ® dermal filler was reconstituted prior to use by the addition of 10 mL of sterile water for injection, as recommended by the manufacturer, resulting in a final concentration of 15 mg/mL of poly-L-lactic acid. Juvederm Volbella ® filler is ready for use and presents a 15 mg/mL of hyaluronic acid concentration. The animals were treated with 20 μL dermal injection of the following treatment options: HA (Juvederm Volbella ® , Irvine, USA), PLLA (Sculptra ® , Berwyn, USA), PBS (control for HA, C HA ), or distilled sterile water (control for PLLA, C PLLA ). All the rats were anesthetized with 4% isoflurane, and kept on 2.5% isoflurane and 2 L/min oxygen anesthesia. Fifteen rats were treated with different treatments or left untreated on each side, resulting in five animals for each treatment at each time point (15,30 and 60 days). Each injection was administered on the right or left flank using a 27-gauge needle (Exel International Medical Products, Los Angeles, CA, USA) and a linear threading technique. 16 Euthanasia was performed by deepening anesthesia according to the following protocol: 90 to 150 mg/kg of sodium thiopental (71-73-8) combined with 10 mg/mL of Lidocaine (137-58-6), intraperitoneally. Full-thickness skin biopsy specimens were collected from the sites using a 11-mm tissue scalpel (Sklar Instruments, West Chester, PA, USA), immediately after the euthanasia, at 15, 30 and 60 days. The biopsy specimens were collected and divided to perform the histological examination and Western blotting assay.
Histological Analysis
A dorsal skin biopsy was obtained from each rat and fixed in 10% neutral-buffered formalin. The dorsal skin was shaved with an electric shaver before the skin samples were collected. A total of 4 μm paraffin sections parallel to the longitudinal axis of the body were stained with hematoxylin and eosin (HE), and mounted on glass slides with biological resin (Permount, Fisher Scientific, Pittsburgh, PA, USA). The histological analysis consisted of assaying qualitatively the distribution of the filler biomaterials among the collagen fibers as well as the presence of inflammatory cells.
Type I Collagen Quantification by Western Blotting
Skin biopsy specimens were pulverized using nitrogen, harvested and homogenized in lysis buffer (50 mM, Tris-HCL pH 7.5, 150 mM NaCl, 1% NP-40, 0.5% deoxycholate, 0.1% SDS) containing protease inhibitors. The Western blotting was carried out as described previously. 13 The primary polyclonal antibody used was anti-type I collagen antibody (1:1000, Abcam, La Jolla, CA, USA). GAPDH primary polyclonal antibody was used as an endogenous control (1:1000; Santa Cruz Biotechnology). A goat anti-rabbit secondary antibody was used for 1 h (1:5000; Sigma). The reaction was revealed with chemiluminescent detection reagents (Pierce) and visualized using digital imaging equipment (ImageQuant LAS 4000, GE Healthcare). Optical density (OD) measurements were performed with NIH ImageJ 1.37 (National Institutes of Health, Bethesda, MD) to analyze the scanned membranes. Data are presented as total protein expression normalized to GAPDH. For accuracy, the experiments were performed twice, in the same conditions, for each sample.
Statistical Analysis
The data were first analyzed for normality, using the Shapiro-Wilk test. Two-way analysis of variance (ANOVA) with post hoc Bonferroni test was applied to all assays performed, at a significance level of 5%. The results were expressed as the mean ± standard deviation.
In vitro Study Cell Proliferation and Viability Assay
At 24 and 48 h, cell proliferation and cell viability were similar among dermal fibroblast cells exposed to different concentrations of HA and the control group (ANOVA Two-Way, p>0.05). At 72 h, cell proliferation was similar among dermal fibroblast cells exposed to different concentrations of HA (ANOVA Two-Way, p>0.05), but superior to the control group (ANOVA Two-Way, p<0.05). At 72 h, the levels of cell viability were similar between the control group and dermal fibroblast cell cultures exposed to HA at 0.5 mg/mL (ANOVA Two-Way, p>0.05) and greater than those ones observed for dermal fibroblast cells exposed to HA at 1 mg/mL, 5 mg/mL, and 10 mg/mL (ANOVA Two-Way p<0.05) ( Table 1). At 24 h, dermal fibroblast cells exposed to PLLA at 0.05 mg/mL and 0.5 mg/mL exhibited low proliferation rate compared with other experiment groups (ANOVA Two-Way, p<0.05). At 48 and 72 h, a significant reduction in cell proliferation was observed for dermal fibroblast cells exposed to different concentrations of PLLA in comparison with the control group (ANOVA Two-Way, p<0.05). At 24, 48, and 72 h, the levels of cell viability were superior in control group compared to those ones observed for dermal fibroblast cells exposed to different concentrations of PLLA (ANOVA Two-Way, p<0.05) ( Table 2).
Type I Collagen Quantification by ELISA
ELISA analyses revealed that culture medium of dermal fibroblast exposed to HA or PLLA, irrespective of the
In vivo Study Histological Analysis
Representative images of dermal skin biopsies after the respective treatments are shown in Figures 2-4. The presence of collagen fibers interspersed with hair follicle, sebaceous glands and blood vessels, characterizing normal dermis, was observed in both control groups (C HA and C PLLA ), and at all the experimental time points (Figures 2-4A-D). The highest HA concentration in the deep dermis ( Figure 2E and F) occurred after 15 days in the HA group; however, the substance was found to be evenly distributed after 30 and 60 days (Figures 3 and 4E and F). No inflammatory process was observed at any of the time points evaluated.
In contrast, an intense inflammatory process was observed in the PLLA group, characterized by multinucleated cells, lymphocytes and macrophages, especially after 60 days. Degradation of the PLLA was observed over time, as identified by the increase in multinucleated giant cells and areas affected by phagocytosis (Figures 2-4G and H). Moreover, innumerous PLLA dermal filler crystals could be seen, especially after 60 days.
Type I Collagen Quantification by Western Blotting
Validation of the histological analysis was achieved by evaluating type I collagen protein content by immunoblotting ( Figure 5). In all the experimental periods, higher type I collagen expression was observed when the dermal fillers were injected, as compared with the respective control groups (C HA or C PLLA ). In both control groups, type I collagen content was found to be significantly lower than that of the respective dermal filler groups. At day 30 and 60, the type I collagen expression was greater in HA group compared to PLLA group (ANOVA Two-Way, p<0.05).
Discussion
The search for non-invasive cosmetic procedures is an on-going activity in medicine. Currently, several dermal fillers considered safe for use are available. However, their indiscriminate use may result in more serious outcomes, and greater patient risk, depending on the individual's systemic conditions. 17 Notably, dermal fillers are extensively used and have proven to be effective in improving appearance and repairing imperfections. 4 Among several available substances, hyaluronic acid (HA) and poly-L-lactic acid (PLLA) have been widely used. However, few in vitro and in vivo studies back up their use, especially if their effectiveness in attenuating imperfections is based on filling action or stimulation of collagen production.
The results obtained in this study confirm the unfavorable effects of PLLA on skin fibroblasts in vitro, and extracellular matrix in vivo after its application. HA supplementation was found not only to promote fibroblast proliferation and viability, but also to increase type I collagen without causing apparent inflammation, compared with PLLA.
In fact, HA is the major component of the skin extracellular matrix, involved in regulating cell proliferation and migration, 10 as well as promoting angiogenesis. 15 As a natural skin component, HA has the advantages of not causing rejection, having low immunogenic action, and possessing viscoelastic and hygroscopic properties. 7 These characteristics promote increased skin hydration when HA is applied as a dermal filling, characterized by excellent mechanical properties and moderate durability. 3 In fact, after 30 days of analysis, the present histological investigation showed that HA occasionally was sparsely distributed, and no inflammatory process was observed at any of the time points evaluated, thus emphasizing its favorable biological properties. In addition to inducing type I collagen formation, HA promoted an increase in These results are in line with those of previous studies, [18][19][20] which suggest that exogenous HA affects fibroblast proliferation and collagen synthesis, by increasing their local volume predictably.
In contrast to the promising achievements of HA dermal filler, PLLA promotes an in vitro decrease in cell proliferation and viability, mainly observed after 72 h. PLLA is a synthetic, biocompatible and biodegradable polymer used in various areas of medicine. 14,21,22 Results of the histological analysis of PLLA conducted in this study confirm that PLLA injection-induced local tissue reaction. An intense inflammatory process ensued from the reaction to a foreign body, represented by innumerous multinucleated cells, surrounded by a fibrous capsule. After 60 days, the presence of PLLA dermal filler was still evident, accompanied by stretching of the skin. These findings reinforce the conclusion that dermal thickness in patients receiving PLLA dermal filler is brought about by local tissue reaction. As soon as the filler is applied, the tissue stretches mechanically, resulting from injection of the fluid and leading to increased edema caused by the fluid carrier. In time, macrophages slowly degrade PLLA into lactic acid microspheres resulting from lactate production, 23 giving rise to many small crystals responsible for fibroblast stimulation and type I collagen production. 8,[24][25][26] Moreover, PLLA particles are flat pointed shape and did not dissolve in distilled water homogeneously. 27 In this respect, the more short-lived effect of HA than PLLA can be observed clinically, thus requiring reapplications of HA to be made at shorter intervals than those of PLLA.
Conclusion
Based on the results of the present study, dermal PLLA filler was found to have a potentially unfavorable effect on the fibroblast phenotype, possibly causing clinical complications. In contrast, HA provides favorable effects on the fibroblast phenotype in vitro and in vivo, including higher cell proliferation and type I collagen synthesis.
|
2020-10-15T05:09:09.561Z
|
2020-09-01T00:00:00.000
|
{
"year": 2020,
"sha1": "e267d4c316dc72124ad03e0b6cfbeb24c35121dd",
"oa_license": "CCBYNC",
"oa_url": "https://www.dovepress.com/getfile.php?fileID=61902",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "e267d4c316dc72124ad03e0b6cfbeb24c35121dd",
"s2fieldsofstudy": [
"Biology",
"Medicine"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
125849408
|
pes2o/s2orc
|
v3-fos-license
|
Azimuthal correlation between jets and ridges
We demonstrate that the trigger-ridge azimuthal correlati on in data can be understood based on a phenomenological model, called the Correlated Emission M del. In this model successive soft emissions due to jet-medium interaction lead to the enhance ment of thermal partons which follow the local flow directions. The enhanced thermal partons are t he source of the ridge particles. The correlation between the flow direction and the trigger direc t on plays a central role in understanding this azimuthal correlation. Features in the ridge yield at various trigger angles as a function of the relative azimuthal angle between the ridge and the tri gge and as a function of the impact parameters are discussed.
Introduction
This talk is on the azimuthal correlation between jets and ridges based on our published work in ref. [1].A typical ridge phenomenon we are interested in is illustrated in the 3d plot of Fig. 1, where the number of dihadron correlation events is plotted as a function of the azimuthal angular difference ∆φ = φ assoc − φ trigger and the longitudinal pseudorapidity difference ∆η = η assoc − η trigger .[2].
There the ridge is closely associated to a jet.It has a relatively long stretch in the longitudinal direction.There is a widening of the azimuthal distribution, up to about ∆φ ∼ 1 rad in the azimuthal direction.For the case illustrated the transverse momentum range is semi-hard.More specifically the trigger range is, 3 < p T trigger < 4 GeV, and the associated particles, p T > 2 GeV.There are also other examples of jet triggered ridges with semihard associated particle momenta.See refs.[3,4,5].
Ridges also appear in the auto correlation data such as the case shown in Fig. 2.Here no distinction between the triggers and associated particles were made.The data illustrated have the transverse momentum range 0.15 < p T < 2 GeV.Notice that here |∆η| extends to about 1.3.
The data are for Au+Au collisions at 130 GeV, from ref. [6].Recently the trigger-ridge azimuthal correlation data became available.This is shown in Fig. 3, from ref. [7].As indicated by the title of this talk, the azimuthal correlation is the main focus of today's presentation.The data shown
PoS(CERP2010)028
Azimuthal correlation between jets and ridges Charles B. Chiu here plays an important role in our investigation.The left figure is for the central region, with the centrality 0-5%.For brevity, hereon this region will be referred to as the C-region and the right figure is for centrality 20-60%, which hereon will be referred to as the noncentral region, or the NC-region.Notice the differences between the two.Compared to the NC case, the C case has a milder drop as the trigger angle φ s increases.They both level off at near 90 ∘ region.We will proceed now to our model.
Our model
Let me begin with a qualitative picture, to see how the azimuthal dependence of interest is to be described within our approach.
A scenario of ridge formation
The processes of interest are from nearside correlation measurements, where the trigger is at mid-rapidity with 3 < p trig T < 4GeV/c and the associated particles with 1.5 < p assoc T < 2 GeV/c.A typical dihadron correlation process begins with a large p T jet from a high energy parton-parton collision where the collision takes place near the surface, say at point P (x 0 , y 0 ).Here the hard (or semihard) parton in the jet exits to form the trigger and the recoiled hard parton moves in opposite direction and is absorbed by the medium.
There are successive soft emissions due to jet-medium interaction.It is the absorption of radiative energy by the medium which leads to energizing the local medium-partons, in turn the generation of the enhanced thermal partons.We identify the enhanced thermal partons as the source for the ridge particles.They are carried by the local transverse flow of the medium.The transverse
PoS(CERP2010)028
Azimuthal correlation between jets and ridges Charles B. Chiu flow direction defines the average direction of the eventual ridge particles.A word of caution is in order here.we do not require thermal equilibrium at early time.The usage of the terms thermal and the enhanced thermal refers to the exponential behaviors of the transverse momentum distributions of the partons at late time just before hadronization.
The correlated emission ansatz
We assume the strength in the emission of ridge particles is correlated to the relative angle between the flow direction and the trigger direction.Let me begin with the geometry of triggers and flows.For the matched case, here enhanced thermal partons which are the potential ridge particles are aligned in the same direction.We assume the reinforcement of the flow enhances the emission of ridge particles.The totally mismatched situation will be the case where the flow direction is perpendicular to the trigger direction.Here potential ridge particles formed at different hard-parton-
PoS(CERP2010)028
Azimuthal correlation between jets and ridges Charles B. Chiu mediun interaction points are emitted along different directions.Due to the lack of coherence, the ridge yield is expected to be suppressed.
The correlated emission ansatz states that the ridge yield favors the matched case and suppresses the mismatched case.Quantitatively this effect is represented by a gaussian function in the angular difference variable, i.e.
where λ is a parameter to be determined.This is a phenomenological formula that cannot be derived from first principles, but has a sound physical basis and will play a central role in our model.For every point (x, y) on the trajectory, the flow direction ψ(x, y) specifies only the average direction of the ridge hadrons.Since there are statistical fluctuations, the magnitude of which depends on how far (x, y) is away from the surface along the direction ψ(x, y).That distance is t ′ .
We introduce another Gaussian form to describe the mentioned fluctuation of the azimuthal angle φ of a ridge particle from the average flow direction where the degree of fluctuation is specified by γt ′ , which is the square of the gaussian width.
Clearly, the farther the emission point from the surface, the wider φ fluctuates from ψ(x, y).
Ridge yield per trigger
The probability of ridge yield at φ initiated from a trigger starting from the interaction point (x 0 , y 0 ) and emerging at angle φ s , is given by where N is an overall normalization constant which will be canceled when we compute the yield per trigger.The variables t, x ξ and y ξ all depend implicitly on the initial coordinates (x 0 , y 0 ).
Here P(x 0 , y 0 ,t) is the probability of detecting a parton emerging from the medium.It is the product of the probability of producing a semihard parton at (x 0 , y 0 ), which is proportional to the product of the longitudinal lengths at that point, L A (x 0 , y 0 )L B (x 0 , y 0 ), and the survival probability S(t), i.e., P(x 0 , y 0 ,t) ∝ L A (x 0 , y 0 )L B (x 0 , y 0 )S(t) . (2.4) The former depends on the nuclear matter density assumed, which we will not detail here.Due to the opaqueness of the dense medium, a sharp suppression factor as t increases is expected.We represent the survival probability as
Comparison with the data
3.1 Model fit to the φ s data.
The expressions given in subsections 2.2 and 2.3 define our model.The main parameter of the model is t 0 which characterizes the thickness of the interaction layer and λ the square of the gaussian width which characterizes the correlation between the trigger direction and the flow direction.
Fig. 5 shows our fit to the data, where t 0 = 0.2 or the interaction layer is about R A /5 and λ = 0.11 or the correlation angle about 20 deg.Notice that for the C case, the model reproduces One can qualitatively understand the contrast in the slopes between the C-case and the NC case from the geometry shown in Fig. 6.The left column is for the C-region and the right column for the NC-region.Consider the situation of the top row.Here φ s ∼ 0 ∘ .For both C and NC cases, the flow direction is more or less aligned with the trigger direction.Comparable yields for C and for NC are expected.The bottom row is at a larger trigger angle, say in the neighborhood of φ s = 70deg.
Due to elliptic geometry for the NC-case, the NC case has a more pronounced mismatch compared to the C-case.Also in this region the NC case has a less matter medium compared to the C case.
PoS(CERP2010)028
Azimuthal correlation between jets and ridges Charles B. Chiu Thus the NC case leads a larger drop in the azimuthal dependence.Notice that at the trigger angle φ s = 90 ∘ , it is the matched case, where the flow aligns with the trigger.The φ s dependence is symmetric about 90 deg, which is responsible for the flatness of the azimuthal dependence near 90 deg.
Comparison with ∆φ data at various trigger angles
Fig. 7b shows the ridge yield as a function of difference ∆φ = φ ridge −φ trigger at the trigger angle φ s = |22 ∘ |.The predicted curve agrees with the data.Instead of combining the ridge contribution for the absolute value of the trigger angles, ±22 deg, Fig. 7a shows the predicted curve for φ s = 22 ∘ case and a separate curve for φ s = −22 ∘ .Notice that there is noticeable shift in the peak positions between the two cases.There is a geometric reason for this shift, which will be discussed in the following section.
The asymmetry parameter
To characterize the variation in the skewness of the curves, we work with the asymmetry parameter defined by For the trigger angle in the range 0 to π/2, Notice that Y + represents the ridge yield for ∆φ ≤ 0. Inspection of the curves shown in Fig. 8 indicates that at within the trigger angle range shown, A ≥ 0. When the trigger angle is 0, there is the symmetry ∆φ = −∆φ , which implies that Y + = Y − , or A=0.By the same token, at φ s = 90 deg there is again the symmetry ∆φ = −∆φ which again leads to A=0.Fig. 9 shows the predicted curves of A versus φ s , where the dashed curve is for the C-region case and the solid curve for the NC-region case.Our work led to the subsequent analysis of the STAR data, the result of which was reported at QM09 [8].Fig. 10 shows a comparison between their data and the CEM predicted curve for the NC-case.One sees that the data confirms the qualitative feature predicted by our model.
R-yield vs b (or Npart) at various trigger angles
Fig. 11 shows the ridge yield as a function of the normalized impact parameter b/R, where R is the effective radius of the colliding nucleus (Au).R=7fm is used.Fig. 11a gives an overview
PoS(CERP2010)028
Azimuthal correlation between jets and ridges Charles B. Chiu trigger may be schematically represented by: ridge-yield (larger b case)=I/(I + εII) where ε is a small number compared to unity.Thus the yield per trigger in the larger b region is greater than that at b=0, which causes a hump in this b region.This is the situation for 0 deg trigger angle.As the trigger angle increases, the bump becomes less and less pronounced.At around φ s = 30 ∘ the hump structure disappears completely, this marks the onset of a smooth monotonic decrease of the yield-curve.It is interesting to see whether future data will confirm the prediction here.
Summary
We see the ridge data have provided strong evidence that the medium response in jet-medium interactions depends on the direction of the transverse flow of the medium.The flow influences the direction in which the loss of energy should go and where the ridge should be formed.CEM uses the presence of the ridge as a means to keep track of the energy loss of the jets going into the medium.We have found that the ridge formation can be strong only within 20 deg around the trigger direction.When the flow is perpendicular to the jet direction, the ridge yield is completely suppressed.
We have shown that the CEM reproduces the φ s dependence of the ridge yield data.Our study allows us to predict the trigger angle dependence of the asymmetry parameter which has subsequently been confirmed by the data.Our study also predicts the impact parameter dependence (or b-dependence) of the ridge yield.Our b-curve averaging over the trigger angles agrees with the data.We have also presented b-curves at various trigger angles for verification in the future.
In this talk our focus has been on the ridge-trigger correlations in the transverse direction.Our investigation on the longitudinal correlation is in progress.After my presentation, there were several followup questions and comments related to the longitudinal correlation.I refer the reader to my contribution of [Note added] to the Saturday QA session, in which the longitudinal correlation problem related to our CEM model is briefly discussed.
Figure 3 :
Figure 3: Ridge structure given in auto-correlation data.No distinction is made between triggers and associated particles.From [6].
Fig. 4
illustrates the situation.The trigger directions are indicated by the thin arrows.a) is for the trigger angle of φ s = 0 ∘ .3 different starting points are indicated.Each has a local flow direction (indicated by the thick green arrow associated with it, along the direction ψ s ).Only the middle one corresponds to the matched case, where the flow direction is aligned with the trigger direction, i.e. ψ = φ .b) is for φ s = 70 ∘ .Here the matched case occurs at the upper one.
Figure 4 :
Figure 4: (Color online) Illustrations of the relationship between the trigger directions φ s in (red) arrows and the flow directions ψ in thick (green) arrows for noncentral collision.(a) Semihard partons at φ s = 0 originated from 3 different points P where only the middle one has matching φ s and ψ that lead to strong ridge, while in (b) for φ s ∼ 70 ∘ only the upper one has matching angles, leading to stronger ridge than in the two lower non-matching cases, but it is weaker than the middle one in (a) because of lower local density at the tip of the ellipse.
Figure 6 :
Figure 6: A comparison of the misalignment situation between the C case and the NC case.Near φ s ∼ 0 region, both dominated by the alignment configuration.At around φ s ∼ 70 ∘ , for the NC case, the elliptic geometry makes the misalignment effect more noticeable, it causes a stronger suppression in the ridge yield.
Figure 7 : 1
Figure 7: (Color online) The data are ∆φ distributions from [7] for 15 < φ s < 30 ∘ at 20-60% centrality for (a) the sum of jet and ridge and (b) ridge alone.The curves are all calculated in the CEM for the ridge distributions only with γ = 1.The dashed and dashed-dotted lines are left-and right-shifted for φ s = ±22 ∘ , respectively.The solid lines are the average over the two signs of φ s .
Figure 8 :
Figure 8: (Color online) The ridge distributions for various positive values of φ s .
Figure 11 :Figure 12 :
Figure 11: (Color online) (a) Ridge yield per trigger vs impact parameter for 5 values of φ s , (b) Y (b), averaged over all φ s , vs impact parameter, and (c) average yield vs N part .The two points in (c) are determined from the data in Fig. 2(a) and (b).
|
2018-12-07T06:16:54.721Z
|
2011-03-01T00:00:00.000
|
{
"year": 2011,
"sha1": "c704730fa86492720b7d333c01d359f96035c465",
"oa_license": "CCBYNCSA",
"oa_url": "https://pos.sissa.it/111/028/pdf",
"oa_status": "HYBRID",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "c704730fa86492720b7d333c01d359f96035c465",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
54798922
|
pes2o/s2orc
|
v3-fos-license
|
On impedance conditions for circular multiperforated acoustic liners
Background: The acoustic damping in gas turbines and aero-engines relies to a great extent on acoustic liners that consists of a cavity and a perforated face sheet. The prediction of the impedance of the liners by direct numerical simulations is nowadays not feasible due to the hundreds to thousands repetitions of tiny holes. We introduce a procedure to numerically obtain the Rayleigh conductivity for acoustic liners for viscous gases at rest, and with it define the acoustic impedance of the perforated sheet. Results: The proposed method decouples the effects that are dominant on different scales: (a) viscous and incompressible flow at the scale of one hole, (b) inviscid and incompressible flow at the scale of the hole pattern, and (c) inviscid and compressible flow at the scale of the wave-length. With the method of matched asymptotic expansions we couple the different scales and eventually obtain effective impedance conditions on the macroscopic scale. For this the effective Rayleigh conductivity results by numerical solution of an instationary Stokes problem in frequency domain around one hole with prescribed pressure at infinite distance to the aperture. It depends on hole shape, frequency, mean density and viscosity divided by the area of the periodicity cell. This enables us to estimate dissipation losses and transmission properties, that we compare with acoustic measurements in a duct acoustic test rig with a circular cross-section by DLR Berlin. Conclusions: A precise and reasonable definition of an effective Rayleigh conductivity at the scale of one hole is proposed and impedance conditions for the macroscopic pressure or velocity are derived in a systematic procedure. The comparison with experiments shows that the derived impedance conditions give a good prediction of the dissipation losses.
Introduction
The safe and stable operation of modern low-emission gas turbines and aero-engines crucially depends on the acoustic damping capability of the combustion system components. Hereby, so called bias flow liner -consisting of a cavity and a perforated face sheet with additional cooling air flow -play a significant role. Since decades the damping performance prediction of these bias flow liner under all possible flow conditions remains a major challenge. However, due to the higher tendency of low-emission, lean burn combustion concepts for combustion instabilities the prediction of the acoustic bias flow liner impedance and therewith its damping performance is a very important prerequisite for the engine design process. Several analytical and semi-empirical models for the impedance description of bias flow liner were developed in the past (see also [1]). This work focuses on the numerical simulation of the acoustic characteristics of bias flow liner applying multi-scale modeling. In principal all theoretical approaches are based on the formulation of the Rayleigh conductivity K R [2,3] which describes the ratio of the fluctuating volume flow Q(t) through a hole to the driving pressure difference P − (t) − P + (t) across the hole: and has the dimensions of length. One major challenge in the model description of the Rayleigh conductivity represents the definition or the specification of the pressure difference since, above and below the perforated liner face-sheet the pressure is not necessarily constant rather a function of the distance from the hole. Here, the present work applying a multiscale asymptotic model will provide an exactly defined solution. More precisely, the Rayleigh conductivity of a single hole in an array of holes is distributed over the whole liner area. In this way the effective Rayleigh conductivity as quotient of the Rayleigh conductivity of one hole and the area A δ of one periodicity cell of the array is introduced that has the dimensions of one over length. Using the effective Rayleigh conductivity the liner impedance can be determined like later shown for example in equation (21).
Methods
We consider acoustic liner that consist of a wall or part of a wall with a periodic dense array of equisized and equishaped holes with an characteristic periodicity that is proportional to a small parameter δ > 0. The holes may not be of cylindrical shape and even tilted in general. For sake of simplicity we consider the perforated wall Ω δ liner with a circular cross-section of inner radius R d , while noting that the proposed procedure to define the Rayleigh conductivity and impedance conditions do not depend on the choice of the cross-section, but only on the hole pattern and hole shape and can be directly transfered to other cross-sections like rectangular.
To derive the impedance conditions we let the parameter δ of the hole period tending zero -so the number holes increases accordingly -while the inner and outer diameter of the cross section are scaled like δ 2 as well as the thickness of the perforated wall, see Fig. 1. As δ → 0, the holes merge and the domain Ω δ liner degenerates to an interface Γ liner on which we will prescibe the impedance conditions representing the correct disspation behaviour of the acoustic liner. For the circular liner the limit interface domain Γ liner is an cylinder of radius R d . As it simplifies the derivation and impedance condition greatly we assume that the area of the periodicity cell of the periodic array A δ = δ 2 .
This liner shall be embedded in a duct domain Ω and the computational domain is Ω δ := Ω \ Ω δ liner for every δ > 0, i.e., , the duct domain without the multi-perforated wall. On this domain we introduce as viscoacoustic model the linearized compressible Navier-Stokes equations in frequency domain in a uniform and stagnant media for a source term f (t, x) = Re(f (x) exp(−ıωt)) with an angular frequency ω > 0: with the acoustic velocity v δ , the acoustic pressure p δ , the mean density ρ 0 > 0, the speed of sound c, the kinematic and secondary viscosities ν(δ), ν (δ) > 0. We scale the viscosities for δ → 0 like δ 4 such that the size of the viscous boundary layers remain asymptotically the same at the scale of a single hole. If the duct is modelled to be of infinite extend then additional conditions at infinity have to be posed, e.g., , for a channel of constant cross section with infinite extend in z ± ∞ these conditions are Moreover, we assume the souce to be located away from the perforated wall such that f = 0 in a neighbourhood.
In the following section we study the solution of the viscoacoustic model in three different geometrical scales beginning at the scale of one hole, pursuing with the scale of one period of the hole array and concluding with the macroscopic scale on which the impedance conditions follow.
Microscopic scale: the near field around one hole
In the vicinity of one hole that tends to a point x Γ on the interface Γ liner we use the local coordinate X := (r, y, z) = ((r − R d )/δ 2 , rθ/δ 2 , z/δ 2 ). As δ → 0, the hole variable X occupies the whole unbounded domain Ω defined by (see Fig. 2a) where Ω hole is the scaled domain representing one hole, and we assume 0 ∈ Ω hole . For instance a vertical cylindrical hole of diameter d 0 δ 2 can be represented by Ω hole = {(r, y, z) ∈ R 3 such that 0 ≤ r ≤ h 0 and y 2 + z 2 < 1 4 d 2 0 }. Close to one hole of the perforated liner, we represent the solution (v δ , p δ ) of (3) as where the near field corrector terms (v 0 , p 0 , p 1 ) do not depend on δ. The scaling of the second corrector for the pressure as δ 2 is due to the associated scaling of the size of the holes. Now, inserting expansion (5) into the viscoacoustic model (3) and identifying formally terms of same powers of δ results first in the fact that p 0 (x Γ , X) is a constant function in X, and then in a product representation of the near-field corrector where c(x Γ ) allows for a slow variation of near field velocity and pressure along the wall.
Here, the color coding corresponds to the amplitude and the arrows to the direction of the velocity.
where ∇, div and ∆ are the gradient, divergence and Laplace operator in X (cf. that act as a excitation from far away and will be used for the matching with the mesoscopic scale (see Sec. 2.2). Here, Γ ± (S) = (r, y, z) ∈ Ω, ±r > r ± and (r − r ± ) 2 + y 2 + z 2 = S 2 , with r − = 0 and r + = h 0 , are the two half-spheres (see Fig. 2a) that are moved towards infinity. Note that in problem (6) the term −ν 0 ∇ divṽ that would appear in the first line cancels out due to the divergence free condition (6b). Moreover, note that the term −ν 0 ∆ṽ can be replaced by ν 0 curl curlṽ and so only the vorticity part of the velocityṽ will exhibit a viscosity boundary layer as we will see later.
Problem (6) is a classical saddle-point problem and admits a unique solution stated by the following Note, that the pressure space V( Ω) allows for a constant behavior towards infinity.
With the near field velocity profileṽ defined by (6) we can define in analogy to the Rayleigh conductivity a posteriori the quantity using the volume flux towards infinity in a symmetric way. Here, n is the outer normal vector. In this way, the quantity k R is a mapping of a constant near field pressure at infinity to the flux at infinity. To see the analogy it suffices to consider time harmonic fields varying like exp(−ıωt), the volume flux Q(t) through the aperture counted positively along the direction of the e r axis to be the same as the volume flux through the surface Γ + (S) (respectively Γ − (S)), counted positively (resp. negatively) along the direction of the normal vector n, and to compare (1) and (8). Note, that the normal component of the near field velocity profile v decays like 1/S 2 towards infinity and combines different behaviour close to and away from the wall (see Fig. 2(c) and (d)). This behaviour can be rigorously justified with similar techniques as in [5,6].
For the usual definition of the Rayleigh conductivity K R it is not evident where the difference of the pressure -as it varies locally -and the volume flux -as in the original acoustic equations the fluid is compressible -shall be evaluated. The quantity k R is, however, clearly defined by (6) and (8) as the near field pressure tends to constant values for |X| → ∞ and as the near field velocity is incompressible. This results from the separation of the effects at the different length scales, namely viscous incompressible behaviour in the vicinity of the holes versus inviscid, compressible behaviour away from them, due to the asymptotic ansatz. As the near field profiles are defined in local coordinates X it has the dimensions of one over length and we denote it as effective Rayleigh conductivity of the liner.
The definition of the effective Rayleigh conductivity k R can be used for inviscid fluids as well for which ν 0 = 0 if the no-slip boundary conditions (6c) are replaced by v · n = 0.
Mesoscopic scale: the hole pattern
Pursuing with the scale of one period of the hole array and in the vicinity of one hole that tends to x Γ , we use the local coordinate X := (R, Y, Z) = ((r − R d )/δ, rθ/δ, z/δ). We consider for fixed δ > 0 the infinite periodicity cell are two semi-infinite parallelepipeds whose opposite lateral faces |Z| = 1 2 √ a and |Y −bZ| = √ a 2 are considered to be identified with each other such that B δ ± and so B δ are topologically equivalent to a torus. With the cross-section of the periodicity cell with X ∈ B δ . Inserting expansion (10) in problem (3) and identifying formally the terms of same powers of δ gives that P δ 0 (x Γ , X) is constant in X and a separation of variables for the mesoscopic corrector as Here, ∇ and div are the gradient and divergence in X. The formal identification of terms of same power in δ can be justified despite the fact that the size of the hole depends on δ as well. For this an additional scale η for the size of one hole has to be introduced that is first considered to be independent of δ due to its different meaning and later fixed to δ 2 . The expansion (10) is then in δ, where the terms of the expansion depend on η. For the brevity of the article we have chosen directly η = δ 2 . Note that (11) is equivalent to an homogeneous Laplace problem with Neumann boundary conditions for the pressure profile P δ , where the velocity profile V δ can be computed from. Following [7, Proposition 2.2], we can therefore state the following Proposition 2.2. For any fixed δ > 0, the kernel of problem (11) is of dimension 2 and spanned by the admits the following limit behaviour: It remains to determine the constant D δ ∞ , where we are in particular interested in its asymptotic behaviour for δ → 0. To obtain this behaviour we will match the mesoscopic functions V δ D and P δ D with the near field profilesṽ andp at half-spheres Γ ± (s δ ) of radius s δ for √ δ < s δ < 2 √ δ centered at the aperture 0. First we note that due to the incompressibility and the limit behaviour of V δ D for its volume flux over the half-spheres it holds ıω 2 Using this equality, definition (8) of the effective Rayleigh conductivity k R , the mesoscopic to microscopic variable change X = X/δ, and matching of the mesoscopic velocity V δ D and the near field velocity profilẽ v we find that By linearity and using definition of problems (6) and (11), the gradient of the mesoscopic pressure P δ D can be matched with the gradient of the near field pressure profile as well. Integrating these gradients, using limit (6d) and Proposition 2.2 leads to This blow up of the coefficient D δ ∞ as δ → 0 in accordance with its numerical computations based on an asymptotic analysis of (3) with only two scales [8], where the hole size is considered not as a scale but as a parameter.
Macroscopic scale and impedance conditions
Finally, away from a vicinity of the layer, the solution (v δ , p δ ) of (3) is represented by Inserting this expansion in problem (3) and making a formal identification in terms of powers of δ gives that (v 0 , p 0 ) is solution of the classical Helmholtz problem and a multiscale analysis [9] for rigid walls leads to the boundary conditions v 0 · n = 0, on ∂Ω .
The limit condition (3d) becomes where T 1 ± is the Dirichlet-to-Neumann operator based on the projection on the outgoing propagative modes, see [10, Eq. (2.7)] and [11]. This problem is completed by jump conditions across the interface Γ liner . To obtain the conditions we match the macroscopic pressure p 0 and flux v 0 · n in a matching zone at distance √ δ to the interface Γ liner to the mesoscopic pressure and velocity functions. For the pressure we find with two functions C N , C D that allow for slow variation along the perforated wall. With the factor δ the limit δP δ D ( x−xΓ δ ) for δ → 0 remains bounded. Subtracting the two limits of (16) for δ → 0 we obtain Taking the gradient in x on both sides of (16) and using (15a), the assumption that f = 0 close to the perforated wall and (11) we find As the two limits for V δ D for R → ±∞ coincide we obtain v 0 · n (x Γ ) = 0, on Γ liner .
Finally, taking the average of (18) and the limit δ → 0 gives in view of (17) the impedance conditions Note, that the impedance conditions do not depend on the pattern of the holes, more precisely on the values a and b (see Fig. 1), but only on their area A δ , namely through ν 0 = ν/A 2 δ in the computation of the effective Rayleigh conductivity k R .
Distinguished limit Note, that the nature of the impedance condition (19b) is due to the choice of asymptotic scales. It represents a distinguished limit meaning that different choice would lead to one of the trivial conditions p 0 (x Γ ) = 0 (transparent wall) or v 0 ·n (x Γ ) = 0 (rigid wall). If we would scale the diameter of each hole like ε(δ) as well as the thickness of the perforated wall such that δ 2 = o(ε(δ)) then we would obtain transparent wall conditions in the limit δ → 0. A contrario, the impedance conditions become rigid wall conditions if we would use the scaling ε(δ) = o(δ 2 ). The choice of asymptotic scales was already stated in [12] for infinitely thin perforated wall and the Stokes flows.
Acoustic Impedance The nature of the impedance conditions is known in the literature: the notion of impedance can be found in the works of Webster in the 1910s [13]. More precisely, he defines the normalized specified acoustic impedance ζ by (note there is a complex conjugate and a different sign due to the different choice of the time-dependency convention) For the derived impedance conditions (19b) and by identification, the normalized specified acoustic impedance for perforated walls is given by The resistance Re(ζ) and the reactance Im(ζ) are positive quantities when k R has a positive real part and a negative imaginary part. Moreover in the inviscid case k R is a positive real number, so that the normalized specified acoustic impedance ζ is purely a reactance.
Formulation in pressure only One can also remark that Problem (15) can be formulated in terms of pressure only: equations (15a)-(15d) give and impedance conditions (19a)-(19b) are written in terms of the pressure as This kind of conditions were proposed for the inviscid case [14], where k R turns out to be the effective plate compliance.
Results and discussion
In this section, we are interested by the numerical computation of the effective Rayleigh conductivity k R , the computation of dissipation losses in acoustic ducts with the impedance conditions and comparison with data from experimental measurements.
Numerical computation of k R
The effective Rayleigh conductivity k R is defined through the solution of the near field velocity and pressure profiles in the unbounded domain Ω around a single hole. To compute k R numerically we truncate the unbounded domain, on which we use the finite element method for discretization and propose an extrapolation procedure to increase the accuracy.
of Ω for a given truncation radius S > 0.5 d 2 0 + h 2 0 (see Fig. 2(a)). It has two artificial boundaries Γ ± (S) that are no boundaries of Ω. We restrict the problem (6) to Ω(S) and ∂ Ω(S) ∩ ∂ Ω, and we approximate the conditions (6d) by settingp |Γ±(S) = ± 1 2 . From the resolution of the truncated problem we compute the approximated Rayleigh conductivity k R (S) taking as well an approximation of (8), namely Its approximated value k R (S) tends to the Rayleigh conductivity k R as 1/S as illustrated in Fig. 4. This first-order convergence can be explained with a rigorous analysis of the solution of problem (6) towards infinity using the Mellin transform [15] and showing that the solution of this problem on Γ ± (S) is a superposition of a radial expansion with respect to 1/S and of a cartesian expansion with terms decaying exponentially with respect to the distance to the boundary. Similar analyses were performed for the Poisson and Helmholtz problems in conical domains with a rough periodic boundary [5] or perforated wall [6]. As, more precisely, the Rayleigh conductivity k R can be expanded in powers of 1/S we use an extrapolation in 1/S of first order approximations k R (S) for different truncation radia S to obtain a second or higher order approximation of the limit value k R .
For the particular case of a straight cylindrical hole that is without loss of generality centered at y = z = 0, the domain Ω(S) is invariant under rotation around the r axis as well as the solution of the problem (6) for the near field profiles. Hence, the finite element method in two dimensions can be used Figure 4: Convergence of the real and imaginary parts of the approximated Rayleigh conductivity k R (S) to its limit value k R = 4.513 − 1.210ı in dependence of the truncation radius S for the liner DC006 at frequency f = 306 Hz (see Table 1). for the numerical resolution in a 2D axis-symmetry setting. To resolve the boundary layer of size ν 0 /ω on the wall boundary (cf. [4, Sec. 3.1]) we use the hp-adaptive strategy of Schwab and Suri [16] (see the mesh shown in Fig. 2(b)). For four liner configurations, see Table 1, from experimental studies [1,17] we have computed the near field velocity and pressure profiles and so the effective Rayleigh conductivity. The relative kinematic viscosity ν 0 is computed as quotient of the kinematic viscosity ν = 1.4660 × 10 −5 m 2 /s of air at 15 • C divided by the period δ to the power of four. In Fig. 2(b) and Fig. 2(c) we illustrate the near field pressure and velocity profilesp andṽ for the liner DC006 at frequency 306 Hz using a truncation radius S = 40. It is visible that the pressure decays almost linearly inside the cylindrical hole, but also the behaviour at distance to the hole. Moreover, the pressure shows close to the rim of the cylinder an edge singularity (i.e., a corner singularity for the 2D axis-symmetric problem) that is resolved numerically by the hp-adaptive refinement strategy. The near velocity profile shows a flux from all sides to and through the hole. It appears that the outward flux of the imaginary part ofṽ over Γ + (S) is negative (resp. positive over Γ − (S)) corresponding to a positive real part of the approximate Rayleigh conductivity k R (S) (see (24)) and so of the Rayleigh conductivity k R . This is in line with the inviscid case, where k R is real and positive. Moreover, we see the higher velocity amplitude inside the hole that decays towards its boundaries. This boundary layer phenomena is more visible for lower frequencies (see Fig. 2(d)), where one also see a local change of the velocity direction on the wall boundary.
In Fig. 5, we plot the effective Rayleigh conductivity k R as a function of the frequency f := ω 2π for different liner configurations given in Table 1. As expected, following the remark on the normalized specified acoustic impedance ζ, the real part of k R is positive and its imaginary part is negative. One can also remark that for liner configurations DC006 and DC007, that have a close value of the porosity σ but quite different hole repartition and hole diameter, their Rayleigh conductivities differ significantly in both their real and imaginary part. In Fig. 6, we show the computed normalized specific acoustic impedance ζ for the liner configuration DC006 in comparison with the Melling model (see [18] and [17,Eq. (12)]), that is given an analytic formula. For the latter an effective kinematic viscosityν(δ) := 2.179ν(δ) is used that shall incorporate also thermal conductivity losses near a highly conducting wall, see [19, p. 239] and [1, p. 62]. We plotted the Rayleigh conductivities obtained from our model with this effective kinematic viscosity. The reactance predicted by the two models are very close, where the resistence differs by up to 20%. The importance of taking the thermal conductivity losses into account will be seen in comparison with the measurements and be discussed later in Sec. 3.3.
Experimental Setup and Analysis
The experimental study is performed in the duct acoustic test rig with a circular cross-section (DUCT-C) at DLR Berlin at ambient conditions. The setup of the test rig is illustrated in Fig. 7. It allows high precision acoustic measurements of the damping performance of various liner configurations, including grazing and bias flow.
The test duct consists of two symmetric measurement sections (section 1 and section 2 in Fig. 7) of 1200 mm length each. They have a circular cross-section with a radius R d of 70 mm. In order to minimize the reflection of sound at the end of the duct back into the measuring section the test duct is equipped with anechoic terminations at both ends (not shown in Fig. 7). Their specifications follow the ISO 5136 standard. The damping module is a chamber of 60 mm. It has a circular cross-section with a radius of 120 mm. A total of 12 microphones are mounted flush with the wall of the test duct. They are installed at different axial positions upstream and downstream of the damping module and are distributed exponentially with a higher density towards the damping module. Two microphones are installed opposite of each other at the same axial position close to the signal source. As evanescent modes become more prominent in the vicinity of the source, their influence is reduced significantly by using the average value of these two microphones for the analysis. This technique helps to reduce the errors for frequencies approaching the cut-on frequency of the first higher order mode and thus, extends the frequency range for accurate results.
At the end of each section a loudspeaker is mounted at the circumference of the duct (A and B in Fig. 7). They deliver the test signal for the damping measurements. The signal used here is a multi-tone sine signal. All tonal components of the signal are in the plane wave range. The signal has been calibrated in a way that the amplitude of each tonal component inside the duct is about 102 dB.
The microphones used in these measurements are 1/4" G.R.A.S. type 40BP condenser microphones. Their signals are recorded with a 16 track OROS OR36 data acquisition system with a sampling frequency of 8192 Hz. The source signals for the loudspeakers are recorded on the remaining tracks. The test signal is produced by an Agilent 33220A function generator. The signals are fed through a Dynacord L300 amplifier before they power the Monacor KU-516 speakers.
For each configuration two different sound fields are excited consecutively in two separate measurements (index a and b). Speaker A is used in the first measurement and in the second measurement the same signal is fed into speaker B. Then, the data of section 1 and section 2 (index 1 and 2) are analyzed separately. This results in four equations for the complex sound pressure amplitudes for each section and measurement for j = a, b:p p + andp − are the complex amplitudes of the downstream and upstream traveling waves. The recorded microphone signals are transformed into the frequency domain using the method presented by Chung [20]. This method rejects uncorrelated noise, e.g., turbulent flow noise, from the coherent sound pressure signals. Therefore, the sound pressure spectrum of one microphone is determined by calculating the cross-spectral densities between three signals, where one signal serves as a phase reference. In our case the phase reference signal is the source signal of the active loudspeaker. As a result we obtain a phase-correlated complex sound pressure spectrum for each microphone signal.
According to Eqs. (25a)-(25b) the measured acoustic signal is a superposition of two plane waves traveling in opposite direction. In order to determine the downstream and upstream propagating portions of the wave in each section, a mathematical model is fitted to the acoustic microphone data. This model Figure 8: Illustration of the sound filed in the duct for measurements A and B by means of the sound pressure amplitudesp, the reflection coefficient r, the transmission coefficient t, and the end reflection r e considers viscous and thermal conductivity losses at the duct wall. They are included in the wave number with the following attenuation factor α as proposed by Kirchhoff [21]: with the duct radius r, the speed of sound c, the kinematic viscosity ν, the angular frequency ω (as in Eq. (3)), the heat capacity ratio γ, and the Prandtl number P r. As a result of this least-mean-square fit, the four complex sound pressure amplitudesp + 1 ,p − 1 ,p + 2 andp − 2 are identified at position z = 0 for both measurements. These sound pressure amplitudes are related to each other via the reflection and transmission coefficients of the test object. This is illustrated in Fig. 8 for the two different measurements A and B. In order to calculate the reflection and transmission coefficients r + , r − , t + , and t − from the sound pressure amplitudes the following four relations can be derived for j = a, b: The equations from both measurements are combined and solved for the reflection and transmission coefficients in downstream and upstream direction, respectively. The advantage of combining the two measurements is that the resulting coefficients are independent from the reflection of sound at the duct terminations. These end-reflections are contained in the sound pressure amplitudes, but do not need to be calculated explicitly. Moreover in the case of a uniform and stagnant flow these coefficients do not depend on the direction we consider, i.e., r − = r + and t − = t + . The dissipation of acoustic energy is expressed by the dissipation coefficient. The dissipation coefficient ∆ can be calculated directly from the reflection coefficient R and the transmission coefficient T via an energy balance R ± + T ± + ∆ ± = 1.
To compute these coefficients, the integration of the acoustic energy flux in a uniform and stagnant flow yields a relation between the acoustic pressure p and acoustic power P quantities(see Blokhintsev [22] and Morfey [23]) : Then, the energy coefficients can be given relative to the pressure coefficients as: where the indices 1 and 2 refer to section 1 and section 2 of the duct as illustrated in Fig. 8. With the energy balance (30) follows the definition of the energy dissipation coefficient This is an integral value of the acoustic energy that is absorbed while a sound wave is passing the damping module. The dissipation coefficient is used to evaluate the damping performance of the test object.
Numerical simulation of dissipation losses
This setup is also simulated numerically using the equivalent problem (22a)-(22b) for the pressure with a source term corresponding to an incoming field p inc (r, θ, z) = exp(ıωz/c) from the left. The scattered field is computed numerically using the mode matching procedure with N = 5 modes [24]: we seek for the scattered field p 0 under the form (see Fig. 9(b)) p 0 (r, θ, z) = p inc (r, θ, z) + inside the waveguide part, and under the form inside the duct part. The pairs (β j , ψ j ) and (β j , ψ j ) are solution of a "2D" transverse eigenvalue problem in the wave-guide and liner parts, using the fact that the source term p inc and the geometry are independent of the angle θ. From the mode matching and assuming that there is only one propagative mode inside the waveguide, i.e., β j ∈ ıR for j = 0, the energy dissipation coefficient is computed as and corresponds to the energy dissipation coefficient D ± (Eq. (33)) if both grazing and bias flows are absent.
(a) Figure 9: Split of the domain Ω into the two semi-infinite waveguides Ω ± and the multi-perforated liner section Ω c . Impedance transmission conditions on the interface Γ liner approximate the behaviour of the many perforations.
