Low expression of GSTP1 in the aqueous humour of patients with primary open‐angle glaucoma

Abstract Primary open‐angle glaucoma (POAG) is characterized by irreversible neurodegeneration accompanied by visual field defects and high intraocular pressure. Currently, an effective treatment is not available to prevent the progression of POAG, other than treatments to decrease the high intraocular pressure. We performed proteomic analysis of aqueous humour (AH) samples from patients with POAG combined with cataract and patients with cataract to obtain a better understanding of the pathogenesis of POAG and explore potential treatment targets for this condition. Samples were collected from 10 patients with POAG combined with cataract and 10 patients with cataract. Samples from each group were pooled. A high‐resolution, label‐free, liquid chromatography‐tandem mass spectrometry‐based quantitative proteomic analysis was performed. In total, 610 proteins were identified in human AH samples from the two groups. A total of 48 up‐regulated proteins and 49 down‐regulated proteins were identified in the POAG combined with cataract group compared with the control group. Gene Ontology (GO) analysis revealed key roles for these proteins in inflammation, immune responses, growth and development, cellular movement and vesicle‐mediated transport in the biological process category. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis indicated the down‐regulated expression of glutathione S‐transferase P (GSTP1) in the glutathione metabolism signalling pathway in the POAG combined with cataract group. Additionally, certain significantly differentially expressed proteins in the proteomic profile were verified by enzyme‐linked immunosorbent assay (ELISA). GSTP1 levels were reduced in the human AH samples from the POAG combined with cataract group, based on the results of ELISA and proteomic profiling. Therefore, GSTP1, a redox‐related marker, may be involved in the pathological process of POAG and may become a treatment target in the future.


| INTRODUC TI ON
Primary open-angle glaucoma (POAG) is defined as a chronic progressive optic neuropathy associated with elevated intraocular pressure (IOP) and visual field defects, and this disorder is one of the leading causes of irreversible blindness worldwide. 1 The dynamic balance of IOP is closely related to drainage of the aqueous humour (AH). A high IOP occurs as the result of impaired AH outflow. 2,3 The AH performs various functions, including maintaining IOP, providing nutrients to the ocular tissues and removing metabolic products from ocular tissues. 4,5 Changes in the protein composition of the AH are closely related to the metabolism of the anterior segmental tissue, such as the trabecular meshwork (TM). 6 The TM acts as a static filter that exerts stable resistance to AH outflow. 7 Abnormal expression of extracellular matrix (ECM) components within the TM microenvironment is associated with impaired AH drainage and leads to an elevated IOP. 8 In addition, abnormal expression of ECM proteins in the TM is related to the oxidative stress response in the TM. 9 Many ocular diseases and related treatments are associated with oxidation and inflammation. [10][11][12][13][14][15] Some studies have also reported markedly elevated levels of oxidative stress markers in the AH from patients with POAG, along with altered expression of markers of antioxidant defences and an enhanced inflammatory response. [16][17][18] Quantitative proteomics based on mass spectrometry (MS) is an important methodology for biological and clinical research. [19][20][21][22] Label-free MS has several advantages, including the lack of a restriction on the sample size, the lack of a requirement for expensive isotope labelling and the ability to detect a wide range of proteins. 23 Comparison and identification of the changes in proteins in the AH using proteomics help researchers detect differentially expressed proteins in patients with different diseases. In this study, we sought to detect differentially expressed proteins and explore the pathological mechanisms of POAG, which will provide new molecular targets for glaucoma treatment.

| Subjects
The Institutional Ethics Committee approved this study, and we ad- Patients with ophthalmic conditions such as uveitis, ocular trauma; a history of ocular surgery within the previous three months; intraocular inflammation; secondary or neovascular glaucoma; or use of topical or systemic corticosteroids were excluded from the study.
AH samples (25-100 μL) were collected through an anterior chamber paracentesis using a 30-gauge needle at the beginning of the surgical procedures from patients undergoing cataract or trabeculectomy surgery. Samples were transported on ice, centrifuged at 704 g for 10 minutes at 4°C and immediately stored at −80°C until further analysis.

| Estimation of the total protein content using the Bradford method
The total protein concentration in the AH was determined using a Bradford assay kit (Thermo Fisher Scientific) according to the manufacturer's protocol. Briefly, 5 μL of AH proteins was added to 200 μL of Bradford reagent. The optical density was measured at 595 nm after a 15 minutes incubation. The protein concentration of each sample was calculated from a standard curve using BSA as the reference.

| Proteolysis
Approximately 5 μg of total AH protein was aliquoted, the volume was brought to 500 μL with 5 mmol/L ammonium bicarbonate (YDYYRCXM-C2018-01/02/03). This work was supported by a grant from Young Medical Talents Program of Tianjin Health  and Family Planning Commission. proteomic profiling. Therefore, GSTP1, a redox-related marker, may be involved in the pathological process of POAG and may become a treatment target in the future.

