PRGCs were isolated using Thy1.2-conjugated magnetic beads from the retinas of 20 male BALB/c mice (weighing 20-30 g, 8-weeks old) obtained from the Laboratory Animal Center of The First Affiliated Hospital of Xinxiang Medical University (Xinxiang, China) as previously described (21). All mice were maintained in a temperature-controlled room (22±2°C) with a 12-h light/dark cycle and a relative humidity of 40-60%, and had free access to food and water. For PRGC culture, tissue culture plates were coated with poly-D-lysine (10 µg/ml; EMD Millipore) and laminin (10 µg/ml; Sigma-Aldrich; Merck KGaA) at 37°C in a humidified tissue culture incubator with 5% CO₂. Primary cells were identified by immunofluorescence staining with neurofilament-L (NF-L; cat. no. 2835; Cell Signaling Technology, Inc.) and reverse transcription-quantitative PCR (RT-qPCR). Western blotting for the expression of the RGC-specific markers, including Brn3a (cat. no. MAB1585 clone 5A3.2; EMD Millipore), thymus cell antigen 1 (Thy-1; cat. no. MAB1406; EMD Millipore) and NF-L was conducted as described previously (22,23). The use of animals and the experimental protocols performed were approved by the Animal Care Committee of The First Affiliated Hospital of Xinxiang Medical University (approval no. 2017-0163) in accordance with institutional guidelines. TMP hydrochloride (100 µM) was purchased from Harbin Medisan Pharmaceutical Co., dissolved in normal saline and diluted immediately prior to each experiment.

The PRGCs were cultured in DMEM supplemented with 10% fetal bovine serum (HyClone; GE Healthcare Life Sciences), 100 U/ml penicillin and streptomycin at 37°C in a humidified atmosphere with 5% CO₂. The PRGCs were treated with different concentrations H₂O₂ (0, 100 200, 400 and 800 µM) at 37°C for 16 h. Cell viability was tested to investigate cell injury induced by H₂O₂ in PRGCs, and 400 µM H₂O₂ was selected to establish the model of oxidative stress injury in PRGCs.

PRGCs were pre-treated with TMP at different dosages (25, 50 and 100 µM) at 37°C for 24 h. The cells were then subjected to oxidative insult with H₂O₂ (400 µM) at 37°C for 16 h and collected for subsequent experiments. Blank cells comprised untreated cells and the control or vehicle was that of cells treated with H₂O₂ or H₂O₂ + DMSO (1% v/v), respectively.

PRGCs (1×10⁴ cells) were seeded into 96-well plates and cultured overnight at 37°C in a humidified tissue culture incubator with 5% CO₂. At the end of treatment, cell viability was determined using Cell Counting Kit-8 (CCK-8) assay (Dojindo Molecular Technologies, Inc.). Briefly, 10 µl CCK-8 reagent (Dojindo Molecular Technologies, Inc.) was added to each well, then the cells were incubated for 3 h at 37°C. The absorbance of the samples was read at 450 nm with a microplate reader (Sunrise™; Tecan Group, Ltd.).

At the end of TMP treatment, the activity of caspase-3 was measured with a caspase-3 assay kit (Abcam) according to the manufacturer's protocols. The samples were analyzed using a microplate reader (Model 680, Bio-Rad Laboratories, Inc.) at 405 nm.

PRGCs were seeded in 6-well plates at a density of 1.0×10⁶ cell/well. Following treatment with TMP, the cells were washed twice with PBS, and then fixed in 70% ice-cold ethanol in PBS at 4°C for 30 min. Subsequently, the cells were stained with 5 µl Annexin V-fluorescein isothiocyanate and 1 µl of propidium iodide (Bio-Science, Co. Ltd.). After incubation for 15 min at room temperature in the dark, cell apoptosis was analyzed with a FACScan flow cytometer (FCM; Beckman Coulter, Inc.). FlowJo software version 7.6.1 (FlowJo LLC, USA) was used to analyze flow cytometry data.