Conclusions
It has been shown that impedance conditions with one numerically computed parameter -the effective Rayleigh conductivity -can predict well the dissipation losses of acoustic liners. The effective Rayleigh conductivity can be obtained by solving numerically an instationary Stokes problem in frequency domain of one hole with a scaled viscosity in an characteristic infinite domain with prescribed pressure at infinity. For the computation the infinite domain is truncated, where we propose approximative boundary conditions on the artificial boundaries and an extrapolation procedure to save computation time. We decoupled in a systematic way the effects at different scales and derived impedance conditions for the macroscopic pressure or velocity based on a proper matching of pressure and velocity at the different scales. In difference to a direct numerical solution for the acoustic liner the overall computation effort is separated into a precomputation of the effective Rayleigh conductivity and a computation of the Helmholtz equation for the pressure with impedance conditions, where no holes have to be resolved anymore by a finite element mesh. The comparison with measurements in the duct acoustic test rig with a circular cross-section at DLR Berlin show that the dissipation losses based on the impedance conditions with effective Rayleigh conductivity are well predicted. The derivation of the impedance conditions do not depend on the cylindrical shape of the liner and can be used for others shapes like rectangular profiles. The procedure for the computation of the effective Rayleigh conductivity can not only be extented to include thermic effects that are currently only heuristically incorporated, but also nonlinear effects inside the hole that lead to an interaction of frequencies.
|
2018-01-16T13:39:07.000Z
|
2018-01-12T00:00:00.000
|
{
"year": 2018,
"sha1": "8c1614fb92d1a5855a88df25d2a46a4bd90d112e",
"oa_license": "CCBY",
"oa_url": "https://mathematicsinindustry.springeropen.com/track/pdf/10.1186/s13362-018-0057-0",
"oa_status": "GOLD",
"pdf_src": "Arxiv",
"pdf_hash": "8c1614fb92d1a5855a88df25d2a46a4bd90d112e",
"s2fieldsofstudy": [
"Physics",
"Engineering"
],
"extfieldsofstudy": [
"Physics",
"Mathematics"
]
}
|
90640577
|
pes2o/s2orc
|
v3-fos-license
|
Probiotic profiling of Leuconostoc species isolated from a traditional fermented cassava product
The various properties of lactic acid bacteria have made them good auxiliary in the manufacturing process in agro food industries and farms. They are widely used as probiotics which can be defined as living microorganisms that have beneficial effects on human health probiotics could replace antibiotic growth promoters in livestock without creating new threats such as that observed with antibiotics. Before their use as probiotics lactic acid bacteria require a perfect knowledge in view to their biochemical and genetic characteristics because it is difficult to differentiate morphologically some Leuconostoc and Lactobacillus strains using morphological characteristics. This study was undertaken in order to evaluate the probiotics potential of Leuconostoc strains isolated from traditional fermented cassava. The results showed that 5 strains of Leuconostoc have antibacterial activity against Staphylococcus aureus (MetiR), Klebsiella pneumoniae (BLSE), Escherichia coli, Salmonella typhimurium and Pseudomonas aeruginosa. The molecular identification of species using the conserved region of the 16S rRNA helped distinguish the species Leuconostoc mesenteroides. All these results showed that the studied Leuconostoc strains could be used as potential probiotics for the biopreservation of various foods.
INTRODUCTION
Probiotics are defined as living microorganisms that have beneficial effects on human health (FAO/ WHO, 2002).Indeed, several studies have demonstrated that probiotics may enhance growth performance, immunity and disease resistance (Saxelin et al., 2005;Ezendam and van Loveren, 2006;Ström-Bestor and Wiklund, 2011).Despite the numerous definitions, the criteria to select probiotic strains are total safety for the host, resistance to gastric acidity and pancreatic secretions, adhesion to epithelial cells, antimicrobial activity, inhibition of adhesion of pathogenic bacteria, evaluation of resistance to antibiotics, tolerance to food additives and stability in the food matrix (Soccol et al., 2010).In addition, the functional properties of probiotics include hypocholesterolemic activity by lowering plasma cholesterol, preventing and treating diarrhoea (Liong and Shah, 2005).The mechanisms by which probiotics exert their beneficial effects on the host include the reduction of luminal pH, competition with pathogens for adhesion sites and nutritional sources, secretion of antimicrobial substances, toxin inactivation, and immune stimulation (Salminen et al., 2004).The most commonly used probiotics are the strains of lactic acid bacteria such as Lactobacillus, Bifidobacterium and Streptococcus (Streptococcus thermophilus); the first two are known to resist gastric acid, bile salts and pancreatic enzymes, to adhere to colonic mucosa and readily colonize the intestinal tract (Fioramonti et al., 2003).
The use of selected probiotics from alternative sources known as "unconventional sources" is likely to increase.These unconventional sources include non-intestinal sources and non-dairy fermented food products, such as traditional fermented foods, traditional fermented drinks, vegetables, and fruit juices (Ramirez-Chavarin et al., 2013;Siddiqee et al., 2013).In Côte d'Ivoire, cassava (Manihot esculenta) root is traditionally fermented into a traditional microbial starter called "mangnan" that is used to prepare a really appreciated food called "attiéké" defined as fermented and steamed semolina cassava (Assanvo et al., 2006;Dje et al., 2008).Studies on this traditional microbial starter showed that the dominant microflora consists of lactic acid bacteria (LAB) (Assanvo et al., 2006).LAB, a group of Gram-positive, non-spore forming, non-motile microorganisms can produce inhibitory compounds such as lactic acid, bacteriocin and hydrogen peroxide preventing the growth of harmful microorganisms.The role of LAB in improving the shelf life and nutritional quality of fermented foods and beverages, controlling diarrhea, as well as their antimicrobial properties have also been established.However, despite an increasing interest in LAB, there is a paucity of literature regarding novel and emerging uses of LAB as probiotics, especially from traditional African fermented foods.Thus, the objective of the current study was to characterize the potential probiotic properties of Leuconostoc species isolated in previous studies (Coulibaly et al., 2016) from traditional fermented cassava.
LAB and indicator strains
The microbial strains used in this study are shown in Table 1.Five (05) LAB belonging to Leuconostoc genus, previously isolated from traditional fermented cassava (Coulibaly et al., 2016)
Antimicrobial activity determination
The method of agar spots as described by Larpent-Gourgaud et al. (1997) was used to evaluate the antimicrobial activity of the selected LAB strains.For this Petri dish, MRS agar was spotted with a 24-h colony of the LAB strain.The plates were seeded at 37°C for 24 h.At the same time, the indicator strains were subcultured in BCC broth for 3 h and then isolated on selective agar and incubated at their optimum growth temperature for 18 h.Each indicator strain was suspended in 2 ml of 0.85% NaCl and then vortexed.The OD was adjusted to 2.5 Mc Farland.Then, the inoculum was obtained by mixing 1 ml of inoculum of each strain in 9 ml of physiological water.The boxes with the spots were inoculated by flooding and the dishes were observed after 24 h of incubation at 37°C.The size of the zones of inhibition produced was measured.
Morphological, physiological and biochemical identification
The Gram characteristics of the isolates were determined using light microscope (Leica DM 1000, France) following staining.LAB are known to be Gram-positive.Cultures were grown in appropriate MRS media at 37°C for 48 h under anaerobic conditions.Cells from fresh cultures were used for Gram staining.
The determination of fermentation profiles (heterofermentative or homofermentative) was performed by inoculating 10 ml MRS broth containing bell Durhams at 28°C for 24 h (Harir et al., 2009).
For the fermentation of sugars, MRS media broth without glucose and supplemented with bromocresol purple as indicator was used (Mannu et al., 2000).For this, 9 ml of the medium is left in test tubes and sterilized for15 min.One milliliter of the sugar solution (10%, p/v) was aseptically added after filtration on a membrane with 0.45 µm porosity.Incubation was performed at 28°C for 48 h.Tested sugars were arabinose, sucrose, fructose, trehalose and esculin (Garvie, 1983).
The method described by Leveau et al. (1991) was used to determine the dihydrolase arginine (ADH), while the ability to grow at 10, 37 and 44°C was determined after incubation of inoculated Petri dishes at these temperatures.
DNA extraction and 16s rDNA amplification
A colony from culture was resuspended in 300 µl pure water in 1.5 ml Eppendorf tubes.DNA isolation and purification were realized using Instagen Matrix kit (Bio-Rad, USA) according to the manufacturer's instructions.PCR targeting the 16s rDNA of LAB were done as described by Bayane et al. (2006).The amplification reactions of 16S rDNA region were realized in a final volume of 50 µl containing 1X Master Mix (5 PRIME HotMasterMix, 5PRIME), 1 µl of DNA template (approximately 50 ng), 1 µM of each primer (Table 2) and PCR grade water.
The amplification was carried out in a 2720 ABI thermalcycler (Applied Biosystems, Syngapore).PCR conditions started with an initial denaturation at 94°C for 2 min; 36 cycles consisting of 1 min denaturation at 94°C for 30 s, annealing at 58°C and elongation at 65°C for 2 min.Then, a final extension for 7 min at 65°C ended the PCR reaction.PCR products were revealed in 1% agarose gel electrophoresis containing ethidium bromide (1 µg/ml).
Sequencing of the amplified DNA and data analysis
PCR products were sent to a Company (GATC BIOTECH, GERMANY) for sequencing.Sequences obtained were compared to those listed in Genbank (National Center for Biotechnology Information) using the nucleotid BLAST 2.5.0 tool (Zheng et al., 2000;Aleksandr et al., 2008).Similarity percentages were determined between the isolated sequences in this study and the closest sequences listed in GenBank.Sequences were considered similar when they have at least 99% percentage of similarity.Phylogenetic constructions were done after re-alignment of the sequences using MEGA 7.0.14(Kumar et al., 2016).The maximum likelihood and UPGMA algorithms (USA) were chosen for trees construction.
Antibacterial activity
Among the nine LAB strains tested for antibacterial activity in solid medium, only 5 strains (BL1, BL7, BL39, BL44, and BL61) showed good activity against all the indicator strains.The inhibition diameters measured are superior or equals to those of reference strains (L.plantarum, E. faecium, and W. confusa) (Table 3).
Inhibition diameter is denoted positive when greater than 8 mm (Schillinger et al., 2001).BL39, BL44 and BL61 strains inhibited the growth of all indicators Gram positive strains Staphylococcus hemolyticus MetiR and Staphylococcus aureus ATCC 25923.
Bacteria of the genus Leuconostoc, in association with the mesophilic LAB are capable of inhibiting the growth of pathogenic microorganisms such as S. aureus, Klebsiella pneumoniae BLSE, Escherichia coli, Salmonella Typhimurium and reference test strains, E. coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853.Indeed, Todorov and Dicks (2005) have shown the growth inhibitory of E. coli and P. aerugionosa by a product of Leuconostoc mesenteroides subsp.mesenteroides.The antagonistic action of LAB against pathogens such as Salmonella, S. aureus, and E. coli was also confirmed by studies of Makras et al. (2006).
The diameters of inhibition of the indicator strains S. aureus ATCC 25923, E. coli ATCC 25922 and P. aeruginosa ATCC 27853 were higher than those of the three reference strains.The strain BL44 showed a strong et al. (2005).
The antagonist effect as shown is this study is related to the biosynthesis of inhibitor compounds observed in strains of lactic bacteria (Servin, 2004).Indeed, organic acids are able to acidify the cytoplasm after dissociation and inhibit the cellular enzymatic activity of acid-sensitive pathogens (Tou et al., 2006;Djéni et al., 2008).This decrease in pH can therefore affect the viability of pathogenic bacteria (Bruno et al., 2002;Servin, 2004).This inhibition effect may be related to a competition with nutrients.An increase in the number of LAB obtained during a probiotic treatment would make it possible to reduce the substrates available for the implantation of pathogenic microorganisms (Fooks and Gibson, 2002).This justify the fact that all bacteriocins produced by LAB antimicrobial activity against Gram + (Dortu and Philippe, 2009).
Leuconostoc producing bacteriocin can be found in different food products including meat, cereals and milk (Hastings et al., 1994;Wulijideligen et al., 2012).
Morphological, biochemical and physiological characteristics
Morphological, biochemical and physiological examinations of the strains with antibacterial activity are shown in Table 4. Bacterial cells were spherical grouped into short chains and diplococcic.All strains were heterofermentative after 24 h.None of strains possessed hydrolase arginine.
Phylogenetic analysis and identification
The PCR targeting the 16 rDNA showed that all the strains that have demonstrated antibacterial activity belong to the lactic bacteria family.Positive samples detected highlighted a 1500 bp PCR product on agarose gel (Figure 1).
Sequences analysis and phylogeny
The 5 Leuconostoc PCR 16 S rDNA products were sequenced.The BLASTn's results confirmed that all the 5 isolates BL1, BL7, BL39, BL44 and BL 61 were genetically related to L. mesenteroides with 99% identity.The multiple sequence alignment of the Ivoirian Leuconostoc strains with other isolated strains from other region of the world show that all the 5 Ivoirian strains presented identical sequence.The genetic tree analysis showed clearly that those strains form a new genetic group related to L. mesenteroides on the basis of Blast results but are genetically distant (Figure 2).The 16S rRNA gene sequences of BL1, BL7, Bl39, BL44, and BL61 where deposited in GenBank nucleic acid sequence database under accession number KM518656 to KM518660.
The evolutionary history was inferred by using the Maximum Likelihood method based on the Tamura-Nei model (Zheng et al., 2000).The tree with the highest log likelihood (-2249.2146) is as shown in Figure 2. Initial tree(s) for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the Maximum Composite Likelihood (MCL) approach, and then selecting the topology with superior log likelihood value.The tree is drawn to scale, with branch lengths measured in the number of substitutions per site (next to the branches).The analysis involved 22 nucleotide sequences.Codon positions included were 1st+2nd+3rd+Noncoding.There were a total of 515 positions in the final dataset.Evolutionary analyses were conducted in MEGA7 (Aleksandr et al., 2008).
Conclusion
In conclusion, it can be stated that strains of LAB that have antibacterial properties belong to L. mesenteroides.In view to their strong inhibitory properties on P. aeruginosa and on methicillin-resistant S. aureus, these strains could be used as biopreservatives of various foods.Furthermore, a better knowledge of other criteria for selecting the probiotics associated with technological properties would make them potential candidates for the formulation of ready to use probiotics
CONFLICT OF INTERESTS
The author(s) have not declared any conflict of interest.
Figure 2 .
Figure 2. Molecular phylogenetic analysis by Maximum Likelihood method of Leuconostoc strains isolated from traditional fermented Cassava.
Table 1 .
LAB and indicators strains.
Table 2 .
Features of primers used for PCR.
Table 3 .
Diameter of inhibition (mm) of indicators strains.
Table 4 .
Biochemical characteristics of the selected Leuconostoc strains.
inhibition (60 mm) of growth of the P. aeruginosa ATCC 27853.All strains diameters inhibition of E. coli and Staphylococcus were similar to those obtained by Labiou
|
2019-01-07T00:45:21.439Z
|
2017-03-14T00:00:00.000
|
{
"year": 2017,
"sha1": "1540311692cc55377d94fdc8fca44edb5ba96732",
"oa_license": "CCBY",
"oa_url": "https://academicjournals.org/journal/AJMR/article-full-text-pdf/FC455AB63154.pdf",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "1540311692cc55377d94fdc8fca44edb5ba96732",
"s2fieldsofstudy": [
"Biology",
"Agricultural And Food Sciences"
],
"extfieldsofstudy": [
"Biology"
]
}
|
55200891
|
pes2o/s2orc
|
v3-fos-license
|
Ir ) rationalism : At the cross-roads of historical and systematic reflection
Copyright: © 2013. The Authors. Licensee: AOSIS OpenJournals. This work is licensed under the Creative Commons Attribution License. This article is dedicated to Ponti Venter for his contribution to the historical roots and systematic implications of philosophical problems. A discussion with him about four decades ago prompted me to investigate the Greek roots of our distinction of thought and being. In the analysis below, a brief sketch was given of the initial identification of thought and being in the thought of Parmenides and the consequences it had for the rationalistic tradition since the Renaissance, particularly in connection with the view that the universe itself has a rational structure. Two options were pursued in our analysis of rationalism: (1) to contrast it with empiricism and (2) to relate it to universality and the problem of what is individual. By distinguishing between conceptual and concept-transcending knowledge, an alternative systematic characterisation of rationalism (and irrationalism) is proposed, namely that it absolutises conceptual knowledge (whilst irrationalism deifies concepttranscending knowledge). This view allows for an acknowledgement of the ontic horizon of human experience, co-constituted by the dimensions of modal aspects and type laws, without elevating human understanding to become the law-giver of the world.
Introduction
It is a privilege to contribute to the Festschrift -a special issue of Koers, dedicated to the philosophical work of Ponti Venter.About 40 years ago we attended the same 'Werk Colleges' (seminars of Professor Van Riessen) in Amsterdam (Free University) as part of the doctoral program that we both followed.What struck me from the beginning was the piercing way in which Ponti always went to the root of an issue and then produced a challenging stance.
Orientation
This contribution aims at exploring something contemplated in a discussion after one of Van Riessen's seminars.It will serve as a starting point for a classical philosophical problem, namely the relationship between thought and being and the facilitating role this issue played in coming to terms with the nature of and difference between rationalism and irrationalism.This angle of approach will pay attention to issues within the domains of systematic philosophy and the history of philosophy.Ponti also grappled with these issues.Our attention will be drawn to the distortions giving birth to rationalistic and irrationalistic orientations.
Scan this QR code with your smart phone or mobile device to read online.
Read online:
philosophy onwards was related to reflections on the nature of space.As an alternative to 'continuous extension' Ponti proposed the idea of 'position' ('place').The immediate effect of this conversation was that it prompted me to return to the sources of Western philosophy, namely to the early Greek philosophers.
The general habit of saying that rationalism is an absolutisation of (human) reason is not very illuminating, because the term 'reason' is not explained.What is reason?And what is rationality?Bernays introduced a meaningful consideration when he claimed that the crucial feature of rationality is found in the conceptual element (Bernays 1974:601).This remark opens up the possibility to account for the nature of concept formation, which, in turn will lead us on the path to a number of related concerns.
It will turn out that concepts are dependent upon distinguishing and identifying universal features.Universality, however, is nowhere encountered 'on its own', for it is always accompanied by what is individual.Yet, if concepts are constituted by universality, then it is evident that the individual side of entities -normally captured in the article 'this' -exceeds the grasp of concepts.The awareness of this chair (its individual side) unbreakably belongs to the being a chair (universal side) of this chair.Since Aristotle, the dominant legacy in Western thought holds that knowledge coincides with conceptual knowledge, which implies that what is individual cannot be known.
The mistaken restriction of knowledge to conceptual knowledge may be seen as a key characteristic of rationalism.Once this content is given to the term rationalism, the natural question of how knowledge is restricted to what is individual should be designated.It will be argued that this is just the other side of the rationalistic coin, because one can define irrationalism as an over-estimation of knowing what is individual and unique.The contrast between universality and what is individual implicitly explores key features of the aspects of number and space, for what is unique and individual highlights the quantitative intuition of being distinct.Within every multiplicity discrete individuals are discernible.Likewise, the awareness of universality rests on our intuition of space ('place' in the above-mentioned conversation between Ponti and myself), in the specified form of all places, everywhere.These distinctions will also appear to be significant for comprehending the elevation of human (conceptual) understanding to be the (formal) law-giver of nature (Immanuel Kant).The law conformity of entities is simply the correlate of the universality of laws for the existence of (natural and social) entities.It will therefore also be argued that order and orderliness (both universal traits) are not themselves rational, although they are intelligible.Historicism relativised universality with its focus on the unique and unrepeatable, thus simply advancing what we shall identify as an irrationalistic alternative to the dominant rationalism of the Enlightenment.
Against this background a few systematic considerations will be introduced, centred in the distinction between two kinds of knowledge: conceptual knowledge and concept-transcending knowledge.Proceeding from a non-reductionist ontology this alternative will attempt to rescue the soundness of these two kinds of knowledge and thus contribute to a more articulated understanding of both rationalism and irrationalism.The aim is thus to arrive at a non-distortive and non-reductionistic ontology.
We proceed now by first looking at some of the Greek roots of the problem of (ir)rationalism.This will provide us with background knowledge for the argument that space does play a key role in concept formation as well as in the definition of rationalism.Without an awareness of space, universality (at all places, wherever, everywhere) will not be understandable.
Space (place), multiplicity and movement
Parmenides, the Greek Eleatic philosopher, focused his philosophical reflections on (static) being and thinking.Regarding being he held the view that a thing is its place.The property 'place' is only applicable to a (material) body and when a body is absent there is no subject to which the feature 'place' can be attributed.This explains why Greek (nature) philosophy did not acknowledge an empty space -if the body, which is supposed to be identical to its place, is not present, then it is plain that there simply is no place (equal to no body).What developed here was a metaphysics of spatial place, identified by Parmenides with thinking: 'for thinking and being is the same' (Diels & Kranz 1959-1960, B Fragment 3).In the second Fragment it is asserted that what is is and that non-being is not.The space metaphysics in the thought of Parmenides acquired a fuller content when being is described as unborn and immutable, for in the now it is together present as a whole, as one, and as something coherent (continuous) (Diels & Kranz 1959-1960, B Fragm. 8:3-8).The remarkable fact is that although being is characterised as a coherent whole, its unity, which is fully homogenous, cannot be divided (Diels & Kranz 1959-1960, B Fragm. 8:23).Parmenides explored this space metaphysics in his attack both on multiplicity and on movement.
Yet it was Zeno, a member of the school of Parmenides, who first realised that wholeness (being a totality) can be divided and in fact entails the feature of an infinite divisibility (see Diels & Kranz 1959-1960, B Fragm. 3).The two parts of this Fragment both proceed from the assumption that there is a multiplicity, but they reach opposite conclusions.
Aristotle explains discreteness in terms of things in succession such that 'there is nothing of their own kind intermediate between them'.That 'which is intermediate between points is always a line' (Phys.231 b 7-10 in Aristotle 2001:316).A line is therefore constituted by the pattern of point-linepoint-line-point-line ….From this Aristotle concludes that a continuous whole is infinitely divisible: 'Moreover, it is plain that everything continuous is divisible into divisibles that are infinitely divisible ' (Phys. 231 b 15-18 in Aristotle 2001:316).
In the first case there must be as many things as there are, in which case their number is limited, and in the second case there are always others in between such that the number of existing things is unlimited.Fränkel is justified in using the whole-parts relation to explain what is here at stake (Fränkel 1968:425ff): if one starts from the (multiple but limited number of) parts and then proceeds to the whole that is constituted by this limited number of parts, then they add up to the whole.But if the different parts are further divided, this process of division can proceed indefinitely, and then the multiplicity is unlimited. 1
Reason and reality
Throughout the history of philosophy and also within the various academic disciplines the dominant theoretical frameworks -since Thomas Kuhn, preferably designated as paradigms -struggled to articulate their respective understandings of the unity and diversity of reality.The word 'reality' is related to the Latin term for a thing (res).One of the misleading consequences of this connection is that existence is sometimes only ascribed to concrete entities (things).Whatever else there may be does not truly 'exist'.This approach surfaced in particular during and after the Renaissance: just compare the representative conviction of Descartes, who holds that number and all universals are modes of thought (Principles of Philosophy, Part I, LVII in Descartes 1965:187).Descartes does acknowledge certain attributes as existing within things, whilst at the same time reifying certain aspects to substances, namely his wellknown res extensa and res cogitans (extended substance and thinking substance).
Whilst the substance concept largely dominated ancient Greek and Medieval philosophy, the rise and development of the modern natural sciences shifted focus towards functional relationships between things.Naturally this was enhanced by the mathematical formulation of natural laws and lawconforming relations, which even prompted scholars to think that the universe itself has a rational structure.
Van Huyssteen refers to Davies who believes that the 'true miracle of nature' is found 'in the ingenious and unswerving lawfulness of the cosmos' (Van Huyssteen 1998:66).Van Huyssteen accepts this view by declaring: 'What is astounding, however, is to what a great extent our world is truly rational, that is, in conformity with human reason (p.68).Three pages further he elaborates on this view: It is indeed fascinating to see, precisely through the fact that the rational nature of our universe is reflected in its basic mathematical structure, that Davies ultimately comes to the point where he has to acknowledge the limits of this reasonableness.(Van Huyssteen 1998:71) 1.Bertrand Russell concedes: 'The relation of whole and part is, it would seem, an indefinable and ultimate relation' (Russell1956:138).
The crucial point in this stance is given in the idea that our world is truly rational in the sense that it is in conformity with human reason.To be in conformity with human reason presupposes that human reason acquired the status of being a law for the world which is conforming to this law.This view asserts that human understanding actually operates as a law (-giver) for the world.Although Van Huyssteen formulated this conviction within the context of his postmodern orientation, it is actually derived from the rationalistic tradition since the Renaissance -the line of Descartes, Hobbes up to Kant.The nominalistic denial of universality outside the human mind caused Descartes to hold that number and all universals are modes of thought (Principles of Philosophy, Part I, LVII inDescartes 1965:187).Kant defended the radical view that human understanding is the law-giver of nature, whilst it was already within the thought of Hobbes that the new (modern) motive of logical creation surfaced.Let us explain this in more detail.
The thought experiment conducted by Galileo eventually contributed to Kant's idea that human understanding is the formal law-giver of nature.In 1638 he imagined a body in motion on a path extended into infinity.According to him -and this argument constitutes his law of inertia -such a body will continue its movement if nothing hampers it.This prompted Immanuel Kant to contemplate the question: how is it possible that one can produce a thought experiment merely proceeding from the spontaneous subjectivity of human thought and then deduce from it a natural law and apply it to moving bodies?Does this experiment entail that human knowledge serves as the law-giver of nature and that our understanding contains structural features preceding our experience?
This is what traditionally was designated as what is a priori to experience, that is, what precedes experience.Kant did not hesitate to affirm the claim that there are a priori elements of human understanding making possible our knowledge of reality.
The position assumed by Kant is an attempt to reconcile the extremes of assigning priority to human thinking or assigning it to sensory experience.An important element in the tradition of rationalism is found in the conviction that there are cognitions that are not derived from the senses since they find their origin purely within conceptual human understanding.Traditionally the basic opposition was between rationalism and empiricism, where the latter gave priority to our experience. 2By implementing the classical opposition between form and matter (content) Kant delivered the matter of human sensibility to a multiplicity of (individual) chaotic impressions (thus accommodating the empiricist tradition of Locke, Berkeley and Hume and continuing the nominalistic stripping of reality from its universal order and law-conforming orderliness), whilst at the same time postulating that the formal side represents something 2.Ponti Venter often refers to both rationalism and irrationalism and his writings evince the consistent interplay of historical and systematic considerations (see, for example, Venter, 1995Venter, , 1999Venter, , 2002Venter, , 2012)).
supra-empirical.This applies first of all to what Kant called the forms of intuition (time and space) and secondly to the 12 categories of human understanding distinguished by Kant.His view that human understanding creates its laws (a priori) not out of nature, but prescribes them to nature (Kant 1961:320; § 36) 3 explains why it is still alleged that the world has a rational structure.Because this view elevates the human subject to the level of law-giver it is also designated as subjectivistic.However, such a view appears to contain an untenable circularity, aptly captured by Roy Clouser: 'Unless there were already laws governing the mind that were not its creations, what would explain the uniformity of the ways the mind imposes laws on experience?' (Clouser 2005:368).
In post-Kantian freedom idealism Fichte, Schelling and Hegel elaborated the rationalistic element by ascribing to human thought the capacity to conceptually bring forth the content of the world independent of experience.According to Hegel the idea is the unity of concept and reality (Hegel 1949:239). 4This position once more evinces the perennial philosophical problem of thought and being.Cassirer remarks that it appears as if the circle of philosophical contemplation found its closure and reached its aim in identifying reality and reason: 'Hegel believed that this is the point where his "Wissenschaft der Logik" is positioned' (Cassirer 1957:10).
Understanding rationalism: Two options Option 1
The one option is to juxtapose thought and experience (thinking and sensing) -when the former acquires priority it is designated as rationalistic and when the latter is given priority it is known as empiricistic.Windelband mentions Hobbes (De corpore, Ch. 6), who accepted the methodological orientation of Galileo, as being a rationalistic in opposition to the empiricism of Bacon (Windelband 1935:327, note 1).This opposition may also be explained with reference to the alternative views of Descartes and Hume.In a typical rationalist mode of thought Descartes holds: 'At all events it is certain that I seem to see light, hear a noise, and feel heat; this cannot be false, and this is what in me is properly called perceiving (sentire), which is nothing else than thinking'(Meditation II; Descartes 1965:90, my italics).The contrasting view of Hume is: 'To hate, to love, to think, to feel, to see; all this is nothing but to perceive' (Hume 1962:113).Either sensing is reduced to thinking or thinking (and everything else) is reduced to sensing.
Although Greek philosophy already gave prominence to reason (nous) the term rationalism was not employed yet.Anaxagoras characterised the eternal and immaterial form origin of the world as the Nous, which has all knowledge and the greatest power.It is not determined by any limits and it 3.Cassirer explains this view of Kant with reference to the following statement of the latter: 'das wir "von den Dingen nur das a priori erkennen, was wir selbst in ihr legen"' [concerning things we can only know a priori that which we ourselves have embedded in them] (Cassirer 1957:239).
4.Hegel phrases it also as follows: 'Idea is the unity of concept and reality, the realized concept as such' (Hegel 1931:155).
is not mixed with germs of matter because it is self-sufficient (Diels & Kranz 1959-1960, B Fragment 6, 11, 12, 14).Later on, when Plato compared the good with the sun, it becomes clear that the ontic forms (eidè) owe their existence, being and knowability to the idea of the good.Viewed in coherence with the nature of the divine workman of Timaeus the idea of the good (agathon) has its seat in the divine demiurg (work master/ craftsman).The active operation of the nous with its focus on ordering is portrayed by Jäger as focused on the relation of the nous to the agathon.This illustrates the influence of Anaxagoras (and Socrates, nous and dynamics): through the primordial design of the idea tou agathou the workman, as the divine nous, is the origin of the eidè, and therefore the formgiver of the world of the senses (see Jäger 1967:106ff). 5 Thomas Aquinas held that the unity of ideas cannot be obtained by assuming just one idea in God.Rather, he holds, the multiplicity of ideas is constituted in such a way that the Divine Mind, as first form, observes the multiple possible ways according to which it can be copied (Questiones Disputate de Veritate, III, 2 in MacKenna 1956).
The first time the word 'rationalist' (rationaliste) was explicitly used was apparently in 1539 when it was already employed in opposition to experience (the empirical: empirique).A rationalist was seen as a person who assigned a greater value to pure thought than to experience.For some time, since the beginning of the 17th century, the term rationalism acquired a more specific theological meaning.An English source from the year 1646 mentions the rationalists as a new sect within the sphere of the Presbyterians and Independents whose adherents accepted their own reason as their point of departure (Böhling 1992:45).
At the time when the term rationalism was liberated from its theological use, it opened the way to describe the great philosophical systems of the 17th and 18th centuries.Simultaneously, atheists and those liberated in their thought introduced the term rationalism during the 19th century as a slogan against the superstition of traditional religions, whilst believing that they had reason and science on their side.Historians now also extrapolate the meaning of this term by generalising it mainly into the epoch of the Enlightenment.Rationalism combines the key elements of the Enlightenment, such as its criticism of understanding, its optimism and its faith in progress (Böhling 1992:46; see also Lecky 1910).
Perhaps this insight may help us also to understand why Habermas, in his desire to make an ongoing plea for the dignity of rationality as it is embedded in communicative action, does not really want to transcend this essential trait of modernity, although he is willing to let go of the optimistic utilitarian spirit that marked its emergence in the period of Enlightenment: 5.In his negative theological approach Plotinus feared that the word demiurg may introduce multiplicity into the absolute One (first hypostasis).As a result he rather applies the term demiurg as designation of the Nous (second hypostasis).See Theiler (1966:127) and also Plotinus (En.1I, 3, 18; II, 9, 6; V, 9, 3; MacKenna 1956).
The concept of modernity no longer comes with a promise of happiness.But despite all the talk of postmodernity, there are no visible rational alternatives of this form of life.What else is left for us, then, but at least to search out practical improvements within this form of life?(Habermas 1994:107).
Earlier in the 20th century we find a remarkable trust in human rationality.After Heidegger moved beyond the rationalism of Husserl, the latter experienced it with a sense of hopelessness positioned by him within the broader context of a crisis of Europe and of the academic disciplines.According to him this crisis is rooted in a misguided rationalism ('einem sich verirrenden Rationalismus'): In order to comprehend what is wrong in the present crisis the concept Europe once again has to be viewed by means of the historical directedness towards the infinite aims of reason; it must be demonstrated how the European world was borne from reason-ideas, that is, out of the spirit of philosophy.The crisis will then clearly emerge as the apparent failure of rationalism.The basis of this failure of a rational culture, however, … is not inherent to rationalism, since it is only found in its externalization, in its decay into naturalism and objectivism.The crisis of European existence provides only two options: the decline of Europe in the alienation from its own rational existential meaning, the decay into an animosity towards the spiritual and a lapse into barbarism, or the rebirth European existence through the spirit of philosophy, particularly through a heroism of reason that will consistently triumph over naturalism.(Husserl 1954:347-348) Option 2 The position of Our age is, in every sense of the word, the age of criticism and everything must submit to it.Religion, on the strength of its sanctity, and law on the strength of its majesty, try to withdraw themselves from it; but by doing so they arouse just suspicions, and cannot claim that sincere respect which reason pays to those only who have been able to stand its free and open examination.(Kant 1961:21) The rise of historicism during the transition from the 18th to the 19th century made possible a new appreciation of what is unique, individual and contingent as an attempt to escape from the grip of conceptual knowledge and its entailed universality.In the Third Part of the Fourth Volume of his work on the problem of knowledge within philosophy and the modern sciences, Ernst Cassirer dedicates the first chapter to the emergence of historicism and the second one to the ideas of B.G. Niebuhr, Leopold von Ranke and Wilhelm von Humboldt (see Cassirer 1957:225ff, 233ff).Cysarz characterises the 19th century as the era of individuality (Cysarz 1983:16).He could have added the issue of continuous change, with reference to Darwin's theory of evolution.
During the 19th century, the tension between the general nature of concepts and what is uniquely historical could not be reconciled.Habermas mentions that, analogous to Peirce, Dilthey also struggled with the relationship between universality and what is individual (see Habermas 1970:200-201).He explains that hermeneutical understanding must grasp an inexpressible individual meaning in categories that are unavoidably general (Habermas 1970:201). 6 Yet, as Cysarz correctly remarks, every (professional) practice must proceed in an individual manner.What is healed in medical praxis is not illness, but the individual sick person, even though this cannot happen outside the matrix of general norms (standards) (Cysarz 1983:13)
The challenge facing a systematic account of rationalism
Our preceding analysis made it clear that an account of the nature of an ism such as rationalism cannot avoid taking into account the relevant historical perspectives.In pursuing this avenue particular distinctions emerged, such as that between the 'rational' and the 'empirical '. Poser (1998) relates this distinction to the difference between Descartes and Leibniz in his discussion of what he calls the 'rationalistic ideal of science'.Leibniz criticised Descartes by introducing his concept of a scientia generalis (general science).However, he articulated a distinction between necessary truths of reason and contingent truths of fact (Leibniz 1976:646;Monadology, 33ff).A similar distinction is found in the thought of Locke, namely that between empirical factual knowledge and knowledge of the necessary eternal relations between ideas (Locke 1964:324-326;IV, i, 9), as well as his introduction of intuition as the foundation of exact scientific knowledge (for instance in the demonstrations in mathematics, cf.Locke 1964:330-331;IV, ii, 14-15).This distinction contradicts his empiricist orientation because with the aid of intuitive demonstration one can arrive at a kind of knowledge that 'is the clearest and most certain human frailty is capable of' (Locke 1964:326).
The position assumed by Leibniz anticipated the views of Kant: just consider the following statement of Leibniz (1965): Now reflection is nothing but an attention to what is in us, and the senses do not give us what we already bring with us.This being so, can we deny that there is a great deal that is innate in our mind, since we are innate, so to speak, to ourselves, and since there is in ourselves being, unity, substance, duration, change, activity, perception, pleasure, and a thousand other objects of our intellectual ideas?And since these objects are immediate to our understanding and are always present, ... why be surprised that we say that these ideas, and everything which depends on them, are innate in us? ' (p. 146, cf. p. 173) On the verge of anticipating Kant's criticism of Hume's empiricism, Leibniz, more than 30 years before the first appearance of Hume's A treatise of human nature (1739), wrote as follows in his New Essays: Now all the examples which confirm a general truth, whatever their number, do not suffice to establish the universal necessity of that same truth, for it does not follow that what has happened will always happen in the same way.(quoted from Stich 1975:45, cf. Kant 1787:5) Paul Bernays, the co-worker of the famous German mathematician David Hilbert, points out that rationality cannot be understood apart from concepts: the 'proper characteristic of rationality' is 'to be found in the conceptual element' (Bernays 1974:601).This view supports what we saw above, namely that in general rationality is crucially connected to conceptual knowledge and to universality.Universality, in turn, cannot be understood apart from its connection to what is individual.
Note that we rather speak of 'what is individual' than of 'individuality' because the latter is still a universal feature.
In being an individual every individual, in a universal way, evinces its uniqueness.We noted that in our everyday experience we fully understand the difference between a chair (a universal feature) and this chair (its individual side).
A systematic perspective
From a historical perspective we have directed our attention to the opposition of thought and experience which, owing to an over-emphasis of the one or the other, may lead to rationalism or to empiricism.Yet, this juxtaposition does not say anything substantial about rationalism as such.For this reason we extended our investigation in order to reflect on the context within which rationalism is positioned.This reflection pointed at the connections between rationality, conceptuality, universality and what is individual.
The tradition of reformational philosophy provides the starting point for an understanding of these issues.For this reason we briefly look at points of connection in this regard in order to advance an alternative approach in terms of the distinction between conceptual knowledge and concepttranscending knowledge.
Within the tradition of reformational philosophy, the emphasis on making sound distinctions created an equally important sensitivity with regard to reductionism in all its negative variants.Both Dooyeweerd and Vollenhoven articulated their views in terms of well-thought-out systematic distinctions and they both attached specific meanings to terms capturing those philosophical stances in which a distorted account is given of universally accessible states of affairs.
Interestingly, Vollenhoven explicitly mentions that rationalists from the 18th century, such as J.G. Walch in his introduction to philosophy of 1727, maintain 'that what he and those who thought like him -the rationalists -had to say was built on the state of affairs' (Vollenhoven 2005:5).
He briefly discusses the slogan of Descartes, cogito ergo sum (I think therefore I am), by pointing out that some are of the 'opinion that this "therefore" denotes a connection of identity'.But according to him this is incorrect, 'for Descartes does not identify being and thinking' since besides thinking he 'also presupposes extension'; therefore, for Descartes 'thinking is only a component of being' (Vollenhoven 2005:10).
Yet Vollenhoven (2005) on the same page points out that: Descartes meant this in a rationalistic way, that is, in the sense that thinking is the essence of being; an opinion that we, of course, reject, as much as we reject the division of being indicated here.(p. 10) This does not entail that Vollenhoven think it is 'rationalistic' 'to subsume thinking and knowing under being, for it makes good sense to speak of a non-thinking and a non-knowing being.There are clearly many things -for example, minerals, plants, and animals -that are but that do not know'.
Vollenhoven therefore concludes negatively 'that knowing is not the same as being' and positively 'that knowing is a component of being' (Vollenhoven 2005:10). 7 In general Dooyeweerd refers to the strict correlation of the law side (cosmonomic side) and factual side of reality in order to explain how he understands both rationalism and irrationalism.In the first Volume of his Magnum Opus, A New Critique of Theoretical Thought (NC), he states that in a rationalistic type of thinking 'the subject-side of reality within the special modal aspects is reduced to the nomosside' (Dooyeweerd 1997:98).In the context of his philosophy of time he rejects both rationalistic and irrationalistic conceptions because the former 'absolutizes the cosmonomic side and the latter the factual-subjective side of time' (Dooyeweerd 1997:28; see also Dooyeweerd 1940:196).
However, Dooyeweerd did not properly distinguish between law and law-conformity (lawfulness).He frequently exchanged these expressions.Earlier we remarked that universal conditions ought to be distinguished from the universal way in which individual entities show that they conform to these conditions or laws.By identifying law and law conformity Dooyeweerd strips factual reality of its universal side.For this reason he often explicitly speaks of the individual factual side.
Although concretely existing things and processes function within all aspects of reality, they at the same time transcend the theoretical grasp of anyone of these modal aspects in which such entities and processes function.We have theoretical access to them through the gateway of the modal aspects which therefore also serve as modes of explanation for our investigations.
Concept formation is made possible by the universality of specific (logically identifiable and distinguishable) features.
The universality here involved could be either that of a universal order for, holding for the type of entities subject to it, or it can refer to the universal orderliness of things, reflecting the universal way in which entities show that they are subjected to the applicable order for their existence.Since conceptual knowledge is tied up with these two forms of universality (order for and orderliness of), the individual side of entities simply transcends the possibilities of conceptual knowledge.When we acknowledge that what is individual exceeds our conceptual grasp, it does not imply that we do not have knowledge of what is individual.But if this knowledge cannot be conceptual, it must be concept transcending.This kind of knowledge may also be designated as idea knowledge.
Proper conceptual knowledge of a triangle or a human being includes those universal features found in all triangles and all human beings.Since this kind of knowledge does not preclude concept-transcending knowledge (idea knowledge), a balanced (and non-reductionist) understanding of reality should acknowledge both.
On the basis of this characterisation and distinction we can now introduce a more advanced definition of rationalism and irrationalism: 1. Rationalism deifies (absolutises) conceptual knowledge.2. Irrationalism deifies (absolutises) concept-transcending knowledge.
An additional perspective on idea knowledge is obtained when the various ways in which terms derived from the different modal aspects of reality are employed.The most basic option is to consider modal (aspectual) terms in their reference to phenomena appearing within the domain of any specific aspect.For example, no one will doubt that our awareness of the 'one and the many' presupposes the meaning of number.Counting a number of things usually follows a sequential pattern that exhibits an order of succession, albeit in a cardinal (how many: one, two three, etc.) or ordinal sense (how many-th: first, second, third, etc.) (see Maddy 1997:17).But when Plato discusses the hypothesis that the One is without multiplicity (subsequently further explored by Plotinus), the numerical term 'one' points beyond the arithmetical aspect towards the origin of the universe.In this case the numerical one is stretched beyond the confines of the quantitative aspect of reality, revealing that it is actually employed in a concept-transcending sense.In a similar way one may reflect on the mere conceptual use of spatial terms and ways in which spatial terms are employed in order to refer to realities exceeding the boundaries of the spatial aspect.
We have noted earlier that although Greek philosophy initially wrestled with an idea of wholeness not allowing division, the ripened conception of If all the divided parts are taken together they constitute a genuine whole or totality.A conceptual use of the terms whole or totality will refer to spatial figures, such as lines or twodimensional figures.Suppose now that we use these spatial terms in order to refer to the existence of something in all it aspects (of which the spatial is but one amongst many others).Then we may speak of that something in its totality.
The referent of this term encompasses more than the spatial aspect alone and therefore represents a concept-transcending use of a modal spatial term.
Although the relationship between what is universal and what is individual may suggest that an over-estimation of either could be captured in the opposition between universalism and individualism, 8 it is preferable to reserve the latter opposition as an indication of alternative basic denominators for the diversity of ontic aspects and entities -alongside ismic orientations such as physicalism, vitalism, psychologism, logicism, economism, aestheticism, and so on.
Dooyeweerd uses the correlation of law side and individual factual side to characterise the difference between rationalism and irrationalism.We noted that he denies universality at the factual side of reality and consistently identifies the orderliness of entities, their lawfulness, law-conformity of factual universality, with the law side of reality.This explains why Dooyeweerd consistently refers to the individual subject side of reality.However, when it is realised that law-conformity is a feature of factual reality, then it is no longer possible to deny the universal side of factual reality.In the atom-ness of an atom this (individual) atom in a (universal) way shows that it is subject to the universal law (conditions) for being an atom (where being an atom represents the universal side of an atom).But then a more appropriate understanding of rationalism is to define it as an absolutisation of conceptual knowledge.Likewise, irrationalism should then be seen as an absolutisation of concept-transcending knowledge.Of course rationalism then cannot any longer be defined as an absolutisation of the law side of reality, as Dooyeweerd did, because universality is also present at the factual side of reality.Furthermore, it is then also insufficient to define irrationalism as an absolutisation of the factual side of reality.
In other words, it seems as though it is more nuanced to advance our proposal, namely to hold that rationalism leaves no room for true concept-transcending knowledge whilst irrationalism leaves no room for genuine conceptual knowledge.In terms of this perspective an acknowledgement of the horizon of human experience and the dimensions of (ontic) modal aspects and type laws is meaningful.It at once avoids elevating human understanding to the a law-giver of the world, which ultimately is a full-blown rationalistic conviction.
Concluding remark
It is remarkable to what an extent an analysis of the nature of rationalism and irrationalism succeeds in bringing widely varying features and theoretical orientations to the fore.The uniqueness of number and space played a key role since the awareness of what is individual derives from the numerical meaning of discreteness (being distinct), whereas the awareness of universality derives from our spatial awareness of place, to be sure, all places, wherever.It appeared that we have to question the long-standing legacy in Western philosophy which restricts knowledge to conceptual knowledge.Knowledge is not blind to what is individual; it is only conceptual knowledge that cannot grasp what is individual.Yet, when we realise that knowledge differentiates into two kinds of knowledge, namely conceptual knowledge and concept-transcending knowledge, it turns out that we can avoid the one-sidedness present in both rationalistic and irrationalistic approaches.
8.
Vollenhoven uses the terms universalism and individualism to indicate an overestimation of individuality and universality, respectively.
Whatever is known 'are always stated and recognized by means of the universal formula [secondary substance].But matter is unknowable in itself' (Metaph.1036 a 1-10 in Aristotle 2001:799).Aristotle therefore ultimately holds that knowledge is restricted to conceptual knowledge, made possible by what is universal.
Husserl points in the direction of the other option open to an understanding of rationalism, which surfaces when we consider the connection between understanding and the difference between what is universal and what is individual.The problem of what is universal relates to the nature of concepts, for our argument is that concepts are bound to universal features.Yet in our everyday experience we always find universality and what is individual side by side (this horse is a horse).Since we do have knowledge of what is individual (we know ourselves in our uniqueness), the restriction of knowledge to conceptual knowledge is clearly problematic.The historically significant lines of development outlined below will elucidate this issue.thatAristotledoesnot have a concept of a concept(Hartmann 1957:101).As general form knowledge, true knowledge cannot be obtained from matter.In his Metaphysics, Aristotle first eliminates all determinations of being and then concludes that matter as such is unknowable.He explains that when it comes to (individual) 'concrete thing[s]', 'of these there is no definition'.Within the development of modern philosophy one may see the 18th century, the era of Enlightenment, as a period in which conceptual rationalism dominated the scene.It echoes the new spirit of rational criticism, exemplified in what Kant said in the Foreword to the first edition (1781) of his Critique of Pure Reason: . This remark highlights an important fact, namely that whatever is universal is always accompanied by what is individual and vice versa.An individual sick person shows, in its being sick, that it shares in the universal feature of sickness.A succinct way to capture this situation is to say that this sick person (individual side) has an illness (universal side).Clearly, sickness, even in its abnormality, displays universal abnormal traits.Disorderliness depends like a parasite on what is normal.
|
2018-12-05T07:23:25.315Z
|
2013-07-03T00:00:00.000
|
{
"year": 2013,
"sha1": "dbb4e87d294e815e72fe48b9318979051a344aa4",
"oa_license": "CCBY",
"oa_url": "https://www.koersjournal.org.za/index.php/koers/article/download/438/2270",
"oa_status": "GOLD",
"pdf_src": "ScienceParseMerged",
"pdf_hash": "dbb4e87d294e815e72fe48b9318979051a344aa4",
"s2fieldsofstudy": [
"Philosophy"
],
"extfieldsofstudy": [
"Art",
"Philosophy"
]
}
|
233346964
|
pes2o/s2orc
|
v3-fos-license
|
XMM-Newton campaign on the ultraluminous X-ray source NGC 247 ULX-1: outflows
Most ULXs are believed to be powered by super-Eddington accreting neutron stars and, perhaps, black holes. Above the Eddington rate the disc is expected to thicken and to launch powerful winds through radiation pressure. Winds have been recently discovered in several ULXs. However, it is yet unclear whether the thickening of the disc or the wind variability causes the switch between the classical soft and supersoft states observed in some ULXs. In order to understand such phenomenology and the overall super-Eddington mechanism, we undertook a large (800 ks) observing campaign with XMM-Newton to study NGC 247 ULX-1, which shifts between a supersoft and classical soft ULX state. The new observations show unambiguous evidence of a wind in the form of emission and absorption lines from highly-ionised ionic species, with the latter indicating a mildly-relativistic outflow (-0.17c) in line with the detections in other ULXs. Strong dipping activity is observed in the lightcurve and primarily during the brightest observations, which is typical among soft ULXs, and indicates a close relationship between the accretion rate and the appearance of the dips. The latter is likely due to a thickening of the disc scale-height and the wind as shown by a progressively increasing blueshift in the spectral lines.
INTRODUCTION
There is a consensus that the majority of ultraluminous Xray sources (ULXs) are stellar-mass compact objects (neutron stars and perhaps black holes) accreting above the critical Eddington rate (see, e.g., King et al. 2001, Poutanen et al. 2007Middleton et al. 2011, Kaaret et al. 2017. The spectral curvature in the X-ray band, with a characteristic downturn above ∼5 keV, and the presence of residuals at energies 1 keV when modelled with featureless continuum models, are two characteristic ULX features (see, e.g., Soria et al. 2004, Goad et al. 2006, E-mail: ciro.pinto@inaf.it Gladstone et al. 2009) that are naturally explained if the primary X-ray photons are seen through a disc wind, as expected for systems accreting in the super-Eddington regime (e.g., Shakura & Sunyaev 1973, Poutanen et al. 2007). The thickness of the wind and, as a result, the "softness" of the observed ULX spectra in the ∼0.1-20 keV band, are likely a function of two main parameters: the mass outflow rate in the wind (which is related to the accretion rate), and our viewing angle, with softer sources being observed closer to the disc plane (see, e.g., Sutton et al. 2013, Middleton et al. 2015a, Pinto et al. 2017). In particular, ULXs with a power-law photon index Γ > 2 in the ∼0.3-5 keV band are empirically classified in a "soft ultraluminous" (SUL) state.
