A total of 108 adult male Sprague-Dawley (SD; 250–280 g; 7–8 weeks) rats were purchased from the Animal Center of Guangzhou University of Chinese Medicine (Guangzhou, China). The experiments were approved by the Care and Use of Experimental Animals Committee of Guangzhou University of Chinese Medicine and performed according to the National Institute of Health Guide for the Care and Use of Laboratory Animals. All animals were allowed to acclimate for 1 week prior to MCAO surgery. Upon receipt and throughout the study, the animals were conditioned to the same environment during the study. Food and water were available ad libitum. Room humidity (50–70%) and temperature (21–25°C) were maintained consistently. In addition, the study room was maintained on a 12-h light/dark cycle.

Under anesthesia (100 mg/kg ketamine and 10 mg/kg xylazine, intraperitoneal injection), rats were subjected to 2 h MCAO surgery using intraluminal suture as described in a previous report (28). Body temperature was kept at ~36.5°C with a heating pad during the surgical process until the animal recovered from anesthesia. The animals were allowed to regain full consciousness in a nursery cage, which was heated (~30°C) and oxygenated prior to returning to their normal cage. After MCAO surgery, rats immediately underwent ICV of miR-335 mimic and inhibitor under anesthesia. The rats were fixed and one small hole was drilled into the skull of the ischemic hemisphere, in order to deliver the miR-335 mimic and inhibitor into the lateral ventricle. The stereotaxic coordinates for ICV injection were as follows: −0.9 mm anterior, 1.8 mm lateral and −3.8 mm from the bregma. A total of 48 male SD rats were randomly divided into four groups (n=12 rats/group): Control group (normal saline), Model group (MCAO + normal saline), Mimic group (MCAO + miR-335 mimic, 100 pmol/rat) and Inhibitor group (MCAO + miR-335 inhibitor, 200 pmol/rat). miR-335 mimic (micrON™ miRNA agomir; 5′-ucaagagcaauaacgaaaaaugu-3′) and miR-335 inhibitor (micrOFF™ miRNA antagomir; 5′-acauuuuucguuauugcucuuga-3′) were purchased from Guangzhou RiboBio Co., Ltd. (Guangzhou, China).

To observe the dynamic alterations in SG formation, levels of apoptosis and infarct volume in the MCAO model, a total of 36 male SD rats were divided into six groups (n=6 rats/group): Control group, 0 h post-reperfusion group, 6 h post-reperfusion group, 12 h post-reperfusion group, 24 h post-reperfusion group and 36 h post-reperfusion group. Rats from each group were sacrificed with CO₂ asphyxiation, and the brains were removed and coronally sectioned into six slices (2 mm/slice). The brain slices were stained in a 2% solution of TTC (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) for 20 min at 37°C; the unstained area was defined as the infarct area and the infarct area volume was calculated with ImageJ software (Version 14.8; National Institutes of Health, Bethesda, MD, USA). Considering edema of the infarct hemisphere, the percentage of infarct area was calculated as follows: Infarct area (%) = [contralateral hemisphere area-(ipsilateral hemisphere area-infarct area)]/contralateral hemisphere area ×100%.

Following 24 h of reperfusion, rats were sacrificed by CO₂ asphyxiation. The cortices were washed with cold normal saline and fixed with 4% paraformaldehyde solution at 4°C overnight. Frozen sections (20 µm thickness) were used in TUNEL and immunohistochemical staining. TUNEL staining was conducted following the manufacturer's protocol (Boster Biological Technology, Ltd., Wuhan, China). The percentage of positive cells in the pathological images captured was calculated with Image Pro Plus software (version 6.0; Media Cybernetics, Inc., Rockville, MD, USA). The histopathological observations were documented using a light microscope (Leica DMI400; Leica Microsystems GmbH, Wetzlar, Germany) at ×400 magnification and photographed. Each experiment was replicated 3 times and the data were presented as the mean ± standard error of the mean (SEM).

