A total of 18 Wistar rats (8 weeks old, male, weighing 220~260 g) were provided by Beijing Vital River Laboratory Animal Technology Co., Ltd. Before the experiments, all rats received 1 week of adaptive feeding, with four to six rats in each cage. Rats were raised with ad libitum access to water and food in a temperature-controlled room (22±2°C) with a 12 h light/dark cycle and a relative humidity of 40-60%. All animal studies were performed following the animal guidelines of the International Association for the Study of Pain (20), and approved by the Ethics Committee of the Second Hospital of Anhui Medical University (approval no. 2019-051; Hefei, China).

For establishment of glaucoma models, rats were anesthetized with intraperitoneal injection of 3% pentobarbital sodium (30 mg/kg; Sigma-Aldrich; Merck KGaA). The right eyes of rats underwent operation. 0.1 ml aqueous fluid was extracted from the anterior temporal horn of iris through a cannula, and 14 rats were slowly injected with 3% compound carbomer solution through the cannula, while 4 rats were treated with the same amount of normal saline. After administration, all rats were treated with 0.5% moxifloxacin three times a day for 3 days. IOP measurement was operated using a tonometer (Mentor O&O, Inc.). IOP was measured at 10 AM and 10 PM every day. A mean IOP was calculated from five automatically averaged measurements. A total of 12 rats with high IOP >30 mmHg were chosen. When the steadily increased IOP was achieved, four glaucoma model rats were injected with 5 µl antagomiR-29b-3p (Shanghai GenePharma Co., Ltd.), whereas the control rats were injected with 5 µl antagomiR-negative control (NC) (Shanghai GenePharma Co., Ltd.). Rats were anesthetized with an intraperitoneal injection of 30 mg/kg pentobarbital sodium and were sacrificed by decapitation.

The right eyeballs of rats were isolated after the rats were sacrificed, and then fixed in 4% paraformaldehyde for 24 h at room temperature. The cornea was incised along the sclera and the iris and lens were removed 30 min later. The retinal tissues were paraffin-embedded and cut into 5-µm sections for H&E staining assay. The sections were stained with hematoxylin for 2 min and stained with eosin for 10 min at room temperature (Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd.). After dehydration with gradient alcohol, the sections were sealed using neutral resin. The morphology of retinal tissues was observed under a light microscope at ×20 magnification (Olympus Corporation), and the results were quantified using ImageJ software (version 1.46; National Institutes of Health). The mean value was obtained from five randomly selected fields.

HTM cells (ScienCell Research Laboratories, Inc.) were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 15% fetal bovine serum, 2 mM L-glutamine, 0.05% gentamicin, 1 ng/ml FGF-2 and 0.25 µg/ml amphotericin B (all from ScienCell Research Laboratories, Inc.) at 37°C with 5% CO₂. The medium was replaced every day when the culture reached 70% confluence. To induce oxidative stress, cells were incubated in DMEM supplemented with H₂O₂ (Beyotime Institute of Biotechnology) at concentrations of 100, 200, 300, 400 µM for 24 h after reaching 85% confluence.

The mature miR-29b-3p mimic (for miR-29b-3p overexpression) and its NC mimic, miR-29b-3p inhibitor (for miR-29b-3p silencing) and its NC inhibitor, small interfering (si)RNA against E3 ubiquitin-protein ligase RNF138 (RNF138; siRNF138, for RNF138 knockdown) and its NC (si-NC) were constructed by Shanghai GenePharma Co., Ltd. The following sequences were included in the present study: NC mimic, 5′-CAG UAC UUU UGU GUA GUA CAA A-3′; miR-29b-3p mimic, 5′-UAG CAC CAU UUG AAA UCA GUG U-3′; NC inhibitor, 5′-UUU GUA CUA CAC AAA AGU ACU G-3′; miR-29b-3p inhibitor, 5′-UAG CAC CAU UUG AAA UCA GUG U-3′; si-NC, 5′-UAG AAG UUA ACU UCA CAG CAU -3′; and siRNF138, 5′-ACA UUU UCU ACA GAA AAC GUG -3′. HTM cells were seeded into the 6-well plates at a density of 1×10⁶ and subsequently transfected with miR-29b-3p mimic/inhibitor (50 nM), NC mimic/inhibitor (50 nM), siRNF138 and si-NC (100 ng) using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) for 48 h. The HTM cells were assigned into the following groups: i) Control group, cells without H₂O₂ stimulation; ii) H₂O₂ group, cells with 300 µM H₂O₂ stimulation; iii) H₂O₂ + NC group, cells with 300 µM H₂O₂ stimulation + NC; iv) H₂O₂ + miR-29b-3p mimic group, cells with 300 µM H₂O₂ stimulation + miR-29b-3p mimic; v) H₂O₂ + miR-29b-3p inhibitor group, cells with 300 µM H₂O₂ stimulation + miR-29b-3p inhibitor; vi) H₂O₂ + siRNF138 group, cells with 300 µM H₂O₂ stimulation + siRNF138; and v) H₂O₂ + miR-29b-3p inhibitor + siRNF138 group, cells with 300 µM H₂O₂ stimulation + miR-29b-3p inhibitor + siRNF138. The time interval between transfection and subsequent experimentation was 48 h.

TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) was employed for the extraction of total RNA from HTM cells. First-strand cDNA synthesis was performed using M-MLV reverse transcriptase (Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. qPCR was conducted using the TaqMan™ Universal Master Mix II (Applied Biosystems; Thermo Fisher Scientific, Inc.) in 7500 Fast Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.). For quantification of miRNAs, cDNA was obtained from 3 µg RNA using a specific stem-loop primer. The 2^−ΔΔCq method (21) was used for quantification and miRNA expression was normalized to U6, while GAPDH acted as the endogenous reference gene for mRNA. The following primers were used for RT-qPCR: miR-29b-3p forward, 5′-UAG CAC CAU UUG AAA UC-3′ and reverse, 5′-GTG CAG GTC CGA GGT -3′; U6 forward, 5′-CGCTTC GGC AGC ACA TAT AC-3′ and reverse, 5′-TTC ACG AAT TTG CGT GTC AT-3′; RNF138 forward, 5′-TAT AGT TCT GGC TCT CTG AA-3′ and reverse, 5′-CTG ACT TGG GTA AAT GCT CAA T-3′; collagen α-1(I) chain (COL1A1) forward, 5′-CCT GTC TGC TTC CTG TAA AC-3′ and reverse, 5′-ATG TTC GGT TGG TCA AAG ATA AAT -3′; collagen α-2(I) chain (COL1A2) forward, 5′-CAG TCG TAT GCG CGT ATA GC-3′ and reverse, 5′-CGT AGT CGT AGC TAG CTA GAG A-3′; and GAPDH forward, 5′-GGA GCG AGA TCC CTC CAA AAT-3′ and reverse, 5′-GGC TGT TGT CAT ACT TCT CAT GG-3′. The thermocycling conditions used 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.

The treated HTM cells were collected and washed with pre-cooling PBS. After centrifugation at 600 × g for 5 min at 4°C, the cells were lysed with radio immunoprecipitation assay buffer (Sigma-Aldrich; Merck KGaA). Then, cell lysates were centrifuged at 12,000 × g for 12 min at 4°C, followed by measurement of proteins in the supernatant using the BCA protein assay (Pierce; Thermo Fisher Scientific, Inc.). After proteins (20 µg per lane) were separated via sodium dodecyl sulfate-polyacrylamide gel electrophoresis on 10% gel, proteins were transferred to polyvinylidene difluoride membranes. Afterwards, the membranes were blocked with 5% non-fat dry milk at 4°C overnight, and incubated with the following primary antibodies at 4°C overnight: Anti-RNF138 (cat. no. K107111P; 1:5,000; Beijing Solarbio Science & Technology Co., Ltd.), anti-Bcl-2 (cat. no. ab185002; 1:1,000; Abcam), anti-Bax (cat. no. ab32503; 1:1,000; Abcam), anti-COL1A1 (cat. no. ab210966; 1:1,000; Abcam), anti-COL1A2 (cat. no. ab96723; 1:1,000; Abcam), anti-ERK (cat. no. ab115799; 1:1,000; Abcam), anti-phosphorylated (p)-ERK (cat. no. ab214036; 1:1,000; Abcam) and anti-β-actin (cat. no. ab179467; 1:1,000; Abcam). The membranes were subsequently cultured with the corresponding secondary antibodies (cat. no. ab205718; 1:5,000; Abcam) for 1.5 h at room temperature. The proteins were detected using the chemiluminescence detection (ECL Plus; Pierce; Thermo Fisher Scientific, Inc.). Semi-quantification was performed using ImageJ software (version 1.46; National Institutes of Health).