K E Y W O R D S
aqueous humour, cataract, GSTP1, inflammation, oxidative stress, primary open-angle glaucoma, proteome (NH 4 HCO 3 ), and the mixture was concentrated using a 3 kD molecular weight cut-off filtration column. Next, a solution containing 100 mmol/L dithiothreitol (DTT) and 50 mmol/L NH 4 HCO 3 was added to the solution and incubated at 60°C for 30 minutes. Subsequently, a solution of 200 mmol/L iodoacetamide and 50 mmol/L NH 4 HCO 3 was added and incubated with the sample in the dark at 37°C for 20 minutes. Trypsin was added to the solution at a 1:50 dilution and incubated at 37°C overnight, after which it was further incubated at 37°C for 20 minutes along with formic acid.
Samples were centrifuged at 15 339 g for 10 minutes to remove the debris, and the supernatant was filtered through a 0.2 μm filter and dried in a speed vacuum. Then, the dried residues in the vials were reconstituted with 2% acetonitrile and 0.1% formic acid, centrifuged, and the supernatant was transferred into total recovery vials.
Trypsin-digested AH proteins from each group (n = 10) of samples were subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

| Nano-HPLC-MS (Q-Exactive) proteomic and data analyses
Samples were subjected to MS analysis (Thermo Fisher). Components obtained by high pH reverse-phase chromatography were resolved with 20 μL of 2% methanol and 0.1% formic acid. Samples were cen-

| Statistical analysis
Protein expression profiles were analysed with MaxQuant software (version 1.6.2.0.). MaxQuant significance A 24 was used to evaluate the significance of differences. Differentially expressed proteins were identified from the raw data (fold change > 1.5 and P < .05).
The data were processed and analysed using GraphPad ® Prism 7 software. Clinical variables and ELISA data were analysed using an unpaired t test followed by the Mann-Whitney U test. Data are presented as means ± SDs, and P < .05 was considered to indicate a statistically significant difference.

| Patient information
According to the inclusion and exclusion criteria, patients with cataract had a turbid lens, normal anterior chamber depth, appropriate cup/disc ratio and regular structure of the optic nerve head.
Patients with POAG combined with cataract displayed slight turbidity of the lens and a normal anterior chamber depth, abnormal cup/disc ratio and irregular optic nerve head. The present study included 10 eyes from 10 patients with POAG combined with cataract and 10 eyes from 10 patients with cataract. The demographic and clinical characteristics are summarized in Table 1. The mean ages of the patients with POAG combined with cataract and the controls were 72.8 ± 2.6 years and 71.7 ± 2.5 years, respectively (P = .923). No significant differences in the sex distribution, axial length, corneal thickness, aqueous depth or visual acuity were observed between the two groups (P > .05). As expected, the POAG combined with cataract group had a higher mean IOP and larger cup/disc ratio than the control group (P = .043 and P = .0005, respectively).

| Data acquisition
The process of the label-free proteomics technology is divided into three main stages: protein sample preparation, MS measurement and data analysis. Ten pooled samples each were available from the POAG combined with cataract group and the cataract group.
The protein concentration was 0.15 µg/µL in the POAG combined with cataract group and 0.07 µg/µL in the control group. Proteins in each group were divided into two subgroups, that is, high-density proteins and low-density proteins, to identify proteins with a low density. The test was repeated twice using the same method. From the heatmap, we concluded that the repeatability of the results was sufficient and that noticeable differences were observed between the two groups. Red represents up-regulated proteins and blue rep-

| Data analysis
All significantly differentially expressed proteins identified in the LC-MS analysis are listed in Table 2, which describes the differentially expressed proteins between the POAG combined with cataract group and the control group. Selected interesting proteins are also labelled in Table 2      as S100-A7 protein (S10A7), S100-A8 protein (S10A8), S100-A9 protein (S10A9), S100-A4 protein (S10A4) and ANXA1, are involved in

| Protein validation with ELISA
AH was collected again from patients with POAG combined with cataract and control patients. The patients' clinical information is shown in Table 3. Figure 6 shows the results of the ELISA verification. Among the proteins, the levels of GSTP1, CRP and TNC were reduced in the AH samples from the POAG combined with cataract group, the GDF11 level was increased in the POAG combined with cataract group, and no significant differences in PLOD1 and TGF-ß levels were observed between the two groups. The GSTP1 expression level was consistent with the proteomic data ( Figure 6).