Supersoft ultraluminous sources
Among the ULXs, a special sub-class is represented by ultraluminous supersoft sources (ULSs), also referred to as sources in the "supersoft ultraluminous" (SSUL) state (Feng et al. 2016). This state is defined by a dominant, cool blackbody component (T bb 140 eV), with very weak or completely absent hard component at higher energies. Observationally, sources in the SSUL state show essentially no photons >1 keV. Their bolometric luminosity is typically a few 10 39 erg s −1 . They should not be confused with the "classical" supersoft sources (see, e.g., Krautter et al. 1996), which are usually interpreted as nuclear burning on the surface of a white dwarf. Classical supersoft sources also have blackbody spectra, but at lower luminosities (L bb 10 38 erg s −1 ), and generally with a smaller bb radius: R bb ∼ 5000 − 10000 km, consistent with a white dwarf, while SSUL spectra can have blackbody radii as large as 10 5 km (Urquhart & Soria 2016).
Blackbody modelling of SSUL spectra at different epochs shows that the characteristic radius is not constant, which rules out a hard surface as the origin of the thermal emission (Feng et al. 2016). It also shows a clear anti-correlation between blackbody radius and temperature, which rules out a standard accretion disc. Instead, such behaviour is consistent with emission from the photosphere of an optically-thick wind. The increase in the photospheric radius corresponds to an enhancement of the wind thickness.
The apparent non-periodic spectral variability in the SSUL state (Liu 2008) is very likely due to a thickening of the wind along our line of sight. One question we want to address is whether the short-term variations in optical depth are just stochastic variability (at a given mass accretion ratė M ) due to the clumpy nature of the wind ("weather", see, e.g., Takeuchi et al. 2013), or instead each variation is driven by a change in the underlyingṀ ("climate"), which then affects the wind density and launching radius.
A deep absorption edge was detected at ≈1.0-1.1 keV in several SSUL sources (e.g., those in M 51, NGC 6946 and M 101: Urquhart & Soria 2016, Earnshaw & Roberts 2017. This feature clearly appears when the source softens, progressively losing all X-ray photons with energy above 1 keV. There are no strong absorption edges predicted at the observed energy of those dips. Viable solutions are blueshifted O viii ionisation edges (871eV at rest), which was observed in novae during their supersoft phase (e.g., Pinto et al. 2012), or a combination of high-ionisation Fe L or Ne ix-x absorption lines with velocities of ∼0.1-0.2c, supporting the case for an optically-thick, mildly-relativistic wind.
NGC 247 ULX-1
Previous X-ray studies (see, e.g., Feng et al. 2016), based on two short XMM-Newton observations, showed that NGC 247 ULX-1 switched from a supersoft state with hardly any flux above 1 keV (in 2009) to a much brighter state (in 2014) consistent with the soft end of the "standard" ULX population (i.e., those with a bright hard X-ray spectral tail).
NGC 247 ULX-1 is also among the most variable ULXs. It exhibits strong dips in its X-ray lightcurve, which last several ks, and during which the flux decreases by an order of magnitude (Feng et al. 2016). Both the low and high flux spectra are characterised by strong spectral features. Pinto et al. (2017) found similarities between the narrow X-ray spectral features of NGC 247 ULX-1 in the high flux state with those of NGC 55 ULX, using observations taken with the high-resolution gratings aboard XMM-Newton. Despite the short duration (33 ks) of the observations available, two remarkable absorption features at 7.5Å and 16.2Å were found along with other weaker features, which can be modelled with absorption from photoionised gas outflowing at ≈ 0.14c. The observation taken in the low-flux (supersoft) state is far too short to provide any useful data. Thus, we proposed and were awarded a 800 ks deep XMM-Newton programme to characterise the properties of the outflows and the origin of the spectral residuals, and correlate them with the continuum flux variability and state changes.
This paper is the first in a series of intriguing results from our XMM-Newton campaign. Here, we focus on the search for wind signatures in the time-average spectrum, taking full advantage of the high-spectral-resolution data, and on the variability of the spectral features around 1 keV. We detail our observing campaign in Sect. 2 and present the results of our spectral analysis in Sect. 3. We discuss the results in Sect. 4, and outline some conclusions in Sect. 5.
NGC 247 XMM-NEWTON CAMPAIGN
We observed the NGC 247 galaxy between December 2019 and January 2020. The roll angle was similar throughout the whole campaign and avoided strong contamination along the dispersed grating spectra from the nearby brightest Xray sources (see Fig. 1 and Appendix A1 for more detail). Seven observations were expected to occur but, owing to an issue with the RGS instrument that occurred during the last observation (id:0844860701), an additional final observation was taken shortly afterwards (id:0844860801) to recover the lost exposure. In Table 1, we report the detail of our observations. We performed the spectral analysis with the SPEX code (Kaastra et al. 1996), we used C-statistics (C-stat, Cash 1979) for spectral fits, which was proved to be efficient in comparing models similarly to χ 2 statistics (Kaastra 2017), and we adopted 1-σ confidence intervals.
Data preparation
In this work we used data from the European Photon-Imaging Camera (EPIC), the Reflection Grating Spectrometer (RGS) and the Optical Monitor (OM) aboard XMM-Newton. The primary science was carried out with the RGS which can detect and resolve narrow spectral features. The broadband cameras (EPIC MOS 1,2 and pn) were mainly used to determine the spectral continuum and cover the hard X-ray band missed by the RGS.
EPIC cameras
We reduced the 8 new XMM-Newton observations with the Science Analysis System (SAS v18.0.0, CALDB available on March, 2020). EPIC-pn and MOS data were reduced with the EPPROC and EMPROC tasks, respectively. Following the recommened procedures, we filtered the MOS and pn event lists for bad pixels, cosmic-ray events outside the field of view, photons in the gaps (FLAG=0), and applied standard grade selections (PATTERN 12 for MOS and PATTERN 4 for pn). We corrected for contamination from high background by selecting background-quiescent intervals on the lightcurves for MOS 1,2 and pn in the 10−12 keV energy band. These lightcurves were extracted in time bins of 100 s and all those with a count rate above 0.4 c/s for pn and 0.2 c/s for MOS were rejected. The MOS 1-2 and pn clean exposure times are reported in Table 1.
We extracted EPIC MOS 1-2 and pn images in the 0.3−10 keV energy range and stacked them with the EMO-SAIC task (see Fig. 1). We also extracted EPIC MOS 1-2 and pn lightcurves from within a circular region of 20" radius centred on the source position (α, δ=00:47:04.0,-20:47:45.7). We used the task EPICLCCORR, which corrects for vignetting, bad pixels, chip gaps, PSF, and quantum efficiency. The background (BKG) lightcurves were extracted from within a larger circle in a nearby region on the same chip. In order to measure the variations in the spectra hardness, we extracted EPIC-pn (for its larger effective area) lightcurves in the soft (0.3-1 keV) and hard (1-10 keV) energy bands (with 1 ks time bins to increase the signal-to-noise ratio of each bin). The PN lightcurves are shown in Fig. 2 (top-left panel) with a length above 800 ks (despite an effective clean exposure of 600ks) due to the large 1ks bin size which does not show the time gaps of high BKG lasting few 100s. The lightcurves were also glued together for displaying purposes but with vertical dotted lines separating them. The hardness ratio was defined as the fraction of hard photons with respect to the total (H/(H + S)). The boundary energy was adopted owing to the strong spectral curvature of supersoft sources exhibited above 1 keV (see Sect. 2.1.4).
We extracted EPIC MOS 1-2 and pn spectra in the same regions used for the lightcurves. The EPIC-pn spectra of the individual observations are shown in Fig. 3. We avoided over-plotting the MOS spectra for displaying purposes.
RGS cameras
The RGS data reduction was performed with the RGSPROC pipeline. We filtered out periods affected by contamination from Solar flares by selecting background-quiescent intervals in the lightcurves of the RGS 1,2 CCD 9 (i.e., 1.7 keV) with a count rate below 0.2 c/s. As usual, Solar flares affected the RGS data on a much lower level than EPIC. The total clean exposure times are quoted in Table 1. We extracted the 1 st and 2 nd -order RGS spectra in a cross-dispersion region of 0.8' width, centred on the source coordinates and the background spectra by selecting photons beyond the 98% of the source point-spread-function. The background regions do not overlap with bright sources. After inspecting the 2 nd -order spectra, we decided not to use them as they were highly affected by the background. We stacked the 1 st -order RGS 1 and 2 spectra from all observations with RGSCOMBINE and the EPIC-pn, MOS 1 and MOS 2 spectra, using EPICSPECCOMBINE. The stacking provided 4 time-averaged high signal-to-noise spectra for RGS, MOS 1, MOS 2, and pn detectors.
All XMM-Newton spectra were grouped in channels of at least 1/3 of the spectral resolution, for optimal binning and to avoid over-sampling, and at least 25 counts per bin, using SAS task SPECGROUP. This has also the advantage to smooth the background spectra in the energy range with low statistics, avoiding narrow spurious features introduced by the background subtraction. This also enabled us to check our results with the χ 2 statistics. The stacked spectra have many counts with the binning affecting only the spectral range at the rather low and high energies (outside the 0.6-1.7 keV RGS band and above 4 keV in EPIC spectra), where line detection is not crucial. We found no significant effect onto our line or continuum modelling by decreasing the binning to just 1/3 of the spectral resolution.
Optical Monitor
We used OM data to search for a possible optical/UV counterpart to NGC 247 ULX-1. To increase the signal-to-noise ration we stacked all the internally aligned full-frame sky images per filter and per observation, using the SAS tool OM-MOSAIC. Each observation contains at least one image in one of these filters: V, UVW1, UM2 and UVW2 and the final total exposures corresponded to 105 ks, 105 ks, 115 ks and 220 ks, respectively. We ran the OMDETECT task on these stacked images with a limit on the detection threshold of 2 σ. At the position of the ULX (Section 2.1.1), no source was detected in any of the filters. This is unsurprising since the ULX region is very close to a bright association of OB stars, which makes such a detection challenging. To derive an upper limit for the ULX emission in these bands, we computed the total background rate for a circular region of 6" around the position of the ULX. This provided 3 σ upper limits for the ULX flux, which are comparable to previous measurements obtained by Feng et al. (2016) using deep observations with Hubble Space Telescope (HST, UV-optical fluxes ∼ 10 −14 erg s −1 cm −2 ). In the far-UV the OM flux upper limits are slightly below the HST detections, suggesting long-term flux variability and that a substantial fraction of the UV flux originates within the accretion disc rather than the stellar companion (see Fig. A3, top panel).
A final OM image obtained by stacking the data from all the filters of all the observations is shown in Fig. 1, bottom panel, where the bright association of OB stars can be seen on the right side of the X-ray source centroid.
Data investigation
The XMM-Newton 0.3-10 keV lightcurve shows a strong dipping behaviour with the source flux approaching zero c/s during time interval of less than 15 ks, as found by Feng et al. (2016), see . The dips have variable duration, with the shortest ones being of a few hundred seconds, which was estimated extracting finer lightcurves (D'Aì et al. in prep). During the dips, the flux drops by an order of magnitude and then returns to the previous level, which makes it difficult to believe it is due to an intrinsic flux change rather than to a temporary obscuration phenomenon. This behaviour causes the multiple peaks present in the histogram of the count rates (Fig. 2, right panel), which disagree with a single-peaked log-normal trend.
The hardness ratio decreases during the dips (Fig. 2, bottom-left panel), which is very similar to the soft source NGC 55 ULX-1 (Pinto et al. 2017). This was interpreted as evidence of temporary obscuration of the inner hot regions from a clumpy disc wind (e.g., Stobbart et al. 2006).
The lightcurve also shows that NGC 247 ULX-1 undergoes a long-term flux variability with observations 4-to-6 exhibiting significantly higher fluxes. Importantly, the frequency of the dips was higher during these observations.
In order to evaluate the source variability during each observation, we calculated the fractional excess variance of the EPIC-pn lightcurve of each observation and the rootmean-square (RMS), following standard formulae (see, e.g., Nandra et al. 1997, Vaughan et al. 2003, and Allevato et al. 2013). We adopted time bins of 1 ks and, as time length, the duration of each exposure (∼ 100 ks, with the exception of obsid:0844860801). The computed RMS values are reported in Table A1 and range from about 10 to 55 %.
The EPIC spectra of the individual observations in Fig. 3 show a variability pattern that is common to ULXs, with the harder band (> 1 keV) exhibiting the largest variation (see, e.g., Middleton et al. 2015a, Brightman et al. 2016, and Walton et al. 2018c. A thorough analysis of the variability pattern involving a careful sampling of time interval with similar flux and hardness ratio, and the study of the power density spectra, will be done in two separate papers (D'Aì et al. in prep, Alston et al. 2021).
SPECTRAL ANALYSIS
In this section we present the spectral analysis of NGC 247 ULX-1. We first show the time evolution of the main spectral residuals around 1 keV through the modelling of the spectral continuum in EPIC spectra from different observations (see Sect. 3.1). Then we will perform a thorough analysis of the high-statistics, time-averaged, stacked spectrum in order to identify the lines in the RGS (Sect. 3.2 and 3.3), to build the spectral energy distribution (SED), and to use physical plasma models for the wind detection and modelling (Sect. 3.4 and 3.5). The final best-fit models are shown in Sect. 3.6 and the statistical significance of our findings in Sect. A5.
ULX spectra require up to three components to obtain a satisfactory description of the spectral continuum. Two blackbody-like components are often used to model the soft (0.3-1 keV) and hard (1-10 keV) X-ray energy bands (see, e.g., Stobbart et al. 2006, Pintore et al. 2015. The availability of high-statistics spectra and broadband data reveals the presence of a third harder component, which dominates the continuum above 10 keV (see, e.g., Walton et al. 2018a). In the framework of super-Eddington accretion, the cooler, soft, component corresponds to the X-ray emission of the wind and the disc around the spherisation radius. The hard component refers to the inner accretion flow. The hard (> 10 keV) tail is either due to Compton scattering in the innermost regions or from an accretion column (see, e.g., Middleton et al. 2015a). The two hard components, especially the hard tail, are weak in supersoft ULXs.
Time evolution of the ∼1 keV residuals
For the modelling of EPIC spectra of individual observations we adopted a simple continuum model consisting of two blackbody (bb) components, which is often used as a proxy for more complex models (see, e.g., Walton et al. 2014, Pinto et al. 2017, Koliopanos et al. 2017, and Gúrpide et al. 2021. We did not model the hard tail in the individual spectra because it is so weak that any model would be highly unconstrained, but for the time-averaged SED modelling we took it into account (see Sect. 3.2). The emission components are corrected for absorption by the Galactic interstellar medium and the circumstellar medium near the ULX using the hot model in SPEX with a low temperature (kT = 0.2 eV, e.g. Pinto et al. 2013), at which the gas is neutral. In the spectral fits, we coupled the column density of the hot model across all observations as it is unlikely that the amount of neutral gas along our line of sight towards the ULX would change on time scales of a few days. In some Galactic Xray binaries (XRB) variable obscuration was found, but it is not clear whether these findings are related to ionised gas (i.e. winds) or uncertainties in the continuum rather than neutral gas (see, e.g., Miller et al. 2009. We simultaneously applied the hot (bb + bb) continuum model to the EPIC MOS 1,2 and pn spectra of all eight observations. The results are shown in Fig. 3 and Table A1. We obtained an average column density of NH = (3.4±0.1)× 10 21 cm −2 . This phenomenological model provides a good description of the broadband spectra, but the C-statistics are high when compared to the corresponding degrees of freedom (Cν ∼ 3−9) due to the well-known strong and sharp residuals in the form of absorption and emission features around 1 keV (see Fig. 3).
It is possible to understand the nature of the residuals by tracking their temporal evolution from one observation to another. In Middleton et al. (2015b) and Pinto et al. (2017), the three main spectral features observed around 1 keV in the spectra of NGC 1313 ULX-1 and NGC 55 ULX-1 were modelled with a positive (emission) and two negative Gaussians (absorption) lines. The availability of deeper observations allow us to fit the three Gaussian components independently, but we fixed the line broadening to zero km/s (i.e. only instrumental broadening) in order to minimise the degeneracy that can be produced by the low spectral resolution of EPIC. We chose three Gaussian lines as previous work on high-resolution RGS data identified Fe / Ne emission lines at 1 keV, O viii absorption lines around 0.7-0.8 keV, and Fe / Ne absorption lines above 1 keV (Pinto et al. 2016). Using only one or two Gaussian lines always resulted in significantly worse fits during alternative tests.
The hot (bb + bb) + (gaus + gaus + gaus) spectral model improves the fits for all observations with respect to the hot (bb + bb) model. In Table A1, we report the best-fitting parameters for each observation. The reduced C-stat are still rather high due to additional residuals that require more complex and physical models. Moreover there is evidence of a weak, broad, hard tail in all spectra above 3 keV, which cannot be explained with atomic lines (Fig. 3).
In Fig. 4, we compare the energy centroids (left panel) and the fluxes (right panel) as measured for the three Gaussian lines in the EPIC spectra of the eight observations. The point size was coded according to the value of their RMS estimate in Sect. 2.1.4. Both the energies and fluxes of the Gaussians lines vary with the time, showing a higher blueshift and lower flux (in absolute value) during the dipping observations with enhanced variability.
Time-averaged continuum: spectral modelling
The RGS spectra of the individual observations do not provide statistics sufficient to detect and resolve the spectral residuals with high significance. The 750 ks RGS 1+2 stacked spectrum instead has a much higher quality and enables line detection, despite the low source flux.
We simultaneously fitted the time-averaged stacked EPIC MOS 1,2 and pn, and the RGS spectra using the absorbed double-blackbody continuum model (hot (bb + bb)) adopted for the spectra of the individual observations. The hard tail above 3 keV is more evident in the stacked data and, therefore, we accounted for it using a third, hotter (kT ∼ 1 keV), blackbody as to mimic additional hard Xray photons down-scattered through the disc photosphere (and/or the wind). The three-blackbody model brings the C-stat from 4671 down to 4488 for four additional degrees of freedom. In all fits the parameters of the blackbody and Table 2. Time-averaged EPIC+RGS spectral fits.
Parameter
Units L X (0.3−10 keV) luminosities are calculated assuming a distance of 3.3 Mpc and are corrected for absorption (or de-absorbed).
The best-fit three-blackbody model and the spectra are shown in Fig. 6. Both models are shown in the SED modelling in Fig. A3.
the ISM absorber were coupled among the EPIC and RGS models. We left the overall normalisations of the MOS 1,2 and RGS models free to vary with respect to pn in order to account for the typical 5-10% cross-calibration uncertainties among their effective areas. Details on the spectral fits for both models are reported in Table 2 and Fig. 5. In order to avoid a crowded plot, we did not include the more noisy RGS spectrum in this plot, while it is shown later in Figs. 6 and 7 where the EPIC data below 1.77 keV was ignored. The blackbody models used so far are the simplest available and allowed us to constrain the parameters. We tested various combination of two-components models to possibly improve the description of the spectral continuum before accounting for narrow features. These consisted of the cool blackbody component plus either a disc blackbody (dbb) or a disc blackbody modified by coherent Compton scattering (mbb) or Comptonisation (comt). These did not provide improvements with respect to the three blackbody model. Similar results were provided by more complex three-component continuum models which anyway over-fit the data and lead to degeneracy among model parameters due to the weak hard (> 1 keV) continuum. Alternatively, one could use the common blackbody plus powerlaw emission model. However, the powerlaw would be very steep with a consequent unphysical divergence in the soft band, which would badly affect the ionisation balance calculation (see, e.g., Pinto et al. 2020a).
Time-averaged continuum: Gaussian line scan
In Fig. 6, we show the stacked RGS and EPIC spectra, indicating the dominant H-/He-like transitions of the X-ray band often found in the spectra of X-ray binaries. A zoom over the RGS spectrum with ad-hoc linear Y-axis can be found in Fig. 7. The RGS exhibits substantial residuals at the same energies as the EPIC residuals but resolves them in a structure of lines, although the former have lower count rates. Strong emission-like features appear near the transition energies of the most relevant neon K and iron L lines. Additional features may be related to N vii and O vii-viii, although the background starts to be important in the RGS below 0.6 keV. Some possible absorption-like features are indicated with vertical dotted lines. The very good agreement between the positions of the RGS, MOS 1,2 and pn residuals rules out instrumental dominant features.
Following the approach used in Pinto et al. (2016) searched for narrow spectral features by scanning the spectra with Gaussian lines. We adopted a logarithmic grid of 1000 points with energies between 0.3 (41Å) and 10 keV (1.24Å). This choice provided a spacing that is comparable to the RGS and EPIC resolving power in the energy range we are investigating (R RGS ∼ 100 − 500 and R EPIC ∼ 20 − 60). We tested line widths (σG=FWHM/2.355) of 100, 250, 500 and 1000 km/s, which are comparable to the RGS resolution. At each energy we recorded the ∆C improvement to the best-fit continuum model and expressed the significance as the square root of the ∆C. This provides the maximum significance for each line (as it neglects the number of trials). We multiplied √ ∆C by the sign of the Gaussian normalisation to distinguish between emission and absorption lines.
In Fig. 8 we show the results of the line scan obtained for the time-averaged stacked RGS+EPIC spectra using the three-blackbody continuum model. We performed the line scan in two ways: the first run using all RGS data (0.3-2 keV) and EPIC data (0.3-10 keV) and the second one ignoring the EPIC data between 0.33 and 1.77 keV, where the RGS effective area is well calibrated. When fitting only RGS in the 0.33-1.77 keV energy band we always fixed the temperatures of the blackbody components to the best-fit values obtained using the EPIC data in the whole 0.3-10 keV. This is due to the low count rate of the RGS spectra that limited our capability to constrain the overall continuum level and shape (see Pinto et al. 2020b).
The line scan of the EPIC + RGS spectra data picked out the strong emission-like features near 1 keV and other two around 0.6 and 1.5 keV. Broad absorption features were also found around 0.7 and 1.2-1.3 keV as previously done in the spectrum of each observation (see Fig. 4 and Table A1).
Owing to the low spectral resolution and high count rate of EPIC, the features appear very broad (∼ 0.1 keV) in the EPIC+RGS Gaussian scan preventing us from identifying them. This becomes easier above 2 keV due to the increasingly higher EPIC spectral resolution. The Gaussian scan performed using only RGS between 0.33 and 1.77 keV resolved the features into a forest of lines. Multiple lines are responsible for the 1 keV emission-like and the 1.2-1.3 keV absorption-like features. Interestingly, most absorption features are consistent with some Lyman α transitions also seen in emission, if we assume a systemic blueshifted absorption of about 0.17c (see vertical ticks in Fig. 8). The 0.6 keV features might either be interpreted as a blueshifted O vii α triplet or redshifted O vii β + O viii Ly α emission lines. The emission lines found between 0.9-1 keV are most likely from Ne ix-x and Fe xviii-xxiv ions. Alternative interpretations correspond to different velocities of the X-ray emitting (and absorbing) plasmas. The use of physical models is necessary to distinguish among the several interpretations. The most simple physical models involve the adoption of plasma in either collisionally-ionisation equilibrium (CIE) or photoionisation equilibrium (PIE).
The single-trial line significance ("σST ") of the individual strongest features is around 5 σ, which of course is smaller if we take into account the look-elsewhere effect. However, plasma models are able to model multiple lines, combining their individual ∆C improvements to the best-fit continuum, and boost the overall significance (see below). Table 2). The bottom panel shows the residuals calculated with respect to the continuum model. The rest-frame energies of the most common strong transitions in the X-ray band (red for RGS band and blue for EPIC) and the absorption features (dotted lines) are also shown. All spectra were rebinned for displaying purposes.
Collisional-ionisation jet modelling
Pinto et al. (2020b) performed automated scan models using either CIE or PIE plasmas. This technique prevents the fits from getting stuck in local minima, although is computationally expensive (lasting a few hours on one CPU).
Collisionally-ionised emitting gas
Following Kosec et al. (2018b) and Pinto et al. (2020b), we performed a multidimensional automated scan with an emission model that assumes collisional ionisation equilibrium (cie model in SPEX). We adopted a logarithmic grid of temperatures between 0.1 and 5 keV (50 points), and lineof-sight velocities, vLOS, between −0.3c (blueshifted jet) and +0.3c (redshifted jet, with steps of 500 km/s). We tested several values of velocity dispersion (from 100 to 10000 km/s), finding comparable results as already shown by the Gaussian line scan in Fig. 8, with the best fit achieved at vRMS ∼ 3000 km/s. Abundances were chosen to be Solar (to limit the computing time) and the emission measure EM = ne nH V was the only free parameter of the CIE in the spectral fit.
We applied the automated routine scanning CIE models onto the NGC 247 ULX-1 time-averaged RGS (0.33−2 keV) and EPIC MOS 1,2-pn (1.77−10 keV) spectra. We adopted the three-blackbody continuum model shown in Sect. 3.2 and Table 2 (see also black line in Fig. 6). The results are shown in Fig. 9 (left panel). The best-fit corresponds to a large improvement with respect to the continuum model (∆C = 82, for 4 additional degrees of freedom) and was achieved for a CIE temperature of 0.9 keV and a small blueshift of around 6500 km/s (∼ 0.022c). These results were driven by the strong lines that can be seen in the RGS spectrum 1 keV in Fig. 8 (bottom two panels).
We have checked our method by testing an identical CIE scan on the well-known Galactic X-ray source SS 433. This source is not ultraluminous in the X-ray band due to obscuration of the accretion disk from local circumstellar gas, but exhibits a persistent and bright (10 40 erg/s) radio jet, which is powering a luminous optical super-bubble (Brinkmann Fabrika 2004;Medvedev et al. 2020). SS 433 is therefore considered to be viewed edge-on. Were it observed face-on, it would likely appear as a ULX (Begelman et al. 2006;Poutanen et al. 2007;Middleton et al. 2018). SS 433 also shows a relativistic jet in the form of blueshifted lines from multi-temperature plasma in collisional-ionisation equilibrium (Marshall et al. 2002
Collisionally-ionised absorbing gas
It is uncommon to adopt absorption models of gas in collisional-ionisation equilibrium in accreting objects as 1) it is difficult to distinguish between photoionisation and collisional ionisation on the sole basis of the dominant resonant absorption lines and 2) we hardly expect any jet to absorb the X-ray source continuum along our line of sight. However, we cannot exclude that shocks are produced by interaction between the ULX wind and the stellar companion or the surrounding bubble (or within the wind itself). Therefore, we also performed a model scan with the hot model in SPEX, which works just like cie but assumes absorbing gas.
In Fig. 9 (right panel) we show the results using the hot model over the same kT range used for the emitting gas, adopting a velocity dispersion of 500 km/s and line-ofsight velocities, vLOS, ranging between −0.3c and zero, i.e. only Doppler blueshifts or outflows rather than inflows. The best-fit solution is obtained for a −0.17c blueshifted with a remarkable ∆C = 48 as suggested by the detection of several negative Gaussians blueshifted by similar values in Fig. 8.
Photoionisation wind modelling
The emission and absorption lines can be produced by winds rather than by jets as expected in the case of super-Eddington accretion discs and, therefore, in ULXs. Accurate photoionisation models require knowledge of the radiation field, i.e. the SED from optical to hard X-ray energies.
SED and photoionisation balance
Following Pinto et al. (2020a,b), we built the time-averaged SED of NGC 247 ULX-1 using data from the XMM-Newton campaign and archival HST observations (as taken from Feng et al. 2016). For issues regarding the non simultaneity of HST and XMM observations, see Appendix A4. For the X-ray band (0.3−10 keV or ∼ 10 16−18 Hz) we used the bestfit three-blackbody continuum model, estimated in Sect. 3.2, Fig. 5 and Table 2. As shown in Sect. 2.1.3, the OM filters were not sensitive enough to detect the optical counterpart, but their flux upper limits in the optical and UV energy bands are however comparable to the HST measurements (see Fig. A3 top panel). We therefore modelled the optical/UV portion of the SED with the two-blackbody model of Feng et al. (2016), which together with the three-blackbody X-ray model formed our five-blackbody SED model.
We can describe the photoionisation equilibrium with the ionisation parameter, ξ, defined as ξ = Lion/(nH R 2 ) (see, e.g., Tarter et al. 1969), where Lion is the ionising luminosity (measured between 13.6 eV and 13.6 keV), nH the hydrogen volume density and R the distance from the ionising source. The ionisation balance was calculating with the SPEX pion model, which calculates the transmission and the emission of a thin slab of photoionised gas, self-consistently.
Following Pinto et al. (2020b), we also computed the stability (or S) curve, which is the relationship between the temperature (or the ionisation parameter) and the ratio between the radiation and the thermal pressure, which can be expressed as Ξ = F/(nHckT ) = 19222 ξ/T (Krolik et al. 1981). The stability curve is shown in Fig. A3 (bottom panel). The branches of the S curve with a negative gradient are characterised by thermally unstable gas. In this work, we assumed that the wind is seeing the same SED that we observe and, therefore, adopted the five-blackbody model SED and ionisation balance. Systematic effects from the SED choice are discussed in Sect. 4 and Appendix A4.
Photoionised emitting gas
Once the SED and the ionisation balance were computed, we scanned the time-average EPIC+RGS spectra with the SPEX pion model with the same multi-dimensional routine used for the cie model in Sect. 3.4.1, and a similar parameter space. We adopted a logarithmic grid of ionisation parameters (log ξ [erg/s cm] between 0 and 6 with 0.1 steps). The only free parameter for the pion is the column density, NH.
Unlike NGC 1313 ULX-1, the RGS spectrum of NGC 247 ULX-1 does not exhibit well resolved emission line triplets. This is perhaps due to the longer integration time required and the variability of the line centroid (see Fig. 4), which could wash out the triplets when stacking all the spectra. Additionally, the crucial O vii complex is affected by the background noise. The lack of He-like triplets means that the volume density and the luminosity of the photoionised gas are degenerate. Fitting both parameters results in poor constraints and much higher computation time. We therefore chose not to fit the volume density and adopted nH = 10 10 cm −3 , which is a lower limit found for NGC 1313 ULX-1 (Pinto et al. 2020b). This would only slightly affect the overall flux and column density of the pion component.
The pion covering fraction is set to zero (i.e pion only produces emission lines) and the solid angle Ω = 4π. Fitting additional parameters such as Ω might provide even better fits but would significantly increase the computing time, without altering the velocity and ionisation parameters.
In Fig. 10 (left panel) we show the results obtained using a pion line width of 1000 km/s. The confidence level (CL) is expressed in σ, which is constrained using Monte Carlo simulations (see Sect. A5). A peak (∆C = 77) corresponding to a solution of blueshift emission is seen around 0.02 − 0.03c in agreement with the CIE model scans (see Sect. 3.4 and Fig. 9). However, the different ionisation balance and types of emission lines in the photoionisation equilibrium detected another, stronger (∆C = 102), peak corresponding to a redshift of ∼ +0.05c.
Photoionised absorbing gas
In principle, we could just use pion for both emission and absorption. However, this model re-calculates the ionisation balance at every iteration and therefore is computationally expensive. Therefore, for the absorbing gas we chose to use the faster xabs model, which is optimized for absorption and adopts the pre-calculated ionisation balance (see Sect. 3.5.1).
The xabs model shares several parameters with pion except the opening angle of the line emission which is zero since no emission is present in this model. We adopted a covering fraction equal to unity in order to avoid degeneracy and reduce the computing time. We calculated the grid of photoionised xabs models in the same way as the pion models, but assuming line-of-sight velocities, vLOS, ranging between −0.3c and zero, i.e. only Doppler blueshifts as for the CIE hot absorption models in Sect. 3.4.2.
In Fig. 10 (right panel) we show the probability distributions from scans of the RGS and EPIC spectra with vRMS = 1000 km/s. As expected, our code confirmed the v = −0.17c solution (∆C = 46).
Final fits with physical plasma models
In order to check the inter-dependence of the emitting and absorbing plasma components and to test for any variations in the values of their parameters we performed two more fits of the RGS and EPIC data (with EPIC still excluded Figure 11. Time-averaged XMM-Newton RGS (0.33−2 keV) and EPIC-pn (1.77−10 keV) spectra. Overlaid are two alternative models (jet -CIE, red line and the wind -PIE, black line). The spectra were regrouped and the plot zoomed onto the RGS data and the 0.4-2 keV energy band for displaying purposes. between 0.33-1.77 keV) adding onto the 3-blackbody continuum two alternative plasma models. The first one was a wind model that used the pion in emission and the xabs in absorption. The second model was an approximation of jets and shocks in the form of cie component in emission and hot component in absorption. The results for the two models are shown in Table 3 and Fig. 11 (zoomed onto the RGS). The absorption components provided very similar results, especially for the column densities and the line-of-sight velocities. As previously noted, the emission components show some differences. The results are discussed in Sect. 4.
The temperatures of the three blackbody components were always fixed to the EPIC 0.3-10 keV fits (Table 2) each time we ignored EPIC data between 0.33-1.77 keV, resulting in parameters consistent with the continuum modelling.
By alternatively excluding one particular plasma component from the spectral model we estimated the relative contribution to the spectral fit and the minimum ∆C-stat improvement of each component for both the wind and jet model (see ∆C e values in Table 3). Following Pinto et al. (2020b), we compared the minimum ∆C-stat values with the Table 3. NGC 247 ULX-1: alternative plasma models.
Model 1 Parameter
Emission Absorption 104 (97) Montecarlo simulations to obtain an approximate estimate of the minimum significance of each wind or jet component (parameter σ e ), which is highest for the emission phases.
Finally, to further test the strength of our results we performed a fit of the RGS and EPIC data including the whole EPIC energy band but fixing the plasma models to the best-fit results of the hybrid RGS (0.33-2 keV) plus EPIC (1.77-10 keV) fits. In Sect. 3.5.3, we noted that the inclusion of the photoionised xabs absorption component significantly decreased the C-stat from 4488 (of the simple 3-blackbody model) to 3033. The addition of the photoionised pion emission component further lowered the C-stat to 2431 (with a χ 2 = 1830 and a total of 1132 degrees of freedom). We performed the same fit by using instead the RGS jet model with the hot collisionally-ionised component fixing the parameters to those in Table 3. This decreased the C-stat from 4488 to 3043, similarly to the photoionised absorber. The addition of a cie emission component (with fixed plasma parameters) implied a final C-stat = 2542, which is slightly worse than pion due to some positive residuals left around 0.9 keV that can also be seen in the RGS spectral modelling in Fig. 11. We notice that all the spectral fits shown here are not formally acceptable (see Table 4), although the winds components provide significant improvements. One reason is the variability of the features, both of their centroids and relative strength (see Fig. 3 and 4). This means that more complex models would be required to correctly fit the lines. On the other hand, the winds are likely multiphase as shown by the low-temperature O vii clearly missed by our single phase model (see Fig. 11). This was already shown in Pinto et al. (2020b) and occurs in SS 433 too (see Appendix A3). Finally, some bad cross-calibration below 0.6 keV between the EPIC and RGS cameras further prevent us from obtaining fully acceptable fits (see Fig. 5).
DISCUSSION
It is still unclear how does the wind vary with the accretion rate and whether it has a major role in shaping ULX spectra. Pinto et al. (2020a,b) showed that the wind evolves with the changes in the continuum from the fainter, harder states to the brighter states, which implies a tight relationship between the source's spectral continuum and wind appearance as observed by comparing winds in different ULXs.
Among the ULXs, the supersoft ultraluminous X-ray (SSUL) sources are particularly fascinating objects. The fact that such sources reach very high luminosities (several 10 39 erg/s) but always exhibiting very soft (kT ∼ 0.1 keV) spectra indicates that they are being observed at moderate-tohigh inclination angles as also suggested by the presence of dips in their lightcurves (see, e.g., Feng et al. 2016). In fact, Urquhart & Soria (2016) modelled the soft X-ray residuals and the ∼ 1 keV drop found in several CCD spectra of ULSs with a model of thermal emission and an absorption edge, which they interpreted as a result of absorption and photon reprocessing by an optically-thick wind which obscures the innermost regions where most hard X-rays are produced.
Time evolution of the 1 keV residuals
The stacked XMM-Newton lightcurve (see Fig. 2) shows different pattern of source variability such as a long-term overall change in the average flux on daily time scales followed by abrupt drops in the flux where the source becomes softer (the dips) on timescales between 100s and a few hours. The dipping activity seems also to enhance during observations with higher flux peaks. The higher flux might be associated with a higher local accretion rate, which then would increase the radiative force and launch optically-thick wind cloudlets in the line of sight, thereby obscuring the innermost regions of the disc responsible for the hard X-ray emission (as suggested by Urquhart & Soria 2016). More insights on the nature of the dips will be provided by Alston et al. (2021). This work shows that the dips in the higher flux observations preferentially occur on 5 and 10 ks timescales, which suggests that they are caused from obscuration at ∼ 10 4−5 RG, where RG is the Gravitational radius (if the timescales are associated with keplerian motion around a NS or a stellar-mass BH). Such a range is comparable to the distance that the 0.17c wind would travel on a time scale of 1 ks, suggesting a possible connection between them.
The high-quality EPIC spectra of the individual observations show a remarkable flux variability in the features around 1 keV (see Fig. 3). In order to quantify such variability, we modelled the two strongest absorption features around 0.7 and 1.2 keV, and the dominant emission-like feature at 1 keV with three Gaussian lines for the EPIC (MOS and pn) spectral of the individual observations. All lines show a distinct pattern with their energy centroids significantly blueshifted during the brightest observations (which also exhibit most dips and the highest variability, see Fig. 4). Interestingly, the fluxes of the high-energy lines (1 and 1.2 keV) significantly decrease during the dipping observations while the 0.7 keV line seems to strengthen (see Table A1). This would either suggest a different location of the three lines, with the 0.7 keV line coming from the outer and less obscured regions, or a change in the ionisation state of the absorber during the high-flux periods. This is similar to what was observed in NGC 1313 ULX-1 (Pinto et al. 2020b, Middleton et al. 2015b. A detailed study of the broadband spectra and residuals evolution will be shown by D' Aì et al. (in prep). The fact that the location and strength of the residuals vary on hourly timescales with the source flux provides strong evidence in support for a disc wind rather than emission from the local ULX bubble or the galactic ISM.
Emission lines
The time-average stacked RGS spectrum showed strong emission residuals near the transition energies of several ions such as O vii-viii, Ne ix-x and Fe xviii (see Fig. 7 and 8). The agreement between RGS and EPIC is corroborated by applying the wind model constrained using only RGS in the 0.33 − 1.77 keV band to the whole EPIC MOS and pn timeaverage spectra (see Table 4). Unfortunately, the He-like emission triples of e.g. O vii and Ne ix are not well resolved likely due to the stacking of RGS spectra from different observations that clearly showed some variability in the line centroid as discussed above. This limited our capabilities of distinguishing between collisional and photoionisation, but the use of full plasma models provided some constraints.
By performing automated searches of plasma models in a large parameter space, we built probability contours for both collisionally-ionised and photoionised plasma emission models. The properties of the line-emitting gas are very similar to those of the Galactic super-Eddington source SS 433 with a low velocity along the line of sight and a mild 1 keV temperature which is expected by the strong Ne K and Fe L emission around 1 keV (see Fig. 9). It is well established that the emission lines of SS 433 are from the jet with the low velocity indicating that the precessing jet was at very high angle, close to 90 degrees, in the analysed observation. If the lines of NGC 247 ULX-1 were also from a variable jet, the observed low velocity would suggest that it is being viewed at high angle in agreement with the presence of dips.
The photoionisation emission models (pion component in SPEX, see Fig. 10) however provided a significantly higher improvement to the spectral fits and a better description of the emission lines (see Table 3 and Fig. 11). This together with the evolution of the lines with the source continuum would favour photoionisation equilibrium similarly to the emission lines in NGC 1313 ULX-1 (Pinto et al. 2020b).
Regardless of the adopted equilibrium state, the luminosity of the line-emission component is remarkably high (L 0.3−10 keV > 10 38 erg/s), similarly to NGC 1313 ULX-1, NGC 5408 ULX-1 (Pinto et al. 2016), NGC 55 ULX-1 (Pinto et al. 2017), NGC 5204 ULX-1 (Kosec et al. 2018a) and other ULXs (e.g., Wang et al. 2019). This is about 2-3 orders of magnitude higher than the emission lines in SS 433 and those producing the winds in classical supergiant X-ray binaries (sub-Eddington neutron stars accreting from supergiant OB stars, e.g. El Mellah & Casse 2017) or the lines from accretion disc coronae of low-mass X-ray binaries (see, e.g., Psaradaki et al. 2018). The luminosity of 1.4 × 10 38 erg/s is instead comparable to the extended X-ray emission recently found around the extremely bright pulsating NGC 5907 ULX-1 (Belfiore et al. 2020), which suggests that the wind might be energetic enough to mechanically drive the ∼ 100-pc super bubbles (see also Pinto et al. 2020a).
Similarly to NGC 1313 ULX-1, the O vii emission lines cannot be reproduced with the emission component responsible for the Fe L and Ne K emission (see Fig. 11). A second component (either photo-or collisionally-ionised) with a low blueshift of 6000 km/s would be required.
Absorption lines
In this work we also reported a highly significant detection of mildly-relativistic, ultrafast, outflows. Both collisional and photoionisation (see Fig. 9, 10 and 11) plasma models identified a high velocity outflow (−0.17c) in the range of the velocities found in other ULX winds.
The ionisation parameter is rather high (log ξ = 4.3) which is not surprising given the soft SED adopted for this source (see Fig. A3). If the wind at the launch is seeing a different SED (e.g., the hot innermost regions presumably obscured along our line of sight) the overall ionisation balance might be significantly different. This subject was extensively discussed in Pinto et al. (2020a) who found larger instability branches in the S curves of harder ULXs. Therefore, as a test, we performed an additional fit with the photoionised pion + xabs wind model (as previously done in Sect. 3.6) by adopting the ionisation balance calculated for the hard state of NGC 1313 ULX-1 in Pinto et al. (2020b) to estimate the systematic effects on the wind parameters. The fit was statistically indistinguishable from the one performed with the ionisation balance computed for NGC 247 ULX-1, with the exception of the ionisation parameters which, as expected, turned out to be significantly lower by about ∆ log ξ ∼ 1.
The absorption lines are generally weaker than the emission lines in the RGS spectrum of NGC 247 ULX-1 which could be due to the low source continuum (Kosec et al. submitted). This was also predicted by Pinto et al. (2017) as the lines are normally seen on top of the continuum from the innermost regions which in this case is likely obscured.
Statistically we cannot distinguish photoionisation from collisional ionisation, but the former is favoured by the photoionised nature of the emitting plasma and the unusual detection of collisionally-ionised absorption in XRB winds.
Accretion disc and wind physics
In the framework of super-Eddington accretion the luminosity scales with the logarithm of the accretion rate in Eddington units (ṁ =Ṁ /ṀE) times the geometrical beaming of the funnel created by the height of the disc around the spherisation radius and by the wind itself (see Fig. 12). Following King & Lasota (2020), the apparent luminosity can be expressed with Lapp = L/b = LE(1 + lnṁ)/b, where L is the intrinsic luminosity, LE the luminosity in Eddington units, and b = 73/ṁ 2 the geometrical beaming.
To estimate the bolometric luminosity of NGC 247 ULX-1 we integrated the broadband SED between 1 eV and 10 keV (or 2.4 × 10 14−18 Hz, see Fig. A3) and obtained 9.4×10 39 erg/s. NGC 247 ULX-1 luminosity could therefore be explained by assuming a black hole accreting above 10 times the Eddington rate or a neutron star accreting abovė m = 25. Atṁ ∼ 10 the spherisation radius, i.e. the base of the wind, would be R sph = 27/4ṁRG = 68RG. Interestingly, this is very close to the escape radius for a −0.17c wind (Re = 2GM/v 2 = 2RGc 2 /v 2 = 73RG), which would indeed suggest that we detected a wind launched from the spherisation radius of a compact object above 10ṀE.
From Eq. (38) in Poutanen et al. (2007), assuming MBH = 10M andṁ = 10, we estimated a temperature for the spherisation radius T sph ∼ 0.3 keV, which is comparable to the warm blackbody component in our fits (see Table 2), with the cooler (∼ 0.1 keV) blackbody associated with the outer disc and, likely, the wind photosphere as suggested by recent work (see, e.g., Qiu & Feng 2021, Gúrpide et al. 2021. We notice, however, that the source is being seen at high inclination with a substantial fraction of the hard X-ray photons obscured by the funnel. The intrinsic luminosity of NGC 247 ULX-1 might therefore be higher than the value estimated above with a higher accretion rate, implying T sph ∼ 0.1 − 0.2 keV, closer to the cooler blackbody component, and a slightly larger R sph . It is also possible that the wind is launched with lower velocity at radii larger than 73RG and it gets accelerated by radiation pressure from the inner accretion flow (see, e.g., Takeuchi et al. 2013).
Similar considerations would apply to a non-magnetar (B 10 12 G) neutron star withṁ = 25 since the spherisation radius (in cm) would be of the same order of magnitude as a 10M black hole as both R sph and T sph scale with thė M and the mass of the compact object, whose trends nearly cancel out. This was briefly discussed in Pinto et al. (2020a).
The kinetic power of the wind can be written as Lw = 1/2Ṁw v 2 w = 2 π mp µ Ω C v 3 w /ξ Lion ∼ 4 × 10 40 erg/s, wherė Mw = 4 π R 2 ρ v 2 w Ω C is the outflow rate, Ω and C are the solid angle and the volume filling factor (or clumpiness), respectively, which were adopted equal to 0.3 as conservative values from MHD simulations of winds driven by radiation pressure in super-Eddington winds (Takeuchi et al. 2013), ρ is the density and R is the distance from the ionising source.
Here we have used the ξ definition to get rid of the R 2 ρ factor where ρ = nH mp µ with mp the proton mass and µ = 0.6 the average particle weight of a highly ionised plasma.
The filling factor of the wind might be much smaller. Using Eq. (23) in Kobayashi et al. (2018) and assuming that the outflow rate is comparable to the accretion rate, we obtain C ∼ 3 × 10 −2 . Systematics would tend to cancel out when also accounting for the uncertainty on the ionisation parameter in the case for a harder SED (∆ log ξ ∼ 1). In the pessimistic case the wind power would be of the order of 10 39 erg/s, which means still high enough to affect the surrounding medium and inflate ISM cavities.
The spectral shape, strong wind features, and presence of dips suggest that NGC 247 ULX-1 is likely being observed at high inclination (see Fig. 12, left panel) where the funnel is already obscuring the innermost, hot, hard X-ray emitting regions. As mentioned in Sect. 4.1, the increase of the average flux level during the intermediate observations (3,4,5) might be caused by a higher local accretion rate. Such a climate change would however affect the properties of both the disc and the wind. The scale-height is already relevant around the Eddington limit (see, e.g., Shakura & Sunyaev 1973, Poutanen et al. 2007. A further increase in the localṀ might push the optically-thick funnel further upwards (see Fig. 12, right panel) thereby obscuring the regions emitting photons with temperatures higher than that of the spherisation radius ( 0.3 keV), causing the very soft dips shown in Fig. 2 (see also Urquhart & Soria 2016).
During the dips the high-ionisation portion of the wind could be hard to see as its absorption lines were primarily affecting the (now) obscured hard X-ray continuum. In fact, the strength of the high-ionisation (1.2-1.3 keV) absorption lines clearly decreases during the dipping observations (see Fig. 4), while the 0.7 keV O viii absorption line seems constant in flux if not even stronger. This might also suggest a stratification in the wind. Outside the dips, an overall increase in the accretion rate would also imply a stronger radiative force and, therefore, a slightly faster wind which seems to be confirmed by the higher blueshift of the lines (see Fig. 3 and 4). The 1 keV emission lines also weaken during the bright / dipping observations, indicating that they should be produced in the inner regions in agreement with their overall larger broadening (see Table 3).
A similar picture was proposed by Guo et al. (2019) who argued that the ∼100s transitions can be explained by the viscous timescale with the X-ray flux variability driven by accretion rate fluctuations (atṁ 10). However, local fluctuations in theṀ might also cause variations in the winds, which could alter the source appearance (Feng et al. 2016).
Although fascinating and self-consistent, this scenario might be not the only one able to explain all the observables.
Additional, alternative, and (ideally) model-independent approaches could be considered. For instance, another phenomenon which might explain the nature of the dips might be the propeller effect due to a strong magnetic field. Such scenario would imply a decreasingṀ and a geometrical beaming to cause the observed brightening. More insights on the temporal evolution of the spectral residuals accounting for the different spectral states that the source shows inside / outside the dips will be given in D' Aì et al. (in prep). Similarly, the Fourier analysis of the characteristics timescales in NGC 247 ULX-1 and the corresponding association with the dipping activity will be shown by Alston et al. (2021). Here, in particular, we argue that the alternation of the dips might be due to azimuthally-dependent structures.
We plan to investigate the variability of the RGS spectral lines to place more constraints onto their nature. However, the low count rate of the grating spectra currently prevent us from trying to study them during the dips and on timescales shorter than 100 ks in the bright states outside the dips. Future missions like XRISM and Athena will revolutionise the study of ULX thanks to their high effective area, high spectral resolution, and low background (see, e.g., Barret et al. 2018, Guainazzi & Tashiro 2018. Pinto et al. (2020b) simulated NGC 1313 ULX-1 microcalorimeter spectra for these two missions and showed 1) how XRISM will strengthen the identification of lines in the 1 − 10 keV band and 2) how Athena / X-IFU will be able to detect winds in observations with just 1 ks of exposure time. The latter is primarily due to the fact that X-IFU will have two orders of magnitude higher effective area than RGS.