Following peroxidase blocking for 30 min and 5% bovine serum albumin (BSA; Sigma-Aldrich; Merck KGaA) blocking for 2 h at room temperature, frozen sections (20 µm) of the brain were incubated with primary antibody (cat. no. ab140595; rabbit anti-TIA1 antibody 1:1,000; Abcam, Cambridge, MA, USA) overnight. Then these sections were incubated with secondary antibody (cat. no. ab6721; horseradish peroxidase-conjugated goat anti-rabbit; 1:5,000; Abcam) for 2 h at room temperature. The expression of TIA1 was detected using a Streptavidin Peroxidase Immunohistochemical staining kit (Tiangen Biotech Co., Ltd., Beijing, China) with DAB staining. The ratio of positive cells was calculated using Image Pro Plus software (version 6.0; Media Cybernetics, Inc.). The histopathological observations were documented by light microscopy (Leica DMI400; Leica Microsystems) at 400× magnification and images were captured. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

A total of 24 male SD rats were randomly divided into control and model groups (n=12 rats/group) to perform miRNA microarray assay and reverse transcription-quantitative polymerase chain reaction (RT-qPCR). The cerebral cortices were removed from the Control and Model group rats for the miRNA microarray assay. Total RNA (600 ng) was extracted from the cortices of the different groups using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). The subsequent steps of the experiment were completed by Guangzhou RiboBio Co., Ltd., according to the manufacturer's protocol. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

Total RNA was extracted from the cerebral cortex using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The concentration and purity of RNA were determined using spectrophotometers (NanoDrop ND-2000; Thermo Fisher Scientific, Inc.). The specific stem-l oop primer of miR-335 was used in RT-PCR and the U6 gene was used as the reference. The following primers were used: miR-335, RT primer 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGATACGACACATTT-3′, forward 5′-CGGCGCTCAAGAGCAATAACGAA-3′, and reverse 5′-ATCCAGTGCAGGGTCCGAGG-3′; U6, RT primer 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACAAAATA-3′, forward 5′-AGAGAAGATTAGCATGGCCCCTG-3′, and reverse 5′-ATCCAGTGCAGGGTCCGAGG-3′. The reverse transcription was performed by the PrimeScript RT reagent kit (Takara, Biotechnology, Co., Ltd., Dalian, China) according to the manufacturer's protocol, at 37°C for 15 min and 85°C for 30 sec. Light Cycler 480 SYBR Green I Master (Roche Diagnostics, GmbH, Mannheim, Germany) was applied for fluorescent qPCR. The amplification process was as follows: Denaturation at 95°C for 10 min, followed by 35 cycles of denaturation at 95°C for 15 sec, annealing at 60°C for 30 sec and extension at 72°C for 1 min. Detection of PCR products was performed using the CFX96 RT-qPCR instrument (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Each experiment was replicated 3 times and the data were presented as the mean ± SEM.

Bioinformatics software (TargetScan, www.targetscan.org; miRBase, www.mirbase.org) was used to predict the target gene of miR-335 and a dual luciferase reporter system was used to confirm the targeting. First pLUC-ROCK2-wild-type (WT) and pLUC-ROCK2-mutant (MUT) plasmid (cat. no. U5679-02, U5679-03; Shenzhen Huaan Ping Kang Bio Technology, Inc., Shenzhen, China) was constructed. Using Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.), PC12 cells [American Type Culture Collection, (ATCC), Manassas, VA, USA] were transfected with pLUC-ROCK2 cloning vectors (200 ng), miR-335 mimics (100 nM) or miR-335 negative control (NC; 100 nM) for 6 h in 24-well plates (5×10⁴ cells per well). After 24 h transfection, the fluorescent value of each well was quantified with Dual-Luciferase^® Reporter assay kit (Promega Corporation, Madison, WI, USA), according to the manufacturer's protocol. In the pLUC-ROCK2 plasmid, the target site of rno (Rattus norvegicus)-miR-335 was inserted into the downstream sequence of Renilla luciferase (hRluc). Firefly luciferase was used as the internal reference and for normalization. PC12 cells transfected with pLUC-ROCK2 vector were used as the positive control. Cells transfected with miR-335 NC and pLUC-ROCK2 were regarded as the NC. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

PC12 cells were purchased from ATCC and were cultured as described previously (29). The serum-free cell model was established by culturing cells in serum-free medium (RPMI-1640 medium; Gibco; Thermo Fisher Scientific, Inc.) for 0, 6, 12, 18, 24 or 30 h. With arsenite stress serving as the positive control treatment, cells were incubated with sodium arsenite (SA; Sigma-Aldrich; Merck KGaA) at the indicated concentrations (0.5 mM) for 50 min as previously described (3). The serum-free cell model was only stimulated with serum-free culture medium and did not include the arsenite stress step. After PC12 cells were stimulated for 0–30 h to ascertain the appropriate duration of stimulation, the expression of TIA1 was detected by immunofluorescent assay as the hallmark of SG production. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