ROS production in PRGCs was analyzed using 2′,7′-dichlorofluorescin-diacetate (DCHF-DA; cat. no. D6883, Sigma-Aldrich; Merck KGaA). Briefly, PRGCs were incubated with 10 µM DCHF-DA for 30 min at 37°C, followed by two washes with PBS. Then, the DCFH-DA staining for the detection of ROS production was observed using a fluorescence microscope (Nikon Corporation). Fluorescence was read at 485 nm for excitation and 530 nm for emission with an Infinite M200 Microplate Reader (Tecan Group, Ltd.).

The concentrations of malondialdehyde (MDA) and superoxide dismutase (SOD) in the conditioned media were analyzed by an MDA Assay kit (cat. no. MAK085) and SOD Assay kit (cat. no. 19160) from Sigma-Aldrich (Merck KGaA) according to the manufacturer's protocols.

Following treatment, total RNA was extracted from retinal tissues or cultured cells using a miRNAeasy mini kit (Qiagen, Inc.) according to the manufacturer's protocols. A total of 200 ng of RNA was reverse-transcribed with a miRNA reverse transcription kit (Qiagen, Inc.) and an mRNA RT kit (Invitrogen; Thermo Fisher Scientific, Inc.), respectively. qPCR was performed using the iTaqTM Universal SYBR Green Supermix (Bio-Rad Laboratories, Inc.) on a 7500HT Real-Time PCR System (Thermo Fisher Scientific, Inc.). Relative expression levels were quantified via normalization to small nuclear (sno)RNA202. The primers employed for qPCR were as follows: miR-182 5′-TGC GCT TTG GCA ATG GTA GAA CTC-3′ (forward) 5′-CCA GTG CAG GGT CCG AGG TAT T-3′ (reverse); miR-34a 5′-ACA CTC CAG CTG GGT GGC AGT GTC TTA GCT-3′ (forward), 5′-CTC AAC TGG TG TCG TGG A-3′ (reverse); miR-150 5′-TCT CCC AAC CCT TGT ACC AGT G-3′ (forward) 5′-CTC AAC TGG TGT CGT G G TA-3′ (reverse); miR-214 5′-ATA GAA TTC TTT CT CCC TTT CCC CTT ACT CTC C-3′ (forward) 5′-CCA GGA TCC TTT CAT AGG CAC CAC TCA CTT TAC-3′ (reverse); miR-137 5′-GCA GCA AGA GTT CTG GTG GC-3′ (forward), 5′-TGG AAC CAG TGC GAA AAC AC-3′ (reverse); miR-210 5′-GTG CAG GGT CCG AGG T-3′ (forward), 5′-CTG TGC GTG TGA CAG CGG CT GA-3′ (reverse); snoRNA202 5′-GTC GTA TCC AGT GCA GGG TCC GAG GTA TTC GCA CTG GAT ACG ACC ATC AG-3′ (RT), 5′-GCA GGG TCC GAG GTA TTC-3′ (forward) and 5′-CTT GAT GAA AGT ACT TTT GA-3′ (reverse). Brn3a 5′-TGG CGT CCA TCT GCG ATC-3′ (forward); 5′-CTC AGG TTG TTC ATT TTT C-3′ (reverse). Thy-1 5′-CGC TTT ATC AAG GTC CTT ACT C-3′ (forward) 5′-GCG TTT TGA GAT ATT TGA A G GT-3′ (reverse). NF-L 5′-ATG CTC AGA TCT CCG TGG AG A TG-3′ (forward); 5′-GCT TCG CAG CTC ATT CTC CAG TT-3′ (reverse). and GAPDH 5′-CAT CAA GAA GGT GGT GAA GCA GG-3′ (forward); 5′-CCA CCA CCC TGT TGC TGT AG C CA-3′ (reverse). The reaction conditions were as follows: 94°C for 5 min, followed by 40 cycles of 94°C for 1 min, 56°C for 1 min and 72°C for 1 min. RT-qPCR assays were performed in triplicate and the changes in expression levels were calculated using the 2^−ΔΔCq method (24).