CONCLUSIONS
Most ULXs are believed to be powered by super-Eddington accreting neutron stars and, perhaps, black holes. The disc is expected to thicken at accretion rates above the Eddington rate and to launch powerful winds through radiation pressure and/or magnetic fields. Evidence of winds has been found in several ULXs through high-resolution X-ray spectrometers. It is yet unclear whether the switch between the classical soft and supersoft state -which is observed in supersoft ULXs -is due to the thickening of the disc and/or the optically-thick part of the wind. In order to better understand such phenomenology and the overall super-Eddington mechanism, we undertook a large observing campaign with XMM-Newton to study NGC 247 ULX-1, which is the brightest (in flux) of all supersoft ULXs.
The new observations showed for the first time unambiguous evidence of a wind in the form of emission and absorption lines from highly-ionised ionic species, with the absorption phase exhibiting a mildly-relativistic outflow (−0.17c) in line with the other ULXs whose grating spectra had sufficient quality to detect and identify spectral lines. Remarkable variability was observed in the source flux with strong dipping activity during the brightest observations, which is typical among soft ULXs such as NGC 55 ULX-1, and indicate a close relationship between the accretion rate and the appearance of the dips. The latter are likely due to a thickening of the disc scale-height and the wind as shown by a progressively increasing blueshift in the spectral lines. Figure 12. A possible scenario for the dips and ULX-ULS transitions in NGC 247 ULX-1. The source is observed at a viewing angle that is high enough that the inner disc is already partly obscured by the wind (soft ULXs, left panel). An increase of the accretion rate pushes up the scale-height of the disc and the optically-thick base of the wind, causing an near-total obscuration of the inner regions and the source appears as an ultraluminous supersoft source (ULS, right panel, see also Pinto et al. 2017, 2020b, Guo et al. 2019).
A1 Nearby bright X-ray source
The RGS extraction region includes a few faint sources with the brightest one being XMMU J004710.0-204708 (X-2 hereafter, see Fig. 1). We extracted its EPIC spectra from all observations and stacked them similarly to ULX-1. The spectrum of X-2 is much flatter than that of the ULX-1 and can be well modelled with a powerlaw model (Γ = 1.60±0.03), a moderate column density NH = (1.0 ± 0.1) × 10 21 cm −2 , and an intrinsic luminosity L 0.3−10keV = (1.45±0.04)×10 38 erg/s (assuming a distance of 3.3 Mpc). This corresponds to the Eddington limit for a Solar-mass star and, given the spectral slope, the source X-2 could be a common XRB near the NGC 247 centre. At 1 keV its spectrum is remarkably featureless and 40-50 times fainter than ULX-1 implying that it will have no significant effects on the RGS spectral lines. Table A1 shows the results of the EPIC spectral modelling and the root-mean square estimated from the EPIC-pn data of each observation (see Sect. 3.1 and 2.1.4).
A3 CIE model scan for the SS 433 RGS spectrum
We analysed the XMM-Newton RGS spectrum of SS 433 from the observation id:0694870201 (2012-10-03), which provides the longest (∼130 ks) and best-exposed RGS spectrum of SS 433. We reduced the RGS spectrum of SS 433 obsid 0694870201 identically to that of the NGC 247 ULX-1 data shown in Sect. 2.1.2. After removing the very little Solar flares we are left with 129.4 ks for both RGS 1 and 2 cameras. We used the 6-25Å range because at higher wavelengths the source emission is significantly absorbed and the background noise dominates the RGS spectrum (see Fig. A1). No significant pile up was found in the RGS spectra.
Strong emission lines were observed close to the restframe energies of the most relevant transitions such as Si xiii, Mg xi, Ne x, Ne ix, Fe xvii, and O viii, which is very similar to NGC247 ULX-1, albeit at higher significance because SS 433 is much closer (∼ 5 kpc) and brigther. We modelled the RGS spectral continuum with an absorbed powerlaw model obtaining results similar to Marshall et al. (2013) and Medvedev et al. (2018) such as a column density NH = (1.14 ± 0.02) × 10 22 cm −2 , a slope Γ = 2.53 ± 0.06 and an X-ray unabsorbed luminosity L [0.3−10 keV] = (1.03 ± 0.05)×10 36 erg/s. We obtained high C-stat/d.o.f = 2393/618 as expected, due to the strong, unmodelled, emission lines.
A4 SED modelling and systematics effects
The non detection of the optical and UV counterpart of NGC 247 ULX-1 (Sect. 2.1.3) prevent us from building a simultaneous multi-wavelength SED for our source. This might have systematic effects on the calculation of the photoionisation balance. Moreover, the flux upper limits obtained with the OM suggest that at least in the far-UV en- Figure A3. SED (top panel) and thermal-stability curve (bottom panel) computed for the time-averaged spectrum of NGC 247 ULX-1. The baseline SED (solid black line) consists of five unabsorbed blackbody models that account for emission in the optical, UV and X-ray bands. An alternative, simpler SED model uses just the X-ray two-blackbody model (dashed-dotted line).
ergy band the source flux was lower than the levels measured in the archival HST observations that we used here. Pinto et al. (2020a) showed that a lower IR-to-UV flux in moderately-hard sources, such as NGC 1313 ULX-1, mainly strengthens thermal instabilities at intermediate temperatures and ionisation parameters. In Fig. A3 we show the SED adopted here (solid black curve) consisting of a 5blackbody model along with the simple 2-blackbody model (dashed-dotted black line, see also Sect. 3.5.1). The lower panel shows the stability curves computed for these models. Some deviations are mainly seen at log ξ from 1.5−2.5, which is well below the values measured in this work (see Table 3). This is not surprising given the shortage of hard X-ray in our spectra which are the primary responsible of thermal instabilities. This suggests that the ionisation balance is not significantly affected even if the optical / UV fluxes were 2 orders of magnitude lower than our assumptions. Occurrences [ν] 4.0σ 4.5σ 5.0σ Log[ν] = 6.57 -0.22*∆C 2 × 10 6 simulations forecast 2 × 10 5 simulations forecast 2 × 10 4 simulations histo-fit
A5 Monte Carlo simulations and significance
The ∆Cmax improvement to the continuum model does not necessarily yield the significance of the corresponding emission or absorption line models. This is due to the large parameters space that was explored and the possibility of detecting random spectral features (the look-elsewhere effect). Among our physical model searches, the one for the absorption lines provided the smallest ∆Cmax due to their strength being lower than that typical of the emission lines. We therefore focused on the results obtained with the xabs component and used them as a proxy for the pion.
Following the method used in Pinto et al. (2020b), we simulated 20 000 RGS and EPIC spectra adopting the 3-blackbody continuum model. Each faked spectrum was scanned with the same xabs grids used in Sect. 3.5.3. The results of our MC simulations are shown in Fig. A4. No outlier was found with ∆C ∆Cmax = 46, which suggests a significance > 4σ for the absorbing gas.
We compared the simulations histogram for NGC 247 ULX-1 with those obtained for different sources by adopting a similar approach: 20k simulations of NGC 1313 ULX-1 (Pinto et al. 2020b), 2k for NGC 5204 ULX-1 (Kosec et al. 2018a) and 50k for the same data with a new, faster, crosscorrelation method (Kosec et al. submitted), 20k for ULX NGC 7793 P13 (Pinto et al. in prep), and 1k for AGN PG 1448 (Kosec et al. 2020). We fit the histograms of the logarithm of the occurrences with straight lines and found an average slope Γ = −0.225 ± 0.015, which agrees with the simulations of NGC 247 ULX-1 (Γ = −0.218 ± 0.006).
We used the best-fit straight lines to estimate the overall shape of the ∆C-stat distribution and forecast the results of larger numbers of simulations thanks to the agreement between the trends from 1 000 to 50 000 simulations. We therefore scaled the histogram fit of NGC 247 ULX-1, assuming a constant slope and multiplying the intercept of the straight line by a number equal to the ratio of the parameter space that we want to forecast for a given number of simulations and the one we obtained with 20 000 simulations. In Fig. A4 we show the predictions for 2 × 10 5 (dash-dotted line) and 2×10 6 (dashed line) simulations. This would suggest 4.5 and 5 σ detection probabilities for ∆C-stat above 35 and 40, respectively, in the data with an uncertainty of 0.2σ according to the spread in the slope of the other histograms.
We finally retrieved the various ∆C-stat values that correspond to confidence levels ranging from 2.0, 2.5, ..., 5.0 σ and plot them as black contours in Fig. 10. The σ contours for the pion model scan were calculated by scaling the parameter space in the histogram of the xabs simulations in the same way used for the forecast.
|
2021-04-23T01:16:14.406Z
|
2021-04-22T00:00:00.000
|
{
"year": 2021,
"sha1": "eb79ca51fbbf591ae8a4a2254edcd3169bee5504",
"oa_license": null,
"oa_url": "https://authors.library.caltech.edu/111640/1/stab1648.pdf",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "eb79ca51fbbf591ae8a4a2254edcd3169bee5504",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
118856638
|
pes2o/s2orc
|
v3-fos-license
|
Energy conditions bounds on f(T) gravity
In standard approach to cosmological modeling in the framework of general relativity, the energy conditions play an important role in the understanding of several properties of the Universe, including singularity theorems, the current accelerating expansion phase, and the possible existence of the so-called phantom fields. Recently, the $f(T)$ gravity has been invoked as an alternative approach for explaining the observed acceleration expansion of the Universe. If gravity is described by a $f(T)$ theory instead of general relativity, there are a number of issues that ought to be reexamined in the framework of $f(T)$ theories. In this work, to proceed further with the current investigation of the limits and potentialities of the $f(T)$ gravity theories, we derive and discuss the bounds imposed by the energy conditions on a general $f(T)$ functional form. The null and strong energy conditions in the framework of $f(T)$ gravity are derived from first principles, namely the purely geometric Raychaudhuri's equation along with the requirement that gravity is attractive. The weak and dominant energy conditions are then obtained in a direct approach via an effective energy-momentum tensor for $f(T)$ gravity. Although similar, the energy condition inequalities are different from those of general relativity (GR), but in the limit $f(T)=T$ the standard forms for the energy conditions in GR are recovered. As a concrete application of the derived energy conditions to locally homogeneous and isotropic $f(T)$ cosmology, we use the recent estimated value of the Hubble parameter to set bounds from the weak energy condition on the parameters of two specific families of $f(T)$ gravity theories.
I. INTRODUCTION
A diverse set of cosmological observations coming from different sources, including the supernovae-type Ia (SNe Ia) [1], the cosmic microwave background radiation (CMBR) [2], and the large-scale structure (LSS) [3] clearly indicate that the Universe is currently expanding with an accelerating rate. A number of alternative models and different frameworks have been proposed to account for this observed late-time accelerated expansion of the Universe. These approaches can be classified into two broad groups. In the first, the framework of general relativity is kept unchanged and an unknown form of matter sources, the so-called dark energy, is invoked. In this regard, the simplest way to describe the accelerated expanding Universe is by introducing a cosmological constant into the general relativity field equations. Although this is entirely consistent with the available observational data, it faces difficulties, including the microphysical origin and the order of magnitude of the cosmological constant. In the second group, modifications of Einstein's gravitation theory are assumed as an alternative for describing the accelerated expansion. Examples of the latter group include generalized theories of gravity based upon modifications of the Einstein-Hilbert action by taking nonlinear functions f (R) of the Ricci scalar R or other curvature invariants (for reviews see Ref. [5]).
An alternative modification of general relativity, known as f (T ) gravity, has been examined recently as a possible way of describing the current acceleration of the Universe [6][7][8]. The origin of f (T ) gravity theory goes back to 1928 with Einstein's attempt to unify gravity and electromagnetism through the introduction of a tetrad (vierbein) field along with the concept of absolute parallelism or teleparallelism [9]. In the teleparallel gravity (TG) theories the dynamical object is not the metric g µν but a set of tetrad fields e a (x µ ), and rather than the familiar torsionless Levi-Civita connection of general relativity, a Weitzenböck connection (which has no curvature but only torsion) is used to define the covariant derivative. The gravitational field equation of TG is then described in terms of the torsion instead of the curvature [10][11][12]. In formal analogy with the f (R), the f (T ) gravity theory was suggested by extending the Lagrangian of teleparallel gravity to a function f (T ) of a torsion scalar T [6,7]. In comparison with f (R) gravity in the metric formalism, whose field equations are of the fourth order, f (T ) gravity has the advantage that the dynamics are governed by second-order field equations.
The fact that f (T ) theories can potentially be used to explain the observed accelerating expansion along with the relative simplicity of their field equations has given birth to a number of papers on these gravity theories, in which several features of f (T ) gravity have been discussed, including observational cosmological constraints [13][14][15], solar system constraints [16], cosmological perturbations [17][18][19], dynamical behavior [20], spherically symmetric solutions [21], the existence of relativistic stars [22], the possibility of quantum divide crossing [23], cosmographic constraints [24], and the lack of local Lorentz invariance [25,27,28] which may give rise to undesirable outcomes from f (T ) gravity [29,30], although suitable tetrad fields can be chosen [31]. For some further references on several aspects of f (T ) gravity theories we refer the readers to Ref. [32].
In the framework of general relativity the so-called energy conditions have been used to derive remarkable results in a number of contexts. For example, the famous Hawking-Penrose singularity theorems invoke the strong energy condition (SEC) [33], whose violation allows for the observed accelerating expansion, and the proof of the second law of black hole thermodynamics requires null energy conditions (NEC) [34,35].
On macroscopic scales relevant for cosmology, the confrontation of the energy conditions predictions with observational data is another important issue that has been considered in a number of recent articles. In this regard, since the pioneering works by Visser [36], a number of articles have been published concerning this confrontation by using model-independent energy-conditions bounds on the cosmological observable quantities, such as the distance modulus, lookback time, and deceleration and curvature parameters [37][38][39][40][41][42][43][44].
Owing to their role in several important issues in general relativity and cosmology, the energy conditions have also been investigated in several frameworks of modified gravity theories, including f (R) gravity [45,46], gravity with nonminimal coupling between curvature and matter [47], Gauss-Bonnet gravity [48], modified f (G) gravity [49], and Brans-Dicke theories [50] (see also the related Refs. [51,52]).
In this article, to proceed further with these investigations on the potentialities, difficulties, and limitations of f (T ) gravity theories, we derive the energy conditions for the general functional form of f (T ) and discuss some concrete examples of these bounds by using observational constraints on the Hubble and the deceleration parameters. The null and strong energy conditions (NEC and SEC) are derived in the framework of f (T ) from first principles, i.e., from the purely geometric Raychaudhuri equation along with the requirement that gravity is attractive. We find that the NEC and the SEC in general f (T ) gravity, although similar, are different from those of Einstein's gravity and f (R) gravity, but in the limiting case f (T ) = T , the standard general relativity forms for theses energy conditions are recovered. The resulting inequalities for the SEC and NEC in the f (T ) gravity framework are then compared with what would be obtained by translating these energy conditions in terms of an effective energy-momentum tensor for f (T ) gravity.There emerges from this comparison a natural formulation for the weak and dominant energy conditions (WEC and DEC) in the context of f (T ) gravity, which also reduce to the standard GR forms for these conditions in the limit f (T ) = T . As a concrete application of the energy conditions for spatially homogeneous and isotropic f (T ) cosmology, we use recent estimated values of the Hubble and the deceleration parameters to set bounds from the WEC on the parameters of two specific families of f (T ) gravity theories.
Our paper is organized as follows. In Sec. II, we give a brief review on the f (T ) theories and derive the field equations. In Sec. III, using the purely geometric Raychaudhuri equations for timelike and null congruences of curves, we derive the SEC and NEC from first principles, and the WEC and DEC through an effective energymomentum tensor. In Sec. IV we use the constraints on present-day values of cosmographic parameters to set constraints on exponential as well as on the Born-Infeld f (T ) gravity from the WEC. Finally, conclusions and final remarks are presented in Sec. V.
II. f (T) GRAVITY THEORY
In this section, we briefly introduce the teleparallel gravity and its generalization known as f (T ) gravity. We begin by recalling that the dynamical variables in teleparallel gravity are the vierbein or tetrad fields, e a (x µ ), which is a set of four (a = 0, · · · , 3) vectors defining a local orthonormal frame at every point x µ of the spacetime manifold. The tetrad vectors field e a (x µ ) are vectors in the tangent space and can be expressed in terms of a coordinate basis as e a (x µ ) = e µ a ∂ µ . The spacetime metric tensor and the tetrads are related by 2 where η ab = diag (1, −1, −1, −1) is the Minkowski metric of the tangent space at x µ . It follows that the relation between frame components, e µ a , and coframe components, e a µ , are given by In general relativity one uses the Levi-Civita connection which leads to nonzero spacetime curvature but zero torsion.
In teleparallel gravity, instead of the Levi-Civita connection, one uses the Weitzenböck connection which is given by An immediate consequence of this definition is that the covariant derivative, D µ , of the tetrad fields vanishes identically. This equation leads to a zero curvature but nonzero torsion.
To clarify the interrelations between Weitzenböck and Levi-Civita connections, one needs to introduce the torsion and contorsion tensors, which are given, respectively, by where above • Γ ρ µν is the Levi-Civita connection. Now, if one further defines the so-called super-potential one obtains the torsion scalar which is used as the Lagrangian density in formulation of the teleparallel gravity theory, which is given by where e = det(e a µ ) = √ −g, κ 2 = 8πG, and G is the gravitational constant. Now, by taking an arbitrary function f of the torsion scalar T , one obtains the Lagrangian density of f (T ) gravity theory, that is Now, by adding a matter Lagrangian density L M to Eq. (11) and varying the resultant action with respect to the vierbein, one obtains the following field equation for f (T ) gravity: and Θ σ a is the energy-momentum tensor of the matter fields.
Here and in what follows we have chosen units such that κ 2 = c = 1.
To bring the field equations (12) to a form suitable for our purpose in the next section. To this end, we first note that if one multiply e −1 g µσ e a ν , both sides of (12), the resultant equation is such that the coefficient of that the term f T takes the form where the relation has been used. On the other hand, from the relation between and Weitzenböck connection and the Levi-Civita connection given by Eq. (7), one can write the Riemann tensor for the Levi-Civita connection in the form whose associated Ricci tensor can then be written as and thus obtain where G µν = R µν − (1/2) g µν R is the Einstein tensor. Finally, combining Eq. (13) and Eq. (18), the field equations for f (T ) gravity Eq. (12) can be rewritten in the form where To close this section, we note that since A µ µ = −(R + 2T ), the trace of Eq. (19), which can be used as an independent relation to simplify the field equation, can be expressed as where B = B µ µ and Θ = Θ µ µ .
A. Strong and null energy conditions
The ultimate origin of strong and null energy conditions is the Raychaudhuri equation together with the requirement that gravity is attractive. The Raychaudhuri equation gives temporal variation of the expansion θ of congruence of geodesics (for a review article see Ref. [57]). For a congruence of timelike geodesics whose tangent vector field is u µ Raychaudhuri equation reads where θ , σ µν and ω µν are, respectively, the expansion, shear, and rotation associated with the congruence defined by the vector field u µ , and R µν is the Ricci tensor. The evolution equation for the expansion of a congruence of null geodesics defined by a null vector field k µ has a similar form as the Raychaudhuri equation (22), but with a factor 1/2 rather than 1/3, and −R µν k µ k ν instead of −R µν u µ u ν as the last term (see Ref. [58] for more details). Thus, its reads where the kinematical quantities θ , σ µν and ω µν are now clearly associated with the congruence of null geodesics. An important point to be emphasized is that Raychaudhuri Eqs. (22) and (23) are purely geometric statements, and as such they make no reference to any theory of gravitation. Now, since the shear is a "spatial" tensor, i.e., σ 2 ≡ σ µν σ µν ≥ 0, from Eqs. (22) and (23), one has that for any hypersurface of orthogonal congruences (ω µν = 0), the conditions for gravity to remain attractive (dθ/dτ < 0) are given by Thus, as long as one can use the field equations of any given gravity theory to relate R µν to the energy-momentum tensor T µν , the above Raychaudhuri Eqs. (22) and (23), along with the requirement that gravity is attractive, lead to Eqs. (24) and (25), which can be employed to restrict the energy-momentum tensors in the framework of the gravity theory one is concerned with.
Equations (24) and (25) are ultimately the SEC and DEC stated in a coordinate-invariant way for an unfixed geometrical theory of gravitation. Hence, for example, in the framework of general relativity, they take, respectively, the forms 3 and which, for example, for a perfect fluid of density ρ and pressure p , i.e., for T µν = (ρ + p) u µ u ν − p g µν , reduce to the well-known forms of the SEC and NEC in general relativity ρ + 3p ≥ 0 and ρ + p ≥ 0 .
B. Energy conditions in f (T ) gravity
According to the previous section the Raychaudhuri equations together with the attractive character of the gravitational interaction give rise to Eqs. (24) and (25), which hold for any geometrical theory of gravitation. In what follows, we maintain this approach to derive the SEC and NEC in the f (T ) gravity context. To this end, we first rewrite the f (T ) field equation (19) in the form Here, Θ µν and Θ denote, respectively, the energy momentum tensor and its trace.
From Eq. (29) and by taking into account the trace equation (21), we have where Now, for the homogeneous and isotropic Friedmann-Lemaître-Robertson-Walker (FLRW) metric with scale factor a(t), i.e., g µν = diag(1, −a 2 , −a 2 , −a 2 ), from Eqs. (6) through (9) along with Eq. (20), we have where a dot denotes derivative with respect to time, H =ȧ/a is the Hubble parameter, and the simplest and suitable tetrad basis was used [31]. Now, for a perfect fluid of density ρ and pressure p, namely for Θ µν = (ρ + p)u µ u ν − p g µν with u µ = (1, 0, 0, 0) , (36) taking k µ = (1, a, 0, 0), we obtain the T µν and its trace T , namely and Thus, from equations (24) and (25) for a general f (T ) gravity, the strong energy condition (SEC) and the null energy condition (NEC) can be, respectively, written as and NEC : We note that the well-known forms for the SEC (ρ+3p ≥ 0) and NEC (ρ + p ≥ 0) in the framework of general relativity can be recovered as a particular case of the above SEC and DEC in the context of f (T ) gravity for the special case f (T ) = T , as one would expect.
To derive the weak and dominant energy conditions (WEC and DEC) in f (T ) gravity, it is important to realize that the above SEC and NEC inequalities [Eqs. (39) and (40)] can also be recast as an extension of the SEC and NEC conditions in the context of general relativity by defining suitably an effective energy-momentum tensor in the context of f (T ) gravity. In fact, in f (T ) gravity theories one can define an effective energy-momentum tensor as 4 from which one defines the effective energy density and the effective pressure in the FLRW by which in turn make apparent that the SEC and NEC given by Eqs. (39) and (40) can be obtained from the corresponding general relativity expressions [Eq. (28)] by using the above effective matter components. Thus, using the effective energy-momentum tensor approach, the weak energy condition (WEC) in f (T ) gravity ( ρ ef f ≥ 0 ) reduce to WEC: Similarly, the dominant energy condition (DEC) in f (T ) gravity ( ρ ef f ≥ | p | ) can be written in the form
IV. CONSTRAINING f (T) GRAVITY THEORIES
The energy conditions (39), (40), (44), and (45) can be used to place bounds on a given f (T ) in the context of FLRW models. To investigate such bounds, we first note that to ensure the positivity of the effective Newton gravity constant, one has f T > 0 [29]. Thus, after some algebra, in terms of present-day values for the cosmological parameters, the energy conditions (39), (40), (44), and (45) can be, respectively, rewritten as SEC: WEC: DEC: where q = −(ä/a) H −2 is the deceleration parameter, and a subscript 0 indicates the present-day value of the corresponding parameter.
To make concrete applications of the above conditions to set bounds on f (T ), we first note that apart from the WEC [Eq. (48)], all the above conditions depend on the current value of the pressure p 0 . Therefore, for simplicity in what follows we shall focus on the observational WEC constraints on f (T ) gravity. Furthermore, we will also take the best fit value H 0 = 0.718 as determined by Cappozzielo et al. [24].
A. Exponential f (T ) gravity As a first concrete example, we shall examine the WEC bounds on the parameter β of the following exponential family of f (T ) gravity theories [7,15,59]: with where the limit β = 0 corresponds to ΛCDM model, Ω m0 is the dimensionless matter density parameter, and T 0 = T (z = 0) is the current value for the torsion scalar. By using T 0 = −6H 2 0 , one finds from (48) the following WEC constraint Now we take Ω m0 = 0.272 +0.036 −0.034 -which arises from the combination of 557 Type Ia Supernovae (SNe Ia) Union 2 set, baryonic acoustic oscillation (BAO), and the cosmic microwave background (CMB) radiation at 95% confidence level -along with the above observational value of H 0 . These values lead to β > −1.256 for the relation (52) to be satisfied. Reciprocally, the inequality (52) is always fulfilled for all values β such that β > −1.256. This makes explicit the constraint on parameter β of the exponential f (T ) gravity [Eq. (50)] for the WEC fulfillment.
B. Born-Infeld f (T ) gravity As the second concrete example, we consider the Born-Infeld (BI) f (T ) gravity given by [53] where ǫ = 4Λ/λ is a dimensionless parameter, Λ is the cosmological constant, and λ is a Born-Infeld-like constant. This gravity theory has been considered in several cosmological contexts, which include the avoidance of singularity in the standard model [54], as a way to an inflationary scenario without inflaton [55], and also to bound the dynamics of the Hubble parameter [56]. Clearly, the BI f (T ) gravity (53) reduces to the standard TG (often referred to as TEGR) when λ → ∞. Here, we focus on the case λ > 0 [53]. In this case, the WEC takes the form This inequality holds for 0 < ǫ < 1 and which makes apparent that the range of ǫ in which the WEC is fulfilled coincides with that of an expanding universe where the cosmological constant is positive (type II of Ref. [53]). Furthermore, by using T 0 = −6H 2 0 , one finds from inequations (55) the WEC lower bound on the parameter λ in the BI teleparallel gravity, namely λ > 12.36 .
V. FINAL REMARKS
Motivated by the attempts to explain the observed accelerating expansion of the Universe with a modifying teleparallel gravitational theory, there have been many recent papers on f (T ) gravity. Despite the arbitrariness in the choice of different functional forms of f (T ), which call for ways of constraining the possible f (T ) gravity theories on physical grounds, several features of f (T ) gravity have been discussed in a number of recent articles.
In this paper we have proceeded further with the investigations on the potentialities, difficulties, and limitations of f (T ) gravity theories by deriving the classical energy conditions in the f (T ) gravity context. Starting from the Raychaudhuri equation along with the requirement that gravity is attractive, we have derived the null and strong energy conditions in the framework of f (T ) gravity and shown that, although similar, they differ from NEC and SEC of general relativity, but in the limiting case f (T ) = T , they reduce to well-known NEC and SEC of Einstein's gravitational theory. The comparison of the NEC and SEC inequalities [Eqs. (39) and (40)] with those which would be obtained by translating these energy conditions in terms of an effective energy-momentum tensor for f (T ) gravity, enabled us to obtain the general expressions for the weak and dominant energy conditions [Eqs. (44) and (45)], which also reduce to the known corresponding energy conditions in general relativity in the limit f (T ) = T .
As concrete examples of how these energy conditions requirements may constrain f (T ) gravity theories, we have discussed the WEC bounds on two different f (T ) families of theories, namely the exponential and Born-Infeld f (T ) gravity theories (Secs. IV A and IV B). To this end, we have used the current observational bounds on H 0 and Ω m0 to show that the WEC are fulfilled for β > −1.256 in the exponential f (T ) gravity, whereas for Born-Infeld f (T ) gravity the WEC fulfillment is guaranteed for any λ > 12.36 such that 0 < ǫ < 1 holds.
Finally, we emphasize that although the energy condi-tions in f (T ) gravity discussed in this paper have wellmotivated physical grounds (the attractive character of gravity together with the Raychadhuri equation), the question as to whether they should be employed to any solution of f (T ) gravity theories is an open question, which is ultimately related to the confrontation between theory and observations. We recall that in the context of Einstein's gravitational theory, this confrontation indicates that all energy conditions seem to have been violated in the recent past of cosmic evolution [37,44].
|
2012-09-24T02:30:08.000Z
|
2012-07-06T00:00:00.000
|
{
"year": 2012,
"sha1": "e37d29585785ad8ba0a74b1342f3b9584cf75e67",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1207.1503",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "e37d29585785ad8ba0a74b1342f3b9584cf75e67",
"s2fieldsofstudy": [
"Physics"
],
"extfieldsofstudy": [
"Physics"
]
}
|
18808770
|
pes2o/s2orc
|
v3-fos-license
|
Cytoplasmic accumulation of NCoR in malignant melanoma: consequences of altered gene repression and prognostic significance
Invasive malignant melanoma (MM) is an aggressive tumor with no curative therapy available in advanced stages. Nuclear corepressor (NCoR) is an essential regulator of gene transcription, and its function has been found deregulated in different types of cancer. In colorectal cancer cells, loss of nuclear NCoR is induced by Inhibitor of kappa B kinase (IKK) through the phosphorylation of specific serine residues. We here investigate whether NCoR function impacts in MM, which might have important diagnostic and prognostic significance. By IHC, we here determined the subcellular distribution of NCoR in a cohort of 63 primary invasive MM samples, and analyzed its possible correlation with specific clinical parameters. We therefore used a microarray-based strategy to determine global gene expression differences in samples with similar tumor stage, which differ in the presence of cytoplasmic or nuclear NCoR. We found that loss of nuclear NCoR results in upregulation of a specific cancer-related genetic signature, and is significantly associated with MM progression. Inhibition of IKK activity in melanoma cells reverts NCoR nuclear distribution and specific NCoR-regulated gene transcription. Analysis of public database demonstrated that inactivating NCoR mutations are highly prevalent in MM, showing features of driver oncogene.
INTRODUCTION
Malignant melanoma (MM) is the most aggressive form of skin cancer accounting for most skin cancer deaths. Worldwide raising MM incidences have been recorded in the last decades, and currently about 48,000 melanoma-related deaths occur every year. MM behavior mainly depends on the stage of disease but it varies from patient to patient, likely reflecting differences in the biological features of the tumor. www.impactjournals.com/oncotarget Three main molecular pathways are found invariably deregulated in melanocytic tumors, including the RAS-RAF-MEK-ERK pathway (through mutation of BRAF, NRAS or KIT), the p16INK4A-CDK4-RB pathway (through mutation of INK4A or CDK4), and the ARF-p53 pathway (through mutation of ARF or TP53). Less frequently, other pathways such as the PI3K-AKT pathway (through mutation of NRAS, PTEN or PIK3CA) and the canonical Wnt signaling pathway (through mutation of CTNNB1 or APC genes) have been also associated with MM [1]. The RAS/RAF/MEK/ERK pathway is a crucial regulator of proliferation and cell survival in different physiologic and pathologic systems, and it is supposed to contribute to tumor progression in part through the activation of other downstream pathways such as NF-κB [2].
The NF-κB transcription factor is a homo-or heterodimeric factor involving five different proteins (p50, p52, p65, c-Rel, and RelB), which plays an essential role in inflammation but also in the regulation of specific cellular functions such as apoptosis, proliferation, cell migration and metastasis. In the absence of external stimuli, NF-κB factors are mostly localized in the cytoplasm bound to the inhibitor of kappa B (IκB) proteins. Phosphorylation of IκB by the IκB kinase (IKK) complex leads to IκB ubiquitination and degradation, resulting in the release of NF-κB that then translocates to the nucleus to activate specific transcription. Thus, detection of nuclear NF-κB in both normal or neoplastic cells is indicative of NF-κB and IKK activation. Constitutive NF-κB/IKK activation has been previously found in different types of cancer including MM [3]. NF-κB plays an important role in preventing tumor cell apoptosis through the induction of anti-apoptotic genes such as the Inhibitors of Apoptosis (IAP) c-IAP1 and c-IAP2, or the melanoma inhibitor of apoptosis (ML-IAP) [4]. NF-κB also controls the expression of several chemokines, interleukin IL-1 and IL-6, the vascular endothelial growth factor (VEGF), as well as other factors that are known to impact in MM progression [4].
Besides their role in NF-κB activation, the IKK kinases regulate other substrates that are directly involved in the control of gene expression such as the silencing mediator of retinoic and thyroid hormone receptors (SMRT) and the nuclear receptor corepressor (NCoR). Both proteins are components of specific multi-protein complexes that contain histone deacetylases (HDACs) and act as critical regulators of gene repression. Alterations in the repressive function of HDAC and NCoR have been found associated to different types of cancer, although little is known on how their deregulation impacts in MM.
Because the IKK kinases promote the exportation of NCoR (cNCoR) and SMRT [5,6] corepressors to cytoplasm in colorectal cancer (CRC) cells, we speculated that IKK activity might also affect NCoR distribution and specific gene transcription in MM, and tumor behavior. In this study, we have determined the subcellular distribution of NCoR, and the presence of nuclear p65-NF-κB as a surrogate marker of IKK complex activation, in a set of human MM samples and melanocytic benign nevi. In addition, we have performed a large-scale gene expression analysis of MM samples with different levels of nuclear NCoR (nNCoR). Our results provide molecular information for the classification of MM subtypes, but also for prognostic and stratification of patients that could benefit for future personalized therapies.
NCoR shows a heterogeneous subcellular distribution in MM samples
By IHC, we found that NCoR protein exhibits a nuclear distribution in melanocytes of nevi samples. However, malignant melanocytes in MM samples showed a variable distribution, which in some cases resulted in the loss of nNCoR, occasionally associated with its cytoplasmic accumulation. More specifically, we observed that all in situ MM samples contained nNCoR similar to benign nevi. In contrast, the number of cells that lost nNCoR at the time of diagnosis correlated with increased tumor staging ( Figure 1A and 1B), whereas the vast majority of metastatic samples disclosed an absent or minimal number of melanocytic cells with nNCoR. To determine the predictive value of loss of nNCoR in the primary tumors, we analyzed NCoR distribution in 63 primary tumor samples from MM patients with different Breslow index using the non-parametric Spearman test. We found an inverse correlation (-0.628) between the percentages of nNCoR and the Breslow index (p < 0.001), with samples showing lower percentage of nNCoR positive cells corresponding to the greater Breslow index ( Figure 1B, Table 1). Next, we analyzed the possible relationship between loss of nNCoR and other prognostic indicators of MM. The analysis demonstrated a significant association between loss of nNCoR localization and higher mitotic index (p < 0.01) and a statistical trend with ulceration (p = 0.051) (see Table 1). However, no differences in other histopathological features (regression or prevalence of inflammatory component), age and gender were recorded between MM with nNCoR or cNCoR. Moreover, we did not detect any significant association between BRAF mutational status and NCoR distribution in 18 samples analyzed (all samples with available material).
Accumulation of cNCoR showed a correlation of 0.551 with p65 nuclear distribution (indicative of IKK activation) ( Figure 1C and 1D), which associated with MM progression. To further validate the relevance of NCoR localization as a predictive biomarker, patients were followed for a median of 59.4 months and analyzed based on the presence of nNCoR. Mortality rate from all causes was significantly higher in those cases with less than 70% of cells containing nNCoR (from now on cNCoR) compared with tumors maintaining nNCoR in more than 70% of cells (from now on nNCoR), with overall survival (OS) at 5 years of 60.8% and 96% respectively. The OS curves of each group, considering a cut-off above and below 70% of cells carrying nNCoR, were significantly different between both groups with a log rank of 14.626 and p < 0.001 (not depicted). Most importantly, the MM specific mortality (DSS) at 5 years was 29% in the cNCoR group compared to 0% in the nNCoR group (log rank 11.568, p = 0.001) (Figure 2A). Considering those patients with skin limited disease at diagnosis (n = 49) significant differences in DSS were maintained (p = 0.010) between nNCoR and cNCoR ( Figure 2B). Disease free survival (DFS) in the cNCoR group was of 67.1% at 3 years and 58% at 5 years while no events were recorded in the nNCoR group (not depicted).
MM show different gene expression profiles depending on NCoR localization
To study the molecular bases that support the different behavior of patients carrying cNCoR and nNCoR tumors, we performed microarray expression analysis of invasive MM tumors with similar clinicopathologic characteristics and comparable mutational status for BRAF, NRAS and KIT (Table 2), which differ in the presence of cytoplasmic (n = 3) or nuclear (n = 4) NCoR. We included in the analysis 3 benign nevi cases in which NCoR was exclusively located in the nucleus. To identify transcripts significantly altered in tumors with nNCoR or cNCoR we fixed a minimum arbitrary fold change of 1.2 and a p-value ≤ 0.05 as the threshold cut-off (submitted to GEO: GSE56494).
An initial analysis of global expression profiles demonstrate that whereas nevi samples clustered together in the Principal Component Analysis space, MM samples segregate randomly, and do not properly differentiate between nNCoR and cNCoR tumors (not depicted). However, supervised analysis of the data identified a 185gene signature that clearly segregated cNCoR from nNCoR in our cohort of MM samples. Interestingly, nNCoR MM samples showed gene expression profiles comparable to the benign melanocytic nevi controls ( Figure 3A). To identify putative targets of NCoR repression that are upregulated during MM progression, we focused on those transcripts that were significantly over-represented in the cNCoR group compared to the other groups. A total of 73 genes were significantly upregulated in these samples including a subset of known cancer-related genes such as IL8 [7], the heat shock proteins HSPA1A/HSPA1B* and HSPA6 [8], the tumor suppressor TFPI2, SEMA3E that is involved in tissue development and cell adhesion [9], the cell cycle regulator MAGEC2/MAGEC3 [10], EYA4 which expression is altered in hepatocellular cacinoma, the protease inhibitor and tumor promoter Serpine 1 [11,12], the matrix metaloprotease inhibitor TIMP3 [13], and the melanoma cell adhesion molecule MCAM [14] (Supplementary Table S1). Functional analysis of an extended list of 654 genes (p < 0.05) performed in Ingenuity Pathway Analysis (IPA) and DAVID revealed that differentially expressed genes in nNCoR and cNCoR clustered into specific functional categories such as Toll-like signaling pathway, Cell Adhesion Molecules or negative regulation of apoptosis ( Figure 3B), among others. Moreover, IPA identified NF-κB as the best candidate complex to connect all the observed changes in gene expression between nNCoR and cNCoR ( Figure 3C).
IKK activity regulates NCoR distribution and gene expression in MM cells
Because we here identify NF-κB pathway as the node connecting the transcriptional program altered in cNCoR, and we previously associated the loss of nNCoR in colorectal cancer with increased activity of the NF-κB kinases, IKK, we aimed to investigate whether IKK regulates NCoR distribution in MM cells.
By immunofluorescence, and using the cell lines SKMEL-37 as experimental model, we found that melanoma cells contain variable levels of nNCoR (5.56 ± 2.40%) when grown under standard conditions. Incubation with the IKK inhibitor BAY-11-7082 for 60 minutes resulted in a partial redistribution of cNCoR into the nuclear compartment, up to 59.08 ± 12.19% ( Figure 4A). By western blot and IF using a specific antibody detecting active phosphorylated IKK (at serines 180 or 181 for IKKα and IKKβ, respectively), we confirmed that BAY11-7082 effectively blocked IKK activation in these cells ( Figure 4B and 4C). Next, we investigated whether nNCoR redistribution, which was achieved after IKK inhibition, was sufficient to revert the basal levels of selected NCoR target genes identified in the microarray screening. By qRT-PCR we found that treatment of SKMEL-37 cells with BAY11-7082 for 60 minutes was sufficient to consistently repress IL8 expression but also other NCoR-regulated genes such as GAGE12J, HSPA6 and SEMA3A ( Figure 4D), that are not direct targets of NF-κB.
NCoR is a putative driver of melanomagenesis
NCoR mutations have previously been found in different malignant disorders [15]. Thus, we investigated the possibility that mutations of the NCoR gene also contribute to tumorigenesis in MM. To that effect, we analyzed the existence of NCoR mutations in 369 MM samples from two public studies (see methods for details) and found 23 missense, 4 synonymous and 1 truncating (stop-gain) mutations. Using the IntOGen-mutations platform that distinguishes cancer driver genes from passenger mutations [16], we tested whether NCoR exhibits any signal of positive selection typical of driver genes across the cohorts of MM tumors studied. We found that the mutations present in the Cancer Genome Atlas MM dataset samples as a group were significantly biased towards high functional impact (p = 0.002) with respect to background mutations in the same MM samples. Specific driver characteristics of NCoR mutations include the homogeneous distribution along the primary structure of the protein (Figure 5), and a high prevalence of missense and truncating (stop-gain) mutations [17] (Supplementary Table S2).
DISCUSSION
Malignant melanoma is an aggressive tumor without curative treatment in advance stages. Genetic alterations that activate the MAPK pathway such as mutation on BRAF, N-RAS, and c-Kit, as well as alteration in the p16-CDK4-Rb-ARF-p53 and PTEN-PI3K function are found to promote tumor development through their interaction with other gene regulatory pathways such as NF-κB, and most of this information is currently used for therapy. However, and despite a significant knowledge that has been recently obtained on the molecular alterations leading to MM, the clinical outcome of MM is very variable among patients, and there are no reliable biomarkers (beyond the Breslow index and the staging) that can predict MM behavior and whether particular individuals will benefit from the available treatments. In this context, there is a need of additional molecular clues that allow patient stratification and facilitates future personalized target treatments for MM.
Previous studies from our group identified SMRT and NCoR proteins as key regulators of specific developmental-related genes (such as Hes1 and Herp2) in the intestinal epithelium. Transcriptional control of Hes1 and Herp2 was totally lost in tumors carrying active IKK/NF-κB. We now found that loss of NCoR-mediated gene repression is not restricted tot colorectal cancer but is also found in MM. Most importantly, whereas in colorectal cancer sample alterations of NCoR distribution affected 100% of the studied samples [6], the cytoplasmic accumulation of NCoR in MM correlated with tumor progression, and it is linked with worse clinical outcome. Thus, evaluation of NCoR distribution in melanoma might provide of additional information about prognosis, which is not applicable to colorectal cancer. The differences in nuclear NCoR maintenance or loss between different tumor types might reflect differences in the mechanisms driving tumor transformation, differences in tumor stage at the time of diagnosis, or differential usage of particular signaling pathways (i.e. the NF-κB/IKK pathway).
Since there were no indications about the biologic significance of nNCoR loss in primary MM, we have here performed a high throughput transcriptome analysis of 4 MM samples with nNCoR and 3 MM samples that had lost nNCoR, but they all shared comparable histopathologic characteristics including the tumor stage (all stage II), the anatomic localization, and the distribution of BRAF, NRAS and Kit mutations (see Table 2). Our results led to the identification of a subset of genes that are significantly upregulated in the MM group without nNCoR (group 1), and some of them had been previously associated with cancer progression in other systems. We propose that this here-identified genetic signature could be refined and used as a predictive tool for MM. Not only this, but most of the identified genes participate of pathways or exert specific functions that are targetable, thus we propose that our information should be used to develop and test novel potential therapeutic agents for MM treatment.
On the other hand, it has been previously shown that the NF-κB pathway is active in different MM subtypes. NF-κB is a family of transcription factors that is composed of five different subunit that associate to form homo-and hetero-dimers. The p65/p50 dimer is the most commonly involved in NF-κB signaling, and for this reason detecting nuclear p65-NF-κB is a reasonable approximation of the activation status of the NF-κB pathway. In our MM series, we found that the presence of nuclear p65-NF-κB in MM cells was associated with advanced disease, although the correlation between nuclear p65-NF-κB and cytoplasmic accumulation of NCoR did not reach statistical significance in these tumors. However, since different NF-κB dimers are induced by IKK in a context dependent fashion and because NF-κB regulation is very complex, we directly tested the possibility that NCoR cytoplasmic export in MM was downstream of IKK. With this purpose, we used melanoma cells that showed variable levels of nNCoR. We found that these cells contain active IKK, and that inhibition of IKK by BAY11-7082 reverted nNCoR distribution leading to the transcriptional repression of specific NCoR target genes, which could have clinical or therapeutic implications. Interestingly, it was previously found that nNCoR loss in neuroblastoma is due to a noncanonical activation of the NF-κB pathway by PEDF [18] that could also be explored in MM.
In conclusion, or results indicate that determination of NCoR distribution in MM can be used as a predictive factor, but most importantly as a criterion for patient stratification and identification of individuals that could benefit from future therapies based on NCoR/IKK function modulation. Moreover, differences in the expression profiles associated with the presence or absence of nuclear NCoR in MM might not only give insight in the pathogenic pathways underlying MM development or progression, but also provide a mechanism-driven base for specific therapeutic intervention of selected patients.
METHODS Patients
A total of 63 patients with primary invasive MM, specifically, 38 males and 25 females, with a median age of 69 years (ranging from 31 to 98) were included in the study. Clinical and pathological data have been gathered from a single institution (Parc de Salut MAR from Barcelona, Spain). Cases analyzed in this study include all patients with invasive MM that were recruited during the period 2000-2009 from Parc de Salut Mar-Biobank, only excluding non-conventional histological variants or doubtful cases of melanocytic proliferations with unknown significance. Cases were not selected on the basis to any specific clinical or histopathological criteria.
Forty-six patients presented a Breslow index > 1 mm (mean: 2.24 mm; median 1.4 mm). Fourteen tumors were ulcerated and 31 showed a mitotic index > 1 mitoses/mm 2 , 18 cases presented with 1 mitosis/mm 2 and no mitoses were seen in the remaining 14 cases. Patients were periodically controlled for a median follow-up of 59.4 months. Forty-nine patients were classified in stage I-II and 14 were in stages III-IV at diagnosis, and from those in which sentinel lymph node biopsies were performed (32 out of 63 patients), 19% (6 patients) presented micrometastases. During the follow-up period, 48 patients did not progress in the disease from the initial stage (IA-IIIA), whereas 15 cases (24%) progressed to more advanced stages. These clinical data are representative of an average MM population.
Immunohistochemistry (IHC) and immunofluorescence (IF)
IHC and IF analysis of the MM samples and cell lines was performed using the following antibodies: Goat polyclonal anti-NCoR (sc-C20) and p65-NF-κB (sc-372) from Santa Cruz biotechnology inc. Antigen retrieval was achieved by boiling the samples for 10 minutes in Tris/EDTA buffer at pH 9.0. Tissue sections were incubated overnight with the corresponding primary antibody (in PBS plus 0.1% BSA), and then extensively washed and incubated with the secondary biotin-labeled antibodies (1:200) at room temperature for 30 minutes. Immunoreactivity was revealed using a streptavidinbiotin-peroxidase complex (sABC-HRP, DakoCytomation K0377; 1:100; DAKO) or Alcaline phosphatase, and developed with diaminobenzidine (DAB) solution (Sigma-Aldrich, Zwijndrecht, The Netherlands) or Fast Red TR (Sigma), respectively. All sections were then counterstained with hematoxylin, and Melan-A was used to confirm the neoplastic (melanocytic) nature of the cells stained for NCoR and p65-NF-κB. Cells were counted directly under a 40× magnification and the percentage of tumor cells that exhibited positive immunoreactivity was determined. At least 100 neoplastic cells were counted for each sample. The mean percentage obtained from 3 independent reads by 2 pathologists was recorded. Nuclear and cytoplasmic staining was evaluated separately. Cell staining scores were analyzed as percentages of cells in each degree of staining. For nuclear NCoR determination only the number (percentage) of nuclei with moderate to intense expression was considered. Intense nuclear staining was considered when it was comparable to the non-tumor samples used as controls. Moderate refers to positivity that was less intense than normal controls. No or light expression was considered when we did not detect a consistent staining at the 40× magnification. Staining scores were compared using Student's t-test or analysis of variance (ANOVA) and non-parametric U-Mann-Whitney was employed when data did not meet the conditions of application of these tests. Likewise comparisons in which subgroups did not reach the minimum sample size Bonferroni corrections were used. Kaplan Meier survival curves were calculated using the SPSS program. The morphological variables and IHC data were recorded and included the histopathological subtype of MM, measurements of invasion (Breslow index and Clark level), the presence of ulceration, mitotic rate, features of regression, intratumor inflammatory infiltrate, the presence of microsatellitosis and vascular invasion. In addition, 40 benign melanocytic lesions (intradermal, junctional and dysplastic melanocytic nevi) were used as controls. Sample size for comparing two means with unilateral contrast, being the percentage of positively stained melanocytes the main variable, was calculated according the algorithm, n = ((za + z2b)^2 *2*s ^2)/Dm^. We used the semi-quantitative Histoscore index: (0x%) + (1x%) + (2x%) + (3x%) to perform parametric statistical tests to determine the association between nuclear NCoR and p65 stainings.
Obtaining and processing somatic mutations datasets
We obtained 51 somatic mutations datasets from the International Cancer Genome Consortium (ICGC), the Cancer Genome Atlas (TCGA), and literature searches, analogously as described [16]. Nine ICGC datasets were downloaded directly from the Data Coordination Centre (DCC) Biomart [19]. Nineteen TCGA datasets were downloaded from the Synapse platform (syn1729383) as MAF files. We obtained thirteen additional datasets from a comprehensive recent study of tumor mutational signatures [20]. Finally, a manual PubMed search allowed us to identify ten other somatic mutations datasets that had been produced by research groups outside these large initiatives. Among these 51 datasets, there were two MM studies, one carried out by the Broad Institute (and obtained via [20]) and a second one developed within the TCGA (downloaded from Synapse) totaling 369 samples.