PC12 cells were grown on glass coverslips in 24 well-plates at 1×10⁵ cells per well. According to the experimental protocol, cells were transfected or co-transfected with miR-335 mimic, miR-335 inhibitor, small interfering (si)-RNA-ROCK2 and pLUC-ROCK2 plasmids, and then stimulated with serum-free medium for 24 h. Following fixation with 4% paraformaldehyde for 30 min at room temperature, PC12 cells were permeabilized with 0.3% Triton X-100 and blocked with 5% BSA. Then cells were incubated with diluted primary antibody (cat. no. 12133-2-AP; rabbit anti-TIA1 antibody; 1:1,000; ProteinTech Group, Inc.) at 4°C overnight. and incubated with the secondary antibody (cat. no. ab150077; Alexa Fluor488-conjugated goat anti-rabbit antibody; 1:5,000; Abcam) for 1 h in the dark. The PC12 cells were rinsed three times with PBS and stained with DAPI solution (BIOSS, Beijing, China) for nuclear labeling. The staining results were observed with a confocal laser scanning microscope (LSM 800; Carl Zeiss AG, Oberkochen, Germany). The formation of SGs was evaluated by calculating the percentage of positive cells with Image Pro Plus software (Version 6.0; Media Cybernetics, Inc.). Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

PC12 cells were grown in 6 well-plates (5×10⁵ cells per well) and divided into different groups including: Control, model, miR-335 mimic, miR-335 inhibitor, siRNA-ROCK2 and NC groups. Each group was transfected with the respective miRNA or siRNA and then incubated with serum-free medium for 24 h except the control group. Control group was incubated with RPMI-1640 medium containing 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.). An Annexin V-FITC Apoptosis Staining/Detection kit (cat. no. ab14085; Abcam) was used to detect the levels of apoptosis. PC12 cells were digested with 0.25% Trypsin and resuspended with 500 µl binding buffer. Annexin V and propidium iodide (5 µl) were then added into the suspension and labeled for 5 min in the dark. The percentage of apoptotic cells was detected using a flow cytometer (AccuriC6; BD Biosciences, Franklin Lakes, NJ, USA). Accuri CFlow software (Version 1.0; BD Biosciences) was used to analyze the data. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

Three different siRNA-ROCK2 inserts were purchased from Guangzhou RiboBio Co., Ltd., which specifically bound and degraded the ROCK2 mRNA. Using Lipofectamine^® 2000, PC12 cells were transfected with siRNA-ROCK2-1, siRNA-ROCK2-2 and siRNA-ROCK2-3 (100 nM) for 6 h in 6-well plates (5×10⁵ cells per well). After 24 h transfection, western blotting and PCR were used to select the most efficient segment. The sequences of siRNA-ROCK2 were as follows: siRNA-ROCK2-1, 5′-CATGGTGCATTGTGATACA-3′; siRNA-ROCK2-2, 5′-CAGAAGTGCAAGTCTATTA-3′; and siRNA-ROCK2-3, 5′-CTCAAGCTGTGAATAAGTT-3′.

The cerebral cortices of the isch-emic brain and PC12 cells were collected for immunoblotting analysis and lysed with radioimmunoprecipitation assay lysis buffer (Thermo Fisher Scientific, Inc.) for 20 min on ice. The cell debris were removed by centrifuging at 12,000 × g for 10 min at 4°C and the supernatant was harvested for further evaluation (stored at −80°C). The bicinchoninic acid method was used to detect protein concentration. Total protein (30 µg) was separated by 10% SDS-PAGE and transferred to 0.45 µm polyvinylidene difluoride membranes (EMD Millipore, Billerica, MA, USA). After blocking with 5% BSA for 2 h at room temperature, the membranes were incubated with primary antibodies (ROCK2, cat. no. ab71598, 1:1,000; TIA1, cat. no. ab140595, 1:1,000; active caspase-3, cat. no. ab2302, 1:1,000; Bcl-2; cat. no. ab692, 1:1,000; GAPDH, cat. no. ab8245, 1:5,000; Abcam) at 4°C overnight. After rinsing with Tris-buffered saline −0.01% Tween-20, the membranes were incubated with the secondary antibodies [horseradish peroxidase (HRP)-conjugated goat anti-rabbit, cat. no. ab6721; HRP-conjugated goat anti-mouse, cat. no. ab6789; 1:5,000; Abcam] for 1 h at room temperature. Enhanced chemiluminescence (ECL) western blotting substrate (Pierce; Thermo Fisher Scientific, Inc.) was applied to visualize the bands. Image acquisition was performed using the Tanon-6200 gel imaging system (Tanon Science and Technology Co., Ltd., Shanghai, China). Image processing software (ImageJ; Versionm 14.8; National Institutes of Health) was used to analyze the images. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