PRGCs (5×10³) were seeded into each well of a 6-well plate for 24 h, and the cells were subsequently transfected with miR-182 mimics, miR-182 inhibitor or the corresponding control vectors synthesized by Shanghai GenePharma Co., Ltd., at a final concentration of 50 nM using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer′s instructions. The sequences are as follows: miR-182 mimics, 5′-UUU GGC AAU GGU AGA ACU CAC ACU-3′; mimics negative control (NC), 5′-UUC UCC GAA CGU GUC ACG UTT-3′; miR-182 inhibitor, 5′-AGU GUG AGU UCU ACC AUU GCC AAA-3′ and inhibitor NC, 5′-CAG UAC UUU UGU GUA GUA CAA-3′. After 24 h post-transfection, the cells were pre-treated with 100 µM TMP for 24 h, then subjected to oxidative insult with H₂O₂ (400 µM) at 37°C for 16 h and collected for subsequent experiments.

TargetScan (version 7.0; www.targetscan.org/) and PicTar (release 2006; https://pictar.mdc-berlin.de) target gene prediction software were used to select MIF as a target gene of miR-182.

The Bcl-2 interacting protein 3 (BNIP3) 3′-untranslated region (UTR) containing complementary sequences or the mutated sequence for the seed sequence of miR-182 was amplified by PCR as described previously (25), and cloned into the firefly luciferase expressing vector, pGL3 (Promega Corporation); the vectors were denoted as wild-type (wt) pGL3-BNIP3-3′-UTR and mutant (mut) BNIP3-3′-UTR, respectively. For the luciferase reporter assay, 293T cells (American Type Culture Collection) were seeded into a 24-well plate (5.0×10⁵/well) and transfected with the wt or mut reporter vector, together with miR-182 mimics or miR-182 inhibitor using Lipofectamine 2000. At 48 h after transfection, the luciferase activity was determined by using a dual-luciferase reporter assay system (Promega Corporation). The pRL-TK plasmid (Promega Corporation) was used as a normalizing control. All experiments were performed in triplicate.

Total protein was extracted from PRGCs or retinal tissues using radioimmunoprecipitation assay lysis (Sigma-Aldrich; Merck KGaA) after treatments. The protein concentration was determined using a BCA protein assay kit (Pierce; Thermo Fisher Scientific, Inc.). Total protein samples (30 µg) were analyzed by 8% SDS-PAGE and transferred to a polyvinylidene fluoride membrane (EMD Millipore). β-actin was used for the normalization of protein expression. The membranes were blocked with 5% non-fat milk at 4°C overnight, and incubated with primary antibodies overnight at 4°C. Primary antibodies against Bcl-2-assoiated X protein (Bax; cat. no. sc-4239, 1:1,000), Bcl-2 (cat. no. sc-176463, 1:1,000), BNIP3 (cat. no. sc-1715, 1:1,000) and β-actin (cat. no. sc-8432, 1:2,000) were purchased from Santa Cruz Biotechnology, Inc., while cleaved (cle)-caspase-3 (cat. no. 9661, 1:1,000), cle-poly(ADP-ribose)polymerase (PARP; cat. no. 5625, 1:1,000), Cytochrome c (cyto c cat. no. 4280, 1:1,000) and Complex IV (cat. no. 4850, 1:1,000) were purchased from Cell Signaling Technology, Inc. After washing with PBS, the membrane was incubated with horseradish peroxidase-conjugated antibody (cat. no. ab205718; 1:2,000, Abcam) for 1 h at room temperature, and the bands were detected with an ECL Advance reagent (GE Healthcare). The intensity of the bands of interest was analyzed with ImageJ software version 1.46 (National Institutes of Health).

After treatments, a commercial Mitochondrial Membrane Potential Assay kit with JC-1 (Invitrogen; Thermo Fisher Scientific, Inc.) was used to determine the mitochondrial membrane potential (ΔΨm) according to the manufacturer's instructions. Briefly, after washing twice with PBS, PRGCs were stained with JC-1 (5 µg/ml) for 15 min at 37°C. The red and green fluorescence intensities were detected (excitation wavelength of 488 nm and emission wavelength at 500 nm) using a Bio-Tek fluorescent microplate reader (BioTek Instruments, Inc.). The ΔΨm of the PRGCs in each treatment group was calculated as the ratio of red to green fluorescence and expressed as a multiple of the level in the control group.

SPSS 13.0 software (SPSS, Inc.) was used to analyze the data. Data were expressed as the mean ± standard deviation of three independent experiments. A Student's t-test was used to analyze differences between two groups. Differences between multiple groups were analyzed by one-way analysis of variance, followed by a Tukey's post-hoc test for multiple comparisons. P<0.05 was determined to indicate a statistically significant difference.