Running the IntOGen-mutations pipeline
The ICGC and mutational signatures datasets were obtained already in the tab-separated format accepted by the pipeline. The MAF files obtained from synapse were parsed and transformed to this tab-separated format. We manually parsed supplementary files of the papers reporting individual studies to extract the lists of somatic mutations detected across tumor samples and then transformed them into tab-separated files. The first part of the IntOGen-mutations pipeline [16] assesses the potential functional impact of somatic mutations detected across the cohort of tumor samples. The Ensembl Variant Effect Predictor (VEP, v.70 script [21] and pre-computed cache files, downloaded from the Ensembl ftp site (ftp://ftp. ensembl.org/pub/), are used to determine the consequences of somatic mutations to annotated protein-coding genes. We employed the results of this first part to compute the number of samples with mutations in NCoR in different datasets -subsequently aggregated per tumor type-, as well as their consequence types. The pipeline also obtains SIFT [22] and PolyPhen2 [23] functional impact scores from the VEP. Pre-computed Mutation Assessor [24] functional impact obtained from its author's webserver (www.mutationassessor.org) during the installation of the pipeline are queried locally during execution. The pipeline implements an expression filter to disregard genes that are not expressed across the tumor types analyzed.
The Oncodrive FM approach has also been described elsewhere [17]. Briefly, Oncodrive FM receives as input the list of synonymous, stop-gained, stop-lost, non-synonymous and frameshift-indel mutations and their corresponding SIFT, PolyPhen2 and Mutation Assessor scores. Then, it assesses whether any gene shows a trend toward the accumulation of mutations with high functional impact as compared to the background distribution of these functional impact scores in all mutations detected across the cohort of tumor samples (FM bias). For each functional impact score method included in the pipeline, the method produces an empirical p-value that evaluates this FM bias. These three p-values are subsequently combined using Fisher's approach to produce one integrated p-value for each gene. To account for possible non-dependence between the three p-values included in the combination, the IntOGen-mutations web discovery tool considers as significant those with a false discovery rate (FDR) below 0.01. We checked specifically whether NCoR exhibits the FM bias in any mutational dataset. (When more than one dataset of the same tumor type was available, we retained only the most significant q-value: see Supplementary Table S2)
Gene ontology analysis of differentially expressed genes
Overrepresentation analysis of Gene Ontology (GO) categories and pathways was used to explore the functions associated with NCoR regulated genes in MM. Gene set enrichment analysis (GSEA) identifies gene sets consisting of co-regulated genes; Gene Ontology (GO) gene sets are based on ontologies and do not necessarily comprise co-regulated genes. Gene sets collected from various sources such as online pathway databases, publications in PubMed, and knowledge of domain experts. The gene set page for each gene set lists its source. Computational gene sets defined by mining large collections of cancer-oriented microarray data. Gene sets are named by GO term and contain genes annotated by that term. (http://www.broadinstitute.org/gsea/msigdb/ collections.jsp#C2); IPA: (http://www.ingenuity.com).
Ethical considerations
The approval for the study was provided by the Comitè Ètic d' Investigació Clínica from Institut Municipal d'Assistència Sanitària (CEIC-PSMAR) and written informed consent was obtained from all patients according to the National and International guidelines (code of ethics, Declaration of Helsinki) and the legal regulations on data privacy (Law 15/1999 of 13 December on the Protection of Personal Data) were considered. All samples stored in the tumor bank (Hospital del Mar in Barcelona, Spain).
Statistical analysis
Non-parametric Kruskal-Wallis test and Mann-Whitney U test were used to determine the association between NCoR and various categorical clinicopathologic parameters. To assess the relationship between NCoR with Breslow index and mitotic rate, Spearman's correlation coefficient was performed. Kaplan-Meier method with a Log-Rank test was used to compare overall survival, disease specific survival and disease free survival among groups. P values less than 0.05 were considered as statistically significant. Analysis performed with SPSS 18.0 (IBM Corp.).
|
2017-04-20T05:09:39.806Z
|
2015-03-19T00:00:00.000
|
{
"year": 2015,
"sha1": "0a1ed365df9acfa82d95db96ebb1b60abe00bcdc",
"oa_license": "CCBY",
"oa_url": "http://www.oncotarget.com/index.php?journal=oncotarget&op=download&page=article&path[]=3252&path[]=7369",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "838e29bf80806ee5af6c1293c4d382ca5f27eae3",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Biology",
"Medicine"
]
}
|
212695839
|
pes2o/s2orc
|
v3-fos-license
|
Combined Approach for Teachers’ Evaluation Aspects Identification Using Dictionary and Patterns Based
Teacher performance evaluation is a common method and often used for evaluates teaching quality in higher education. With the rapid growth of opinion mining technique. Aspect-based opinion mining application has been possibly employed to extraction and summarization of students' comments for teacher evaluation. However, to automated teacher evaluation features identification from a large number of students' comments collection is very hard work. This study has the goal to address this problem. The main objectives of the proposed method are: (1) to identify teacher evaluation aspects, (2) to compare the efficiency of dictionary based, patterns based and the combination of them, and (3) to enhance the accuracy result in the teachers’ evaluation aspects identification from the unstructured text of students' feedbacks. The students' feedbacks were collected by questionnaires and the dataset was constructed manually with a total of 4,496 sentences from 300 undergraduate student responses in 10 subjects by purposive sampling and the collection of positive and negative sentences from 30 participants group interviewed in the workshop. Both approaches were applied to identify the frequency teachers' evaluation aspects. The experimental results found that our proposed approach provided reasonably more accurate results, the combination approach enhanced a significantly average of precision and recall. For future work, we focus on the application of new linguistic patterns and non-frequency aspects in order to increase the accuracy result. Keywords—aspects identification, lexicon relation, linguistic pattern, opinion mining, teacher evaluation.
I. INTRODUCTION
With the rapid growth of the internet and social media, people can discuss and share their opinions about several issues in forums, blogs, microblogs, and each other social network sites. Recently, with a large number of those textual information. Many scholars tried to analyze and apply them into more valuable issues such as product reviews, tourism, political, stock price, medical and et al. However, it is very difficult to read and analyze those unstructured data. Therefore, the automated tools for discovery those hidden opinions and summarized them into the usable forms are required [1], [2], [3]. Since 2006 Pang & Lee proposed the technique namely Opinion mining or sentiment analysis to deal with their problems for the extraction and summarization of people's opinions from a large volume of unstructured texts [4]. It is the field of automatic extraction the evaluation information from subjective text and the computational analysis about people's attitudes, opinions, appraisals, emotional and sentiments which express in text [1], [3].
In recent year, sentiment analysis was grown up. Therefore, it has been widely used in the evaluation of products and services from customer reviews and it has been applying to the evaluation of political, tourism, medical and other areas [5], [6], [7], [8], [9]. However, with the rapid growth of sentiment analysis technique. The main objective of this technique is to discover opinions and sentiments which express in text, and then classify their polarity. The classification process was divided into three levels: (1) document-level aims to classify an opinion document as expressing a positive or negative opinion, (2) sentence-level aims to classify sentiment expressed in each sentence, and (3) aspectlevel aims to classify the sentiment with respect to the specific aspects of entities. The opinion holders can define different opinions for different aspects of the same entity [2], [10].
Regarding the teacher evaluation from the students' feedback, most scholars focused to numerical students' feedbacks analysis using the statistical technique while some scholars have been done on students' text comments by applying sentiment analysis to analyze the student's feeling and their opinion about the particular teacher. As mentioned in [11] the authors employed the sentiment analysis to study the student's perception. The result shown the word "confuse" from 60 percentages of all student comments. Then he decided to improve his teaching style and repeat this section again. Moreover, in our previous paper the result indicated that it might be possible to convert from qualitative to a quantitative type of teacher evaluation by performing lexicon-based sentiment analysis [12]. As mentioned above this indicate that it might be possible to apply this technique to teacher evaluation. Aspect-based opinion mining application has been possibly employed to extraction and summarization of students' comments for earth teacher evaluation aspects. However, the analysis of students' text comment is difficult and implicates various stages to get summarize results. It consists of three core subtasks: (1) identifying teacher evaluation aspect that student commented on, (2) determining sentiment polarities on teacher evaluation aspects, (3) generating the teacher evaluation summary [1], [13].
In this paper, we focus on the identifying of teacher evaluation aspects from student commented. We studied the efficiency of dictionary based, patterns based and the combination of them to enhance the teachers' evaluation aspects identification from the unstructured text of students' feedbacks. The rest of this paper is divided as follows: Section II describes the related works done previously, Section III describes the proposed methodology, Section IV describes the research results and Section V describes the conclusion and future work.
II. RELATED WORK
Sentiment analysis is implemented to explore the hidden knowledge for evaluation. In educational domain, it was applied to explore the answers relevant to student opinion from open-ended questions in the evaluation process. We discovered seven works that specified this idea as follows.
First, in [14] proposed the system for analyzed and summarized the student feedbacks about each topic. According to [8] studied the course evaluation form student comments using sentiment dictionary to identify the frequency words and sentiment words, calculated the sentiment scores and represented with tag clouds. Moreover, [15] proposed the system for analyzed and summarized the student feedbacks from SMS and calculate the sentiment scores. Their system represented the output in a graph. On the other hand, [16] used the lexicon resource to study the students drop out behavior in Massive Open Online Course (MOOC). However, this work has some limitation. The sentiment word polarity was predicted based on the lexicon recourse of product reviews. In [17] proposed a combined method between Spanish lexical based sentiment analysis and machine learning techniques to analyze the students' feedback on Facebook. The results suggested that it is possible to perform sentiment analysis to analyze the students' feedback in Facebook with high accuracy. However, this work still has some limitation, all the words tagged as the same polarity get the same score. Similar to [18] proposed the construction of their teaching evaluation lexicon resource. In this work, the weight score of terms as defined by the experts with the ranged from 1.00 to 1.00. They employed three machine learning algorithms in their experimental in order to perform the sentiment classifications with a 97% highest accuracy of SVM. This proposed method can address the problem of automated sentiment orientation polarity definition in teaching evaluation, but it was constructed in Thai language. According to [19] study sentiment analysis about faculty evaluation. They considered Noun and Adjective extraction. In this study the frequency features and opinion words extraction from students' feedback using two pattern mining algorithms; e.g., Apriori and Generalized Sequential Pattern (GSP). The experimental results indicated that GSP is more efficient than Apriori for frequent features and opinion word extraction.
As mentioned above, in educational domain the application of sentiment analysis was used in various objectives; e.g., faculty evaluation, teaching evaluation, course-online evaluation and, teacher evaluation. It is possible to perform sentiment analysis in students' comment. Current researchers in this area focus on aspect-based sentiment analysis for extract and summarize the opinion about each teacher's evaluation aspects. The target of automatic sentiment analysis is improving the better accuracy result of teacher evaluation aspects identification, sentiment classification, and summarization. Therefore, in this study we proposed the new method to enhance teacher evaluation aspects identification.
III. PROPOSED METHODOLOGY
In this section, we described the overview of our proposed method. In order to automatic identify teachers' evaluation features from students' comments, we divided this section into three tasks as follows; 1) Setting teacher evaluation criteria 2) Data Sets and 3) Identifying Teachers' Evaluation Features.
3.2.Data Sets
In this study, based on the teacher evaluation criteria above the positive and negative students' feedbacks were collected from two sources as present in Table 2. Then the teachers' evaluation dataset was constructed manually from those data. We describe the stages of dataset construction as follows.
3.3.Stage 1: Questionnaires
The open-end questionnaires were designed based on the teacher evaluation criteria in Table 2. Then the students' feedbacks were collected from 10 subjects of Computing Faculty in Universiti Teknologi Malaysia during the second semester of 2015. In this stage, the purposive sampling was done to collect a small dataset from 300 undergraduate student responses with a total of 3,296 sentences.
3.4.Stage 2: Group Interview
To construct the bigger dataset, the open-end questionnaires were designed based on 16 teachers' evaluation sub-aspects from Table 2 and the 30 undergraduate students in each faculty were invited as the workshop participants. Then the positive and negative students' feedbacks regarding those teachers' evaluation sub-aspects were collected from them by group interviewed. In this stage, we could collect 1,200 positive and negative sentences regarding each teacher evaluation aspects. Example of positive and negative feedbacks as shown in Table 3.
3.5.Stage 3
In this stage, based on text data from those previous stages. The teachers' evaluation small dataset was constructed manually with a total of 4,496 sentences.
3.6.Stage 4
To construct the annotated dataset, then we tagged the teacher evaluation in each sentence from the collection of students' comments manually. In order to annotate the positive, negative sentences and the opinion target in each sentence. The amount of work is reasonable. • The grading system was well designed.
•
The material is easy to understand.
• He cannot explain well. • Bad approach in delivery lecture. • Too many assignments given at once. • The grading system was confused.
•
The teaching materials were not up-to-date.
3.7.Pre-processing
In this stage, we applied Python NLTK to prepare students' comments corpus from a collection and perform some data preprocessing to prepared text data for teacher evaluation aspects extraction as follows; -Split document to sentences -Converting to lower case -Removing punctuation -Removing numbers -Stripping white spaces -Removing stop words -Stemming
3.8.Identifying Teachers' Evaluation Aspects
In order to identify teachers' evaluation aspects from students' comments. In this paper, we describe the detail of the teacher evaluation aspects extraction approaches as follows.
1.
Dictionary-Based Approach Based on the teacher evaluation criteria in Table 2, the teacher evaluation items were collected as a small set of teacher evaluation opinion target words manually, and then based on dictionary-based approach a small set of seed teacher evaluation opinion target words was used to grow this set by searching their semantic relation from WordNet [32] using Python NLTK [33] for their semantic relation of synonyms (Syn), antonyms (Anto), hyponyms (Hypo) and hypernyms (Hyper). The newly found words are added to the seed list of teacher evaluation aspects words list and stop when no more new words are found. This newly words list was constructed as the lexicon resource for identify teachers' evaluation aspects from students' comments [34], [35], [36], [37], [38], [39].
2.
Patterns-Based Approach Since the opinionated sentences consisting of opinion targets and opinion words. Many scholars have used the sequences of noun and adjectives to identify opinion targets. This linguistic pattern called base noun phrase has been employed by various research work. In this paper we employed linguistic patterns in [40]
3.9.Experiments
In the stage of identifying teachers' evaluation aspects from students' comments in this study, the experiment was designed and conducted by three automated techniques. We describe the detail of those three different teachers' evaluation aspects extraction approaches as follows.
Using Dictionary-Based Approach
In this stage, we employed a simple technique of dictionary-based approach to identify teachers' evaluation aspects to constructed teachers' evaluation aspects words list. Then using the constructed words list to identify teacher evaluation aspects from students' comments and compared with the annotated sentences. The experimental was set up by using four different types of semantic relation words, e.g. (1) Syn, (2) Syn+Anto, (3) Syn+Anto+Hypo, and (4) Syn+Anto+Hypo+Hyper.
Using Patterns-Based Approach
In this stage, we employed the patterns-based approach to identify teachers' evaluation aspects by using Python NLTK, POS tagging and linguistic patterns [40] to identify teacher evaluation aspects from students' comments and compared with the annotated sentences. The experimental was set up by using three different patterns, e.g. (1) BNP, (2) dBNP, and (3) bBNP.
Using Combined Approach
Based on the previous stages above in this stage, we employed the combination of dictionary and patternsbased approach to identify teachers' evaluation aspects from students' comments and compared with the annotated sentences. The experimental was set up by using three different types of the combination of dictionary and pattern-based approach, e.g. (1) Syn+Anto+Hypo+Hyper+BNP, (2) Syn+Anto+Hypo+Hyper+BNP+dBNP, and (3) Syn+Anto+Hypo+Hyper+BNP+dBNP+bBNP as shown in Figure 1.
3.10.Evaluation
In order to evaluate our proposed teacher evaluation aspects identification algorithm performance, we perform the standard evaluation measures. To calculate Precision, Recall, Accuracy, and F-score all parameters was set up as follows; -True Positive : Number of extracted aspects which are target aspects. Based on the experimental results, in order to investigate and compare the efficiency between three different teacher evaluation aspect identification approaches, e.g (1) dictionary based, (2) pattern based and (3) combination approach as present in Table 6. As indicated in Table 5, the results obtained by the dictionary-based approach. We found that the using of Syn+Anto+Hypo+Hyper has higher accuracy than other semantic relations, it indicated that the using of semantic relation and the hybrid of them can improve the accuracy of teacher evaluation aspects identification. However, it was found that dictionarybased approaches which depend on semantic relation was not completed to identify teachers' evaluation opinion targets. There is some limit to identify the semantic relation between the similar of a single word and compound words, e.g. 'Teaching', 'Teaching style', 'Teaching material', 'Material'.
From the results obtained by the pattern-based approach. We found that the using of bBNP has the highest accuracy and all linguistic patterns have higher accuracy than each semantic relation, it indicated that the using of linguistic patterns can improve teachers' evaluation aspects identification. However, there is some limit to identify the semantic relation and lexicon relation between the other related opinion targets, e.g.
The syllabus is not clear. (#syllabus) The room is very hot. (#room) This course is very interesting. (#course) As mentioned above, syllabus, room and course are Noun that we obtained the pattern-based approach to identify teachers' evaluation aspects. However, these are not the opinion target in teacher evaluation domain. It was found that the patterns-based approach which depends on semantic relation were not completed to identify teacher evaluations' aspects.
From the results obtained by using the combination of dictionary and pattern-based approach. We found that Syn+Anto+Hypo+Hyper+BNP+dBNP+bBNP is outperformed. This indicated that the combination technique can improve the identification of both semantic relation and lexicon relation between the related teacher evaluation opinion targets.
V. CONCLUSION AND FUTURE WORK
In this paper, we explored the identification of teacher evaluation aspects from students' comments using dictionary and patterns-based approach. It was found that both dictionary-based approach which depends on lexicon relation and the patterns-based approach which depends on semantic relation were not completed to identify teacher evaluation opinion targets by its' technique. While the combination of them can improve the identification of both semantic relation and lexicon relation between the related teacher evaluation opinion targets. For future work, we focus on the application of new linguistic patterns and non-frequency aspects in order to increase the accuracy result.
|
2020-02-27T09:09:24.561Z
|
2019-07-13T00:00:00.000
|
{
"year": 2019,
"sha1": "ba37c90b565a7bc4756e8c25ee2d6e94e1a7ce4f",
"oa_license": null,
"oa_url": "https://doi.org/10.25124/ijies.v3i02.40",
"oa_status": "GOLD",
"pdf_src": "MergedPDFExtraction",
"pdf_hash": "8248bd1af2bbc500473cae5474c9c32477bdfdd7",
"s2fieldsofstudy": [
"Computer Science",
"Education"
],
"extfieldsofstudy": [
"Computer Science"
]
}
|
220521860
|
pes2o/s2orc
|
v3-fos-license
|
Comparison of DNA Extracted from Pediatric Saliva, Gingival Crevicular Fluid and Site-Specific Biofilm Samples
(1) Introduction: Due to the non-invasive nature of saliva, many methods have been used to isolate and collect DNA from saliva samples for microbial screening. Many oral microbes also inhabit the oral biofilm, which may represent significantly different microbial constituents that may contribute to oral health and disease, including caries and periodontal disorders. Moreover, the biofilm may vary within the same patient at different sites. Few studies have evaluated the comparison between DNA isolated from saliva and DNA from site-specific biofilm, with virtually no studies addressing this analysis among pediatric patients. (2) Methods: An existing repository of paper point derived biofilm, gingival crevicular fluid (GCF), and unstimulated saliva samples previously collected from pediatric patients (n = 47) was identified. DNA was isolated from biofilm sites (tongue, upper buccal molar, mandibular lingual incisor), and GCF and saliva were used for quantitative DNA comparison using a phenol:chloroform extraction. A quantitative and qualitative analysis was performed using the NanoDrop 2000 spectrophotometer using absorbance readings at A230 nm, A260 nm and A280 nm. (3) Results: These data demonstrated the successful isolation of DNA from all of the patient samples, with the highest concentrations observed among unstimulated saliva (4264.1 ng/μL) and the lowest derived from GCF (1771.5 ng/μL). No differences were observed between males and females or minorities and non-minority patients. In addition, comparison of the overall concentrations of DNA obtained from adult samples was slightly higher than, but not significantly different from, the concentrations obtained from pediatric samples (p = 0.2827). A real-time quantitative qPCR screening revealed that all of the samples evaluated harbored bacterial and human DNA of sufficient quantity and quality for a molecular screening greater than the limit of detection (ΔRn = 0.01). (4) Conclusions: Many methods are currently available to provide the sampling and screening of saliva and specific sites within the oral cavity, but the validation and comparison of simple and low-cost methods, that include paper point sampling and unstimulated saliva collection, may suggest these methods and protocols provide sufficient DNA quality and quantity for molecular screening and other comparison applications. In addition, although heterogeneity will be a constant and consistent feature between patient samples, standardized methods that provide similar and consistent DNA from various oral sites may provide needed consistency for screening and molecular analysis.
Introduction
Due to the non-invasive nature of saliva collection, many methods have been used to isolate and collect DNA from saliva samples for microbial screening [1][2][3]. Screening methods for these samples have included highly specific and technical procedures involving immunofluorescence microscopy, proteomic analysis and next-generation sequencing [3][4][5][6]. Among the most sensitive, rapid, reliable and widely available methods for oral microbial screening is the polymerase chain reaction (PCR) [7,8]. Although a few studies have made detailed analyses from DNA isolated from saliva and other oral fluids, such as gingival crevicular fluid (GCF), as well as oral biofilm comparisons and analyses [9,10], to date there have been virtually no comparisons with oral site-specific biofilms to determine the comparative quality and quantity for a subsequent microbial PCR screening [11].
Many oral microbes also inhabit the oral biofilm, which may represent significantly different microbial constituents that may contribute to oral health and disease [12,13]. The production and growth of oral biofilms are implicated in oral diseases, including caries and periodontal disorders [14,15]. Therefore, the sampling and analysis of oral biofilms may also represent a significant source of readily available, non-invasive data regarding oral health and disease [16,17].
Although a previous study from this group evaluated the comparison between DNA isolated from saliva or GCF with DNA from site-specific biofilm in adults [11], virtually no studies addressing this analysis among pediatric patients have been completed. Based upon this lack of analytic data, the primary goal of this study was to make a quantitative and qualitative comparison to determine the suitability of the DNA from these oral samples for microbial PCR screening.
Study Approval
This was a retrospective analysis of previously collected saliva and oral samples. The protocol for this study was reviewed and approved exempt by the University of Nevada Las Vegas Office for the Protection of Research Subjects (OPRS) on 6 February 2015 under Protocol#1502-506M. The original protocol OPRS#1305-4466M "The Prevalence of Oral Microbes in Saliva from the UNLV School of Dental Medicine Pediatric and Adult Clinical Population" was approved on 31 May 2013. These protocols (and OPRS approval) comply with the Health Information Portability and Accountability Act (HIPAA) standards for the privacy of individually identifiable health information (Privacy Rule), issued by the United States Department of Health and Human Services (HHS).
Original Sample Collection
From the original protocol, patients (or parents/guardians if the patient was under 18 years of age) were asked for their voluntary participation. Any patient or parent/guardian that declined to participate was excluded. Patients were asked to provide informed consent. If patients were under 18 years of age, patients were asked to provide pediatric assent with the parent/guardian providing informed consent. No patients were given money or services for their participation.
Each patient was then given a sterile saliva collection tube and were requested to produce up to five mL of unstimulated saliva. Gingival crevicular fluid was acquired from the buccal gingival crevice of the maxillary central incisor using sterile paper points (Size 40 Quality Endodontic Distributors, QED) for a standardized three minutes. Biofilm samples were acquired from the dorsum of the tongue, supragingival biofilm on the buccal surface of the first upper molar (maxillary), and lingual surface of the central incisor (mandibular), as previously described [11]. Each sample was stored on ice and transferred to a biomedical biosafety level 2 (BSL-2) laboratory for long-term storage and processing. Samples from each patient were given a randomly generated, non-duplicated number to avoid any specific patient information or other identifying information from being associated with each sample. Only basic demographic information, such as age, sex and race/ethnicity, was noted.
DNA Isolation
Samples were thawed and DNA was isolated using the phenol:choloform extraction method with TRIzol reagent (Invitrogen) specifically designed to isolate high-quality nucleic acids (RNA, DNA) from tissues or fluids, as previously described [18,19]. In brief, 100 µL of thawed saliva was removed for analysis. For each paper point sample, 100 µL of sterile 1X phosphate buffered saline (PBS) was added to each paper point containing sample collection tubes and was vortexed to release any microbial or human cells. This solution was added to 300 µL of TRIzol reagent prior to incubation. After five minutes, 200 µL of chloroform was added and incubated for an additional five minutes. Each sample was subsequently centrifuged at 12,000× g, or the relative centrifugal force (RCF), for fifteen minutes at 4 • C.
The nucleic acid-containing phase was transferred to a new microcentrifuge tube with the addition of 100% isopropanol, which was gently mixed to precipitate the nucleic acids from the solution. These samples were then centrifuged for an additional five minutes to pellet the DNA. The supernatant was aspirated and the pellet washed with 100% ethanol and was centrifuged for an additional five minutes. The ethanol was aspirated and the DNA pellet was resuspended in 100 µL of DNA rehydration solution for analysis. Negative controls for liquid samples, including saliva and GCF (distilled water), as well as for paper points (sterile/blank paper points), were used where appropriate.
DNA Analysis
The concentration and purity of the DNA was determined using the NanoDrop spectrophotometer (ThermoFisher) with absorbance readings at 260 and 280 nm. An estimation of the DNA concentration can be obtained using the absorbance at A260 nm and the correction for turbidity at A320 nm, adjusting for the dilution factor. An estimation of the purity can be determined from the ratio of A260 nm and A280 nm with the corresponding dilution factor. High quality DNA will correspond with a ratio of A260:A280, ranging between 1.7-2.0. Absorbance at A230 was also measured to provide the A260:A230 ratio-commonly accepted as an accurate measure of the residual chemical contamination from nucleic acid extraction procedures. Expected values typically range from approximately 2.0 to 2.2 for "pure" nucleic acid samples acceptable for qPCR analysis and screening.
qPCR Screening
Screening for the presence of microbial and human DNA was accomplished using a quantitative polymerase chain reaction (qPCR) using primers for bacterial 16S rRNA and human glyceraldehyde 3-phosphate dehydrogenase (GAPDH), with specifications that included an initial incubation at 50 • C for two minutes, followed by denaturation at 95 • C for 10 min, and 25 cycles: denaturation at 95 • C for 15 s and annealing at 51 • C for one minute. The reaction included: 15 µL TaqMan universal PCR master mix (Applied Biosystems), 0.6 µL of the following primers at a concentration of 10 uM from Eurofins MWG Operon (Huntsville, AL), and 0.75 µL of probe resulting in a final probe concentration of 0.2 uM, with 2 µL of the DNA samples. Sterile, nuclease-free distilled water from Promega (Madison, WI, USA) was used to adjust the final reaction volume to 30 µL. Each screening was performed in duplicate.
The primers synthesized from Eurofins MWG Operon (Huntsville, AL, USA) were:
qPCR Analysis
Results from the real-time qPCR screening were analyzed using the delta (∆) normalized reported (Rn) value or ∆Rn. To obtain this value, the fluorescent signal from each qPCR reaction is normalized to the signal from the passive (unincorporated) reference dye by dividing the qPCR signal by the passive reference dye signal. The delta Rn value is then obtained by finding the difference between the normalized Rn value minus the baseline signal generated by the specific qPCR instrument. This analysis can be used to reliably calculate the magnitude of the specific signal produced by any given set of qPCR conditions above the threshold limit of detection (∆Rn = 0.01).
Statistical Analysis
DNA measurements involve continuous parametric data, therefore descriptive statistics regarding these data were given where appropriate (mean and the standard error of the mean (SEM)). Graphic analysis of data was visualized using box-and-whisker plots, which included using the Tukey method to plot the interquartile range (IQR) between the 25th and 75th percentiles, as previously described [11]. Any differences between the concentrations or purity between the samples or oral sites were measured using Student's t-tests, which are appropriate for the analysis of parametric data. Due to the confounding influence of multiple two-way t-tests, the analysis of variance (ANOVA) was used to confirm all statistical analyses.
A demographic analysis of the study sample was performed using simple descriptive statistics (raw number and percentage of the overall sample) for comparison with overall clinic averages using the categorical variables (sex, race/ethnicity). Differences between categorical variables can be most appropriately analyzed using Chi Square (χ 2 ) analysis, reporting the degrees of freedom (d.f.) and resulting p-values.
Results
A quantitative and qualitative analysis of DNA was successfully completed from all the cryopreserved patient samples, n = 47 ( Table 1). The oral sites with the overall highest DNA concentrations were derived from unstimulated saliva (average 4264.1 ng/µL), which were significantly different from all other sites (p = 0.0348). All three of the sites sampling biofilm (the dorsum tongue, upper buccal and mandibular lingual) had averages (2382.4 ng/µL, 20148.4 ng/µL and 2428.6 ng/µL, respectively) that were not significantly different from one another (p = 0.5376). The oral site with the overall lowest DNA concentrations were derived from the gingival crevicular fluid (GCF; average 1771.5 ng/µL), which was not significantly different from those derived from the biofilm (p = 0.4805). In addition, DNA purity was measured using A260:A280 ratios, which was fairly consistent among all the sampling sites with averages ranging between 1.71 and 1.82. Because some individual samples had A260:A280 ratios as low as 1.55, absorbance at A230 was also measured to provide the A260:A230 ratio-commonly accepted as an accurate measure of the residual chemical contamination from nucleic acid extraction procedures. These results demonstrated a fairly low residual contamination, with values ranging from 1.92 to 2.01 that closely approximates the expected values between 2.0 and 2.2 for "pure" nucleic acid samples acceptable for qPCR analysis and screening. To visualize these data, box-and-whisker plots were created to evaluate the minimum, maximum, first and third quartile ranges ( Figure 1). These data clearly demonstrate the similarity between the median and means of the non-salivary samples revealed by the data from Table 1. In addition, this demonstrated further similarity between the first quartile ranges of the tongue, upper buccal, mandibular lingual and GCF (713.4-1057.0 ng/µL) and third quartiles ranges (2214.6-3055.1 ng/µL), which were markedly different from those of unstimulated saliva (1588.2-6198.8 ng/µL).
Methods Protoc. 2020, 3, x FOR PEER REVIEW 6 of 11 Figure 1. Box-and-whisker plot analysis of DNA from oral sampling sites. The Tukey method used to plot the interquartile range (IQR) between the 25th and 75th percentile reveals similarities between the DNA concentrations derived from biofilm and GCF, which were significantly lower than those derived from unstimulated saliva (p = 0.0348).
To evaluate whether the results obtained from the cryopreserved samples were similar to the results obtained from the fresh samples taken during the same study [11], the average concentrations from the original samples (fresh) were plotted against those from the current study (frozen) to determine if the equivalent saliva sample volumes yielded similar amounts of DNA ( Figure 2). These data demonstrated that, although the overall concentration of DNA obtained from the original (fresh) saliva samples was slightly higher, this was not significantly different from the concentrations obtained from the pediatric samples (cryopreserved: 4264.1 ng/μL, original (fresh): 4809.3 ng/μL (p = 0.5389)). In addition, comparison of the DNA obtained from gingival crevicular fluid also exhibited similar results, which were also not significantly different (cryopreserved: 1762 ng/μL, original (fresh): 2639.5 ng/μL (p = 0.2253)). Box-and-whisker plot analysis of DNA from oral sampling sites. The Tukey method used to plot the interquartile range (IQR) between the 25th and 75th percentile reveals similarities between the DNA concentrations derived from biofilm and GCF, which were significantly lower than those derived from unstimulated saliva (p = 0.0348).
To evaluate whether the results obtained from the cryopreserved samples were similar to the results obtained from the fresh samples taken during the same study [11], the average concentrations from the original samples (fresh) were plotted against those from the current study (frozen) to determine if the equivalent saliva sample volumes yielded similar amounts of DNA ( Figure 2). These data demonstrated that, although the overall concentration of DNA obtained from the original (fresh) saliva samples was slightly higher, this was not significantly different from the concentrations obtained from the pediatric samples (cryopreserved: 4264.1 ng/µL, original (fresh): 4809.3 ng/µL (p = 0.5389)). In addition, comparison of the DNA obtained from gingival crevicular fluid also exhibited similar results, which were also not significantly different (cryopreserved: 1762 ng/µL, original (fresh): 2639.5 ng/µL (p = 0.2253)).
To evaluate whether the results obtained from the cryopreserved samples were similar to the results obtained from the fresh samples taken during the same study [11], the average concentrations from the original samples (fresh) were plotted against those from the current study (frozen) to determine if the equivalent saliva sample volumes yielded similar amounts of DNA ( Figure 2). These data demonstrated that, although the overall concentration of DNA obtained from the original (fresh) saliva samples was slightly higher, this was not significantly different from the concentrations obtained from the pediatric samples (cryopreserved: 4264.1 ng/μL, original (fresh): 4809.3 ng/μL (p = 0.5389)). In addition, comparison of the DNA obtained from gingival crevicular fluid also exhibited similar results, which were also not significantly different (cryopreserved: 1762 ng/μL, original (fresh): 2639.5 ng/μL (p = 0.2253)). The concentrations from the surface biofilm samples from the frozen and original (fresh) samples were graphed for comparison ( Figure 3). These data demonstrated that the overall DNA concentration isolated from the original (fresh) biofilm samples was higher than, and significantly different from, the concentrations obtained from the cryopreserved samples (p = 0.0308). More specifically, the average DNA concentrations from the tongue biofilm, upper buccal molar and lingual incisor from frozen samples (2382 ng/µL, 2048 ng/µL, 2428 ng/µL, respectively) were significantly different from those of the original (fresh) samples (3049 ng/µL, 2838.1 ng/µL, 2894.9 ng/µL, respectively (p = 0.0308)).
A real-time qPCR was performed to evaluate the presence of bacterial (16S rRNA) and human (GAPDH) DNA from the salivary and GCF isolates (Figure 4). The qPCR screening revealed that all the samples evaluated harbored bacterial and human DNA of sufficient quantity and quality for a molecular screening greater than the limit of detection (∆Rn = 0.01). No significant differences in the qPCR signal intensity for 16S or GAPDH were found between the cryopreserved and original samples (p = 0.7401).
Finally, to determine if there are any study-specific biases within the original sample collection, simple descriptive statistics of the demographic characteristics were assembled for analysis (Table 2). These data demonstrated a significantly lower proportion of females in the study sample versus the overall percentage of females from the pediatric clinic population, from which they were originally obtained (36.2% and 52.8%, respectively (p = 0.0012)). In addition, the proportion of samples derived from non-minority (white) patients was significantly lower than the overall percentage from the pediatric clinic population (4.25% and 24.7%, respectively (p = 0.00018)). specifically, the average DNA concentrations from the tongue biofilm, upper buccal molar and lingual incisor from frozen samples (2382 ng/μL, 2048 ng/μL, 2428 ng/μL, respectively) were significantly different from those of the original (fresh) samples (3049 ng/μL, 2838.1 ng/μL, 2894.9 ng/μL, respectively (p = 0.0308)). A real-time qPCR was performed to evaluate the presence of bacterial (16S rRNA) and human (GAPDH) DNA from the salivary and GCF isolates (Figure 4). The qPCR screening revealed that all the samples evaluated harbored bacterial and human DNA of sufficient quantity and quality for a molecular screening greater than the limit of detection (ΔRn = 0.01). No significant differences in the qPCR signal intensity for 16S or GAPDH were found between the cryopreserved and original samples (p = 0.7401). Finally, to determine if there are any study-specific biases within the original sample collection, simple descriptive statistics of the demographic characteristics were assembled for analysis (Table 2). These data demonstrated a significantly lower proportion of females in the study sample versus the overall percentage of females from the pediatric clinic population, from which they were originally obtained (36.2% and 52.8%, respectively (p = 0.0012)). In addition, the proportion of samples derived from non-minority (white) patients was significantly lower than the overall percentage from the pediatric clinic population (4.25% and 24.7%, respectively (p = 0.00018)).
Discussion
Although evaluations of DNA isolated from saliva or GCF and site-specific biofilm have been performed in adults [11], the primary goal of this study was to make a quantitative and qualitative comparison to determine the suitability of the DNA from pediatric oral samples for microbial PCR screening. These data provide the first quantitative and qualitative data that suggest sampling of the oral biofilm among pediatric patients using paper points may provide consistent and high quality results from a variety of patients and oral site locations-a novel finding that may expand the range of institutions and research groups that may contribute to the advancement of oral health research [20,21].
In addition, only a very limited number of studies have provided an in depth analysis and comparisons of pediatric and adult biofilm samples [22,23]. This study provides direct comparisons of DNA concentrations not only from unstimulated saliva and GCF, but also from the biofilm derived from several distinct oral sites. Although many efforts have been made to describe changes to the oral biofilm over time in either pediatric or adult patients, these studies may be both cost prohibitive and time consuming-which may delay the discovery and development of relevant findings to help reduce the burden of oral disease [24,25].
As the need for standardized approaches becomes more evident, the validation of standardized protocols and methods using commonly available dental office materials becomes a research priority [26]. This study provides preliminary evidence that contributes to the development of these protocols and methods and also provides comparisons of pediatric and adult samples from different oral biofilm sites. However, this study sample was limited in number due to financial and other site-specific limitations. Future research in this area should include larger comparison samples that may provide more robust evidence for quantitative and qualitative evaluation.
Conclusions
Many methods are currently available to provide the sampling and screening of saliva and specific sites within the oral cavity, but the validation and comparison of simple and low-cost methods that include paper point sampling and unstimulated saliva collection may suggest that these methods and protocols provide sufficient DNA quality and quantity for molecular screening and other comparison applications. In addition, although heterogeneity will be a constant and consistent feature between patient samples, standardized methods that provide similar and consistent DNA from various oral sites may provide the needed consistency for screening and molecular analysis.
Author Contributions: All authors have read and agree to the published version of the manuscript. K.K., J.M. and S.M. were responsible for the overall project design. J.E. and R.D. were responsible for data generation and analysis. All authors contributed to the writing or editing of this manuscript.
|
2020-07-15T13:05:57.911Z
|
2020-07-09T00:00:00.000
|
{
"year": 2020,
"sha1": "49fb4afef185345eb3cbb2b3a9b05848883235a1",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2409-9279/3/3/48/pdf",
"oa_status": "GOLD",
"pdf_src": "PubMedCentral",
"pdf_hash": "4a6edf5f0895f9a174d5912c0f4017aa5c80aa25",
"s2fieldsofstudy": [
"Medicine",
"Biology"
],
"extfieldsofstudy": [
"Chemistry",
"Medicine"
]
}
|
149453483
|
pes2o/s2orc
|
v3-fos-license
|
Application of the Non-Contact Video Gauge on the Mechanical Properties Test for Steel Cable at Elevated Temperature
As the limit of traditional contact measurement, it is difficult to precisely measure the steel cables twisted by a branch of wires especially at elevated temperature. In this paper the strain-stress relationships of S355 and S690 structural steel, 1860 MPa steel cable twisted by seven wires have been measured by the strain gauge, extensometer and non-contact video gauge at ambient temperature and elevated temperature, respectively. Comparison of the stress-strain curves gotten by different measuring technology, it indicates that the non-contact video gauge can provide a more efficient and reliable database than the strain gauge as well as extensometer, especially at an elevated temperature. It is worth noting that the non-contact video gauge can capture not only the full range of stress-strain curves of steel cables, but is also efficient for the specimens with a complex shape.
Introduction
In order to improve the reality of a structural response for pre-tensioned steel structures exposed to fire by using numerical simulation, it is very important to capture the full range of the stress-strain relationship at elevated temperature for steel cables, which are always with a complex shape. It is well known that the strain gauge can accurately capture the stress-strain relationship for steel material at ambient temperature, but cannot at an elevated temperature. In previous test, most of the stress-strain relationships of steel material have been tested by extensometers at an elevated temperature. The brittle ceramic arms of an extensometer had to be removed before the ultimate strength in order to prevent being broken due to the limit measuring range. Thus, it can only capture the partial range of the stress-strain curve before the ultimate strength at an elevated temperature [1]. Although the external mechanical device with linear variable displacement transducer can trace the full range of the stress-strain curve at an elevated temperature, the strain resulted from this kind of measurement device is the average strain, which is different from the engineering strain. Furthermore, it is difficult to grip a group of wires in a tensile test due to the complex surface of a steel cable twisted by a branch of wire. The displacement between the front end of an arm and the surface of the specimen always occurred, especially for the test object with a complex geometrical surface such as a steel cable [2].
Recently, a charge-coupled device camera (CCDC) system, which has been improved by the digital image correlation method (DICM), has been used to measure the strain in the test coupon subjected to tension. A computer program based on this correlation method was developed for the Appl calculation of the displacement components and the deformation gradients of an object surface due to deformation [3,4]. Several experiments were performed to demonstrate the viability and accuracy of the correlation method.
To form a gauge length is the primary step to collect test data by CCDC. Hu et al. [5] presented a novel laser-engraving technology to create speckles on the surface of a metallic specimen, which will form gauge length and be kept up till the melting temperature of metallic materials. This kind of speckles was also applied to investigate the elastic modulus of tungsten material up to 1000 • C. However, it is not economical and convenient to mark gauge length for steel cable specimens compared with a high temperature paint sparkler.
In 1996, Lyons et al. [4] proposed that the thermal expansion tests of standard inconel specimens up to 650 • C was conducted by a combined dual element optical fiber lamp and digital camera. The normalized cross interaction function was developed to determine the value of thermal expansion of specimens. In 2010, Pan et al. [6] reported that the blue LED light and band-pass filter imaging system can overcome the decorrelation effect due to high bright light at elevated temperature. This kind of technology was used by CCDC to investigate the thermal expansion of Cr-Ni austenitic stainless steel up to 1200 • C. In 2012, Gales et al. [7] carried out a series of tensile tests on pre-stressing steel at an elevated temperature by a steady method and transient method, respectively. It indicates that the novel digital image correlation method can measure the strain of tendon at an elevated temperature. In 2015, Yang et al. [8] investigated the mechanical properties of Gh738 material at an elevated temperature using the digital image correlation method.
The CCDC is also used to investigate the mechanical response of welded joints, concrete members and structural members at ambient temperature and elevated temperature respectively [9][10][11][12][13][14].
In order to calibrate the CCDC system, which can be used to measure the steel cables with a complex shape till rupture at an elevated temperature, a different measurement method was compared with each other in the present study as there is no previous study that can be referenced directly for this proposal. Finally, the present study conducted the test to capture the full range of stress-strain curves at an elevated temperature, which can improve the accuracy of the fire resistance analysis for pre-tensioned steel structures in the civil engineering area.
Basic Principle of Digital Image Correlation Method
From then, now, the previous studies have contributed to update the algorithm of the digital image correlation method (DICM), which is the core technology of the CCDC system, and improved the accuracy of the measurement by CCDC system. Hereby, the basic principle of the CCDC system was introduced as below.
Digital image correlation method is an optical measurement based on digital image processing and numerical calculation. According to the basic principle of DIMC, speckle pattern is taken as the information resource of the deformation for an object [15,16]. Figure 1a displays the initial speckle pattern and Figure 1b shows the speckle pattern after deformation. Comparison of Figure 1a,b, the geometric coordination of a moving point, P, can be traced by using the digital image correlation method.
Appl. Sci. 2019, 9, x 2 of 14 calculation of the displacement components and the deformation gradients of an object surface due to deformation [3,4]. Several experiments were performed to demonstrate the viability and accuracy of the correlation method.
To form a gauge length is the primary step to collect test data by CCDC. Hu et al. [5] presented a novel laser-engraving technology to create speckles on the surface of a metallic specimen, which will form gauge length and be kept up till the melting temperature of metallic materials. This kind of speckles was also applied to investigate the elastic modulus of tungsten material up to 1000 °C . However, it is not economical and convenient to mark gauge length for steel cable specimens compared with a high temperature paint sparkler.
In 1996, Lyons et al. [4] proposed that the thermal expansion tests of standard inconel specimens up to 650 °C was conducted by a combined dual element optical fiber lamp and digital camera. The normalized cross interaction function was developed to determine the value of thermal expansion of specimens. In 2010, Pan et al. [6] reported that the blue LED light and band-pass filter imaging system can overcome the decorrelation effect due to high bright light at elevated temperature. This kind of technology was used by CCDC to investigate the thermal expansion of Cr-Ni austenitic stainless steel up to 1200 °C . In 2012, Gales et al. [7] carried out a series of tensile tests on pre-stressing steel at an elevated temperature by a steady method and transient method, respectively. It indicates that the novel digital image correlation method can measure the strain of tendon at an elevated temperature. In 2015, Yang et al. [8] investigated the mechanical properties of Gh738 material at an elevated temperature using the digital image correlation method.
The CCDC is also used to investigate the mechanical response of welded joints, concrete members and structural members at ambient temperature and elevated temperature respectively [9][10][11][12][13][14].
In order to calibrate the CCDC system, which can be used to measure the steel cables with a complex shape till rupture at an elevated temperature, a different measurement method was compared with each other in the present study as there is no previous study that can be referenced directly for this proposal. Finally, the present study conducted the test to capture the full range of stress-strain curves at an elevated temperature, which can improve the accuracy of the fire resistance analysis for pre-tensioned steel structures in the civil engineering area.
Basic principle of Digital Image Correlation Method
From then, now, the previous studies have contributed to update the algorithm of the digital image correlation method (DICM), which is the core technology of the CCDC system, and improved the accuracy of the measurement by CCDC system. Hereby, the basic principle of the CCDC system was introduced as below. Digital image correlation method is an optical measurement based on digital image processing and numerical calculation. According to the basic principle of DIMC, speckle pattern is taken as the Based on the digital image correlation method, shown in Figure 2, a charge-coupled device camera has been used by the CCDC system to capture a series of speckle patterns and transferred to a data collector. In the meantime, the computer software, which developed on the base of the digital image correlation method, analyzed the database efficiently, and output the transient strain of a specimen. The measurement range of the non-contact video gauge is up to a strain of 100%, which can capture the full range of strain till specimens fails. The high quality video also can be recorded to review the test process once more. The practice of the non-contact video system is shown as Figure 3.
Appl. Sci. 2019, 9, x 3 of 14 information resource of the deformation for an object [15,16]. Figure 1a displays the initial speckle pattern and Figure 1b shows the speckle pattern after deformation. Comparison of Figures 1a,b, the geometric coordination of a moving point, P, can be traced by using the digital image correlation method. Based on the digital image correlation method, shown in Figure 2, a charge-coupled device camera has been used by the CCDC system to capture a series of speckle patterns and transferred to a data collector. In the meantime, the computer software, which developed on the base of the digital image correlation method, analyzed the database efficiently, and output the transient strain of a specimen. The measurement range of the non-contact video gauge is up to a strain of 100%, which can capture the full range of strain till specimens fails. The high quality video also can be recorded to review the test process once more. The practice of the non-contact video system is shown as Figure 3. In present study, the CCDC pictured with 2452 pixels × 2056 pixels and a constant aperture of 1:1.8. The object distance of the special lens was 174 mm for material tests, and the magnification ratio of the lens was 0.338. The maximum frequency of data collection was 15 Hz. The resolution ratio of displacement was 0.012 μm and that of the strain was 5 με.
Calibration Tensile Test at Ambient Temperature
Charge-coupled device camera information resource of the deformation for an object [15,16]. Figure 1a displays the initial speckle pattern and Figure 1b shows the speckle pattern after deformation. Comparison of Figures 1a,b, the geometric coordination of a moving point, P, can be traced by using the digital image correlation method. Based on the digital image correlation method, shown in Figure 2, a charge-coupled device camera has been used by the CCDC system to capture a series of speckle patterns and transferred to a data collector. In the meantime, the computer software, which developed on the base of the digital image correlation method, analyzed the database efficiently, and output the transient strain of a specimen. The measurement range of the non-contact video gauge is up to a strain of 100%, which can capture the full range of strain till specimens fails. The high quality video also can be recorded to review the test process once more. The practice of the non-contact video system is shown as Figure 3. In present study, the CCDC pictured with 2452 pixels × 2056 pixels and a constant aperture of 1:1.8. The object distance of the special lens was 174 mm for material tests, and the magnification ratio of the lens was 0.338. The maximum frequency of data collection was 15 Hz. The resolution ratio of displacement was 0.012 μm and that of the strain was 5 με.
Calibration Tensile Test at Ambient Temperature
Charge-coupled device camera In present study, the CCDC pictured with 2452 pixels × 2056 pixels and a constant aperture of 1:1.8. The object distance of the special lens was 174 mm for material tests, and the magnification ratio of the lens was 0.338. The maximum frequency of data collection was 15 Hz. The resolution ratio of displacement was 0.012 µm and that of the strain was 5 µε.
Uniaxial Tensile Tests on S355 Structural Steel
Both of the strain gauge and CCDC were used to investigate the stress-strain relationship of S355 structural steel. The dimension of the specimen for a standard uniaxial tensile test is shown as
Uniaxial Tensile Tests on S355 Structural Steel
Both of the strain gauge and CCDC were used to investigate the stress-strain relationship of S355 structural steel. The dimension of the specimen for a standard uniaxial tensile test is shown as Figure 4. The standard uniaxial tensile tests were conducted on a servo-hydraulic testing machine, with a maximum stroke displacement of 75 mm and a capacity of 500 kN, shown in Figure 5. There are two groups of specimen measuring by the strain gauge and CCDC system, respectively. For CCDC system measuring, the first step is to sprinkle black paint on the surface of each specimen to reduce the metallic brightness, and then, markers with white spots painting on a black color background were prepared within the middle part of a specimen surface of about 30 mm length. On the other side of the same specimen, a strain gauge with the measuring range of 100,000 με was pasted. The specimen was lightly tensioned to about 1 kN to eliminate any mechanical relaxation. Then, the tensile specimen was tested until failure at a straining rate of 0.0005 /s, in keeping with the strain rates set out in the EN 6892-1 [17]. As shown in Figure 6, the S355 steel specimen experienced a ductile model with necking phase. The broken cross section in the middle of length formed a cone-shape with cross section area gradually reduced towards the broken surface.