HTM cell viability was assessed using an MTT assay. After transfection and H₂O₂ stimulation, cells (1,000 cells per well) were seeded in the 96-well plates, followed by incubation with 50 µl MTT (5 mg/ml; Sigma-Aldrich; Merck KGaA) for 4 h. Then, 150 µl dimethylsulfoxide (Sigma-Aldrich; Merck KGaA) was used to dissolve the formazan deposited in cells for 10 min. The absorbance at 490 nm was detected using a microplate reader (Bio-Rad Laboratories, Inc.).

After transfection and H₂O₂ stimulation, HTM cells were washed in PBS and digested using EDTA-free trypsin. Then, cells were suspended in binding buffer at a density of 1×10⁶ cells/ml. Following which, 5 µl Fluorescein isothiocyanate-labeled Annexin V and 100 ng propidium iodide were added to the 500 µl suspended cells, followed by incubation for 15 min at room temperature in the dark. Cell apoptosis was determined using the Attune™ NxT Flow Cytometer (Invitrogen; Thermo Fisher Scientific, Inc.) and the apoptosis rate was analyzed with FlowJo software (version 10.0; FlowJo LLC). Apoptosis rate = apoptosis rate in quadrant (Q)2 + apoptosis rate in Q3.

ROS levels in HTM cells were examined using a ROS assay kit (cat. no. S0033M; Beyotime Institute of Biotechnology). After transfection and H₂O₂ stimulation, HTM cells were washed in PBS and subsequently cultured in DMEM containing 10 µM 2,7-dichlorodihydrofluorescein diacetate (DCFHDA) for 20 min at 37°C. After washing with PBS twice, the cells were treated with 0.25% trypsin-EDTA, followed by centrifugation at 800 × g for 6 min at 37°C. Then, cells were resuspended in 500 µl PBS. The fluorescence intensity, regarded as ROS generation, was measured by an Infinite M200 microplate reader (485 and 525 nm emission).

Targets of miR-29b-3p were predicted using miRanda version 3.3a (http://cbio.mskcc.org/microrna_data/miRanda-aug2010.tar.gz), microT-CDS DIANA Tools version 5.0 (http://diana.imis.athena-innovation.gr/DianaTools/index.php?r=microT_CDS/index) and RNA22 version 2.0 (https://cm.jefferson.edu/rna22/Interactive/).

The wild-type (Wt) and mutated (Mut) sequences of the 3′UTR of RNF138 complementary to miR-29b-3p were subcloned into the pmirGLO-luciferase plasmids (Promega Corporation) to generate Wt RNF138 (RNF138-Wt) and Mut type RNF138 (RNF138-Mut). RNF138-Wt and RNF138-Mut reporters were co-transfected with NC or miR-29b-3p mimic into 293T cells (American Type Culture Collection) using Lipofectamine 3000. The interaction between miR-29b-3p and 3′-UTR of RNF138 was detected using a Luciferase Assay system (Promega Corporation) 48 h post-transfection. Relative luciferase activity was defined as the ratio of firefly luciferase activity to Renilla luciferase activity.

Data are presented as the mean ± standard deviation of three independent experiments and analyzed using the SPSS 21.0 software (IBM Corp.). Comparisons between 2 groups were analyzed using unpaired Student's t-test, while comparisons among ≥3 groups were analyzed using one-way analysis of variance. Dunnett's or Tukey's post hoc tests were used following one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference.

As shown in Fig. 1A, in the control group, there was no significant difference between the mean and max IOP; however, the max IOP was significantly higher than the mean IOP in the glaucoma group. Moreover, RT-qPCR revealed a higher expression of miR-29b-3p in the glaucoma group compared with the control group (Fig. 1B). As presented in Fig. 1C, miR-29b-3p expression levels were decreased in rats following injection with antagomiR-29b-3p compared with the NC group. Next, the functions of antagomiR-29b-3p in a glaucoma model were investigated. H&E staining revealed that the thickness of each layer of the retinal, including the outer nuclear layer (ONL), inner nuclear layer (INL) and ganglion cell layer (GCL), were significantly thinner in the glaucoma model compared with the control group (Fig. 1D and E), and transfection with antagomiR-29b-3p increased the thickness.