At days 3 and 7 of the first passage, PRGCs were collected for immunofluorescence analysis of an RGC marker. Membranous staining of NF-L was observed in the soma and neurites (Fig. 1A). To verify the purity of the isolated PRGC population, the expression levels of NF-L, Thy-1 and Brn3a (RGC markers) were detected by RT-qPCR. As presented in Fig. 1B-D, the expression levels of these markers were significantly increased in the isolated PRGCs compared with in retinal tissue. Similar results were observed via western blotting (Fig. 1E). These data demonstrated that the isolated neoplastic stromal cell populations were not contaminated by histiocytes or giant cells.

H₂O₂ is widely used to establish an oxidative injury model using RGCs to mimic the development of glaucoma in vitro (26,27). As presented in Fig. 1F, H₂O₂ significantly suppressed the viability of PRGCs in a dose-dependent manner; however, no significant difference in viability between 400 and 800 µM was observed. Based on previous reports (17,28,29), we used 400 µM H₂O₂ to generate a cell model of oxidative injury. To determine whether TMP promotes the growth of PRGCs, we treated cells with 25, 50 and 100 µM TMP for 24 h, followed by 400 µM H₂O₂. Subsequently, cell viability was evaluated by a CCK-8 assay. The results showed that TMP alleviated H₂O₂-induced suppression of cell viability, and this effect was does-dependent (Fig. 1G). Further investigation revealed that H₂O₂ significantly increased the activity of caspase-3 compared with in the control group, whereas TMP attenuated the promoting effects of H₂O₂ on caspase-3 activity (Fig. 1H). Consistent with these results, TMP also significantly attenuated H₂O₂-induced cell apoptosis (Fig. 1I). Of note, 100 µM TMP was used for subsequent experiments and this concentration has also been used previously (17,28,29). Collectively, these results indicated that the oxidative injury model of PRGCs was successfully established and TMP treatment could improve H₂O₂-induced cell damage.

It has been acknowledged that glaucoma is associated with oxidative stress, which can lead to the apoptosis of RGCs (4,5). MDA is a biomarker of oxidative stress of cells; SOD is the main antioxidative enzyme and can resist the damage of oxygen-free radicals to cells (30). Therefore, in the present study, the levels of ROS, MDA and SOD were examined to assess the protective effects of TMP in H₂O₂-treated PRGCs. As presented in Fig. 2A and B, H₂O₂ treatment significantly increased intracellular ROS levels compared with in Blank cells. However, pretreatment with TMP significantly decreased ROS levels in a dose-dependent manner. It was also observed that H₂O₂ treatment significantly increased the levels of MDA and decreased the levels of SOD compared with that in Blank cells, but these effects were significantly attenuated by TMP (Fig. 2C and D). Our findings suggested that TMP attenuated H₂O₂-induced cell damage by reducing oxidative stress.

Recently, ROS has been reported to alter the expression of certain miRNAs (31,32). Given TMP suppressed the accumulation of ROS, we hypothesized that TMP exerts its protective effects against H₂O₂-induced cell injury via the regulation of miRNAs. Thus, we investigated the regulatory effects of TMP on the changes in the expression of miRNAs, which have been reported be altered in response to ROS exposure previously (33). The results of RT-qPCR showed that H₂O₂ significantly increased the expression of miR-214 and miR-210, and decreased the expression of miR-182 and miR-137, compared with control group, which corresponds with previous studies (34-37). Interestingly, the expression levels of miR-150 and miR-34a were markedly unaltered after H₂O₂ treatment, which differs with previous reports (38,39); this may be due to the use of different cell lines. Of note, the expression of miR-182 in H₂O₂-treated PRGCs was significantly downregulated compared with the control, but increased in response to TMP; however, no significant changes were observed in the expression of miR-214, miR-137 and miR-210 following TMP treatment (Fig. 3A-F). These findings prompted us to investigate the role of miR-182 in TMP-mediated protective effects against H₂O₂-induced cell damage.