Uniaxial Tensile Tests on S355 Structural Steel
Both of the strain gauge and CCDC were used to investigate the stress-strain relationship of S355 structural steel. The dimension of the specimen for a standard uniaxial tensile test is shown as Figure 4. The standard uniaxial tensile tests were conducted on a servo-hydraulic testing machine, with a maximum stroke displacement of 75 mm and a capacity of 500 kN, shown in Figure 5. There are two groups of specimen measuring by the strain gauge and CCDC system, respectively. For CCDC system measuring, the first step is to sprinkle black paint on the surface of each specimen to reduce the metallic brightness, and then, markers with white spots painting on a black color background were prepared within the middle part of a specimen surface of about 30 mm length. On the other side of the same specimen, a strain gauge with the measuring range of 100,000 με was pasted. The specimen was lightly tensioned to about 1 kN to eliminate any mechanical relaxation. Then, the tensile specimen was tested until failure at a straining rate of 0.0005 /s, in keeping with the strain rates set out in the EN 6892-1 [17]. As shown in Figure 6, the S355 steel specimen experienced a ductile model with necking phase. The broken cross section in the middle of length formed a cone-shape with cross section area gradually reduced towards the broken surface. There are two groups of specimen measuring by the strain gauge and CCDC system, respectively. For CCDC system measuring, the first step is to sprinkle black paint on the surface of each specimen to reduce the metallic brightness, and then, markers with white spots painting on a black color background were prepared within the middle part of a specimen surface of about 30 mm length. On the other side of the same specimen, a strain gauge with the measuring range of 100,000 µε was pasted. The specimen was lightly tensioned to about 1 kN to eliminate any mechanical relaxation. Then, the tensile specimen was tested until failure at a straining rate of 0.0005 /s, in keeping with the strain rates set out in the EN 6892-1 [17]. As shown in Figure 6, the S355 steel specimen experienced a ductile model with necking phase. The broken cross section in the middle of length formed a cone-shape with cross section area gradually reduced towards the broken surface. The stress-strain curves were obtained by a tensile test with strain gauge and CCDC system, respectively, as shown in Figure 7a. The elastic modulus can be determined by the tangent modulus of the initial linear phase of the stress-strain relationship. The effective yield strength is the intersection point of the stress-strain curve and the proportional line offset by 0.15% strain. The ultimate strength is the maximum stress in the stress-strain curve, and the fracture strain is dependent on the specimen broken. The mechanical properties of S355 steel are listed in Table 1. The stress-strain curves were obtained by a tensile test with strain gauge and CCDC system, respectively, as shown in Figure 7a. The elastic modulus can be determined by the tangent modulus of the initial linear phase of the stress-strain relationship. The effective yield strength is the intersection point of the stress-strain curve and the proportional line offset by 0.15% strain. The ultimate strength is the maximum stress in the stress-strain curve, and the fracture strain is dependent on the specimen broken. The mechanical properties of S355 steel are listed in Table 1. The stress-strain curves were obtained by a tensile test with strain gauge and CCDC system, respectively, as shown in Figure 7a. The elastic modulus can be determined by the tangent modulus of the initial linear phase of the stress-strain relationship. The effective yield strength is the intersection point of the stress-strain curve and the proportional line offset by 0.15% strain. The ultimate strength is the maximum stress in the stress-strain curve, and the fracture strain is dependent on the specimen broken. The mechanical properties of S355 steel are listed in Table 1. Test results indicate that the ultimate strength obtained by different measurements was the same. As shown in Figure 7b, the elastic modulus and yield strength, which obtained by strain gauge were smaller than those obtained by CCDC, and the deviation was 1.34% and 1.55%, respectively, as listed in Table 1. The test process displayed that the tensile specimen also experienced a large strain from the ultimate strength till fracture. As the limited of measuring range, the strain gauge can only capture a partial range of the stress-strain curve before ultimate strength. Comparing the strain gauge, the CCDC system can obtain a full range of stress-strain curve until broken as shown in Figure 7a.
Therefore, it is more efficient to use the CCDC system than strain gauge to measure the specimens with large deformation through the test process.
Uniaxial Tensile Tests on Steel Cable
The dimension of a steel cable specimen with ultimate strength of 1860 MPa is shown in Figure 8. The nominal diameter is 15.2 mm and cross section area is 139 mm 2 . The process of marker painted is the same as mentioned in Section 3.1. Shown in Figure 9, the CCD camera was settled right ahead of the specimen in the distance of 1.5 m to form a measure plan and supplement brightness using a white lamp. Shown in Figure 10, an extensometer with the gauge length of 50 mm was settled in the mid-length of the same specimen. A couple of strain gauge was pasted on the surface of the mid-length. Thus, the stress-strain relationship of the steel cable was measured by three methods, i.e., strain gauge, extensometer and CCDC system, at ambient temperature. The standard tensile test was carried out by a servo-hydraulic test machine with a maximum stroke displacement of 75 mm and a capacity of 500 kN. There were four specimens repeatedly tested in a tensile test to get the average stress-strain curve as shown in Figure 10. carried out by a servo-hydraulic test machine with a maximum stroke displacement of 75 mm and a capacity of 500 kN. There were four specimens repeatedly tested in a tensile test to get the average stress-strain curve as shown in Figure 10. carried out by a servo-hydraulic test machine with a maximum stroke displacement of 75 mm and a capacity of 500 kN. There were four specimens repeatedly tested in a tensile test to get the average stress-strain curve as shown in Figure 10. Shown in Figure 11, the stress-strain curves were obtained by different measuring methods. The effective yield strength, elastic modulus, ultimate strength and rupture strength were derived from each stress-strain curve with the same method as mentioned in Section 3.1. The values of the mechanical properties of steel cables obtained by different measuring methods are listed in Tables 2-4. There is a blank in Table 2 as a strain gauge is difficult to paste on the surface of a wire with 5 mm diameter, one of strain gauges split away off the surface of a wire and induced a blank in Table 2. As the arm of an extensometer cannot grab the steel cable tightly enough, there was the slipping displacement between the surface of the arm and steel cable. Thus, there are a few of blanks in Table 3. Comparison of the values of elastic modulus and effective yield strength, which were obtained by Shown in Figure 11, the stress-strain curves were obtained by different measuring methods. The effective yield strength, elastic modulus, ultimate strength and rupture strength were derived from each stress-strain curve with the same method as mentioned in Section 3.1. The values of the mechanical properties of steel cables obtained by different measuring methods are listed in Tables 2-4. There is a blank in Table 2 as a strain gauge is difficult to paste on the surface of a wire with 5 mm diameter, one of strain gauges split away off the surface of a wire and induced a blank in Table 2. As the arm of an extensometer cannot grab the steel cable tightly enough, there was the slipping displacement between the surface of the arm and steel cable. Thus, there are a few of blanks in Table 3. Comparison of the values of elastic modulus and effective yield strength, which were obtained by an extensometer and CCDC system, showed that they matched well with only 1.13% and 0.6% deviation, respectively. However, the values of elastic modulus and effective yield strength measured by a strain gauge were larger than those obtained by the extensometer and CCDC system. For an elastic modulus obtained by strain, the deviation was 2.83% and 3.99% compared to that by an extensometer and CCDC system. For the effective yield strength, the deviation was 16.6% and 17.1%, respectively. an extensometer and CCDC system, showed that they matched well with only 1.13% and 0.6% deviation, respectively. However, the values of elastic modulus and effective yield strength measured by a strain gauge were larger than those obtained by the extensometer and CCDC system. For an elastic modulus obtained by strain, the deviation was 2.83% and 3.99% compared to that by an extensometer and CCDC system. For the effective yield strength, the deviation was 16.6% and 17.1%, respectively. Figure 11. Comparison of stress-strain curves of pre-stressed steel cables measuring by different measurement methods.
As the strain gauge was pasted along twist wires, the strain measured by a strain gauge is off the longitudinal axis, which is smaller than the axis strain in the longitudinal direction. Thus, the smaller strain would induce the larger elastic modulus and yield strength. It also indicated that strain gauge cannot be suitable to measure the stress-strain curve of steel cables, which were twisted by a group of wires. As the strain gauge was pasted along twist wires, the strain measured by a strain gauge is off the longitudinal axis, which is smaller than the axis strain in the longitudinal direction. Thus, the smaller strain would induce the larger elastic modulus and yield strength. It also indicated that strain gauge cannot be suitable to measure the stress-strain curve of steel cables, which were twisted by a group of wires.
As listed in Table 3, the extensometer cannot captured the ultimate strength and rupture strain as its arm must be removed before the ultimate strength due to limited travel distance. Thus, an extensometer cannot capture the full range of stress-strain curves in a material properties test, especially for steel cables. The rupture strain of steel cables measured by the CCDC system was smaller than that by a strain gauge with 2.55% deviation. It also indicated that the strain gauge could only obtain stress-strain curve of one wire in a steel cable. Thus, the CCDC system was more efficient and accurate to measure the mechanical properties of steel cables than other methods. Table 3. Mechanical properties of pre-tension steel cables obtained by extensometer. Table 4. Mechanical properties of pre-tension steel cables obtained by CCDC system.
Uniaxial Tensile Tests on S690 Structural Steel
The dimension of a S690 structural steel specimen for a standard uniaxial tensile test at an elevated temperature is the same as shown in Figure 4. The standard uniaxial tensile tests at an elevated temperature were also conducted on a servo-hydraulic testing machine, with a maximum stroke displacement of 75 mm and a capacity of 500 kN as shown in Figure 12. A specimen of S690 high-strength steel was installed and heated by a tubular high-temperature furnace as shown in Figure 13. The heat part was a split-tube furnace with a three-zone configuration and a side view window. A type K thermocouple was mounted at the centre of each zone to measure the heating temperature. The specimen was heated to 400 • C at the heat rate of 10 • C/min and subject to tension [18]. Both the extensometer and CCDC system were used to measure the strain in the specimen under the applied tension force.
The dimension of a S690 structural steel specimen for a standard uniaxial tensile test at an elevated temperature is the same as shown in Figure 4. The standard uniaxial tensile tests at an elevated temperature were also conducted on a servo-hydraulic testing machine, with a maximum stroke displacement of 75 mm and a capacity of 500 kN as shown in Figure 12. A specimen of S690 high-strength steel was installed and heated by a tubular high-temperature furnace as shown in Figure 13. The heat part was a split-tube furnace with a three-zone configuration and a side view The dimension of a S690 structural steel specimen for a standard uniaxial tensile test at an elevated temperature is the same as shown in Figure 4. The standard uniaxial tensile tests at an elevated temperature were also conducted on a servo-hydraulic testing machine, with a maximum stroke displacement of 75 mm and a capacity of 500 kN as shown in Figure 12. A specimen of S690 high-strength steel was installed and heated by a tubular high-temperature furnace as shown in Figure 13. The heat part was a split-tube furnace with a three-zone configuration and a side view As shown in Figure 13, the arms of the extensometer are ceramic rods with quartz chisel ends, which passed through the side entry port and contacted the test coupons. It is specified for a gauge length of 25 mm and could travel 2.5 mm. Thus, the maximum measurable strain was 10%. During testing, the displacements of the test coupons were directly detected and transmitted by the rods to the strain gauges placed within the extensometer body. A cooling fan was mounted close to the extensometer to maintain the temperature of the extensometer lower than 150 • C. There were four specimens repeatedly tested in a tensile test to measure the average high temperature stress-strain curve. Two specimens measured by an extensometer, and the other by the CCDC system.
The same as for the marker paint process in Section 3.1. For the high temperature test, the high temperature paint was used to make markers on the surface of test specimens.
Both the tests measured by the extensometer and CCDC system were conducted on the S690 steel. The target temperature was up to 400 • C at the heating rate of 10 • C/min and held for 30 min to obtain uniform temperature distributions in both the furnace and the test specimens. The strain rate was 0.003 /min [18]. For the CCDC system, the white light lamp was used to supply the brightness of the specimen, which was mounted in the furnace. The function of the exposure compensation was used to help the CCD camera eliminate the extra exposure due to a high temperature.
Both the tests measured by the extensometer and CCDC system were conducted on the S690 steel. The target temperature was up to 400 °C at the heating rate of 10 °C /min and held for 30 min to obtain uniform temperature distributions in both the furnace and the test specimens. The strain rate was 0.003 /min [18]. For the CCDC system, the white light lamp was used to supply the brightness of the specimen, which was mounted in the furnace. The function of the exposure compensation was used to help the CCD camera eliminate the extra exposure due to a high temperature. The tests observed that the high strength steel S690 experienced a necking at 400 °C . As shown in Figure 14, the broken cross section area gradually reduced towards the broken surface.
The stress-strain curves of S690 at 400 °C was shown in Figure 15, and the mechanical properties was listed in Table 5. The tests observed that the high strength steel S690 experienced a necking at 400 • C. As shown in Figure 14, the broken cross section area gradually reduced towards the broken surface.
The stress-strain curves of S690 at 400 • C was shown in Figure 15, and the mechanical properties was listed in Table 5. The mechanical properties listed in Table 5, i.e., elastic modulus, yield strength and ultimate strength of S690 steel at an elevated temperature obtained by the CCDC system was slightly larger The mechanical properties listed in Table 5, i.e., elastic modulus, yield strength and ultimate strength of S690 steel at an elevated temperature obtained by the CCDC system was slightly larger than those by the extensometer. The deviation between them was 3.3%, 1.8% and 0.49%, respectively.
Since the ceramic arm of the extensometer has to be removed away from the test specimen before the rupture of the test specimen due to its limited travel distance as mentioned above; the extensometer reading can only cover the partial range of the stress-strain curve at an elevated temperature. The CCDC system can obtain the full range of the curve. A good fit between the extensometer and CCDC readings is generally observed as shown in Figure 15 and listed in Table 5. It is worth noting that the S690 steel experienced a long strain history after the ultimate strength at an elevated temperature. This ductility behavior of the S690 high strength steel at an elevated temperature is important to evaluate the mechanical response of structural members exposed to fire.
The test results showed that both the CCDC system and extensometer could capture the stress-strain curves accurately. However, the extensometer could only capture the partial range of the stress-strain curve due to its limited travel distance. The CCDC system was more efficient than the extensometer to obtain the full range of stress-strain curves, which is a benefit for analysis on the mechanical response of structural members exposed to fire.
Uniaxial Tensile Tests on 1860 MPa Pre-Tensioned Steel Cable
As the calibration tests have conducted as above, the CCDC system was used to measure the strain in the steel cables at different temperatures. The uniaxial tensile test process is the same as those illustrated in the Section 3.1. Three specimens were repeatedly tested at each temperature and the average of stress-strain relationships was obtained by the CCDC system as shown in Figure 16. Comparison of test curves and those resulted by EN1992-1-2 (2005) [19] indicated that the strain after the ultimate strength was ignored by EN1992-1-2 at an elevated temperature. The full range of stress-strain curve showed that steel cable specimens experienced a large strain after the ultimate strength, which would induce the large deformation of a pre-tensioned steel cable at an elevated temperature. Thus, the used of CCDC system could measure the stress-strain curves accurately and efficiently.
Uniaxial Tensile Tests on 1860 MPa Pre-Tensioned Steel Cable
As the calibration tests have conducted as above, the CCDC system was used to measure the strain in the steel cables at different temperatures. The uniaxial tensile test process is the same as those illustrated in the Section 3.1. Three specimens were repeatedly tested at each temperature and the average of stress-strain relationships was obtained by the CCDC system as shown in Figure 16. Comparison of test curves and those resulted by EN1992-1-2 (2005) [19] indicated that the strain after the ultimate strength was ignored by EN1992-1-2 at an elevated temperature. The full range of stressstrain curve showed that steel cable specimens experienced a large strain after the ultimate strength, which would induce the large deformation of a pre-tensioned steel cable at an elevated temperature. Thus, the used of CCDC system could measure the stress-strain curves accurately and efficiently. Figure 16. Stress-strain relationships of high tensile strength steel cables at elevated temperature.
Cautions
As the CCDC system is an optical measurement technology based on the digital image correlation method, the speckles distribution, random vibration of servo-hydraulic testing machine can induce the deviation of test data in a limit range. The use of a high-temperature filter, exposure compensation system and rigid tripod can decrease the deviation. Meanwhile, the data collection
Cautions
As the CCDC system is an optical measurement technology based on the digital image correlation method, the speckles distribution, random vibration of servo-hydraulic testing machine can induce the deviation of test data in a limit range. The use of a high-temperature filter, exposure compensation system and rigid tripod can decrease the deviation. Meanwhile, the data collection frequent should be high enough to get more numbers of test data in each group. The more high data collection frequent reached, the more accurately the average value can be gained to improve the smoothness of data curves.
Conclusions
This paper investigated a series of calibrate standard uniaxial tension test at ambient temperature and an elevated temperature, respectively. The following conclusions can be drawn: • At ambient temperature, the data measured with the strain gauge and CCDC system fit well before the ultimate strength in a uniaxial tensile test of structural steel. The strain gauge could not capture the full range of stress-strain curves, but the CCDC system could.
•
Compared to the strain gauge and extensometer method, the CCDC system not only could measure the test data accurately as other measurement methods, but also was able to capture the full range of the stress-strain relationship curves for steel cables with a complex shape.
•
The use of high temperature paint could make speckle to form on the gauge length, which could be captured by a CCD camera. The high temperature filter also could improve the efficient workability of the CCDC system for mechanical properties tested at an elevated temperature.
|
2019-04-27T04:16:07.500Z
|
2019-04-23T00:00:00.000
|
{
"year": 2019,
"sha1": "596423055d49b758857dd3165b781c09858a50cf",
"oa_license": "CCBY",
"oa_url": "https://www.mdpi.com/2076-3417/9/8/1670/pdf",
"oa_status": "GOLD",
"pdf_src": "Unpaywall",
"pdf_hash": "596423055d49b758857dd3165b781c09858a50cf",
"s2fieldsofstudy": [
"Materials Science"
],
"extfieldsofstudy": [
"Engineering"
]
}
|
225141622
|
pes2o/s2orc
|
v3-fos-license
|
Macrocycles-assisted polymeric self-assemblies fabricated by host–guest complexation and their applications
Macrocycles exhibit unique features with prospective applications due to their inherent structural features combined with rigidity and functionality. Developing macrocycles-assisted polymeric self-assemblies (MPs) is one of the promising ways to convert macrocyclic polymers into functional materials with responsiveness, endowing them with realistic features such as self-healing, good processability, and recyclability. The present review focuses on the rational design of MPs, over the past decade, with realistic potential applications. The constructions of MPs mainly focuses on the following facets: (i) those assembled from macrocyclic side-chain polymeric hosts interacting with small molecular/polymeric organic and inorganic guests, and (ii) those assembled from macrocyclic monomers interacting with side-chain polymeric guests. These resultant MPs include the dynamic cross-linking of different copolymeric hosts and/or guests, which further leads to the formation of supramolecular self-assembled nanomaterials with unique functionalities. We also discuss the fabrication of stimuli-responsive MPs that are influenced by various external stimuli, and the resultant supramolecular host–guest complexation-mediated reversible self-assemblies are also highlighted. Furthermore, a wide range of applications of MPs, including self-healing materials, adhesives, adsorbents, drug delivery, and smart windows, and their modes of assemblies are elaborately presented. As far as we know, this is the first review particularly focusing on the side-chain polymeric host–guest interactions and their self-assemblies for diverse applications, whereas linear polymers and post-polymerizations are not included. This review summarizes the fundamental design concepts in the fabrication of polymeric backbone-assisted MPs, which provides new directions for future research in the field of supramolecular chemistry and materials chemistry.
Introduction
The progression of new macrocycles-assisted polymeric selfassemblies (MPs) has been playing a key role in our lives and science.Motivated by the exciting functionalities offered by MPs, researchers have strived to develop novel polymeric selfassembled nanomaterials with desired structural features and properties. 1,2J.-M.Lehn first designated the concepts of supramolecular polymers-the productive combination between supramolecular chemistry and polymers. 3,4][7][8][9][10][11] This universal methodology permits the fabrication of supramolecular selfassembled nanomaterials.The inherent dynamic properties of MPs are highly advantageous; they offer easy structure adaptation by fine-tuning the assemblies of the constituent polymeric units.MPs are comprised of elegant functional moieties that are involved in cross-linking via host-guest interactions, charge transfer interactions, hydrophobic effects, van der Waals forces, electrostatic interactions, hydrogen bonds, and p-p stacking.
Specifically, MPs are constructed from the noncovalent crosslinking of organic/inorganic guests with the existing polymeric hosts, thus endowing them with structural flexibility, functional diversity, and ease of modification. 12The resultant MPs built from various polymeric backbones with well-designed structural configurations retain amphiphilic subunits that can further selfassemble into nanoaggregates in either water or suitable organic media. 13Nanoaggregates so formed are heavily influenced by coordinating or resisting forces arising at the interface between MPs and the surrounding environment. 14However, MPs containing functional and tunable polymeric subunits enhance the covalent/noncovalent amphiphilic interactions, which in turn strengthen the stimuli-responsiveness of the systems. 15][18] The macrocycle-assisted host-guest strategy is unique, owing to its self-complementary interactions and directional properties.The host-guest interactions have played a crucial role not only in supramolecular chemistry but also in the field of polymer and materials chemistry.0][21][22] The geometric dimensions of their core cavities non-covalently interact with specific guests, resulting in the formation of supramolecular polymeric materials.][25][26][27][28][29] Highly complicated supramolecular structures, such as supramolecular amphiphiles, supramolecular polymers, and molecular machines, have been successfully developed [30][31][32][33] based on hostguest complexation using the aforementioned macrocycles.Macrocycles can be broadly classified into two categories based on their solubility.Firstly, CD-and CB-based materials, such as nanoaggregates, supramolecular polymers, and hydrogels are mostly explored in aqueous media. 34,357][38] Since the neutral states of the latter macrocycles are insoluble in water, organic solvents have been used to prepare supramolecular polymers and gels from their native states.Further, upon modification of these receptors with more suitable polar substituents, their solubility in water can be enhanced, and the resulting amphiphilic polymers involved in the self-assembly process can produce nanoaggregates in aqueous solution.
0][41][42][43] The binding affinity of the host-guest complexations can be fine-tuned by various stimuli-responsive factors, such as light, enzyme, redox, pH, temperature, reactive oxygen species (ROS), and ions.Thus, based on the variety of stimuli-responses and the mode of reversibility of the host-guest interactions, MPs can be specifically designed and constructed towards intended applications.In the present review, the cutting-edge research on the construction of various MPs by the supramolecular approach is highlighted.The noncovalent interactions between side-chain polymeric hosts and/or guests, with various applications are discussed in detail.However, the topics related to the discussion of linear polymers and post-polymerizations, and their roles in fabricating MPs are excluded.The construction of MPs with their cores built upon five important macrocyclic hosts, namely CEs, CDs, CAs, CBs, and PAs, is the main focus.Discussions on MPs-controlled topological structures achieved by various facile chemical routes and the functioning of MPs in bulk materials are presented. 44,45The main theme of this review includes the construction of MPs by host-guest complexation and their applications in the fields of drug delivery, construction of hydrogels, development of self-healing materials and adhesive materials, as well as some other successful applications, including ionic conductivity, molecular recognition, adsorbent/separation, and smart windows.This review also covers the current challenges that should be considered for the efficient construction of MPs.Such discussions enliven the understanding of the reader and strengthen some of the fundamental concepts governing the construction of MPs, which would provide some clues and new directions for researchers in the fields of supramolecular and materials chemistry.
Crown ether-based MPs
Crown ethers (CEs) are comprised of a ring with various ether groups and were the first generation of synthetic macrocycles, known to be the birth of supramolecular chemistry. 46These macrocycles were developed by the cyclization of polyether units, which consist of more than two 'O' atoms with two or more 'C' atoms around each 'O' atom.These multi-dentate macrocycles exhibit a strong interaction with various guests, while the macrocycles containing 'O' atoms facilitate the resulting polarity and H-bonding. 47,48Most of the CEs, such as 12-crown-4 and 15-crown-5 are water-insoluble, while some of the water-soluble CEs, such as 18-crown-6, and 24-crown-8 display weak interactions with guests.Such weak interactions may be attributed to the presence of water molecules that limit the ion-dipole interaction between the CEs and guests.It may further be noted that water-soluble CEs alter their critical entropy level from the flexible to less pre-arranged polyether units.CEs-based flexible chains are well-known ionophores that can easily bind with several metal ions while exhibiting less toxicity to anti-parasitic, antibacterial, and anti-cancer treatments. 49Moreover, supramolecular interactions between CEs and different guests (secondary ammonium salts and paraquat derivatives) have been utilized for the fabrication of supramolecular nanostructures.During this decade, based on CEs derivatives, various fascinating topological structures, which include supramolecular polymers, amphiphiles, rotaxanes, polypseudorotaxanes, and catenanes, have been developed.However, investigations on CEs-based polymeric self-assemblies are few.Therefore, the construction of CEsfunctionalized supramolecular self-assemblies driven by polymeric CEs with electron-withdrawing guests is essential and remains a serious challenge to the modern scientific community.
In 2010, Chen and Wang et al. 50developed supramolecular dendritic polymers from the polymeric crown ethers (pCE) and three different protonated ammonium guests (G1-G3).The reversible self-assemblies of dendritic polymers were achieved by the addition of trifluoroacetic acid (TFA) or triethylamine (TEA) (Fig. 2).The percentage of guest (G1) complexation with pCE was estimated to be around 90% and declined to 85% after three cycles; these results are quiet similar to G2 and G3.On further comparison with coil polymers, it was observed that dendronized polymeric chains more facilely access guest motifs as compared to coil polymers.Generally, amine-functionalized fluorophores acting as guests usually cause fluorescence quenching of the system due to the intramolecular charge transfer (ICT) process.Therefore, the amino group was protected in the respective guests and recovered the fluorescence.The protonation/deprotonation of the guests on the supramolecular complexes was achieved by TFA/TEA titration experiments.This approach has been adopted by many research groups involved in the preparation of supramolecular polymeric gels by CEs-functionalized monomers/copolymers crosslinking with monomeric/dimeric secondary ammonium salts.Hence, the prepared CEs functionalized polymeric gels were investigated for their self-healing properties and stimuli-responsiveness under different stimuli such as cation, anion, and pH-changes. 16Han et al. 51 constructed thermo-and ion-responsive supramolecular hydrogels by incorporating CEs into the poly(N-isopropylacrylamide) (PNIPAM) polymer.The swelling ratio (SR) of the copolymeric hydrogel was decreased in comparison to the pristine PNIPAM.When the hydrogel was further exposed by K + ion, the SR of the hydrogel increased considerably in comparison to that in aqueous solution, owing to the ion recognition ability of CEs.
Cyclodextrins-based MPs
Cyclodextrins (a-, b-, and g-CD) are a family of cyclic oligosaccharides that consist of D-glucose linked by a-1,4-glucosidic units, comprised of 6 to 8 D-glucopyranose residues.The toroidal shape of CDs has primary hydroxyl groups in the lower end with a reduced cavity size diameter as compared to secondary hydroxyl groups in the upper end.The bristling hydroxyl groups present at the outer cavity were more polar, while the inner cavity exhibited a nonpolar nature.3][54][55][56][57][58][59][60] Moreover, CDs have not only been used for host-guest complexations, they have also been employed as outstanding building blocks to fabricate supramolecular self-assembled nanomaterials. 61Different CDs have specific properties to form inclusion complexes with target guests and these abilities predominantly depend on the cavity size of the CDs. 62For instance, dodecyl guests can form different stoichiometric complexes with a-CD, b-CD, and g-CD, respectively, which further cause a reduction in the cavity size.Ph-ADA showed better binding affinity (7.6 Â 10 5 M À1 ) with b-CD in comparison to ADA (0.4 Â 10 5 M À1 ) since the hydrophobic Ph-ADA can gain more control over b-CD with stronger binding interactions. 63The nature of the binding interactions and their reversible assemblies can be regulated by external stimuli (pH, light, and redox) and competitive hosts/ guests.These approaches have been generically applied to fabricating various kinds of MPs with excellent embedded functionalities.
Adamantyl derivatives as guests for the construction of CD-derived polymeric assemblies
Molecular recognition-based self-assemblies are achieved at the molecular level and a similar mechanism at the macroscopic level is extremely hard to realize due to weaker selective interactions.Inspired by this approach, Harada et al. 64 developed a well-defined macroscopic assembly using CD-functionalized polymeric gels (a-CD and b-CD) with different hydrocarbonderived polymeric guests (adamantyl (Ad) (G4), n-butyl (n-Bu) (G5), and t-butyl (t-Bu) (G6), respectively) (Fig. 3).Interestingly, they demonstrated that the polymeric host and guest gel pieces could selectively adhere alternately on their surfaces via size-and shape-selective molecular host-guest interactions, producing macroscopic structures with sizes ranging from millimeter (mm) to centimeter (cm).For example, pieces of gels (a-CD, b-CD, G5 and G6) were mixed and shaken in water, producing the specific ordered macroscopic assemblies (a-CD stuck to G5 (K a = 5.7 Â 10 1 M À1 ), and b-CD stuck to G6 (K a = 1.7 Â 10 2 M À1 )).Similarly, b-CD and G4 were adhered to form large aggregates, and there were no interactions with the same pieces of gels, revealing that host-guest interactions could occur at the molecular level as well as at the macroscopic level.G4 had a better binding affinity with b-CD (K a = 1.5 Â 10 3 M À1 ) than a-CD (K a = 9.8 Â 10 1 M À1 ), and their reversible assemblies were achieved at high temperature (90 1C) with soluble competitive hosts/guests.
Bernard and Stenzel et al. 65 reported the fabrication of comb-shaped supramolecular polymers via Ad-end-functionalized reversible addition-fragmentation chain transfer (RAFT) agent and b-CD-grafted polymers in aqueous media, and their reversible assembly was achieved by competing host analysis.Becer et al. 66 prepared a thermoresponsive triblock copolymer by RAFT polymerization.This polymer could form inclusion complexes with mono-and hepta-mannosylated-CDs to further assemble into stable micelles (T 4 LCST (lower critical solution temperature)) covered with high carbohydrate density.These micelles were further utilized for the binding of concanavalin A (Con A) without affecting the micelle bilayer, but the micelle structure was disassembled by adding competitive CD or lectin.In 2018, both Ad-and Fc-containing stimuli-responsive block copolymers (BCP) were prepared by Gu et al. 67 Supramolecular self-assembled micelles were prepared by mixing BCP and b-CD (equimolar to Ad units), and the obtained micelles were responsive to redox-controlled reversible assembly by using FeCl 3 as an oxidant and GSH as a reductant in water.Moreover, oxidation-triggered benzocaine (BA) release (in vitro) from the BA-loaded spherical micelles (encapsulation efficiency = 4.3%) was observed in a slow and durable manner (50%, 40 h).Similarly, a poly(norbornene)based BCP side-chain comprised of Ad and n-Bu was achieved by ring-opening metathesis (ROMP) and supramolecular selfassembled micelles were obtained via host-(b-CD)-guest (BCPs) complexations. 68
Azobenzene derivatives as guests for the construction of CD-derived polymeric assemblies
Photoresponsive supramolecular hydrogels have been constructed with the help of photo-functional structural transformations of azobenzene guests with CDs through non-covalent interactions.Harada et al. 69 prepared both a curdlan-functionalized a-CD and an azo-derivative-functionalized side-chain polymer (G7), for the formation of a cross-linked supramolecular hydrogel (a-CD@G7) (1 : 1).Upon irradiating the formed hydrogels with UV light (365 nm), the trans-isomer was converted into the cis-isomer of the azo-polymer (trans/cis = 12 : 88), and the supramolecular hostguest complexation was demolished.Thus, reversible gel-sol phase changes were observed by the irradiation of light (365 nm/ 430 nm) or by heating (Fig. 4).Similarly, Zhang et al. 70 reported that b-CD mediated supramolecular self-assembled spherical micelles (b-CD@Azo) and their reversible assemblies were achieved by light irradiation (UV/visible, 30 min).
Red-and near-infrared (NIR) light-responsive supramolecular assemblies are more efficient than the UV-responsive ones.Therefore, Wu et al. 71 constructed red-light-responsive supramolecular hydrogels by the combination of both tetra-orthomethoxy-substituted azobenzene (mAzo) and b-CD-functionalized polymers, and reversible assemblies (gel/sol) were obtained by red/ blue (625 nm/470 nm) light irradiation (30 min).Furthermore, this hydrogel was utilized as a protein (bovine serum albumin (BSA)) carrier and red-light assisted controlled protein release (B83%, 60 min, 60 mW cm À2 ) from the gel was observed in deep tissue.An alternative approach was introduced by Armes and Tian et al., 72 in which the rate of thermo-responsive supramolecular morphological changes from vesicles to worms and/or spheres (20 to 2 1C) could be achieved by host-guest interactions between b-CD-functionalized BCP and azobenzene-linked mPEG (methoxy polyethylene glycol).The rate of morphological transitions was faster in the case of supramolecular vesicles (d = 100 nm, 70 min at 2 1C) owing to its reduced geometric packing of the supramolecular assemblies as compared to the b-CD-functionalized vesicles alone (d = 340 nm, 300 min at 2 1C).
Ferrocene derivatives as guests for the construction of CD-derived polymeric assemblies
Ferrocene-based supramolecular materials have dual functionalities, i.e., a redox switching and self-healing nature.Inspired by the above functions, Harada et al. 73 designed the supramolecular hydrogels (b-CD@G8) composed of multiple crosslinking between poly(acrylic acid)-functionalized b-CD and ferrocene (G8) polymers, which further produced the redox switching (sol-gel) and self-healing materials (re-adhesion) (Fig. 5).The b-CD@G8 hydrogel showed the highest elasticity (176 Pa) as compared to other combinations (o50 Pa), and the reduced form of Fc showed higher affinity for b-CD, due to its hydrophobic property, in comparison with the oxidized form (Fc + ).For reversible assembly (gel to sol), a competitive host (b-CD (excess))/guest (Ad) was used to cut off/rejoin the b-CD@G8 hydrogel (recovered strength = 84%, 24 h).Similar work on the construction of redox switchable supramolecular graft polymers via host-guest complexation between Fc-copolymers and end-chain functionalized b-CD polymer was reported by Ritter and Barner-Kowollik et al. 74 In this system, the host-guest assembly in aqueous solution produced mono-and multi-core micelle-aggregates (d = 10 nm and 300 nm) and the reversible assembly/disassembly could be easily achieved by redox reactions (cyclic voltammetry (CV)) and a competing guest (potassium 1-adamantylcarboxylate (Ad-COOK)).The visual recognition of the supramolecular self-assemblies in aqueous solutions is an interesting phenomenon in supramolecular chemistry.Ritter and Barner-Kowollik et al. 75 prepared supramolecular graft polymers via b-CD end-functionalized polymeric host interactions with phenolphthalein-based polymeric guests.Upon complexation of b-CD with phenolphthalein, the pink color of phenolphthalein changed to colorless and its absorption band (560 nm) gradually decreased upon increasing the b-CD concentrations (1 to 100 equiv.) in basic media.This is probably because of the re-lactonization of the phenolphthalein moiety in the brush-like supramolecular assembly.
To enhance the anticancer efficiency, drug solubility was improved by encapsulating the drugs within the b-CD cavity.Thus, Kim et al. 76 developed a stable nano-assembly based on polymeric b-CD (pCD) and polymeric paclitaxel (pPTX) via multivalent host-guest complexation between pCD and pPTX.In comparison, the multivalent pCD@pPTX binding interaction was higher (10 4 ) than the monovalent paclitaxel (PTX)/b-CD interaction, which extended the blood passage time and hindered the early drug release.Activator protein 1 (AP-1) with efficient targeting effects towards the interleukin-4 receptor was then incorporated into the fabricated supramolecular nano-assembly, making the nanomedicine (AP-1-pCD@pPTX) able to precisely release PTX to MDA-MB-231 cells via receptorassisted endocytosis (Fig. 6).In this system, PTX was linked to the building blocks of the polymer via ester bonds, which is responsible for the successful drug delivery by enzymatic degradation in cells.AP-1-pCD@pPTX could considerably suppress tumor growth and improve the survival rate of tumorbearing mice.The anticancer efficiency of this nanomedicine was increased as compared to that of the currently available (Taxol), playing a vital role in cancer treatment.
A thermoresponsive supramolecular system was constructed by Zhu et al. 77 through noncovalent interactions between cholic acid functionalized copolymer (CA) and b-CD (K s = 4.07 Â 10 3 M À1 ) and the phase transition temperature of this system could be reversed by the addition of a competing guest (K-Ad, K s = 3.98 Â 10 4 M À1 ).They further fabricated a supramolecular hydrogel (b-CD@CA) 78 by mixing water-soluble CA and b-CD copolymers, and the stability of this hydrogel was increased (cross-linking density) at a high molar fraction of CA (G 0 = 217 Pa at 8.3 wt%, [CA]/[b-CD] = 1).The dynamic reversible assembly of the hydrogel (gel-sol) was further evaluated by adding a competing guest (K-Ad), as well as by heating.
Wang et al. 79 developed an alternative approach for the preparation of an air-stable supramolecular complex (solid/solution) through the complexation of the air-sensitive CpFe(CO) 2 (Fb) pendant-containing polymer (G9) with b-CD.The formed b-CD@G9 complex produced an intramolecular dimerization with the nearby Fp pendant, exhibiting improved stability (Fig. 7).In contrast, the b-CD@G9 decomposed upon the removal of CD via the competing guest (Ad-COOH).Interestingly, this metal complex system exhibited a high response towards light (LED) and could thus be successfully utilized for CO release and antimicrobial materials.
Calixarene-based MPs
Calix[n]arenes (CAs) and their functionalized derivatives consist of phenolic rings attached via methylene bridges, and they possess distinctive conformations and cavities to accommodate various guests. 80In particular, CA [4] is one of the well-known macrocycles in supramolecular chemistry, and various analogous of macrocycles (oxacalixarenes, thiacalixarenes, azacalixarenes, calixpyrrole, etc.) have been synthesized by replacing the methylene bridges with some other units.Well-defined structural modifications of CAs have been employed as specific hosts for the recognition of target guests and have been utilized for various applications. 81CAs-functionalized polymers have played a vital role in the construction of supramolecular self-assemblies due to its rigid structure, producing highly ordered nanostructures with enhanced stability. 82To enhance their sustainability, water-soluble CAs were developed with suitable substituent groups (ammonium, phosphate, sulfonate, peptide, etc.) on the upper rim, which could be used as templates for the fabrication of various MPs nanoarchitectures by interacting with hydrophobic guests in aqueous media. 83n 2010, Haino et al. 84 developed a supramolecular crosslinked assembly (M n = 32 000, PDI = 2.81 at 0.5 equiv. of CA) by Fig. 6 Structures and schematic illustrations of the AP-1-b-CD@pPTX nano-assembly-assisted paclitaxel (PTX) drug delivery. 76sing a fullerene (C60) grafted polymer (G10) (M n = 13 000, PDI = 1.38) and CA [5]-functionalized homoditopic cross-linker (interchain manner) as shown in Fig. 8. Their noncovalent cross-linking stability was evaluated in different solvents (toluene 4 chloroform 4 o-dichlorobenzene).G10 showed a particle-like morphology (d = 10-30 nm), which changed into an ordered fibrous network structure (pitch = 19.6AE 1.6 nm) of the cross-linked CA[5]@G10.A few research groups [85][86][87] also reported the preparation of CA [4]-based polymeric derivatives, which were further used as resin/adsorbent materials for the selective recognition and removal of toxic heavy metal ions (Pb 2+ , Cu 2+ , Cd 2+ , Hg 2+ , Co 2+ ) and alkali metals (Na + , K + , Cs + ) in aqueous solution.
Aseyev et al. 88,89 synthesized photo-responsive poly(azocalix-[4]arene)s that were functionalized with different alkyl/tetraethylene glycol chains on the lower rim, and studied their dynamic host-guest interactions with different alkyl chainsubstituted pyridinium guests, respectively.To increase the sustainability, Taylor et al. 90 prepared a water-soluble supramolecular polymeric system from a sulfonated CA [4] polymer and methyl viologen (MV 2+ ) guest via non-covalent interactions, and examined their fluorescence and electrochemical stimuliresponsiveness.
In 2016, Huang and Sessler et al. 93 developed an interesting approach for the fabrication of a 'Texas-sized' molecular box-based supramolecular self-assembly through the host-guest interaction between a Texas-sized (Ts) macrocycle and an anionic polymer (G11) in a monotonic manner (1 : 1).By tuning the Ts ratio with respect to the anionic G11, various morphologies including tubules, vesicles, fusion (strings of beads), worm-like micelles, and spherical micelles could be obtained as shown in Fig. 10.This supramolecular self-assembled system exhibited responsiveness to both pH and biological molecules (adenosine-5 0 -triphosphate (ATP), adenosine-5 0 -diphosphate (ADP) and adenosine-5 0 -monophosphate (AMP)), and it could be further used for various cargo loading/releasing applications in living cells (HUVEC and HepG2 cell lines).Recently, various polymers bearing calix [4]pyrrole and tetracationic 'Texas-sized' molecular box moieties have been applied to extract inorganic and organic anions, and adhesive polymeric materials related applications, which were also highlighted in the reported reviews. 94,95
Cucurbituril-based MPs
In 1905, Behrend et al. 96 first synthesized cucurbiturils (CBs) from acid-catalyzed condensation reactions of urea, glyoxal, and formaldehyde, and their complete structures were elucidated by Mock and Freeman et al. using X-ray crystallography. 97im and Day et al. [98][99][100] synthesized and isolated the homologues of CB[n] (n = 5, 7, 8, and 10).In the case of CB [5] to CB [8], the cavity diameter increased from B4.4 Å to B8.8 Å, corresponding to the diameter of the portal (B2.4 Å to B6.9 Å).The cavity sizes of CBs resemble that of the cyclodextrin (CD) derivatives (a-, b-, and g-CD), but the symmetric pumpkin-like structure of CBs is far different from the CD derivatives.The CB portals are decorated with electron-rich carbonyl groups, thus CBs can noncovalently interact with various sizes of positively charged/neutral guests to form supramolecular host-guest complexations via hydrogen bonding, charge-dipole, and the hydrophobic/hydrophilic effect. 101For example, (i) CB [5] with a smaller cavity can complexes with larger sized cations (NH 4+ and Pb 2+ ) by suspending on the portals; 102,103 (ii) CB [6] can form inclusion complexes with hydrophobic neutral guests (tetrahydrofuran and benzene), protonated amines (diaminoalkanes (K a 4 10 5 M À1 ) and p-methylbenzylamine (K a B 10 2 M À1 ); [104][105][106] (iii) CB [7] can interact with naphthalene, protonated adamantanamine, MV 2+ , Fc, and carborane, respectively, thus leading to the formation of 1 : 1 stoichiometric complexes; 107 (iv) CB [8], with a large cavity, is involved in the complexation with large-sized guests (cyclen, cyclam, and their respective metal complexes).Interestingly, the larger cavity volume of CB [8] (479 Å 3 ) has been utilized as a ''molecular handcuff'' to accommodate two guests (electron-deficient and electron-rich) within its cavity to form a stable ternary complex (1 : 1 : 1) (10 15 M À2 ) via multiple noncovalent interactions. 108,109Due to the unique nature of CB [8], it has been exploited to construct MPs that have been used for various applications.
Another approach introduced by Ikkala and Scherman et al. 112 was the construction of cellulose nanocrystal (CNC)-based nanocomposite hydrogels, which were based on the supramolecular cross-linking of the three-component recognition of the CNC bearing Np polymer (G14) with the MV 2+ -functionalized polymer (G15) and CB [8] in aqueous solution (pH = 0.5).The formed nanocomposite hydrogels showed rapid sol-gel transition (o6 s), high storage modulus (G 0 4 10 kPa) and rapid self-healing properties facilitated by reinforced colloidal stability with three-component complexations (1 : 1 : 1) at lower pH (Fig. 12).Abell and coworkers 113 prepared supramolecular ternary complex-assisted hydrogel beads through the non-covalent interactions of CB [8] with MV 2+ and Np by using microdroplets as templates.FITC-dextran (cargo) was loaded on the hydrogel beads and could be potentially utilized for controlled cargo release by adding a competitive guest (1-adamantylamine).Likewise, Cao et al. 114 fabricated self-healing supramolecular hydrogels by the cross-linking of CB [8] with naphthaline containing sidechain copolymers to form a ternary complex (1 : 2).
6][117] For instance, Weck et al. 118 reported the directional host-guest self-assembly of cyanine dyes-functionalized colloidal particles with telechelic BCPs attached to patchy particles.These colloidal particles could directionally assemble into longer and branched chains based on the selective heteroternary complexation of polymeric guests (MV 2+ and azobenzene) within the inner cavity of CB [8].Later, Chen and coworkers 119 fabricated an amphiphilic supramolecular brush copolymer by the ternary complexation of CB [8], 4,4 0 -bipyridinium-functionalized copolymer (G16) and PEGylated Np (G17).This supramolecular polymer could further selfassemble into supramolecular spherical NPs (SNPs, d = 40 nm) with the aggregation-induced emission (AIE) effect.These SNPs were further used for DOX encapsulation (encapsulation efficiency = 11.7%) in aqueous solution, while the DOX-loaded SNPs showed fluorescence switching phenomenon (fluorescence-off) due to the formation of a Fo ¨rster resonance energy transfer (FRET) system (Fig. 13).Upon adding Na 2 S 2 O 4 (5 mg mL À1 ), these DOXencapsulated SNPs exhibited efficient anticancer drug release (77%, 36 h), with a clear observation of the fluorescence turn-on mechanism.
Typically, the branched polymeric topology can modify the chain relaxation, increasing the toughness and mechanical strength of the hydrogels.Based on the above approach, Scherman et al. 120 stated the fabrication of supramolecular hydrogels from the CB [8]-threaded highly branched polymer (HBP-CB [8]) (M w = 3.0 AE 1.5) Â 10 6 g mol À1 ) and linear hydroxyethyl cellulose-functionalized naphthalene (G18) (2 wt% of each).In this system, they abridged the synthetic route from three components to two components via the mechanical locking of CB [8] into the branched polymeric backbones.Compared with the three-component complexation (G19@CB[8]@G18, weak hydrogel, shear rate = 100 Pa s), the two-component strategy led to the formation of a strong hydrogel HBP-CB[8]@G18 with high viscosity (shear rate = 2000 Pa s) (Fig. 14).Briefly, compared with the linear G19@CB[8]@G18, HBP-CB[8]@G18 showed strain hardening due to the crosslinking of the intrachain to the interchain in the branched polymers.
Pillararenes-based MPs
Pillar[n]arenes (PAs) are 'rising star' macrocyclic hosts in supramolecular host-guest chemistry.The pillar-shaped structure is constructed by the methylene bridges at the para positions of the functionalized aromatic rings. 121Compared with other macrocyclic hosts, PAs have unique functionalities with vast advantages that include the following: (i) the capability of selective guest binding owing to its highly rigid and symmetrical structure; (ii) facile functionalization at all the rims or selective rims with one or two substituents employed for the perfection of supramolecular host-guest complexation-mediated superstructures; (iii) good solubility in both organic and aqueous solutions, making PAs a good supplement for other water-soluble macrocyclic hosts with analogous cavity size. 122In particular, pillar [5]arene (P[5]A) and pillar [6]arene (P[6]A) possess higher electron density than calix-shaped structures, and thus they are accountable for the excellent binding affinity with electron-withdrawing or neutral guests.Therefore, PAs function as potential candidates for the fabrication of diverse supramolecular self-assemblies in the field of rotaxanes, 123 hybrid materials, 124 liquid crystals, 125 hydrogels, 126 polymers, [127][128][129] chemosensors, 130,131 light-harvesting systems, 132,133 drug deliveries, 134 and catalysts. 135umerous homologous PAs have been developed via elegant synthetic routes. 136Notably, P[5]A and P[6]A derivatives have been widely employed in the construction of different supramolecular host-guest assemblies due to their facile functionalization and suitable cavity size (P[5]A = 4.7 Å, P[6]A = 6.7 Å).Owing to the cavity-dependent inclusion in these structures, P [6]A can interact with larger-sized guests to form supramolecular complexes in comparison with P [5]A.The binding affinity of watersoluble PAs with guest molecules is greatly enhanced due to the introduction of cationic or anionic moieties on the PA backbone, leading to the formation of supramolecular amphiphiles.As an example, water-soluble PAs with carboxylate, imidazolium, and trimethylammonium functional groups can offer an appropriate binding environment to different guests via non-covalent/ electrostatic interactions. 137Stimuli-responsive MPs have also been constructed by the utilization of stimuli-responsive polymeric hosts and/or guests, affording supramolecular micro-and nanostructures with superior functionalities.
Wang and Hu et al. 138,139 developed side-chain polypseudorotaxanes from the P[5]A-polymeric host and n-octylpyrazinium hexafluorophosphate guest.The disassembly of the polypseudorotaxane system was achieved by adding halide ions, which could be used as an anion-responsive fluorescent sensor.Huang et al. 140 studied the effects of LCST in a thermoresponsive polymer by PA-based supramolecular host-guest interactions.The LCST behaviour and solubility of the MV 2+ -based polymer increased when it formed an inclusion complex with suitably-sized P [6]A due to the inhibition of interpolymer aggregation.Similarly, Yu et al. 141 reported the thermoresponsive supramolecular system formed by P [10]A and the MV 2+ -based polymeric guest, and studied their reversible assembly by introducing a competitive guest (1,10-phenanthrolinium).