| D ISCUSS I ON
In recent years, the application of omics technology in the field of biomedical research has become increasingly widespread, which has enhanced the data output capability of life science research.
Proteomics is an important approach used to explore differentially expressed proteins related to various diseases and has been widely applied in glaucoma, cataract, corneal lesions, macular degeneration, and uveitis and other diseases. [25][26][27][28][29][30][31] Compared with labelled protein profiling, unlabelled MS has the advantages of a lack of limitation on the sample size, the low cost of isotope labelling and the ability to detect a wide range of proteins. 32 The dynamic balance of the AH is closely related to IOP, which is an important risk factor for POAG. Changes in AH components also reflect tissue metabolism and pathological processes in the anterior segment of the eye. At the same time, because AH is relatively easy to obtain and convenient to store, it is a better sam-  of their enzyme activity. A meta-analysis on various GST mutations discovered that a polymorphism in the GSTP1 gene was significantly F I G U R E 3 GO analysis of significantly differentially expressed proteins between the POAG combined with cataract group and cataract group. The GO analysis mainly includes three aspects: cellular component (CC), molecular function (MF) and biological process (BP). In the cellular component category, the significantly differentially expressed proteins were mainly enriched in the cytoskeleton and nucleus and mainly located in the extracellular region, cytoplasm, membrane and mitochondrion. In the molecular function category, the significantly differentially expressed proteins were mainly enriched in rRNA binding, protein binding or bridging. Numerous proteins engaged in metal ion binding and receptor binding. Some also participated in signal transduction activity, transmembrane transporter activity and integrin binding. In the biological process category, significantly differentially expressed proteins were enriched in cell death, division, adhesion, growth, metabolism, cell-to-cell signal and cell motility. Many more proteins were involved correlated with increased POAG risk in a Caucasian population. 46 The GSTM1 null/GSTP1, Ile/Val or Val/Val genotypes were associated with increased IOP and more advanced defect of the right eye optic nerve and visual field. 47 The frequency of the GSTT1 and GSTP1 mutation was not statistically different between POAG patients and healthy controls group based on genomic DNA from peripheral blood. 48 In conclusion, the relationship between GSTP1 and POAG remains undetermined. Further, the above-mentioned studies made their conclusions based on gene level from patients' blood.
Our research found that GSTP1 protein expression was decreased ELISAs were not supported by our Nano-HPLC-MS data, this study provides a potential target or a meaningful insight into the association between POAG and GDF11, as well as CRP regulation, which is also of considerable significance.
Compared with a single multi-facility study, we further examined the literatures in search for proteins with functions similar to those we found. They analysed aqueous humour, used different testing methods, and studied those of different races. From these studies, we noted the differentially expressed proteins in POAG related to oxidative stress. The GST family was also implicated in these studies.
Other proteins, such as those involved in neuronal protection, have also attracted our attention. We believe that inflammation, oxidative stress and neuroprotection are candidate signalling pathways for future therapeutic development.
All patients were enrolled in strict accordance with the inclusion criteria to reduce bias. Based on clinical reality and medical ethics, we were unable to collect AH samples from healthy people at the clinic. Therefore, in this experiment, patients with POAG combined with cataract were enrolled as the study group and compared with patients with cataract alone as the control group. We must not completely ignore the changes in the AH of the cataract group compared with the healthy group. In addition, the medication status of the patients in the POAG group should not be ignored. Different patients took different medications but generally used ß-adrenergic receptor agonists and neuroprotective drugs. Drug interference cannot be ruled out, mainly for reasons of patient safety.
An imbalance in redox reactions is caused by various factors, leading to increased inflammation. Other factors promote changes in the levels of ECM proteins and increases in the ECM cross-linking of TM cells, resulting in blocked AH release, increased IOP and retinal nerve disruption, which eventually lead to the onset or progression of POAG. GSTP1, a redox-related protein, is expected to become a target for preventing the onset of or treating glaucoma.
Our follow-up study will focus on the biological functions of GSTP1 in TM cells under oxidative stress.

CO N FLI C T S O F I NTE R E S T
The authors confirm that no conflicts of interest exist.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.