Role of vasorin, an anti‐apoptotic, anti‐TGF‐β and hypoxia‐induced glycoprotein in the trabecular meshwork cells and glaucoma

Abstract Glaucoma, one of the leading causes of irreversible blindness, is commonly associated with elevated intraocular pressure due to impaired aqueous humour (AH) drainage through the trabecular meshwork. The aetiological mechanisms contributing to impaired AH outflow, however, are poorly understood. Here, we identified the secreted form of vasorin, a transmembrane glycoprotein, as a common constituent of human AH by mass spectrometry and immunoblotting analysis. ELISA assay revealed a significant but marginal decrease in vasorin levels in the AH of primary open‐angle glaucoma patients compared to non‐glaucoma cataract patients. Human trabecular meshwork (HTM) cells were confirmed to express vasorin, which has been shown to possess anti‐apoptotic and anti‐TGF‐β activities. Treatment of HTM cells with vasorin induced actin stress fibres and focal adhesions and suppressed TGF‐β2‐induced SMAD2/3 activation in HTM cells. Additionally, cobalt chloride‐induced hypoxia stimulated a robust elevation in vasorin expression, and vasorin suppressed TNF‐α‐induced cell death in HTM cells. Taken together, these findings reveal the importance of vasorin in maintenance of cell survival, inhibition of TGF‐β induced biological responses in TM cells, and the decreasing trend in vasorin levels in the AH of glaucoma patients suggests a plausible role for vasorin in the pathobiology of ocular hypertension and glaucoma.

the balance between aqueous humour (AH) secretion by the ciliary epithelium and AH drainage through the conventional or trabecular pathway consisting of the trabecular meshwork (TM) and Schlemm's canal, and the non-conventional pathway consisting of the ciliary muscle, supraciliary and suprachoroidal spaces. 6,7 Impairment in the conventional AH outflow pathway has been recognized to be the main cause for elevated IOP in glaucoma, and experimental elevation of IOP has been demonstrated to induce glaucoma. 6,7 However, the development of novel efficacious and mechanism-based IOP lowering therapies is currently hampered by our limited understanding of the aetiological mechanisms involved in ocular hypertension. 5 Alterations in levels of various external factors including TGFβ, endothelin-1, connective tissue growth factor (CTGF), lysophosphatidic acid (LPA), TNFα, autotaxin and extracellular matrix in the AH of glaucoma patients have been found to be associated with elevated IOP in glaucoma patients. [6][7][8][9][10][11][12][13] Several of these external factors have been demonstrated to regulate AH outflow through the trabecular pathway in different experimental models. [6][7][8][9] Additionally, increased cell death and loss of TM cells in the trabecular pathway have been found to be associated with ocular hypertension in glaucoma. 14,15 Improvements in proteomics technology have enabled the identification of novel and differentially regulated secreted proteins in AH derived from subjects with different types of glaucoma, indicating the involvement of several external cues in the homeostasis of AH outflow and IOP. 6,16,17 In this study, the glycoprotein vasorin was identified in almost every sample of non-glaucomatous human AH (>10 samples tested) we analysed by proteomics analysis.
However, we have no knowledge regarding the role and significance of vasorin in AH, and its possible involvement either in the physiology of the TM or in the aetiology of ocular hypertension.
Vasorin, a cell surface single-pass transmembrane glycoprotein with an estimated molecular mass of ~110 kDa and consisting of 673 amino acid residues, has been shown to be abundantly expressed by vascular smooth muscle cells and moderately by several other tissues and organs. 18,19 The extracellular portion of vasorin contains several tandemly organized leucine-rich repeat regions, an epidermal growth factor-like repeat, a fibronectin type III domain and a short intracellular carboxy terminal peptide with no known sequence homology with other proteins. 18 Vasorin, a developmentally regulated protein, has been shown to regulate various cellular activities including proliferation, differentiation, migration, angiogenesis, folliculogenesis and calcification and to be involved in the pathobiology of fibrosis, nephropathies and tumorigenesis. [18][19][20][21][22][23][24] Vasorin has also been demonstrated to bind TGFβ, and block TGFβ biological activity 18,25 and regulate Notch1 signalling by interacting with Numb and preventing degradation of Notch1. 20 ADAM17 (a disintegrin and metalloprotease 17) has been shown to cleave and release the extracellular portion of vasorin as a soluble and active protein. 25 Only the soluble/extracellular form of vasorin has been shown to interact with and trap TGFβ and augment TGFβ activity under the deficiency of ADAM17. 25 Moreover, hypoxia-induced expression of vasorin suppresses TNFα mediated apoptosis via regulating mitochondrial thioredoxin activity in mouse embryonic fibroblasts. 21 While vasorin knockout mice exhibit fertility defects, not much additional knowledge is currently available regarding the physiological role of vasorin. 19,21 In this study, we determined the levels of vasorin in the AH of POAG patients, expression, secretion and release of the soluble form of vasorin by HTM cells, effects of vasorin on TGF-β2 mediated cellular events and in TNFα induced TM cell death, to gain insights into the role of this glycoprotein in TM biology, AH dynamics and IOP.