In 2015, Huang et al. 142 reported a dual thermo-and photoresponsive supra-amphiphilic polypseudorotaxane, which was constructed by hydrophobic interactions between P [7]A and an azobenzene polymer (G20) as shown in Fig. 15.This supraamphiphilic polypseudorotaxane could self-assemble into vesicles (d = 260 nm) with reversible morphological transformations by tuning the solution temperature (LCST) or irradiating with light.Calcein was successfully loaded into these supramolecular vesicles to further investigate the controlled release via irradiation or thermal treatment.For thermal stimulus (60 1C), 90% of calcein was released within 10 h due to the structural disruption into solid NPs.Upon irradiation (365 nm), only 70% of the calcein was released due to the existence of the cis-trans isomerization equilibrium.They also fabricated DOX-loaded SNPs for self-imaging drug delivery (in vitro/in vivo) with the help of the AIE effect. 143ang et al. 144 developed a redox responsive supramolecular gel-like network (pFc + -pP[6]A) through the combination of P [6]A with ferrocene-functionalized copolymers.Interestingly, the oxidized form of ferrocene (pFc + ) displayed a larger binding affinity with pP [6]A than pFc.The redox-responsive assemblies can be controlled by various external stimuli, such as chemical redox reagents and competing hosts.Liao et al. 145 prepared a multi-responsive supramolecular polymeric gel through hostguest interactions between pP [5]A and the bis(pyridinium) dication guest (MV 2+ ).The gel to sol transition was achieved by adding a competitive host/guest or by heating.The freeze-dried gel displayed a halo peak at the wide-angle region (2y = 25.21)measured by wide-angle X-ray scattering (WAXS), which may be attributed to the ordered structure of supramolecular complexes and p-p stacking of P [5]As.When this hydrogel was heated to 50 1C, the peak intensity decreased, revealing the disassembly of the supramolecular gel networks.
Stimuli-responsive materials play a significant role in the development of smart windows due to the self-adjustment of the window transparency through external environmental stimuli, thus creating a comfortable indoor environment for advanced lifestyles.Inspired by smart materials, Wang and coworkers 146 synthesized color-switchable thermochromic hydrogel materials based on the host-guest interactions of ethylene glycolfunctionalized P[6]A (EGP[6]A) and the polymeric ferrocene derivative (G21).Due to the efficient EGP [6]A-G21 complexation, the swelling ratio of the EGP [6]A-G21 hydrogel drastically increased to 150% as compared with the dry G21-gel.The thermo-responsiveness of the EGP[6]A-G21 hydrogel was examined by increasing/decreasing the temperature (40 1C/25 1C), leading to reversible opaque (4T cloud )/transparency (oT cloud ) transformation.The reversibility and stability of hydrogels were efficiently maintained for more than 100 cycles without the leakage of EGP [6]A from the hydrogel.On the other hand, the redox responsiveness of the EGP[6]A-G21 hydrogel was achieved by introducing chemical reagents ((NH 4 ) 2 S 2 O 8 /N 2 H 4 ÁH 2 O) or through a more favourable electrochemical approach to achieving the warm and cool colour switching (Fig. 16).They have also reported the construction of supramolecular hydrogels with controlled shrinking behaviour (89.2% in wt), which was achieved by noncovalent interactions between water-soluble P[5]A (WP[5]A) and quaternary ammonium-functionalized copolymer. 147iodegradable polymeric assemblies have played a vital role in reducing antimicrobial resistance and prevent their accumulation.In 2019, Gao and coworkers 148 developed a supramolecular polymeric system (CP [5]A@G22) from the anionic carboxylatopillar [5]arene (CP[5]A) and quaternary ammonium polymer (G22).This system showed better antibacterial activity in Gram-positive and methicillin-resistant Staphylococcus aureus (S. aureus) (MRSA).In particular, CP [5]A@G22 material was involved in the selective membranolysis of S. aureus rather than E. coli (Fig. 17).The bacterial elimination mechanism was further explained by z-potential measurements of CP [5]A@G22 with S. aureus, which showed an insignificant change (À38.5 to À41.7 mV); however, in the case of CP [5]A@G22 with E. coli, an obvious z-potential shift was observed (À45.3 to À29.6 mV).The above result indicated that the anionic PA and a thicker lipopolysaccharide-linked membrane in E. coli inhibited the binding of CP [5]A@G22 to the bacterial surface, thus leading to a selective antibacterial potency toward S. aureus.The ionic conductivity of the supramolecular complex (pP [5]A@Py) showed greater performance than the complex of pP [5]A-copolymer@Py.This may be ascribed to the regularity in its structure, which further enhanced the conductivity in the system.Moreover, the glass transition temperature (T g ) exhibited a progressive decline in the supramolecular polymer in comparison with the individual polymer, indicating that the improved flexibility of the supramolecular polymer could prevent its leakage from the system.Therefore, we conclude that this supramolecular material exhibits good compatibility and efficient tuneable conductivity upon increasing the ratio of the ionic guest.
Summary and outlook
Over the past decade, significant efforts have been made to achieve efficient and novel functional materials in the field of MPs.These efforts and innovative results indicate the depth and achievement of MPs in spreading their influences in supramolecular and materials chemistry.In the present review, significant properties and potential applications of MPs have been highlighted, which include stimuli-responsive hydrogels, nanocarriers for drug delivery, air-stable metal complexes for CO delivery, adsorbent/adhesive materials for the sensing and removal of toxic cations/anions, and smart windows, etc.Despite the unquestionable success achieved with MPs, there is still of the need for cutting-edge research that is yet to be industrialized using MPs with advanced functionalities and futuristic applications.Hence, this review provides some clues about the construction of supramolecular self-assemblies via macrocycles/small molecules-functionalized polymeric building blocks, tasks for current researchers, and guidelines for future challenges.
MPs with dynamic functionalities can provide a complementary approach to traditional assemblies.In comparison, low molecular weight supramolecular assemblies have lower stability than polymeric assemblies, thus hampering their full realization of realistic applications.Therefore, macrocyclic polymers noncovalently interacting with small molecular guests functionalized by side-chain polymers can provide a route for achieving dynamic MPs with high stability.This noncovalent strategy slows down the molecular exchange via multiple interactions and their polymeric entanglement is responsible for better stability than the analogs of small molecules.Furthermore, the loading of enzymes and other biomaterials into the MPs can be more facilely achieved than covalent chemistry and these materials have great potential applications for long-term drug delivery and therapy.
In the past decades, more MPs have been constructed and utilized for clinical applications, although some significant issues need to be resolved before the clinical application of supramolecular systems, such as stability, sensitivity, specificity, reliability of the platforms, therapeutic efficacy, excretion, and immunological reactions, etc. 150,151 On the other hand, the fabrication of MPsbased nanosystems with well-defined features, such as size, shape, stability, surface tension, charge, payload and aggregation number are rarely reported and are foremost in the upcoming research.Also, the expansion of nanomedicines with excellent targeting ability and therapeutic efficiency-related investigations are crucial for biomedical remedies.
Undoubtedly, researchers will face many problems that need to be actively addressed; these are related to diverse fields of applications and may be overcome by utilizing the scope of chemistries, new methodologies, and materials available.Moreover, the discovery of new macrocycles with fascinating properties, including inexpensive and facile synthetic routes with reasonable yields, could also be evaluated for replacing available MPs.To predict the properties of newly designed MPs, experimental and theoretical analyses are necessary for the further growth of this field.Moreover, in the case of biomedical applications, MPs disassemble into various parts and try to determine the ultimate destiny of hydrophobic polymeric units.
Progress in the design and construction of fluorescent selfassembled hybrid materials as nanocarriers for therapeutic applications is forthcoming.For example, MPs loaded with new components, including gold nanoparticles (AuNPs), gold nanorods (AuNRs), magnetic NPs, upconversion NPs, and quantum dots, which can produce fluorescent self-assembled hybrid nanomaterials with photodynamic/photothermal properties, stimuli-responsiveness, target sensing, and other bio-related features, will be a success.State-of-the-art studies on MPs, MPsfunctionalized hybrid nanomaterials and their latent practical applications present challenges.These challenges can be fruitfully addressed by the blending of rational materials design and Ultimately, by merging the scope of macrocyclic polymers and membrane technology, size-and shape-dependent polymeric membranes containing macrocycles can be produced, which will allow for the separation/removal of various analytes, such as cations, anions, small molecules, and complex mixtures.Considering these advantages, these tunable membranes can be used to capture and remove organic/inorganic pollutants from aqueous/organic solvent mixtures with greater accuracy and reliability than commercially available polymeric membranes.These strategies can produce prospective applications for these MPs-based materials, which fulfil the global challenges faced by our modern society.
Fig. 1
Fig. 1 Graphical representation of the fabrication of different macrocyclesassisted polymeric self-assemblies based on host-guest complexation and their potential applications.
Fig. 4
Fig. 4 Structures and schematic representation of the interactions of the a-CD (cylindrical shapes) with G7 (yellow shapes) upon irradiation with UV (365 nm) and visible light (430 nm) or heating at 60 1C.Reproduced from ref. 69 with permission from John Wiley & Sons, Inc., Copyright 2010.
Fig. 9
Fig. 9 Structure and schematic representation of the (a) C4P-assisted anion recognition and acid-induced disassembly.(b) Anion removal and (c) regeneration of C4P from aqueous solution.Reproduced from ref. 92 with permission from John Wiley & Sons, Inc., Copyright 2018.
Fig. 11 (
Fig.11(a) Schematic representation and structures of the G12 (AuNP functionalized with a mixture of 3 and 4) and G13.(b) Illustration of the microcapsule formation process from the initial droplet stage to the dehydrated stable capsules.111
Fig. 13 (
Fig. 13 (a) Chemical structures and schematic representation of CB[8], G16 and G17, and the preparation of SNPs and DOX-loaded SNPs.(b) Illustration of the imaging-guided drug delivery.Reproduced from ref. 119 with permission from the American Chemical Society; Copyright 2017.
This journal is © The Royal Society of Chemistry 2020 Mater.Adv., 2020, 1, 2646--2662 | 2659 Recently, Xie and Liao et al. prepared a supramolecular polymeric system based on the non-covalent interactions between a pyridinium guest (Py) and a P[5]A-derived polymeric host (pP[5]A).
Fig. 17
Fig. 17Structures and illustrations of G22 and CP[5]A.Preparation of the supramolecular material (CP[5]A@G22) and its antibacterial activity for MRSA-infected wound healing treatment.Reproduced from ref. 148 with permission from John Wiley & Sons, Inc., Copyright 2019.
Fig. 17Structures and illustrations of G22 and CP[5]A.Preparation of the supramolecular material (CP[5]A@G22) and its antibacterial activity for MRSA-infected wound healing treatment.Reproduced from ref. 148 with permission from John Wiley & Sons, Inc., Copyright 2019.
|
2020-10-28T18:55:05.993Z
|
2020-11-16T00:00:00.000
|
{
"year": 2020,
"sha1": "94bbbd4bde406e113a977c47d4160aa0ace40a67",
"oa_license": "CCBYNC",
"oa_url": "https://pubs.rsc.org/en/content/articlepdf/2020/ma/d0ma00625d",
"oa_status": "GOLD",
"pdf_src": "Anansi",
"pdf_hash": "ca364ec39515585bf058230350d3db7740740dac",
"s2fieldsofstudy": [
"Materials Science",
"Chemistry"
],
"extfieldsofstudy": [
"Materials Science"
]
}
|
250770905
|
pes2o/s2orc
|
v3-fos-license
|
Limits on the Auroral Generation of H$_3^+$ in Brown Dwarf and Extrasolar Giant Planet Atmospheres with Keck/NIRSPEC
The molecular ion H$_3^+$ is a potentially powerful tracer of the ionospheres and thermal structures of Jovian planets, but has never been detected in a planetary mass object outside of the solar system. Models predict that H$_3^+$ emission driven by EUV flux and solar wind on hot Jupiters, or by powerful aurorae on brown dwarfs, will be between $10^2$ and $10^5\times$ more intense than that of Jupiter. If optimal conditions for the production of emission do exist, the emission may be detectable by current ground-based instruments or in the near future. We present the first search for H$_3^+$ line emission in brown dwarfs with Keck/NIRSPEC $L^\prime$ high-resolution spectroscopy. Additionally, we survey stars hosting giant planets at semi-major axes near $0.1-0.2$ au, which models suggest may be the best planetary targets. No candidate H$_3^+$ emission is found. The limits we place on the emission of H$_3^+$ from brown dwarfs indicates that auroral generation of H$_3^+$ in these environments likely does not linearly scale from the processes found on Jupiter, plausibly due to deeper atmospheric penetration by precipitating auroral electrons. Detection of H$_3^+$ emission in brown dwarfs may be possible with the James Webb Space Telescope (JWST), or future thirty-meter class telescopes.
INTRODUCTION
A major goal in exoplanet studies is understanding the diversity and evolution of planetary atmospheres, including atmospheric responses to variable stellar environments and loss rates over time. Within this goal, describing the thermodynamic properties and energy budgets that govern atmospheric temperatures, chemistry, and expansion are paramount. As the interface between an atmosphere and its surrounding space, the upper atmospheres of planets, including the thermosphere, ionosphere, and exosphere, are excellent laboratories to probe atmospheric thermodynamics as they are more sensitive to changes in energy flow than lower atmospheric regions.
In the past decade, planetary upper atmospheres outside our solar system have begun to be probed by spectroscopic observations of hot Jupiters and brown dwarf planetary analogues. Notably, atmospheric expansion and hydrodynamic escape, as well as thermospheric and ionospheric layers, temperatures, and winds, have all been measured or inferred for exoplanets using tracers such as He I and Na I, among others (see Vidal-Madjar et al. 2011;Spake et al. 2018;Seidel et al. 2020;Cauley et al. 2021;Allart et al. 2019). Recently, observations of ionospheric species have been described as evidence for the possible first detection of an exoplanetary magnetic field (Ben-Jaffel et al. 2021).
A direct and potentially significant tracer of the ionosphere, which has yet to be detected in an extrasolar planetary mass object, is the molecular ion H + 3 . H + 3 is expected to form readily in the upper atmosphere via the ionization of H 2 by extreme-ultraviolet radiation (EUV, 10 − 100 nm), or by energetic electrons precipitating along field lines during aurorae: Thus, H + 3 is expected to only exist in concentration above the homopause where atmospheric mixing is weak and abundances of heavier species are low. In stellar atmospheres, H + 3 is not expected to be abundant except in the coolest, metal poor stellar atmospheres (Tennyson & Miller 2001), such that any detection in a stellar environment can be attributed to a planetary ionosphere.
Within the thermospheres and ionospheres of giant planets and brown dwarfs, H + 3 is expected to be a dominant source of radiative cooling that scales with incoming energy flux, sometimes called the "H + 3 thermostat" (Maillard & Miller 2011). Observations of Jupiter's auroral regions have shown that H + 3 rotational-vibrational emission in the near-and mid-infrared effectively balances variable energy input from EUV flux, the solar wind, and particle precipitation. Its role as a coolant is expected to continue for extrasolar giant planets, at least up to high thermospheric temperatures less than ∼10,000 K, above which the formation will be suppressed by the dissociation of H 2 (Koskinen et al. 2007). In our own solar system, observations of H + 3 in Jupiter, Saturn, and Uranus have been a useful diagnostic of upper atmospheric temperatures, densities, wind dynamics, and cooling, as well as magnetosphere-atmosphere interactions (see a review by Miller et al. 2020).
Theoretical predictions for H + 3 emission in close-in giant-planet and brown dwarf atmospheres agree that the emission is likely to be orders of magnitude larger than that emitted by Jupiter. This increase can be driven by elevated EUV flux compared to Jupiter in the case of giant exoplanets, or by powerful magnetic fields and associated aurorae for brown dwarfs or exoplanets with strong magnetospheres. While there is consensus that H + 3 emission is an important thermal mechanism, exactly how much stronger emission is to be expected in these environments is uncertain. Early predictions suggested that a planet like τ Boo b may have an H + 3 emission luminosity upwards of 10 17 W, roughly 10 5 × that of Jupiter (Miller et al. 2000), with most of that energy in concentrated in a few lines. Subsequent models, however, have been conservative, with Koskinen et al. (2007) predicting a maximum total luminosity for any planet of 10 15 W and Chadney et al. (2016) predicting maximum total luminosities of around 10 16 W for a planet like HD 209458 b between 0.1 and 0.2 au from a young, active K or M dwarf. An additional complication is that individual line strengths do not grow linearly with the total intensity as the thermosphere becomes hotter, as more molecular transitions become populated at higher excitation temperatures. Furthermore, none of these models account for auroral contributions to H + 3 generation, which may boost the H + 3 luminosity depending on the auroral energy deposition.
Free-floating brown dwarfs may represent the best environment for successful H + 3 detection as a number of them possess evidence of powerful aurorae (10 4 × more energetic than Jupiter's) in the form of electroncyclotron maser (ECM) emission and, in a few cases, synchronous variable Balmer emission (Hallinan et al. 2015;Pineda et al. 2016). The auroral current is postulated to be driven by either the break between corotating and subrotating magnetic field lines outside the dwarf (Nichols et al. 2012), or by interaction with a brown dwarf satellite (Hallinan et al. 2015). By assuming that the strength of the aurorae corresponds to a linear increase in the number of precipitating electrons, and that the energy of the electrons do not increase significantly, Helling & Rimmer 2019 have suggested that H + 3 emission will similarly be near or above 10 4 × that of Jupiter (10 16 -10 18 W). Pineda 2017 argues that if brown dwarf aurorae operate analogously to that of Jupiter, then H + 3 L emission of H + 3 could carry up to 10× more energy than optical Balmer emission that has been associated with aurorae for some nearby brown dwarfs (up to 10 18 W). These works suggest that detection of H + 3 emission from a free-floating brown dwarf with current instruments is plausible.
A number of attempts have previously been made to detect H + 3 emission in hot Jupiters, most of which have focused on observing the fundamental ro-vibrational H + 3 lines in the L band around 4µm (Brittain & Rettig 2002;Goto et al. 2005;Shkolnik et al. 2006;Laughlin et al. 2008;Lenz et al. 2016). Shkolnik et al. (2006) carried out a survey of six systems hosting extrasolar giant planets (EGPs) with the CSHELL instrument on NASA IRTF and set the lowest detection limit so far of 6.3 × 10 17 W for GJ 436 at 1σ significance. Most recently, Lenz et al. (2016) observed the HD 209458 system with VLT/CRIRES using occultation spectroscopy to attempt to reduce the effect of the stellar background. They achieved a 3σ detection limit of 5.3 × 10 18 W for the Q(3,0) line at 3985.5 nm, but were significantly hampered by poor weather.
There has been one observational attempt to detect H + 3 emission in brown dwarfs. Pineda (2017) searched for H + 3 lines in K band, medium resolution spectroscopy of brown dwarfs with known ECM radiation using Keck/MOSFIRE. They did not detect any excess infrared emission attributable to H + 3 , but did not set emission limits for these objects. Within their discussion, they suggest high-resolution L-band spectra to confirm their results, as only about 10% of H + 3 emission is expected in K, with the remaining energy emitted mostly in L (Neale et al. 1996). Therefore, a high-resolution search for emission lines in the L band to set observational limits is of high interest despite the added challenge of increased thermal background when observing at longer wavelengths.
In this paper, we describe a search for H + 3 emission in 5 brown dwarfs and 5 systems hosting giant planets using Keck/NIRSPEC. We describe our target selection methodology in Section 2. Section 3 outlines our observations of the 10 targets and our data reduction process. In Section 4, we present results and set upper limits on emission for each of our targets. We interpret our nondetections in the context of atmospheric and magnetospheric models for brown dwarfs and giant planets in Section 5, and finally discuss future paths for detection with observatories such as JWST.
SURVEY SCOPE AND TARGET SELECTION
The primary goal of our study is to attempt to constrain the upper atmospheric and auroral chemistry of brown dwarfs by setting observational emission limits for H + 3 . As an additional goal, we attempt to detect or set upper limits for H + 3 in giant exoplanets at distances of 0.1 to 0.2 au from solar or later type dwarfs, which models suggest are the most promising planetary targets (Koskinen et al. 2007;Chadney et al. 2016). In determining a number of targets to observe, we attempted to balance the integration time per target with sample size such that we would reasonably avoid non-detection due to small number statistics.
To select our brown dwarf targets we relied upon radio observations of ultra-cool dwarfs summarized in Williams (2018). At present, approximately a dozen brown dwarfs have been observed emitting ECM radiation, which is thought to be produced by auroral processes and is similar to auroral radio emission observed from the poles of Jupiter. We chose the closest brown dwarfs with observed ECM radiation, and which span the spectral range from the end of the main sequence to T dwarfs. H + 3 emission has only been previously modelled in one brown dwarf (2MASSI J1835379+325954) (Helling & Rimmer 2019). Therefore, we decided to select targets with diverse spectral types and therefore atmospheric conditions rather than those that necessarily had the most intense ECM radiation, since the optimal atmospheric conditions for high H + 3 emission in brown dwarfs is still mostly unexplored. We also selected the ultra-cool dwarf binary system 2MASS J07200325-0846499 (Scholz's star) as it is possible the binarity helps stimulate aurorae and H + 3 generation as Io does with Jupiter by producing an interacting plasma (Connerney et al. 1993), although it is not clear if the separation in this system (a ∼ 2.2 au, Dupuy et al. 2019) is close enough for a similar phenomenon to occur. We do not independently consider the inclination or rotation rates of the dwarfs, although inclinations have been measured for some dwarfs (e.g. Vos et al. 2017) and may effect aurorae observability (although this has not been quantified). Our four selected brown dwarf systems can be found in Table 1.
The motivation for an additional search for H + 3 in exoplanets is that of three previous observational attempts, two have focused on only one ultra hot Jupiter each (τ Boo b, Laughlin et al. 2008, andHD 209458 b, Lenz et al. 2016). While detection of H + 3 in any planet is at the current limits of technical capability, models published after these observations (Chadney et al. 2016) suggest that the thermospheres of these specific EGPs are very likely too hot to contain significant H + 3 abundance. Therefore, we decided to include a number of planetary systems that are cooler, yet still have a high ionizing flux from their host star To select planetary targets, we used the NASA Exoplanet Archive (DOI 10.26133/NEA12). Confirmed planets were sorted based on semi-major axis, bolometric flux received, mass (or M sin i), and distance from Earth. We excluded planets with masses less than that of Saturn, where the thermospheric composition is expected to begin to change from being hydrogen and helium dominated. Planets with semi-major axes less than 0.07 au were excluded based on the probability that their thermospheres are too hot to contain H + 3 through previously described dissociation processes. Otherwise, increased proximity to the star and bolometric flux received were considered to make targets better candidates, as those planets are simply exposed to more ionizing radiation. In reality, it is the EUV, XUV, and X-ray flux received that is most important for H + 3 production, not the bolometric flux, but EUV and XUV are not measured for most exoplanet systems -due to a lack of observatories in this wavelength region and because of absorption from the interstellar medium -so the less-informative bolometric flux is used as a proxy. Systems with multiple EGPs were preferred even if some of those giant planets are not necessarily good candidates for detection. Since signal-to-noise of our spectra is of central importance to emission limits, apparent magnitude was used as a primary selection criterion. It is for the same reason that all of our targets are non-transiting systems detected through radial velocity measurements, as known transiting planets farther than 0.1 au from their host stars are still generally much farther from Earth due to the lower probability of having a transiting orbital inclination at increased semi-major axis. The fact that they are non-transiting precludes the possibility of using more sophisticated observational techniques like occultation spectroscopy to search for H + 3 . Our five selected planetary targets can be found in Table 2. A comparison of the estimated EUV flux that some of these planets receive compared to that of Jupiter and a few other exoplanetary systems is shown in Figure 1. This figure illustrates that we are probing a parameter space between the extremes of thermospheres that are too hot, and too little ionizing EUV radiation to produce observable H + 3 . EUV flux estimates come from the X-exoplanets database (Sanz-Forcada et al. 2011), which uses coronal modeling to predict EUV if no observational data are available. These values should be viewed as order-of-magnitude. In particular, the EUV flux received by planets around GJ 876 is probably an order-of-magnitude higher as the star is an active M dwarf.
As a third target class, we chose to observe the white dwarf, brown dwarf (WD-BD) binary GD 1400. This novel target class may combine the benefits of planetary and brown dwarf targets for H + 3 emission. The brown dwarf may have a strong magnetic field and aurorae, but it also receives strong EUV flux from the white dwarf. The white dwarf is dimmer than main-sequence stars in mid-IR, so the system will have a lower stellar contrast. Unfortunately, GD 1400 is very faint despite being the closest WD-BD binary known, and is a thus an extremely challenging target for L-band high-resolution spectroscopy.
All together, we selected 10 promising targets for the detection of H + 3 in the atmosphere of a planetary mass object. Highest observational priority was given to the brown dwarf 2MASSI J1835379+325954, as it has strongest evidence for an aurora is the only brown dwarf with a published model for H + 3 . Otherwise, observing time was based primarily on object observability and scheduling constraints for a given night. Observations of these objects are described in the next section.
OBSERVATIONS AND DATA REDUCTION
We observed a total of two and a half nights with Keck/NIRSPEC (McLean et al. 1998;Martin et al. 2018;López et al. 2020) on the 10 targets described in Section 2. Individual observations occurred on 2020 August 27th, September 3rd, October 6th, and 2021 March 3rd, UT.
Spectra were obtained using the high-resolution mode with the KL filter. The echelle and cross-disperser angles were set to 62.25 • and 33.72 • respectively. This setup places six H + 3 lines within order 19 of the echellogram with a wavelength range of approximately 3.94 to 4.02 microns. These are the same lines included in previous searches by Shkolnik et al. (2006) and Lenz et al. (2016), and which have also been observed prominently in Jupiter (e.g. Maillard et al. 1990). Of these six lines, the Q(1,0) and Q(3,0) transitions, at 3953.0 and 3985.5 Note-Stellar spectral types, distances, and magnitudes come from the SIMBAD astronomical database (Wenger et al. 2000). Radial velocity references: (1) Note-Stellar spectral types, distances, and magnitudes come from the SIMBAD astronomical database (Wenger et al. 2000). Stellar radial velocity references: (1) nm, respectively are expected to be the most intense, with the Q(1,0) being most intense at lower rotational and vibrational temperatures akin to Jupiter (∼500 K) and Q(3,0) becoming dominant and higher temperatures (∼2000 K) (Neale et al. 1996). Other transitions in the R and P branch do reach similar intensities as these lines, however, the region around 4 microns is the densest re-gion of prominent lines across low and high excitation temperatures, and other prominent lines could not be observed with a single echelle setting. Each night, we chose the slit width to be comparable to the average seeing in order to balance target throughput with minimizing sky background. The best seeing we achieved on any night was around 0.4", so only the 0.432"×12" (R∼ 25, 000) and wider slits were used. There was no appreciable cloud cover on any night listed above. The precipitable water (PW) ranges for our nights were 1 to 1.2 mm for 27 Aug 2020, 0.6 to 1 mm for 03 Sep 2020, 4 to 6 mm for 06 Oct 2020, and 0.5 to 0.65 mm for 03 Mar 2021.
All observations used a standard ABBA-nodding pattern along the slit that allows removal of thermal background by A-B pair subtraction. At the beginning and end of observing each science target, we observed at least one spectroscopic standard star of A0 or adjacent spectral type, and at similar elevation to the target, for use in correcting telluric lines. Exposure times were chosen for each target to be safely below detector non-linearity at ∼ 20, 000 ADU. Except for the brightest stellar targets, the exposure time is generally limited by the thermal background itself with an exposure of 30 seconds per co-add nearing non-linearity.
Observation parameters for each target are detailed in Tables 1 and 2. As discussed in Section 2, our observation strategy attempted to balance sensitivity with number of targets observed. Thus, no target is observed on sky for longer than 2.5 hours, corresponding to a maximum integration time of about 90 minutes with readout and nod overheads. Due to target visibility and scheduling constraints, most targets are not observed to this limit.
Initial data reduction steps, including calibration, order rectification, trace fitting, and spectral extraction were carried out using a modified version of the python pipeline described by Piskorz et al. (2016). Modifications were made to bad pixel correction to prevent clipping with our data. This pipeline uses optimal extraction methods described in Horne (1986). Stated simply, optimal extraction improves spectral signal-to-noise by weighing pixels based on the wavelength averaged PSF. Reduction steps are carried out individually for each nod pair for bright targets. For dimmer targets, it was sometimes necessary to combine adjacent pairs together, essentially as a co-add, so that the spectral trace could be well-fit. Our faintest target, the WD-BD binary GD 1400, is sufficiently dim that nearly all pairs must be added to confidently fit the trace.
Raw 1-D science spectra are continuum normalized before being combined into a median spectrum. Continuum normalization is achieved by smoothing the spectrum with a Gaussian filter then constructing a polynomial fit between local maxima, which trace the average continuum level, that is then divided out. This normalization routine changes the systematic structure of the spectra, but not at the scale of the expected H + 3 emission lines.
Wavelength calibration is performed by a 2nd-order fit of sky emission line positions in the observation data to the same lines in atmospheric transmission models by Lord (1992). Line centers are determined by a Gaussian fit. We find that even with the spectrograph angles set to the same positions each night, there is still a variation of about 10 angstroms in the wavelength position of a given pixel between nights. The average root-meansquare (RMS) deviation of measured sky line positions compared to our preferred wavelength solutions is 1.1 pixels, with the most accurate wave solution for HD 217107 (RMS∼ 0.37 pixels) and the least accurate for LSPM J0036+1821 (RMS∼ 1.6 pixels). For reference, the expected FWHM of any H + 3 line is ∼ 4 pixels. We use ESO's molecfit tool for telluric correction of the science spectra Kausch et al. 2015). Molecfit creates a synthetic transmission spectrum using a radiative transfer code and local observatory and meteorological data. Synthetic telluric correction is preferable to using standard stars alone as the synthetic spectrum is theoretically noiseless whereas standard stars necessarily contain photon noise in addition to any systematic errors that can exist in both synthetic and observed spectra. For each target, we generate a synthetic telluric spectrum by fitting a molecfit spectrum to the corresponding standard star. The atmospheric model used is the molecfit default. The exception is that we additionally fit the molecule NO 2 . The most significant absorption features observed within our wavelength range are caused by NO 2 . This coincidence is beneficial, as NO 2 is expected to have less atmospheric temporal variation compared to other molecules, notably H 2 O. For planetary targets, we take the additional step of using the best fit telluric model from the standard stars and using it as the start for an additional molecfit refinement using the science spectrum itself. This step is not possible for the brown dwarfs as their intrinsic spectra contain too many absorption lines for molecfit to be able to reliably disentangle the telluric features. An additional benefit of using molecfit that we utilize is a final refinement of the wavelength solution based on the synthetic model.
As a test for systematic errors that molecfit might introduce to our spectra, we experimented with performing fits using different standard stars for the target GJ 876 (multiple standard stars were observed around this data set). After comparing fits from 3 different standard stars fit separately, there are no systematic differences of large enough magnitude or at a spectral scale to alter the interpretation of the corrected spectrum. Therefore, while some systematic errors are likely introduced by molecfit, we do not believe they could be responsible for an erroneous positive or negative detection.
For our planetary targets, we expect that the radial velocity does not vary significantly compared to our resolution over our roughly two hour periods of observation. Therefore, we expect the H + 3 line positions to remain effectively stationary so that no spectral shift and add technique is necessary.
In the next section, we describe our search for H + 3 lines within the data, and our method for setting observational limits for H + 3 .
4. LIMITS ON H + 3 EMISSION After data reduction, we search for candidate H + 3 lines within our spectra, and in the absence of any, set limits on the line luminosities that we would have expected to detect given our spectral signal-to-noise.
For all our targets, we elect to take the simple approach of a direct, visual search near the anticipated H + 3 line positions based on the object radial velocities and after correcting for Earth's barycentric velocity. For exoplanetary systems, we also need to account for the radial velocity of each planet, which is dependent on the orbital inclination and phase at observation. Fortunately, most of these exoplanets have relatively well characterized orbits with known inclinations and precise ephemeris data. For example, the uncertainty in the orbital positions of GJ 876 b and c are less than 1%. In cases where the inclination of the planet is not known, we simply calculate the range in line positions given a minimum and maximum possible RV at the time of observation.
Theoretically, we could perform a cross-correlation approach for planetary targets that would be independent of the known planetary RV, either by correlating a H + 3 line model with our combined spectra, or by correlating our spectra at either end of a target's observation sequence to try to catch the slight change in RV of the planet over that time period. We do not believe, however, that either of these options would result in an increased sensitivity. The primary reason is that we only expect a few lines to have intensities that are greater than our spectral noise, and cross-correlation techniques perform best when many lines are available. Also, models of H + 3 line intensities from the ionospheric environment are dependent on non-LTE considerations. If our line models do not account for this correctly it could lead to erroneous results. As for measuring the change in the planetary RV, the maximum change in the RV of any planet over the observation period is on the order of ∼ 1 km/s, or ∼ 0.5Å, which is at the edge of our resolution capability. Finally, by correlating spectra at different points in the observation we necessarily reduce the signal-to-noise of the individual spectra, which makes it less likely that the H + 3 emission will rise above the stellar and thermal background and successfully be correlated.
Spectra for the targets 2MASSI J1835379+325954 and GJ 876 are shown in Figures 2 and 3, with wavelength ranges set to expected line locations. The top panel shows the science and telluric model produced by molecfit before the telluric correction has been applied, and the bottom panel shows the spectrum after the telluric model has been divided out of the science spectrum. Note that while the brown dwarf spectra are not as high signal-to-noise as the stellar spectrum, the observed variability is partly due to closely spaced absorption lines from various species in the brown dwarf atmosphere, and is not purely noise. Spectra for all other targets are provided in appendix figures.
We do not detect any candidate H + 3 emission in any of our brown dwarf or planetary targets. While some targets have a local maxima near one or two line positions, these are all low significance. Specifically, while 2MASSI J1835379+325954 has several low significance maxima at predicted line locations, it does not have any peak at the Q(1,0) line, which is expected to be the second strongest observed line in our wavelength range at a wide range of excitation temperatures.
To calculate emission limits for each target, we follow a similar procedure as in previous searches by Shkolnik et al. (2006) and Lenz et al. (2016). Our limits are calculated individually for the Q(1,0) and Q(3,0) lines, since these are expected to be the most intense. First, we calculate the standard error of the normalized mean science spectrum in each wavelength bin. Next, if the predicted line location is precisely known (as in the case for brown dwarfs and planets with known inclinations) we calculate both the median of the standard error and the standard deviation of the spectrum within the expected FWHM of an H + 3 line, approximately 4 pixels. To ensure our sensitivity estimate is conservative, we use whichever value is larger as uncertainty in the spectrum at that location. In almost all cases, the standard deviation of the spectrum in the binned wavelength region is higher than our estimate of the standard error from individual exposures within the same region. If the line locations do not have a precise prediction, we calculate the same metrics within the entire region where the line may appear. The theoretical line width of an H + 3 line is calculated from the instrument profile (an output of molecfit) and doppler broadening. Despite the fact that all of our brown dwarf targets are likely rapid rota- tors (P < 3 hrs), H + 3 emission from auroral generation is expected to be concentrated towards the poles (as in Jupiter, see latitude profile in Drossart 2019) such that at our resolution, the line width is still dominated by the instrument profile. We assume that our synthetic telluric spectrum is noiseless, which is one of the major benefits of using a synthetic model. Next, we estimate the L magnitude for each target (if no L photometric measurements are available) using K-L colors derived from (Bessell & Brett 1988) for main-sequence stars or by using colors from brown dwarfs with similar spectral types to our targets published in Golimowski et al. (2004). This estimated L magnitude, and its associated uncertainty, are then converted to a flux density in the region of our H + 3 lines, and then to a luminosity density given the target parallax. We scale our spectral uncertainty to the luminosity within the line profile and assume that we would follow-up any peak greater than 3σ significance to get our average line sensitivity for the target.
Our brown dwarf sensitivities are plotted in Figure 4. Planetary sensitivites are shown in Figure 5 against sensitivities from previous observational attempts and limits from theoretical studies. Brown dwarf sensitivities are roughly an order of magnitude or more stringent than any upper limits previously set for giant exoplanets, which is primarily a result of their low luminosity. We achieve the highest sensitivity for any planetary target so far for GJ 876 c, with a sensitivity of 2.2 × 10 17 W for the Q(1,0) line. All calculated sensitivities are compiled in Tables 2 and 1.
DISCUSSION
In this section, we discuss the physical interpretation of our non-detection for the properties of brown dwarf and giant planet atmospheres and magnetospheres. We will examine brown dwarfs and giant planets separately, beginning with brown dwarfs.
Auroral Mechanisms in Brown Dwarf Atmospheres
Our upper limits for H + 3 emission from brown dwarfs are the lowest limits for any planetary mass objects outside of the solar system by more than an order of magnitude. This reduction in upper limits is possible due to the lower intrinsic luminosity of brown dwarfs compared to bright host stars, and by the relative abundance of nearby brown dwarfs to our own solar system compared to suitable exoplanetary targets. Importantly, we have corroborated the results of Pineda (2017) that no H + 3 is detected in the atmospheres of free-floating and binary brown dwarfs that are known to possess strong magnetic fields and probable aurorae. Our observations are complementary in that we have observed the wavelength region of the fundamental H + 3 ro-vibrational spectrum at high-resolution in the L band, whereas their study observed the wavelength region of the overtone spectrum at medium-resolution in K band. As Pineda (2017) notes, around 90% of the H + 3 energy in the Jovian atmosphere is emitted in L band lines while the total K band H + 3 emission in all lines is thought to be at similar luminosities to auroral Hα emission. Hα emission has been detected and posited to have an auroral origin for 2MASSI J1835379+325954, which was observed in their study and in ours. Figure 4 shows that the previously observed Hα luminosity is well above our upper limits for H + 3 , however, individual H + 3 line luminosity's in L band may fall below our upper limits by a factor of several depending on excitation temperatures. Nevertheless, the fact that we have not detected H + 3 emission in this target, if the detected Hα is indeed auroral, suggests that the auroral processes that generate H + 3 on Jupiter may not be completely analo-gous to the process that occurs on brown dwarfs. The idea that the physical process is likely to be different is consistent with other observational work by Saur et al. (2018), who found that the UV spectrum of 2MASSI J1835379+325954 is dissimilar to the UV spectrum from Jupiter's aurora and is at least two orders of magnitude fainter than might be expected from a Jupiter-like auroral mechanism. One of the most likely explanations posited by both Helling & Rimmer (2019) and Pineda (2017) is that the electron beams on brown dwarfs have a higher energy distribution than those on Jupiter. Simplistically, one might assume that stronger auorae are due to a greater number of electrons rather than electrons with a different energy distribution. This assumption is justified to an extent because if electron energies are shifted high enough, then the electrons will become relativistic and emit synchrotron radiation rather than ECM, yet strong ECM radiation is still observed from these objects. However, if the electrons do have moderately higher energies (but are not relativistic) they would ionize H 2 at a greater depth in the brown dwarf atmosphere where H + 3 is quickly destroyed by neutral hydrocarbons or H 2 O before it has time to emit. We estimate that our upper lim- Blue lines represent sensitivity limits placed in this work for brown dwarfs. The red/black line is the observed luminosity of optical Balmer emission for the brown dwarf 2MASSI J183537+32594 (Hallinan et al. 2015) for which total H + 3 emission is thought to have similar or higher luminosity depending on auroral mechanisms (Bhardwaj & Gladstone 2000). The green arrow is our estimated Q(3,0) emission luminosity from 2MASSI J1835379+325954 based on its relative cyclotron maser emission intensity compared to Jupiter and assuming the same auroral electron energy distributions as Jupiter. We estimate that JWST/NIRSPEC can reach this 10 16 W sensitivity limit to lines around 4 microns for a 2000 K brown dwarf with ∼ 1 hour of exposure time.
its lack an order of magnitude to sufficiently probe this hypothesis for our targets. If the total H + 3 luminosity of Jupiter (Lam et al. 1997) is multiplied by the relative ECM emission of 2MASSI J1835379+325954 compared to Jupiter (∼ 10 5 ×), and the rotational and vibrational temperatures of H + 3 are assumed to be 2800 K, then the Q(3,0) emission luminosity expected from 2MASSI J1835379+325954 is around 10 16 W. The same calculation for our other brown dwarf targets yield similar gaps between our upper limits and the calculated Q(3,0) luminosity, as our targets with lower upper limits all have lower observed ECM radiation intensities. Therefore, an increased sensitivity by a factor of several is required to begin constraining electron distributions to higher energies than that of Jupiter. More than two orders of magnitude decrease in upper limits would be required to fully exclude this hypothesis as individual auroral electrons in a brown dwarf atmosphere could possess 10 2 × more energy than those of Jupiter before becoming relativistic (an increase from ∼ 5 keV to ∼ 500 keV, Gérard et al. 2009). Regardless, if electron energies do increase, Helling & Rimmer (2019) has suggested to instead ob-serve reaction productions of H + 3 , such as H 3 O + . At present, there is no calculated H 3 O + line list appropriate for an observational attempt. An additional prediction of higher energy electrons is that it will cause more auroral energy to be radiated in the radio, but less in UV and infrared, which could be consistent with surveys of brown dwarfs with radio emission (Gustin et al. 2013).
Another likely explanation, which is at least of some effect, is that the detection of H + 3 is hampered by the higher thermospheric temperatures of brown dwarfs. Individual line strengths and total luminosity do not scale linearly with temperature, as higher temperatures allow more excitation modes. The thermospheres of our observed brown dwarfs are certainly hotter than that of Jupiter, at least up to the effective temperature of ∼2800 K in the case of 2MASSI J1835379+325954 (Berdyugina et al. 2017). This increase in temperature means that although the targeted Q(1,0) and Q(3,0) lines will still be among the most intense, it is no longer valid to assume that those lines will carry most of the H + 3 emission luminosity. Instead it will be carried by many lower inten- Blue/grey dots represent sensitivity limits placed in this work for extrasolar giant planets. Square points represent sensitivity limits placed by other works. X points represent theoretical predictions for extrasolar giant planets. Note that one observed system can contain multiple giant planets. The green line represents the predicted sensitivity by Lenz et al. (2016) of the E-ELT in 6.5 hours of exposure time using an instrument similar to CRIRES (currently on the VLT) to observe the GJ 86 system. Recent models (Chadney et al. 2016) predict that the grey shaded region is the distance from a host star where a planet will reach a maximum H + 3 luminosity for a Sun-like or smaller stellar host.
sity lines due to the increase in available excited states. Other explanations are that non-LTE effects are reducing emission intensities or simply that detection requires observing variable aurorae when they are at maximum intensity. Non-LTE effects are certainly important to consider when modelling H + 3 , however, most models of H + 3 in giant planets have accounted for it and have not shown significant drops in predicted H + 3 luminosity because of it (Koskinen et al. 2007;Chadney et al. 2016). While the fact that we have observed multiple targets increases the chance of observing an aurora near a peak of activity, future searches for H + 3 could attempt to simultaneously measure auroral activity by other observables (e.g. ECM radiation) to place H + 3 limits in the context of the activity at the moment of observation.
In future studies, targeting white dwarf -brown dwarf binaries will be an interesting avenue to pursue in addition to free-floating brown dwarfs. Our search for H + 3 from the ground is primarily limited by Earth's atmosphere and background, and observing dim targets (L > 14 mag) past 2.5 microns is difficult with highresolution spectroscopy. Our limits for GD 1400 are closer to those of our planetary targets than of our other observed brown dwarfs. Very few WD-BD binaries are known nearby to Earth (GD 1400 being the closest to our knowledge), and realistically it is not possible to target them efficiently with a high-resolution spectroscopic search. WD-BD binaries should be revisited with futurespace based observation or with thirty meter telescopes.
Extrasolar Giant Planets
Our sensitivity limits for extrasolar giant planets between 0.1 to 0.2 au from their host stars largely lands us in the same regime as previous observations by Shkolnik et al. (2006), Laughlin et al. (2008), and Lenz et al. (2016). We have achieved the lowest upper limits so far for any extrasolar giant planets with our target GJ 876 by a factor of a few. The value of our study therefore comes primarily from probing a few new systems farther from their stars than have previously been explored. These planets may have cooler thermospheres that allow H + 3 to form. They are also less likely to be tidally locked and therefore may have stronger magnetic fields that could power observable H + 3 emission (although un-likely to be at kilogauss levels). Our observations principally rule out strong auroral processes on these planets that could drive H + 3 production, or if strong aurorae do exist, they are not driving an H + 3 emission increase that might naïvely by expected, possibly for the same reasons as have been discussed for brown dwarfs.
Prospects for Future H + 3 Detection in Brown Dwarfs
Comparing our upper limits for H + 3 emission in brown dwarfs to those of giant exoplanets shows the challenge in dealing with photon noise from a host star. Because of this challenge, future detection has the highest likelihood in brown dwarfs rather than giant exoplanets. Our observations likely set the order-of-magnitude limit achievable with current ground-based instruments for brown dwarfs, at least within one night of observation. Focusing only on one brown dwarf, it is reasonable that luminosity limits of 10 16 W could be achieved with a single night of observations by Keck/NIRSPEC or a similar instrument. At this limit, it may be possible to place lower limits on the energy distribution of precipitating electrons in the brown dwarf atmosphere. Lenz et al. (2016) previously investigated upper limits that might be achievable for giant planets with the planned E-ELT. Their predicted upper limit of 10 16 W in 6.5 hours of exposure time on the exoplanet system GJ 86 with an instrument like CRIRES on the VLT is shown as part of Figure 5. By simple extrapolation, it seems plausible that the E-ELT could reach limits of around 10 15 W in the same amount of time if observing a brown dwarf rather than GJ 86.
In the more near future, it will be possible to observe brown dwarfs in the near-infrared from space using the recently launched JWST. The greatly reduced thermal background is a significant benefit for potential searches for H + 3 in L and M bands. The drawback is that JWST/NIRSPEC is medium resolution (R∼2,700), compared to the high-resolution capability of groundbased instruments. This means more blackbody radiation is collected per spectral bin without more H + 3 emission, except for closely spaced lines that are relatively low intensity.
To estimate the ability of JWST/NIRSPEC to detect H + 3 emission from brown dwarfs, we reverse our steps from Section 4, going from an emission luminosity to a required signal-to-noise for detection, then use the online JWST Exposure Time Calculator (ETC) (Pontoppidan et al. 2016) to find the time needed for JWST to reach it. We do these steps using the 2000 K low-temperature PHOENIX model (Phillips et al. 2020) provided as a default option within the ETC and determine the needed exposure for three high intensity lines at different wavelengths representative of the range of the fundamental band of H + 3 . These lines are R(3,-3) at 3533.6 nm, Q(3,0) at 3985.5 nm, and P(6,6) at 4874.4 nm. We find that to reach a limit of 10 16 W requires a total exposure time of around 1 hour for the R(3,-3) and Q(3,0) lines, however, only ∼ 10 minutes is required to reach that same limit for the P(6,6) line. All 3 of these wavelength regions can be explored simultaneously with the F290LP filter. Thus, despite lower spectral resolution than ground-based spectrographs, JWST offers significant opportunity to search for H + 3 emission in brown dwarfs and constrain their auroral processes and subsequent atmospheric effects.
CONCLUSIONS
In this study, we have presented a search for H + 3 emission from brown dwarfs with evidence of aurorae and select systems hosting giant planets at semi-major axes of 0.1 to 0.2 au. The key results are as follows.
1) We do not detect any H + 3 emission, but set the first upper limits for H + 3 emission from free-floating brown dwarfs, which, to our knowledge, have never been searched with high-resolution spectroscopy.
2) Our upper limits for H + 3 emission from brown dwarfs range between 2.7 and 9.3 × 10 16 W, within the upper range of a plausible detection based on models.
3) Our non-detection in brown dwarfs suggests that the aurora-like processes occurring in these objects are probably not analagous to those of Jupiter, namely that electron beams are not able to generate a H + 3 density proportional to the energy of these phenomena. An order-of-magnitude increase in sensitivity is needed to probe this hypothesis further. 4) We set the lowest upper limit for H + 3 emission in a giant exoplanet yet, with a limit of 2.2 × 10 17 W. Despite these limits, we are not able to place additional constraints on giant exoplanet atmospheres or magnetospheres. 5) We suggest that brown dwarfs are the best targets for future H + 3 detection attempts, and show that JWST will be able to reach emission limits around an orderof-magnitude deeper than current ground-based instruments with equal exposure time.
nership among the California Institute of Technology, the University of California and the National Aeronautics and Space Administration. The Observatory was made possible by the generous financial support of the W. M. Keck Foundation.