Previous studies have been reported that miR-182 possesses potent antioxidative and antiapoptotic activity in a variety of diseases (33,35). To investigate whether ectopic expression of miR-182 involves the protective effects of TMP on H₂O₂-induced cell damage, miR-182 inhibitor or miR-182 mimics were transfected into PRGCs prior to H₂O₂ and TMP treatment. As presented in Fig. 4A, miR-182 expression was significantly decreased and increased after miR-182 inhibitor or miR-182 mimics transfection in PRGCs, respectively. It was observed that miR-182 knockdown significantly alleviated the protective effects of TMP in H₂O₂-treated PRGCs, by reducing cell viability and inducting apoptosis (Fig. 4B and C). The intracellular ROS levels in the H₂O₂ + TMP + miR-182 inhibitor group were significantly higher than H₂O₂ + TMP group (Fig. 4D). In addition, as shown in Fig. 4E and F, the levels of MDA were significantly increased and the levels of SOD were decreased in the H₂O₂ + TMP + miR-182 inhibitor group, compared with that of the H₂O₂ + TMP group. These findings suggested that miR-182 knockdown suppressed the protective effects of TMP in H₂O₂-induced injury.

To explore the molecular mechanism by which miR-182 functions in the protective effects of TMP in H₂O₂-induced injury, we performed TargetsScan and PicTar analyses to predict the target genes of miR-182 and identified BNIP3 as a potential target of miR-182, with the target site located in the 3′-UTR (Fig. 5A). To validate this bioinformatic predication, we established the luciferase reporter plasmids containing the wt or mut 3′-UTR segments of BNIP3. The luciferase reporter assay revealed that miR-182 mimics significantly inhibited the luciferase activity compared with the mimic NC, while miR-182 inhibitor significantly enhanced the luciferase activity compared with the inhibitor NC (Fig. 5B). Additionally, miR-182 mimics or inhibitor did not markedly affect luciferase activity when the targeted BNIP3 sequence was mutated in the miR-182-binding site (Fig. 5B). To further confirm that BNIP3 expression is regulated by miR-182, BNIP3 protein expression was analyzed by western blotting. We found that the expression levels of BNIP3 were notably downregulated by miR-182 mimics, but markedly enhanced by miR-182 inhibitor (Fig. 5C). Taken together, these findings indicated that miR-182 inhibit the expression of BNIP3, further suggesting that the miR-182/BNIP3 axis may be involved in the protective effects of TMP in H₂O₂-induced injury.

BNIP3 is a well-known effector of the mitochondria-mediated apoptosis, which induces the formation of the pathological mitochondrial permeability transition pore (40,41). Thus, we sought to determine whether TMP could regulate the mitochondrial apoptotic pathway via the miR-182/BNIP3 axis. Western blotting was performed to detect the expression levels of apoptosis-related proteins in PRGCs transfected with miR-182 inhibitor prior to H₂O₂ and TMP treatment. As presented in Fig. 6A, H₂O₂ treatment significantly increased the expression of pro-apoptotic proteins (BNIP3, Bax, cle-caspase-3 and cle-PARP) and decreased that of anti-apoptotic Bcl-2, compared with the control group. Of note, the levels of these pro-apoptotic proteins were suppressed, whereas the anti-apoptotic protein was upregulated after TMP pretreatment. However, these effects of TMP were attenuated by miR-182 inhibitor. cyto c release from mitochondria into the cytosol is a critical event in apoptosis (42). Therefore, we further detected the effects of TMP on cyto c release. The data showed that H₂O₂ treatment induced the release of cyto c from the mitochondria, which was attenuated by TMP. However, the inhibitory effects of TMP were significantly reversed by knockdown of miR-182 (Fig. 6B). In addition, whether TMP could reduce the H₂O₂-induced Δψm loss was investigated as dysregulated mitochondrial function causes the loss of Δψm. As presented in Fig. 6C, H₂O₂ treatment resulted in decreased red/green fluorescence ratio, which indicated that Δψm had dissipated, whereas TMP pretreatment attenuated H₂O₂-induced Δψm loss. However, knockdown of miR-182 significantly suppressed the promoting effects of TMP on Δψm. These results suggest that TMP may attenuate H₂O₂-induced PRGC apoptosis via the miR-182/mitochondrial apoptotic pathway.