| Aqueous humour collection
Aqueous humour samples were collected at the initiation of cataract or glaucoma surgery. A tuberculin syringe with a 30-gauge needle was inserted into the anterior chamber through a limbal paracentesis tract at the start of the surgery, and approximately 30-100 μl of AH was slowly aspirated. The AH samples were transferred from the syringe to a 1.5 ml Eppendorf tube and centrifuged at 1000×g for 10 min at 4°C. The supernatant obtained from the AH samples was collected and stored at −80°C until further use.

| Mass spectrometry
Ten µl of AH samples was solubilized in 2% sodium dodecyl sulphate, 100 mM Tris-HCl (pH 8.0), reduced with 10 mM dithiothreitol, alkylated with 25 mM iodoacetamide and subjected to tryptic hydrolysis using the HILIC beads SP3 protocol (ReSyn Biosciences, Gauteng, South Africa) as described by Hughes et. al. 26 The resulting peptides were analysed with a nanoAcquity UPLC system (Waters The PCR products were sequenced to confirm the identity of amplified DNA.

| Enzyme-linked immunosorbent assay (ELISA)
A human vasorin ELISA kit (MyBioSource, San Diego, CA) was used to determine the levels of vasorin in human AH (duplicates of 10 µl AH were used from each sample), using the manufacturer's protocol which included appropriate standards and background controls.

| Immunohistochemistry
To determine the distribution profile of vasorin in the conventional AH outflow pathway, tissue sections derived from formalin-fixed, paraffin-embedded human donor eye whole globes (90 year old) were immunostained with a vasorin antibody as we described previously. 29 Briefly, 5μm thick tissue sections were deparaffinized and rehydrated using xylene, absolute ethyl alcohol and water.  Table S1). Slides were washed and incubated with Alexa Fluor-488 goat anti-mouse secondary antibody (Table S1) for two hrs at room temperature. Immunostained slides were viewed and imaged using a Nikon Eclipse 90i confocal laser-scanning microscope (Nikon Instruments, Melville, NY, USA). Immunohistochemistry analyses were carried out in duplicates and included a negative control with no primary antibody.

| Immunofluorescence
Human TM cells grown on gelatin (2%)-coated glass coverslips were fixed with 4% paraformaldehyde, permeabilized, blocked and immunostained for vasorin using mouse monoclonal antibody alone or together with TOM20 (Table S1). Additionally, serum starved TM cells (24 h) treated with vasorin (10 ng/ml or 50 ng/ml) for 24 h were stained for F-actin with phalloidin-Tetramethylrhodamine B isothiocyanate and immunostained for vinculin with a mouse monoclonal anti-vinculin antibody (Table S1). TM cells treated with CoCl 2 (0.4 mM) for eight and 24 h were fixed as mentioned above with 4% paraformaldehyde and immunostained for vasorin using an anti-vasorin antibody and appropriate secondary antibodies conjugated with Alexa fluorophores 488 or 568, as we described previously. 30 In the above described analyses, samples were incubated with primary antibodies for 24 h at 4°C and with secondary antibodies for two hrs at room temperature. Cell nuclei were counterstained with Hoechst. Finally, coverslips were mounted onto glass slides and imaged using a Nikon Eclipse 90i confocal laser-scanning microscope.

| Immunoblot analysis
Cells were homogenized at 4°C in hypotonic buffer ( Table S1 for details). Membranes were washed with TBS buffer containing 0.1% Tween-20 and incubated at room temperature with appropriate secondary antibodies for two hrs. Immunoblots were developed by enhanced chemiluminescence, followed by scanning and analysis using ChemiDoc Touch imaging and Image Lab™ Touch Software (Bio-Rad Laboratories) respectively.
Densitometry analysis was carried out using Image J software.

| Cell viability
To determine whether vasorin inhibits TNFα induced apoptosis Phase contrast images of TM cells exposed to the above treatments were captured using a Zeiss Axiovision microscope.

| Statistical analysis
Statistical analysis was performed using GraphPad Prism version 7 for Windows (GraphPad Software, La Jolla, CA, USA). All cell biology data represent the average values (mean ± standard error of mean) from at least four independent experiments unless otherwise mentioned. The Mann-Whitney U-test was employed for comparing vasorin levels in human AH samples from glaucomatous and nonglaucomatous groups. Student's t-test and one-way analysis of variance (ANOVA) were used for statistical comparisons between any two groups and three groups respectively. A p value of >0.05 was considered statistically significant.

| Vasorin is a common constituent of human AH
To identify secretory proteins including vasorin that play a poten-

| Decreased levels of vasorin in the AH of POAG patients
Since vasorin is known to be involved in several diseases, 19 we determined the levels of vasorin in AH derived from twentyone POAG patients and twenty age-and gender-matched non-glaucoma (cataract) patients by ELISA. 621.89 ± 6.91 pg/ml) were marginally (by 4.4%) but significantly (p > 0.05) lower than those in non-glaucoma patients (649. 58 ± 8.79 pg/ml) based on the Mann-Whitney U-test ( Figure 1C).