The authors wish to recognize and acknowledge the very significant cultural role and reverence that the summit of Maunakea has always had within the indigenous Hawaiian community. We are most fortunate to have the opportunity to conduct observations from this mountain. The top portion of each panel shows the continuum normalized science and telluric model produced by molecfit together, while the bottom portion shows only the science spectrum after the telluric model has been divided out. The inclination of HD 192263 b is not known, so the line positions show a range from the minimum to maximum possible planetary RV. Noise from the telluric spectrum is obviously a limiting factor. In future work, the telluric star will be replaced by a synthetic spectrum. The upper limits for H + 3 emission for HD 192263 from the Q(1,0) and Q(3,0) lines are 2.3 × 10 18 W and 3.2 × 10 18 W respectively. Figure 11. Spectrum of υ Andromedae. This star hosts at least three giant planets, υ Andromedae b, c, and d. Each panel is zoomed into the expected position of one or more H + 3 emission lines. The top portion of each panel shows the continuum normalized science and telluric model produced by molecfit together, while the bottom portion shows only the science spectrum after the telluric model has been divided out. The expected H + 3 line positions are shown separately for each planet. υ Andromedae d is not plotted as it is not considered a good candidate for H + 3 detection. The upper limits for H + 3 emission for υ Andromedae b from the Q(1,0) and Q(3,0) lines are 2.9 × 10 18 W and 8.7 × 10 18 W respectively. The corresponding upper limits for υ Andromedae c are 2.5 × 10 18 W and 1.0 × 10 19 W.
|
2022-07-22T09:21:24.050Z
|
2022-06-07T00:00:00.000
|
{
"year": 2022,
"sha1": "2ac0c6b415ecbea691a16ab043a0c1a2f0cd4f56",
"oa_license": "CCBY",
"oa_url": "https://doi.org/10.3847/1538-3881/ac7718",
"oa_status": "GOLD",
"pdf_src": "Arxiv",
"pdf_hash": "b674a08c75eee9f69ec84bfd97021d8b47394393",
"s2fieldsofstudy": [
"Physics",
"Environmental Science"
],
"extfieldsofstudy": [
"Physics"
]
}
|
14984695
|
pes2o/s2orc
|
v3-fos-license
|
Central effects of clozapine in regulating micturition in anesthetized rats
Background We previously showed that systemic administration of the atypical neuroleptic clozapine in the rat altered a number of urodynamic variables and inhibited the external urethral sphincter. Since clozapine acts at several receptor types both at the periphery and the central nervous system, the site of action remained uncertain. Therefore, the purpose of this study was to determine the effects of central administration of clozapine on the bladder and the external urethral sphincter during cystometry and to examine differences in spinal versus supraspinal administration. We extended our observations by delivering clozapine centrally in anesthetized rats instrumented with either an intrathecal (L6-S1 spinal segment) or an intracerebroventricular (lateral ventricle) catheter. Results Clozapine decreased micturition volume and increased residual volume possibly by acting at a supraspinal site. Expulsion time and amplitude of the high frequency oscillations were reduced by clozapine possibly by acting at a spinal site. Bladder capacity was increased after central clozapine but probably due to a peripheral effect. Clozapine acting at spinal and supraspinal sites increased pressure threshold. Contraction time and peak pressure were not affected by clozapine. The EMG from the external urethral sphincter was also reduced following clozapine centrally and suggests a spinal and a supraspinal site of action. Conclusions The results from the present study suggest that spinal and supraspinal central sites mediate clozapine's action in inhibiting expulsion parameters and the external urethral sphincter of the rat. Therefore, the reduction in the voiding efficiency observed after clozapine appears to be mediated by spinal and supraspinal sites.
Background
Clozapine is a widely used atypical neuroleptic with affin-ity for multiple receptors, including dopamine, serotonin, alpha adrenergic, muscarinic and histaminergic receptors [1,2]. Although clozapine is effective in the treatment of schizophrenia refractory to traditional antipsychotic medication, it also has a number of significant side effects ranging from the potentially fatal but rare agranulocytosis to weight gain, constipation, seizures and urinary incontinence [3]. Urinary problems have been reported in the clinical literature with incontinence present in up to 44% of patients [4] and enuresis in 27% of patients [5]. Since clozapine has potent anti-muscarinic and anti-alpha adrenergic effects [2], it has been proposed that peripheral effects on the lower urinary tract might be responsible for the micturition disturbances [6]. Incontinence has also been reported following therapy with other atypical neuroleptics such as risperidone [7][8][9] and recently, olanzapine [10].
We have previously shown that intravenous administration of clozapine to anesthetized rats [11,12] altered several micturition parameters including a decreased amount of volume voided per micturition and a concomitant increase in the residual volume. In addition, clozapine also inhibits the external urethral sphincter. Other, newer, atypical neuroleptics (e.g. olanzapine and risperidone; [13]) were also found to have similar effects but with differing potency compared to clozapine. Since risperidone has little or no anti-muscarinic activity, a primary antimuscarinic peripheral effect may be ruled out as the cause of micturition disturbances following risperidone administration, and suggest a possible central role for atypical neuroleptics in regulating micturition.
The purpose of the present study was to determine the effects of clozapine administered centrally at two different sites, spinal (L6-S1 spinal segments) vs. supraspinal (lateral ventricle), on micturition and the external urethral sphincter during cystometry in the anesthetized rat. By limiting the application site to a specific area of the central nervous system and comparing the results with our previous findings after systemic administration [12] it might be possible to determine whether the effects of clozapine are mediated at a central or peripheral level. Moreover, a comparison of the two central routes might indicate whether the central effect involves spinal or supraspinal structures. We report in this study that most of the effects of clozapine on micturition are mediated by its central effects. Furthermore, there are differences in spinal versus supraspinal effects. (Table 1I) Figures 1 and 2 show representative examples of the effects of clozapine on single cystometry in anesthetized rats after intrathecal (i.t.; L6-S1 spinal segment) and intracerebroventricular (i.c.v.; lateral ventricle) administration, respectively.
Effects of central administration of clozapine on urodynamic parameters during single cystometry
Bladder capacity was increased by clozapine given i.t. or i.c.v. only at the highest dose tested (Fig 1G; 2G). After 50 nmoles of clozapine i.t. the bladder capacity was 0.47 ± 0.09 ml, compared to 0.29 ± 0.027 ml after vehicle administration ( Fig 3A). Similarly, after 50 nmoles of clozapine i.c.v the bladder capacity was 0.51 ± 0.11 ml, compared to 0.25 + 0.027 ml after vehicle administration ( Fig 3A). Micturition volume was significantly decreased after 5.0 and 50 nmoles of clozapine i.t. (Fig 3B) to 0.11 ± 0.031 and 0.08 ± 0.019 ml, respectively compared to 0.18 ± 0.022 ml after vehicle. In the case of intracerebroventricular administration, 0.5 nmoles of clozapine decreased the micturition volume to 0.11 ± 0.02 ml compared to 0.16 ± 0.018 ml after vehicle ( Fig 3B). The effect of 0.5 nmoles of clozapine i.c.v. on micturition volume was significantly different from the effect of the same dose given i.t. Further reductions in micturition volume were observed after 50 nmoles of clozapine i.c.v.
Residual volume was significantly increased after 5.0 and 50 nmoles of clozapine i.t. to 69 ± 6.2% and 79 ± 4.3%, compared to 34 ± 7.5% after vehicle administration ( Fig 3C). Clozapine i.c.v. resulted in a significant increase in the residual volume at all doses, including 0.5 nmoles. Clozapine also decreased the amplitude of (and in some animals even abolished) the high frequency oscillations (HFO) after i.t. administration (Fig 1; 5C). After 5.0 nmoles of clozapine the HFO were 0.4 ± 0.23 compared to 1.5 ± 0.1 mm Hg after administration of vehicle. In 4 of the 6 animals in this group, 5 nmoles of clozapine i.t. abolished the HFO. The reduction in amplitude observed at this dose after i.t. administration of significantly different from the effects observed after the same dose i.c.v. ( Fig 5C). A greater dose of clozapine did not cause a larger effect although the amplitude of the HFO after 50 nmoles of clozapine i.t. was still significantly reduced (0.5 ± 0.19 mm Hg) when compared to vehicle administration. Clozapine i.c.v decreased the amplitude of the HFO only at 50 nmoles (0.6 ± 0.2 mm Hg compared to 1.6 ± 0.11 mm Hg after vehicle administration; Fig 2), and the HFO were abolished in only 1/6 of the animals in this group.
Figure 1
Representative traces of the effects of increasing dose of clozapine administered intrathecally (over the L6/S1 spinal segments) on the cystometrogram and external urethral sphincter of anesthetized rats (n = 6). Top panel shows bladder pressure during filling (0.11 ml/min) while bottom panel shows the integrated EMG recorded from the external urethral sphincter. Panels A and B also show some of the urodynamic parameters measured during cystometry: BC = amount of fluid infused into the bladder to elicit a contraction; PT = pressure at which contraction begins; PP = peak pressure during contraction; CT = contraction time; HFO = high frequency oscillations recorded during expulsion; ET = time between peak pressure and end of high frequency oscillations. The EMG activity was examined by dividing the contraction into 3 phases (Table 1I) Clozapine i.t. did not have an effect on the activity of the EUS during Phase 1 (the rising phase during a contraction), and only the highest dose of clozapine i.c.v (50 nmoles) was found to decrease the activity of the EUS ( Fig 6A) to 62 ± 14.9% of that seen during administration of vehicle.
Effects of central administration of clozapine on the activity of the external urethral sphincter during single cystometry
Phase 2 of the EUS EMG (occurring during the time of HFO) was decreased by clozapine either i.t or i.c.v. 5 and 50 nmoles of clozapine i.t. decreased the EMG to 37 ± 24 and 40 ± 14% of that observed during administration of vehicle, respectively ( Fig 6B). In fact, in 4/6 animals, 5.0 nmoles of clozapine i.t. abolished the bursting pattern of the EMG observed during phase 2 ( Fig 1F). Clozapine administered i.c.v at a dose of 0.5 nmoles decreased the EUS EMG to 68 ± 13.9% of the activity observed after vehicle administration ( Fig 6B). Larger doses (5 and 50 nmoles) of clozapine i.c.v further reduced the EUS activity during this phase to 54 ± 8.4% and 26 ± 9.9%. Clozapine i.c.v, however, was not observed to abolish the bursting pattern during this phase (Fig 2), except in 1/6 animals.
Phase 3 of the EUS EMG (recorded during the falling phase of a bladder contraction) was not affected by clozapine administered intrathecally ( Fig 6C). However, the largest dose of clozapine given i.c.v (50 nmoles) decreased the activity to 50 ± 14.9% of the activity observed during vehicle administration.
Finally, the amplitude of the bursts observed during phase 2 in the EUS EMG were decreased to 32 ± 20.5% by 5.0 nmoles of clozapine i.t. (Fig 6D). A similar dose of clozapine i.c.v had no effect. However, 50 nmoles of clozapine i.c.v. decreased the amplitude of the bursts to 65 ± 17.2% of that observed during vehicle administration.
Discussion
In the present experiments, central application of clozapine (intrathecally over the L6/S1 spinal segments or intracerebroventricularly into the lateral ventricle) resulted in a number of changes in the urodynamic parameters of anesthetized rats. The major effect of clozapine was to decrease the voiding efficiency of the bladder by inhibiting expulsion parameters, such as micturition volume, residual volume and expulsion time. In addition, the activity of the EUS also decreased upon central application of clozapine. One problem in delivering substances centrally is the possibility that peripheral spread may confound the results. Since clozapine crosses the blood-brain barrier readily [14], the maximum dose selected in the present experiment was restricted to the first appearance of a decrease in blood pressure as an indication of possible
Figure 3
Effects of increasing doses of clozapine administered intrathecally (i.t.) or intracerebroventricuarly (i.c.v.) on bladder capacity, micturition volume and residual volume. For comparison purposes, the effects of intravenous admnistration of clozapine (data from [12]) are also shown. We compared the effects obtained in the present study against those observed previously after intravenous administration [12] expecting that central effects would require significantly less application of clozapine than systemic administration. We considered a minimum difference of 10× magnitude in the central vs. peripheral dose that elicited a significant effect as an indication of possible central action. Our doses of 0.5, 5 and 50 nmoles of clozapine correspond to serum levels of 8, 80 and 800 ng/ml (assuming a blood volume of 8% of body weight) and are in the range of therapeutic levels (260-387 ng/ml [5,15]).
A second problem in administering a substance at a particular central location is that of redistribution to other parts of the central nervous system. Since application of a substance at the lumbosacral spinal level may travel to the brain after some time and vice versa, we established two criteria to help determine a possible central site of action: 1) the first dose to elicit a significant effect by either i.t. or i.c.v administration; 2) significant differences at the same dose but different routes (i.t. vs i.c.v). [12]. Theses changes are plotted at the dose (mg/kg) that first yielded significant results. The route that we consider most likely to be the site of action for clozapine is italicized.
Bladder capacity was increased to a similar degree by an equivalent dose of clozapine, regardless of route (Fig 7A; 3A), making a determination of most likely site of action difficult. Given that the peripheral effects at low doses were similar to the effects after the highest central dose of clozapine (by either route) it appears likely that this effect is due to peripheral actions of clozapine. Bladder capacity was reported to increase following i.c.v. administration of muscarinic antagonist (atropine [16]) or i.t. alpha2 antagonists (yohimbine [17]; atipemazole [18]). Given clozapine's strong affinity for muscarinic and alpha2 receptors [2] it is possible that central effects are also contributing to the increase in bladder capacity.
Micturition volume, on the other hand, showed a reduction after smaller doses of clozapine i.c.v or i.t. when compared to the i.v. dose (Fig 7B;3B). Therefore, central effects of clozapine in controlling micturition volume are likely. Furthermore, when comparing the effects of i.t. vs. i.c.v administration, a 5 nmoles dose of clozapine i.c.v. elicited a reduction in micturition volume that was significantly greater from the effects observed after the same dose i.t. (Fig. 3B). Therefore, supraspinal effects of clozapine in regulating micturition volume are likely with spinal effects perhaps contributing. Clozapine i.t. also decreased the micturition volume, however the dose that first showed a significant effect was higher than the i.c.v. but smaller than the i.v. dose. Spinal antagonism of either alpha1 or alpha2 adrenergic receptors was reported to increase micturition volume [17] in anesthetized rats. However, intrathecal atipamezole (alpha2 antagonist) increased residual volume in awake rats [18]. Therefore, it is difficult to interpret clozapine's effects on micturition volume in terms of spinal alpha adrenoceptor antagonism and possibly the supraspinal effects predominate.
Clozapine given i.t. or i.c.v. also increased the residual volume (Fig 7C; 3C). Since the effects were observed at doses that were lower than those after i.v. administration, a peripheral effect of clozapine to increase residual volume appears unlikely. The smallest dose of clozapine given i.c.v. (0.5 nmoles) increased residual volume to 160% (Fig 7C), suggesting that supraspinal effects predominate with possible contributing effects from spinal sites. Ishiura et al. [16] reported a decrease in voiding efficiency (comparable to an increase in residual volume) following atropine i.c.v. or i.t. suggesting that muscarinic receptors at supraspinal and spinal sites are involved and may account for our effects after clozapine i.t. or i.c.v. A decrease in residual volume has been reported following spinal antagonism of alpha2 receptors with yohimbine [17] but atipamezole produced an increase in the residual volume [18], similar to our findings with clozapine i.t.
Clozapine increased pressure threshold after i.c.v. or i.t. administration only after 50 nmoles (Fig 7D; 4A). It should be noted that both central doses were at the range observed to result in cardiovascular changes and therefore the possibility of peripheral leakage of clozapine must be considered. Still the effective central doses are approximately 14× less than the first dose observed to produce significant results intravenously, and suggests a possible central site of action.
Peak pressure was not changed by clozapine after either route in the present study after central administration, consistent with our findings afer intravenous administration [12]. Ishiura et al. [16] reported a decrease in maximal voiding pressure (equivalent to our peak pressure) after atropine i.v., i.c.v or i.t. in awake rats undergoing continuous cystometry. Given that clozapine has a relatively high affinity for muscarinic receptors [1,2] it is surprising that we have not seen an effect on peak pressure. Clozapine, and also olanzapine [11,13] were able to decrease the contraction amplitude after electrical stimulation of the pelvic nerve. Since the contraction pressures observed during cystometry were smaller than those observed after electrical stimulation of the pelvic nerve (and against a closed urethra) it is possible that the antimuscarinic effects of clozapine on bladder contraction pressure are not detected because during cystometry maximal bladder pressures are not necessary for emptying.
Contraction time was not affected by clozapine i.c.v or i.t. (Table 1 and 2; Fig 5A) however it was clearly decreased after intravenous administration [12]. Therefore, we sus- pect that solely peripheral effects may explain the effects of clozapine on contraction time. Antimuscarinic agents have been show to decrease contraction time (e.g. atropine [19]) and since clozapine has high affinity for muscarinic receptors it is likely that peripheral anti-muscarinic effects of clozapine are responsible for the reduction in contraction time after systemic administration.
Expulsion time was reduced by clozapine after i.t. or i.c.v. administration at much smaller doses than were observed to produce similar results i.v. (Fig 7E; 5B). Therefore, cen-tral effects of clozapine are likely responsible for the reduction in expulsion time. Since the intrathecal effects were greater than those observed after i.c.v. administration it is possible that a spinal action predominates with possible contribution from supraspinal sites.
The amplitude of the high frequency oscillations observed during the expulsion time in the rat micturition cycle [20,21] was reduced by clozapine i.t. at a dose of 5 nmoles (Fig 8F). This dose was 10× smaller than the first i.c.v. dose observed to have a significant effect and approxi-
Figure 6
Effects of clozapine i.t. or i.c.v. on the EMG recorded from the EUS during cystometry in anesthetized rats. Data from i.v. administration of clozapine [12] are presented for comparison. A) Clozapine i.t. had no effect on the integrated EMG recorded during phase 1 of the contraction. Only the highest dose of clozapine (50 nmoles) i.c.v. decreased the EMG during this phase to 62%; B) Clozapine i.t. or i.c.v. decreased the integrated EMG during phase 2 of the contraction. All doses of clozapine i.c.v. significantly reduced the EMG whereas only 5 and 50 nmoles i.t. resulted in significant reductions. C) Only the highest dose of clozapine i.c.v (50 nmoles) decreased the EMG during this phase of the contraction. Clozapine i.t. had no effects. D) The amplitude of the individual bursts of EMG occurring during phase 2 of the contraction were reduced by clozapine i.t and i.c.v. However, 5 nmoles of clozapine i.t. produced a reduction of EMG that was significantly different from the effects of the same dose given i.c.v. Only the highest dose (50 nmoles) was effective in reducing the EMG when administered i.c.v. V= vehicle administration (saline) for i.t. or i.c.v. groups; (V) = vehicle administration for i.v. group. * = p < 0.05; ** = p < 0.01; # = p < 0.05 when compared to the effects of the same dose, different route. Summary of the effects of clozapine given i.t., i.c.v. or i.v. (data from [12]) on urodynamic parameters and the activity of the EUS during cystometry in anesthetized rats. The first dose to yield a significant effect has been plotted for all 3 routes. Doses have been converted to mg/kg for ease of comparison. Peak pressure data are not shown since no significant differences were found with clozapine after any route. A) Bladder capacity is increased to a similar extent by comparable doses of clozapine, regardless of the route. Therefore, a peripheral effect may be predominating. In the case of micturition volume (B) and residual volume (C), both clearly show a central effect since the first dose to yield a significant effect is 142× smaller than the i.v. dose. Both of these parameters show a possible supraspinal site of action. D) Pressure threshold increased only after the highest dose of clozapine i.c.v or i.t., however, that dose was still 14× smaller than the first dose i.v., suggesting a possible central effect. E) Expulsion time shows a 142× difference between the central dose and the peripheral dose. In addition, intrathecal administration shows a larger effect than after i.c.v., suggesting a possible spinal site of action. F) The effect of clozapine on the amplitude of the HFO appears to be mediated by a spinal site. G) Phase 1 of the EMG was decreased after equivalent doses of clozapine i.v. or i.c.v. Intrathecal administration had no effect. H) Phase 2 of the EMG (where the bursting occurs) clearly showed a central effect since a much smaller dose (142×) was required to produce an effect. In addition, the reduction after i.c.v. was greater than after i.t. therefore suggesting a possible supraspinal site of action. I) Phase 3 of the EMG was not affected by clozapine intrathecally. Only the highest dose of clozapine given i.c.v decreased the EMG during this phase. That dose was 14× smaller than the first dose to produce a significant effect after i.v. administration suggesting a possible supraspinal site of action. J) The amplitude of the individual bursts during phase 2 of the EMG was decreased by central effects of clozapine at smaller doses than those observed i.v. Also, the intrathecal administration was much more effective in reducing the amplitude (abolished in 4/6) animals, suggesting a possible spinal site of action.
I.T.
I.V. J mately 142× smaller than the first i.v. dose to show a significant effect. Therefore, a spinal site mediating the effects of clozapine on the amplitude of high frequency oscillations appears likely. Previous results [12] suggest that D2 receptors modulate the amplitude of the HFO, since raclopride (selective D2 antagonist) decreased the amplitude of the HFO by 30%.
In addition to clozapine's effects on urodynamic parameters, the EMG recorded from the external urethral sphincter also showed changes after clozapine i.t. or i.c.v. The EMG during phase 1 of the contraction only showed a reduction after the largest dose of clozapine i.c.v.. This dose is equivalent to our lowest dose i.v. and in fact all doses i.v. previously showed a significant reduction [11]. Intrathecal administration had no effect on the EMG at phase 1.
The EMG during phase 2, corresponding to the period of high frequency oscillations was decreased by clozapine after i.t. or i.c.v administration at doses that were much lower than those necessary to produce an effect after i.v. Therefore, the effects of clozapine in reducing the EMG during this phase appear to be central in origin. Since the first significant effect was obtained after i.c.v. administration (Fig 7H) it is possible that supraspinal effects predominate with spinal sites contributing.
The EMG during phase 3 (closing phase) of the contraction was decreased only after the highest dose of clozapine (50 nmoles) i.c.v. (Fig 6C). This dose was still 14× smaller than the first significant dose i.v. and therefore a central effect of clozapine is probable.
Finally, the amplitude of the individual bursts of EMG recorded from the external urethral sphincter during phase 2 of the contraction also decreased after clozapine i.t. or i.c.v. (Fig 6D). Since the doses that yielded significant reductions were smaller than those observed after i.v. admnistration, a central effect is likely. Moreover, since 5 nmoles of clozapine i.t. produced a significant reduction when compared to the same dose i.c.v., a spinal effect of clozapine in mediating the reduction of the burst amplitude of the EMG is possible.
In the anesthetized rat, the external urethral sphincter contracts and relaxes during the expulsion phase in a manner that is complimentary to the high frequency oscillations observed in the bladder pressure record [19] whereas in humans the external sphincter not active during voiding. Pharmacological blockade of the external urethral sphincter in the rat resulted in decreased micturition volume and increased residual volume [20][21][22][23] suggesting that the activity of the external urethral sphincter contributes to efficient voiding in the rat. Therefore, cen-tral administration of clozapine, by reducing the activity of the EUS, contributes to the decrease in voiding efficiency by reducing micturition volume and increasing residual volume. Alpha1 adrenergic antagonists have been shown to inhibit pudendal reflexes in anesthetized cats [24][25][26]. However, a systemic dose of prazosin (alpha1 antagonist) did not inhibit the EUS EMG activity during high frequency oscillations in the rat [27]. Thus, although alpha1 antagonism has been shown to modulate pudendal reflexes in the cat, their role in modulating the activity of the EUS during micturition in the rat appears unclear.
In summary, our results in the present experiments suggest that most of the effects of clozapine on urodynamic parameters can be ascribed to central effects. Expulsion parameters, such as micturition volume, residual volume, expulsion time, and amplitude of the high-frequency oscillations, appear to be mediated by the central action (spinal or supraspinal) of clozapine. The activity of the EUS also appears to decrease after central application of clozapine. Therefore, central effects of clozapine appear to decrease the voiding efficiency of the bladder in the rat. Contraction time clearly showed a peripheral effect only, whereas changes in bladder capacity were difficult to explain from central effects and probably reflect peripheral effects of clozapine.
Clozapine is metabolized mainly at the liver resulting in several metabolites [5,14]. One of the major metabolites, N-desmethylclozapine has been shown to have pharmacological activity both in vitro [28] and in vivo in rats [29]. In addition, N-desmethylclozapine is found in large concentrations in the serum of schizophrenic patients [5,15] and in rats [30]. The contribution of N-desmethylclozapine to clozapine's central effects has been questioned recently, since the levels of N-desmethylclozapine in the brain were much lower than those of clozapine [30] suggesting that N-desmethylclozapine does not cross the blood-brain barrier as readily as clozapine. Since we observed effects from central application of clozapine, we consider it unlikely that the effects of metabolites contributed significantly. However, whether any of the major clozapine metabolites also have a role in regulating micturition remains to be determined.
Conclusions
Atypical neuroleptics are useful in treating patients that are refractory to "traditional" antipsychotic medication and produce fewer extrapyramidal side effects. However, other side effects still occur with varying severity and frequency [31] and continue to pose a challenge to effective treatment. Urinary disturbances as a result of clozapine therapy have been well documented, and include inconti-nence, enuresis and urgency [4,5,32]. Other atypical antipsychotics, such as risperidone [7] and olanzapine [10] have been reported to produce urinary incontinence.
We have previously shown [11][12][13] that systemic administration of clozapine, olanzapine and risperidone to anesthetized rats reduced voiding efficiency. Risperidone had smaller maximal effects than olanzapine and clozapine and had no direct (peripheral) inhibitory effects on the amplitude of bladder contractions. In the present study we show that clozapine acts at supraspinal and spinal sites to inhibit certain urodynamic parameters and the external urethral sphincter of the rat resulting in decreased voiding efficiency.
If these effects also occur in patients, they may contribute to the urinary disturbances reported following clozapine therapy. The exact receptor types (or combination of receptor types) responsible for clozapine's central effects on micturition were not investigated in the present study. However, isolating particular receptors that contribute to the effects of clozapine might be useful in designing neuroleptics that may avoid these side effects or in providing an adjunct therapy to relieve some of the side effects.
Surgical procedures
The experiments were conducted in compliance with the USDA Animal Welfare Act and amendments thereto and the revised Guide for the Care and use of Laboratory Animals DHEW (NIH) and were approved by the Animal Studies Subcommittee of the Bay Pines Veterans Administration Medical Center. Surgical procedures have been described in detail elsewhere [12]. Rats (female Sprague-Dawley; n = 16; 230-270 g; Harlan; IN) were anesthetized with halothane and placed on a heating pad. A catheter (PE-50) was introduced into the jugular vein to administer urethane (1.1 g/ kg) over a period of 20 minutes while decreasing the level of halothane to prevent respiratory depression. Rats were instrumented with either an intrathecal cannula placed over the L6/S1 spinal segment or a cannula into the right lateral ventricle.
An incision was made over the dorsal aspect of the neck and the overlying muscles were retracted to expose the atlanto-occipital membrane. An intrathecal catheter (PE 10) was introduced through a small slit in the atlanto-occipital membrane and positioned over the L6/S1 spinal cord segments [33]. Saline soaked gelfoam was placed around the catheter and the neck muscles and skin were sutured. Spinal segmental location of the catheter was verified post-mortem by performing a laminectomy.
Following a small craniectomy, a stainless-steel cannula (27 ga) was placed in the lateral ventricle at the following coordinates: AP -1.0, ML = 1.2; V= 3.2 mm [34]. The cannula was held in place with skull screws and dental acrylic. At the end of the experiment, 5 µl of fast-green (1%) was infused through the cannula while observing the CSF though a small slit in the atlanto-occipital membrane. Almost immediate visualization of the fast-green in the fourth ventricle was taken to indicate appropriate placement of the cannula into the lateral ventricle.
After an abdominal incision, both ureters were tied distally and cut centrally and allowed to drain onto cotton gauzes that were directed outside the animal. A catheter (PE-90) was introduced into the bladder dome and tied in place with a purse string suture. A catheter (PE-50) was introduced into the right femoral artery for blood pressure recording. Stainless-steel wires (0.003 in.; A-M Systems; WA) insulated except at the tip were introduced into the external urethral sphincter for EMG recording (DAM 50; WPI; bandwidth= 3 to 3 kHz; gain: 1000-10,000).
Urodynamic studies
In pilot animals, the dose and duration of effects was determined by administering clozapine (saline, 0.5,5,50,100 nmoles) at 10 min intervals either intrathecally (n = 2) or intracerebroventricularly (n = 2) during continuous cystometry (infusion rate = 0.11 ml/min) while recording bladder pressure, external urethral sphincter EMG, and blood pressure. Fifty (50) nmoles of clozapine administered by either route resulted in a decrease in arterial pressure (mean decrease in MAP was 20 ± 2 and 15 ± 3 mm Hg for intrathecal and intracerebroventricular administration, respectively. The onset time ranged from 1-1.5 min from the start of the infusion). Since clozapine has alpha1 antagonist effects [2] it is possible that the blood pressure decreases were due to spread of clozapine into the periphery following central administration. Doses smaller than 50 nmoles did not elicit a drop in arterial pressure. In addition, 50 nmoles represents a dose close to the smallest intravenous dose used previously (0.1 mg/kg [12]). Therefore, during single cystometry, the dose range was limited to 0.5 to 50 nmoles in order to reduce possible peripheral spread of clozapine.
Single cystometry studies were conducted as follows. The bladder was emptied and allowed to equilibrate to air pressure for 5 minutes before beginning each cystometrogram. Room temperature saline was infused into the bladder (0.11 ml/min) while recording bladder pressure and the infusion was stopped when a contraction occurred. Volume expelled was determined by placing cotton gauze at the urinary meatus and weighing before and after micturition. External urethral sphincter EMG (EUS-EMG) was recorded throughout the cystometrogram and for some time after the filling had stopped. Increasing doses of clozapine (Sigma, vehicle, 0.5, 5, 50 nmoles in a volume of 5 µl; followed by a 7 µl saline wash) were administered through the intrathecal catheter (n = 6; Mean weight = 245 gm) or through the intracerebroventricular cannula (n = 6; Mean weight = 245 gm) at approximately 10-minute intervals. Clozapine was dissolved in a minimal amount of 0.1 N hydrochloric acid, and brought up to volume with saline (final pH = 6). Cystometrograms were started approximately 3 minutes after each drug administration. At the end of the experiment the rat was euthanized with an overdose of urethane (3.0 mg/kg; i.v.).
Data analysis
Bladder pressure, EUS-EMG and blood pressure during the single cystometrograms were displayed in an electronic chart recorder (RC Electronics; Goleta, CA) and analyzed off-line (Dataview, W.J. Heitler, U. St Andrews, Scotland). The following parameters were examined from the cystometrogram as described in detail earlier [12]: bladder capacity (amount of fluid infused to elicit a contraction); micturition volume (amount of fluid expelled); residual volume ([bladder capacity-micturition volume]/ [bladder capacity] ×100); pressure threshold (pressure at which contraction begins); peak pressure (maximal pressure during contraction); contraction time; expulsion time (time between peak pressure and end of high frequency oscillations); amplitude of high frequency oscillations. The EMG activity was examined by dividing the bladder contraction into three phases in a modification of the technique of Chien et al. [19] a contraction phase (phase 1); an expulsion phase (phase 2) and a closing phase (phase 3).
The raw EMG was rectified, integrated (0.5 second bin) and the area under curve of the EMG corresponding to each phase of the bladder contraction was measured (Sigma Scan/Image; Jandel Scientiflcs, San Rafael, Ca). Drug effects for the EMG were calculated as percent of control.
Values are presented as Mean + S.E.M. Repeated measures ANOVA (GB Stat; Dynamic Microsystems; MD) were performed on all parameters and when statistical significance (p<0.05) was obtained, comparisons between control and different drug dosages and between different routes (i.t. vs. i.c.v.) were made using Fisher's protected t-test [35] .
|
2014-10-01T00:00:00.000Z
|
2002-03-07T00:00:00.000
|
{
"year": 2002,
"sha1": "89b99deb60141b7f2bfdc57822291fef35556ee7",
"oa_license": null,
"oa_url": "https://bmcpharma.biomedcentral.com/track/pdf/10.1186/1471-2210-2-6",
"oa_status": "BRONZE",
"pdf_src": "PubMedCentral",
"pdf_hash": "2c2bfb59e945d35c9ab771b6a23d3d89171827b2",
"s2fieldsofstudy": [
"Medicine",
"Psychology"
],
"extfieldsofstudy": [
"Medicine"
]
}
|
772186
|
pes2o/s2orc
|
v3-fos-license
|
Stanley depth of the integral closure of monomial ideals
Let $I$ be a monomial ideal in the polynomial ring $S=\mathbb{K}[x_1,...,x_n]$. We study the Stanley depth of the integral closure $\bar{I}$ of $I$. We prove that for every integer $k\geq 1$, the inequalities ${\rm sdepth} (S/\bar{I^k}) \leq {\rm sdepth} (S/\bar{I})$ and ${\rm sdepth} (\bar{I^k}) \leq {\rm sdepth} (\bar{I})$ hold. We also prove that for every monomial ideal $I\subset S$ there exist integers $k_1,k_2\geq 1$, such that for every $s\geq 1$, the inequalities ${\rm sdepth} (S/I^{sk_1}) \leq {\rm sdepth} (S/\bar{I})$ and ${\rm sdepth} (I^{sk_2}) \leq {\rm sdepth} (\bar{I})$ hold. In particular, $\min_k \{{\rm sdepth} (S/I^k)\} \leq {\rm sdepth} (S/\bar{I})$ and $\min_k \{{\rm sdepth} (I^k)\} \leq {\rm sdepth} (\bar{I})$. We conjecture that for every integrally closed monomial ideal $I$, the inequalities ${\rm sdepth}(S/I)\geq n-\ell(I)$ and ${\rm sdepth} (I)\geq n-\ell(I)+1$ hold, where $\ell(I)$ is the analytic spread of $I$. Assuming the conjecture is true, it follows together with the Burch's inequality that Stanley's conjecture holds for $I^k$ and $S/I^k$ for $k\gg 0$, provided that $I$ is a normal ideal.
Introduction and preliminaries
Let K be a field. Throughout this paper, the polynomial ring K[x 1 , . . . , x n ] in n variables over the field K is denoted by S.
Let M be a finitely generated Z n -graded S-module. Let u ∈ M be a homogeneous element and Z ⊆ {x 1 , . . . , x n }. for all Z n -graded S-modules M. For an introduction to Stanley depth, we refer the reader to [16].
Let I ⊂ S be an arbitrary ideal. An element f ∈ S is integral over I, if there exists an equation The set of elements I in S which are integral over I is the integral closure of I. It is known that the integral closure of a monomial ideal I ⊂ S is a monomial ideal generated by all monomials u ∈ S for which there exists an integer k such that u k ∈ I k (see [7,Theorem 1.4.2]).
Remark 1.1. Let I be a monomial ideal and let G(I) = {m 1 , . . . , m s } be the set of minimal monomial generators of I. For every 1 ≤ i ≤ s, there exists integer k i ≥ 1 such that m k i i ∈ I k i . Let k = lcm(k 1 , . . . , k s ) be the least common multiple of k 1 , . . . k s . Now for every 1 ≤ i ≤ s, we have m k i ∈ I k and this implies that u k ∈ I k , for every monomial u ∈ I. It follows that for every monomial u ∈ S, we have u ∈ I if and only if u k ∈ I k .
The ideal I is integrally closed, if I = I, and I is normal if all powers of I are integrally closed. By [22,Theorem 3.3.18], a monomial ideal I is normal if and only if the Rees algebra R(I) is a normal ring.
Apel [1] proved that for every monomial ideal I ⊂ S, the inequality sdepth(S/I) ≤ sdepth(S/ √ I) holds (see also [11]). It is clear that for every monomial ideal I ⊂ S, we have I ⊆ I ⊆ √ I and therefore √ I = √ I and hence sdepth(S/I) ≤ sdepth(S/ √ I). Now it is natural to ask about the relation of sdepth(S/I) and sdepth(S/I). There is no general inequality between sdepth(S/I) and sdepth(S/I), as the following examples show.
be a monomial ideal in the polynomial ring S = K[x 1 , x 2 , x 3 ]. Then one can easily check that I = (x 2 1 , x 2 2 , x 1 x 2 ). The maximal ideal m = (x 1 , x 2 , x 3 ) of S is an associated prime of S/I and therefore [10, Proposition 1.3] (see also [1]) implies that sdepth(S/I) = 0. Since m is not an associated prime of S/I, it follows from [4, Proposition 2.13] that sdepth(S/I) ≥ 1. Thus in this example sdepth(S/I) < sdepth(S/I).
Then the maximal ideal m = (x 1 , x 2 , x 3 ) of S is not an associated prime of S/I and therefore [4,Proposition 2.13] implies that sdepth(S/I) ≥ 1. On the other hand by [13,Theorem 2.4], m is an associated prime of S/I and therefore using [10, Proposition 1.3] (see also [1]), it follows that sdepth(S/I) = 0. Thus in this example sdepth(S/I) > sdepth(S/I). Examples 1.2 and 1.3 show that there is no general inequality between the Stanley depth of S/I and the Stanley depth of S/I. However, we prove that for every monomial ideal I ⊂ S there exist integers k 1 , k 2 ≥ 1, such that for every s ≥ 1, the inequalities sdepth(S/I sk 1 ) ≤ sdepth(S/I) and sdepth(I sk 2 ) ≤ sdepth(I) hold (Corollary 2.10). In particular Ratliff [20] proves that for every ideal I in a commutative Noetherian ring S, the asymptotic set of associated primes of integral closure of powers of I is a subset of the asymptotic set of associated primes of powers of I. We use Corollary 2.10 to give a new proof for Ratliff's theorem in the case of monomial ideals (Theorem 3.2).
We also prove that for every monomial ideal I ⊂ S, the inequalities sdepth(S/I k ) ≤ sdepth(S/I) and sdepth(I k ) ≤ sdepth(I) hold for every integer k ≥ 1 (Theorem 2.1). This implies that for every normal monomial ideal I, there exists k, such that sdepth(S/I k ) = sdepth(S/I sk ), for every integer s ≥ 1.
In Section 2, we present a conjecture, regarding the Stanley depth of integrally closed monomial ideals. In order to do this, we need to introduce some notation and well known results.
Let I be a monomial ideal of S with Rees algebra R(I) and let m = (x 1 , . . . , x n ) be the graded maximal ideal of S. Then the K-algebra R(I)/mR(I) is called the fibre ring and its Krull dimension is called the analytic spread of I, denote by ℓ(I). This invariant is a measure for the growth of the number of generators of the powers of I. Indeed, for k ≫ 0, the Hilbert function H(R(I)/mR(I), K, k) = dim K (I k /mI k ), which counts the number of generators of the powers of I, is a polynomial function of degree ℓ(I)−1. Let I ⊂ S, be a monomial ideal. An ideal J ⊆ I is called a reduction of I, if JI t = I t+1 , for some integer t ≥ 1. It is known by [9, Corollaries 8.2.5 and 8.3.9] that for every ideal I and every reduction J of I, the inequality ht(I) ≤ ℓ(I) ≤ µ(J) holds, where µ(J) denotes the number of minimal generators of J.
Let I ⊂ S be a monomial ideal. A classical result by Burch [5] says that By a theorem of Brodmann [2], the quantity depth(S/I k ) is constant for large k. We call this constant value the limit depth of I and we denote it by lim k→∞ depth(S/I k ). Brodmann improves Burch's inequality by showing that with equality if the Rees algebra R(I) is a normal ring. In Section 2, we conjecture that for every integrally closed monomial ideal I, the inequalities sdepth(S/I) ≥ n − ℓ(I) and sdepth(I) ≥ n − ℓ(I) + 1 hold. Assuming the conjecture is true, it follows together with the Burch's inequality that Stanley's conjecture holds for I k and S/I k for k ≫ 0, provided that I is a normal ideal.
Stanley depth and integral closure of monomial ideals
Let I be a monomial ideal. As the first result of this paper, we compare the Stanley depth of the integral closure of I and the Stanley depth of the integral closure of powers of I. Theorem 2.1. Let J ⊆ I be two monomial ideals in S. Then for every integer k ≥ 1 Proof. Let u ∈ S be a monomial. Then u ∈ I if and only if u s ∈ I s , for some s ≥ 1 if and only if u ks ′ ∈ I ks ′ , for some s ′ ≥ 1 if and only if u k ∈ I k . By a similar argument u ∈ J if and only if u k ∈ J k . Now consider a Stanley decomposition Thus for each monomial u ∈ I \ J, we define Z u := Z i and t u : where the sum is taken over all monomials u ∈ I \ J. For the converse inclusion note that for every u ∈ I \ J and every h ∈ K[Z u ], clearly we have uh ∈ I. By the choice of t u and Z u , we conclude u k ∈ t u K[Z u ] and therefore ∈ J k and as argument above shows, uh / ∈ J.
where the sum is taken over all monomials u ∈ I \ J. Now for every 1 ≤ i ≤ m, let Without lose of generality we may assume that U i = ∅ for every 1 ≤ i ≤ l and U i = ∅ for every l + 1 ≤ i ≤ m. For every 1 ≤ i ≤ l, let u i be the greatest common divisor of elements of U i . Note that where the second sum is taken over all monomials u ∈ U i . Since for every It follows that Next we prove that for every 1 ≤ i, j ≤ l with i = j, the summands u i K[Z i ] and u j K[Z j ] intersect trivially. By contradiction let v be a monomial in The following corollaries are immediate consequences of Theorem 2.1 The following example from [6] shows that the inequalities of Corollary 2.3 do not necessarily hold, if I is not a normal ideal. and sdepth(I) ≥ n − ℓ(I) + 1 hold. These inequalities have been proved for some special classes of monomial ideals. In [17], the authors prove that if I ⊂ S is a weakly polymatroidal ideal (see [7,Definition 12.7.1]), which is generated in the same degree, then sdepth(S/I) ≥ n−ℓ(I) and sdepth(I) ≥ n − ℓ(I) + 1. In [18] the authors study the Stanley depth of powers of edge ideal of forest graphs. Let G = (V, E) be a forest graph with n vertices and p connected components and let be the edge ideal of G. Then sdepth(S/I(G) k ) ≥ p, for every integer k ≥ 1 ([18, Theorem 2.7]). But it is known and easy to prove that for every forest with n vertices and p connected components, ℓ(I(G)) = n − p (see [23], page 50 for more details), which means that sdepth(S/I(G) k ) ≥ n − ℓ(I) for every integer k ≥ 0.
The following example shows that these inequalities do not hold for an arbitrary monomial ideal. The ideal I, in Example 2.5, is not integrally closed. In fact the author has no example of integrally closed monomial ideals, for which the inequalities sdepth(S/I) ≥ n − ℓ(I) and sdepth(I) ≥ n − ℓ(I) + 1 do not hold. Therefore he presents the following conjecture. Assuming the conjecture is true, it follows together with Burch's inequality that Stanley's conjecture holds for I k and S/I k for k ≫ 0, provided that I is a normal ideal.
It is clear that for every monomial ideal I in S, the Stanley depth of I is at most n. Hence the Stanley depth of infinitely many powers of I is constant. The following corollary gives a refinement of this fact in the case of normal ideals.
Corollary 2.7. Let I ⊂ S be a normal monomial ideal. Then the following statements hold.
(i) There exists an integer k ≥ 1 such that for every integer s ≥ 1, we have sdepth(I k ) = sdepth(I sk ). (ii) There exists an integer k ≥ 1 such that for every integer s ≥ 1, we have sdepth(S/I k ) = sdepth(S/I sk ).
Proof. (i) Let k ≥ 1 be an integer such that sdepth(I k ) = min t {sdepth(I t )}. Since I k is a normal ideal, Corollary 2.3, implies that for every integer s ≥ 1, we have and thus by the choice of k, sdepth(I k ) = sdepth(I sk ).
(ii) The proof is similar to the proof of (i).
Let I be a monomial ideal. In the following theorem we compare the Stanley depth of I and the Stanley depth of powers of I. We will use this result in Section 3, to give a new proof for a result of Ratliff in the case of monomial ideals (see Theorem 3.2).
Theorem 2.8. Let I 2 ⊆ I 1 be two monomial ideals in S. Then there exists an integer k ≥ 1, such that for every s ≥ 1 . Proof. Note that by Remark 1.1, there exist integers k 1 , k 2 ≥ 1, such that for every monomial u ∈ S, we have u k 1 ∈ I k 1 1 (resp. u k 2 ∈ I k 2 2 ) if and only if u ∈ I 1 (resp. u ∈ I 2 ). Let k = lcm(k 1 , k 2 ) be the least common multiple of k 1 and k 2 . Then for every monomial u ∈ S, we have u k ∈ I k 1 (resp. u k ∈ I k 2 ) if and only if u ∈ I 1 (resp. u ∈ I 2 ). Hence for every monomial u ∈ S and every s ≥ 1, we have u sk ∈ I sk 1 (resp. u sk ∈ I sk 2 ) if and only if u ∈ I 1 (resp. u ∈ I 2 ). Now we prove that for this choice of k and for every s ≥ 1 sdepth(I sk 1 /I sk 2 ) ≤ sdepth(I 1 /I 2 ), and this proves our assertion. Let be a Stanley decomposition of I sk 1 /I sk 2 , such that sdepth(D) = sdepth(I sk 1 /I sk 2 ). By the argument above, for every monomial u ∈ I 1 \ I 2 , we have u sk ∈ I sk 1 \ I sk 2 . Now for each monomial u ∈ I 1 \ I 2 we define Z u := Z i and t u := t i , where i ∈ {1, . . . , m} is the uniquely determined index such that u sk ∈ t i K[Z i ]. It is clear that where the sum is taken over all monomials u ∈ I 1 \ I 2 . For the converse inclusion note that for every u ∈ I 1 \ I 2 and every h ∈ K[Z u ], clearly we have uh ∈ I. By the choice of t u and Z u , we conclude u sk ∈ t u K[Z u ] and therefore u sk h sk ∈ t u K[Z u ]. This implies that u sk h sk / ∈ I sk 2 and as argument above shows and by the choice of k, we have uh / ∈ I 2 . Therefore where the sum is taken over all monomials u ∈ I 1 \ I 2 . Now for every 1 ≤ i ≤ m, let Without lose of generality we may assume that U i = ∅ for every 1 ≤ i ≤ l and U i = ∅ for every l + 1 ≤ i ≤ m. For every 1 ≤ i ≤ l, let u i be the greatest common divisor of elements of U i . Note that where the second sum is taken over all monomials u ∈ U i . Since for every u ∈ U i , It follows that Next we prove that for every 1 ≤ i, j ≤ l with i = j, the summands u i K[Z i ] and u j K[Z j ] intersect trivially. By contradiction let v be a monomial in which is a contradiction, because m i=1 t i K[Z i ] is a Stanley decomposition of I sk 1 /I sk 2 . Therefore is a Stanley decomposition of I 1 /I 2 which proves sdepth(I 1 /I 2 ) ≥ min l i=1 |Z i | ≥ sdepth(I sk 1 /I sk 2 ). We illustrate the procedure of the proof of the Theorem 2.8 in the following example.
Proposition 2.13] implies that sdepth(S/I 2t ) ≥ 1, for every integer t ≥ 1. But as we mentioned in Example 2.4, sdepth(S/I) = 0. This shows that there does not exist any integer k such that for every s ≥ 1 the inequality sdepth(S/I sk ) ≤ sdepth(S/I) holds The following corollaries are immediate consequences of Theorem 2.8 and Corollary 2.10. 3
. An application of Stanley depth
Let I be a monomial ideal of the polynomial ring S = K[x 1 , . . . , x n ]. In this section we will examine the sets of associated primes of the powers of I, that is, the sets Ass(S/I k ) = {P ⊂ S : P is prime and P = (I k : c) for some c ∈ S}, k ≥ 1.
Since I is a monomial ideal of a polynomial ring S, the associated primes will be monomial primes, which are primes that are generated by subsets of the variables, see [7,Corollary 1.3.9]. In [3], Brodmann showed that the sets Ass(S/I k ) stabilize for large k. That is, there exists a positive integer N 1 ≥ 1 such that Ass(S/I k ) = Ass(S/I N 1 ) for all k ≥ N 1 . We denote the set Ass(S/I N 1 ) by Ass ∞ (S/I). Ratliff studied the set of associated primes of the integral closure of powers of ideals. By his results [19,20], one has that the sets Ass(S/I k ) form an ascending chain which stabilizes for large k. Thus, there exists N 2 ≥ 1 such that Ass(S/I k ) = Ass(S/I N 2 ) for all k ≥ N 2 . We denote the set Ass(S/I N 2 ) by Ass ∞ (S/I). The set Ass ∞ (S/I) is nicely described in [14]. It is known [20, Theorem 2.8] that the inclusion Ass ∞ (S/I) ⊆ Ass ∞ (S/I) holds for any ideal I of a commutative Noetherian ring (see [15,Proposition 3.17] for additional details). As an application of Corollary 2.10 we give a new proof for this result in the case of monomial ideals. To the best of my knowledge, this would be the first application of Stanley depth.
First we need to introduce some notation and basic facts. Let P = (x i 1 , . . . , x ir ) be a monomial prime ideal in S, and I ⊆ S any monomial ideal and let L = [n] \ {x i 1 , . . . , x ir }. We denote by I(P ) the monomial ideal in the polynomial ring S(P ) = K[x i 1 , . . . , x ir ], which is obtained from I by applying the K-algebra homomorphism S → S(P ) with x i → 1 for all i ∈ L. It is known that ([8, Ass(S(P )/I(P )) = {Q ∈ Ass(S/I) : We use this simple fact for proving Theorem 3.2. We also need the following simple lemma. Proof. Let N 1 , N 2 ≥ 1 be two integers such that Ass(S/I N 1 ) = Ass(S/I N 1 +k ) and Ass(S/I N 2 ) = Ass(S/I N 2 +k ), for every integer k ≥ 1. Let P ∈ Ass(S/I N 2 ) be a monomial prime ideal of S. Then by [8, Lemma 1.3], we have P ∈ Ass(S(P )/I N 2 (P )) = Ass(S(P )/I N 2 (P )) = Ass(S(P )/I(P ) N 2 ), where the first equality follows from Lemma 3.1 and the second equality is trivial. Since P is the maximal ideal of S(P ), It follows from [10, Proposition 1.3] (see also [1]) that sdepth S(P ) (S(P )/I(P ) N 2 ) = 0. By Corollary 2.10, there exists an integer l ≥ N 1 such that sdepth S(P ) (S(P )/I(P ) l ) ≤ sdepth S(P ) (S(P )/I(P ) N 2 ), and therefore sdepth S(P ) (S(P )/I(P ) l ) = 0.
|
2012-11-17T17:10:19.000Z
|
2012-05-31T00:00:00.000
|
{
"year": 2013,
"sha1": "d9fbd126933402673fd0442da75b2c25d51b0342",
"oa_license": null,
"oa_url": "http://arxiv.org/pdf/1205.6971.pdf",
"oa_status": "GREEN",
"pdf_src": "Arxiv",
"pdf_hash": "d9fbd126933402673fd0442da75b2c25d51b0342",
"s2fieldsofstudy": [
"Mathematics"
],
"extfieldsofstudy": [
"Mathematics"
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.