HTM cells were treated with different concentrations of H₂O₂ (100, 200, 300 and 400 µM) for 24 h and concentration of H₂O₂ at 0 µM was used as the control. The data from MTT assay revealed that 100-400 µM of H₂O₂ inhibited cell viability in a concentration-dependent manner compared with the control (Fig. 2A). Since 300 µM of H₂O₂ decreased >50% of cell viability, 300 µM H₂O₂ was used for induction of oxidative damage in further assays. Subsequently, the apoptosis rate and ROS generation of HTM cells were detected. Flow cytometry results showed that the H₂O₂ group exhibited a significant increase of cell apoptosis (Fig. 2B). Similarly, the reduced level of Bcl-2 (anti-apoptotic marker) and the elevated level of Bax (pro-apoptotic marker) were revealed in the H₂O₂ group (Fig. 2C), implying that the stimulation of H₂O₂ promoted HTM cell apoptosis. As presented in Fig. 2D, H₂O₂ stimulation increased the level of ROS compared with the control.

As presented in Fig. 3A, miR-29b-3p expression was at a higher level in the H₂O₂ group than in the control group. miR-29b-3p expression was increased by transfection with miR-29b-3p mimic compared with the control group. Under transfection with miR-29b-3p inhibitor, miR-29b-3p expression was 34% of the control group. MTT assay revealed that the H₂O₂ + miR-29b-3p mimic group displayed decreased cell viability, while the H₂O₂ + miR-29b-3p inhibitor group showed the opposite result (Fig. 3B). Flow cytometry showed that the apoptosis rate of HTM cells was higher in the H₂O₂ + miR-29b-3p mimic group, and lower in the H₂O₂ + miR-29b-3p inhibitor group, compared with the H₂O₂ + NC group (Fig. 3C). Likewise, Bcl-2 expression was downregulated and Bax was upregulated in the H₂O₂ + miR-29b-3p mimic group, whereas the miR-29b-3p inhibitor exerted opposite effects (Fig. 3D). Furthermore, the H₂O₂ + miR-29b-3p mimic group exhibited an increased level of ROS and the H₂O₂ + miR-29b-3p inhibitor group presented the opposite trend (Fig. 3E). Additionally, the expression levels of COL1A1 and COL1A2 were determined by RT-qPCR and western blotting. As presented in Fig. 3F and G, the mRNA and protein levels of COL1A1 and COL1A2 were higher in the H₂O₂ groups than in the control group. miR-29b-3p mimics increased COL1A1 and COL1A2 levels, whereas the miR-29b-3p inhibitor decreased COL1A1 and COL1A2 levels in HTM cells under oxidative injury.

As displayed in Fig. 4A, there were 9 potential targets of miR-29b-3p. RNF138 expression showed the most significant downregulation after miR-29b-3p overexpression using miR-29b-3p mimics (Fig. 4B). The predicted binding sites between miR-29b-3p and RNF138 are exhibited in Fig. 4C. To confirm the direct binding between miR-29b-3p and RNF138, a luciferase reporter assay was performed. The luciferase activity of RNF138-Wt was reduced by miR-29b-3p mimic, while that of RNF138-Mut reporters was not significantly altered in response to the overexpression of miR-29b-3p (Fig. 4D), indicating that RNF138 3′ UTR was directly targeted by miR-29b-3p. In addition, the downregulated expression of RNF138 in HTM cells was found in the H₂O₂ group compared with control group (Fig. 4E). RNF138 expression was decreased in the H₂O₂ + miR-29b-3p mimic group and elevated in the H₂O₂ + miR-29b-3p inhibitor group (Fig. 4E). Western blotting showed that miR-29b-3p overexpression decreased the level of RFN138 protein, whereas miR-29b-3P downregulation had the opposite result (Fig. 4F).

The silencing efficiency of RNF138 was confirmed via RT-qPCR and western blotting (Fig. 5A). Knockdown of RNF138 significantly inhibited viability (Fig. 5B), promoted apoptosis (Fig. 5C and D), increased ROS level (Fig. 5E) and ECM deposition (Fig. 5F) in HTM cells under oxidative injury. Moreover, siRNF138 significantly rescued the effects of miR-29b-3p inhibitor on H₂O₂-stimulated HTM cells.

As presented in Fig. 6, the protein level of p-ERK was lower in the H₂O₂ group than in the control group. Compared with the H₂O₂ + NC group, the H₂O₂ + miR-29b-3p inhibitor group displayed increased ratio of p-ERK/t-ERK, whereas the H₂O₂ + siRNF138 group showed the opposite results. Moreover, the effect of miR-29b-3p inhibitor on the ERK pathway was significantly rescued by siRNF138.