No correlation was found between patients AH vasorin levels and
IOP in the POAG group ( Figure S1A). Vasorin levels in aqueous humour of POAG and non-glaucoma patients were also not significantly correlated with patient age ( Figure S1B).

| Human TM cells express and secrete vasorin
Vasorin has been demonstrated to be abundantly expressed in vascular smooth muscle cells. 18 Since TM tissue exhibits smooth muscle-like properties and regulates AH outflow and IOP, 33 we investigated the expression and secretion of vasorin by HTM cells.
RT-PCR analysis confirmed the expression of vasorin, based on the results derived from two different TM cell strains and using two different sets of PCR primers. Figure

| Regulation of the secreted form of vasorin in TM cells
To gain insight into the mechanisms regulating production of the secreted/soluble form of vasorin in HTM cells, we evaluated the expression of ADAM17, which is known to cleave the extracellular part of vasorin to generate the secreted form of the protein. 25 We con-  Figure 3B). Figure 3D shows the GelCode blue stained protein profile of CM derived from both control and BB-94 treated TM cells, with the boxed protein band used as a loading control.

| Vasorin induces actin cytoskeletal reorganization and cell adhesive interactions in TM cells
The extracellular region of vasorin contains several motifs including tandemly organized leucine-rich repeat regions, an epidermal growth factor -like repeat and a fibronectin type III domain,  Figure 4D depicts the fold change in levels of p-Pax and pMYPT1 under vasorin treatment, with GAPDH used as loading control. Total MLC and GAPDH were used as loading controls, respectively, to calculate fold changes in pMLC and total MLC levels ( Figure 4D). Collectively, these results argue a role for soluble vasorin in influencing the contractile and cell adhesive characteristics of TM cells.

| Vasorin-mediated mitigation of TGF-β2-induced changes in TM cells
One of the well-recognized biological activities of vasorin is trapping and blocking of TGFβ activity, 18,25 and suppressing the TGFβ-induced epithelial-to-mesenchymal transition (EMT). 25 Since  Figure 5A, B) resulted in significant suppression of these changes. The levels of total SMAD2 and SMAD3 were found to be unchanged in the above described samples ( Figure 5A, B). These results imply the ability of vasorin to block TGFβ induced changes in TM cells.

| Vasorin suppresses TNF-α induced cell death in TM cells
Having  Figure 7A, B). Addition of vasorin to TM cells together with TNFα and cyclohexamide resulted in a significant decrease in cell death relative to that observed in TM cells treated with TNFα and cyclohexamide ( Figure 7A, B), revealing that vasorin protects TM cells from TNFα-induced cell death. The lower panels in Figure 7A show phase contrast images of HTM cell morphology under the above described treatments. Relatively, the TNFα and cycloheximide treated cells exhibit notable changes in cell morphology with punctate appearance compared to control and vasorin supplemented cells.

| DISCUSS ION
While AH outflow through the trabecular meshwork, and IOP are modulated by a variety of extracellular inputs including physiological growth factors whose dysregulation is associated with ocular hypertensive glaucoma, 7,11 we are yet to establish a complete understanding regarding identity of the external cues regulating AH drainage and IOP. This study identifies vasorin as a common constituent of AH, demonstrates that expression of this glycoprotein which is expressed and secreted by TM cells, is induced in response to hypoxic conditions, and that vasorin stimulates actin cytoskeletal reorganization, focal adhesion formation and contractile changes in TM cells.
Importantly, vasorin appears to block the biological effects of TGFβ The well-recognized physiological characteristics of vasorin include a role in regulation of TGFβ activity and inhibition of TNFα induced apoptosis. 18,21 Vasorin has been demonstrated to directly bind TGFβ and block TGFβ signalling activity including the fibrotic response and EMT in different cell types. 18,25,38 Importantly, elevated levels of TGFβ is not only a common finding in the AH of POAG patients, but increased levels of TGFβ are recognized to elevate IOP by stimulating fibrogenic activity and transdifferentiation of TM cells into extracellular matrix producing fibroblast-like cells. 11,27,35 In this study, vasorin was found to block the biological activity of TGF-β2 as evidenced by suppres- Additionally, TGFβ-induced fibrogenic activity has been shown to be augmented in the absence of vasorin, indicating the importance of varosin expression in regulation of TGFβ activity. 18,25,38 Vasorin has also been demonstrated to suppress hypoxia-and TNFα-induced apoptosis partly through regulation of thioredoxin anti-oxidative activity and generation of reactive oxygen species by mitochondria. 21 Since elevated levels of TNFα in the AH and

ACK N OWLED G EM ENTS
We thank Ms. Iris Navarro for her help with obtaining patient consent for the collection of AH samples from glaucoma and cataract patients, and Harold P Erickson, Ph.D., Duke University, for providing fibronectin antibody. This study was supported by the grants from the National Institutes of Health (R01EY018590 and R01-EY028823 (PVR)).

CONFLICTS OF INTEREST
The authors have no financial and/or non-financial interests in relation to the work described.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request from the authors.