According to the manufacturer's protocol of the Direct Magnetic IP/Co-IP kit (Pierce; Thermo Fisher Scientific, Inc.), the protein extracted from the rat brain tissues was quantified by a bicinchoninic acid assay following addition of phosphatase and protease inhibitors. A total of 500 µg protein (1–2 mg/ml) was incubated with 25 µl TIA1-coupled magnetic beads for 2 h at room temperature on a rotator. After washing and eluting the binding beads, TIA1 protein was collected and a western blotting assay was used to detect the phosphorylation levels. ECL western blotting substrate (Pierce; Thermo Fisher Scientific, Inc.) was used to visualize the bands. Image acquisition was performed with the Tanon-6200 gel imaging system (Tanon Science and Technology Co., Ltd.). Image processing software (ImageJ; National Institutes of Health) was used to analyze the images. Each experiment was replicated 3 times and the data are presented as the mean ± SEM.

The experimental data are presented as the mean ± SEM. SPSS 22.0 software (IBM Corp., Armonk, NY, USA) was used for statistical analysis. The figures were all produced using GraphPad Prism (Version 6.0; GraphPad Software, Inc., La Jolla, CA, USA). Student's t-test was used to analyze the difference between two groups. The comparison between multiple groups was analyzed using one-way analysis of variance followed by Tukey's multiple comparisons test as the post hoc test. P<0.05 was considered to indicate a statistically significant difference. Each experiment was replicated 3 times.

To evaluate the optimal point of reperfusion in the MCAO model, rats subjected to MCAO (n=36 rats, 6 rats/group) were sacrificed at 0, 6, 12, 24 or 36 h following reperfusion. The brain tissues were harvested and infarct volume was detected by TTC staining. SG formation and apoptosis levels were detected by immunohistochemical and TUNEL staining in frozen brain sections, respectively. The results revealed that SG formation peaked at 6 h postreperfusion, then progressively decreased until 24 h and increased at 36 h (Fig. 1A). Apoptosis levels and infarct volume exhibited similar patterns during different time points following reperfusion, which peaked significantly at 24 h post-reperfusion and reduced at 36 h post-reperfusion (P<0.0001; Fig. 1B and C). At 24 h post-reperfusion, SG formation was at the lowest level, whereas apoptosis and infarct volume were at the highest levels. However, at 36 h post-reperfusion, SG formation was increased, whereas apoptosis and infarct volume were decreased; this may be due to the recovery ability of the rats. These data suggested that MCAO injury may reduce SG formation and elevate apoptosis levels at 24 h. In addition, 24 h post-reperfusion was the optimal time point to investigate the association between SG formation and apoptosis.

A total of 24 male SD rats were randomly divided into control and model groups (12 rats/group). The MCAO model was established using the suture method. Following 24 h of reperfusion, all of the rats were sacrificed by CO₂ asphyxiation, the detailed procedure were performed as described previously (30). The relative expression of miR-335 was measured by RT-qPCR (Fig. S1A). Neurological deficit scoring was performed to evaluate the model. The whole brains were removed for TTC staining. Compared with the control group, the infarct volume and behavioral score significantly increased in the model group (P<0.001 and P=0.0026; Fig. 2A and B), which verified the success of MCAO model establishment. Representative TTC-stained coronal sections of each group were presented in Fig. 2C. An miRNA microarray assay was performed to determine the miRNAs with specific expression in the ischemic cortices of rats. When compared with the control group, there were 20 miRNAs which included 4 upregulated miRNAs and 16 downregulated miRNAs that exhibited differential expression in the model group. miR-335 exhibited the most obvious differential expression in downregulated miRNAs (Fig. 2D). This experimental result was confirmed by the qPCR assay of multiple miRNAs (miR-335, miR-361-3p, miR-290, miR-292-3p, miR-296-5p and miR-3473) in the two groups. Compared with the control group, the miR-335 expression of the model group was significantly decreased (P<0.0001; Fig. 2E). These findings suggested that miR-335 expression was reduced in MCAO rats.

To investigate the protective effect of miR-335 in vivo, 48 male SD rats were randomly divided into 4 groups (control, model, miR-335 mimic and miR-335 inhibitor groups; 12 rats/group). The MCAO model was established using the suture method in rats. miR-335 mimics or the miR-335 inhibitor were immediately injected into the left lateral ventricle of the rats in the corresponding groups following neurological deficit scoring at 0 h. Following 24 h of reperfusion, all of the rats were sacrificed and brain tissues were harvested for the subsequent analyses. The TTC staining (Fig. 3A) and neurological deficit scores were used to evaluate the MCAO model. Compared with the model group, the infarct volume and behavioral score significantly decreased in the mimic group (P=0.0029 and P=0.0104, respectively; Fig. 3B and C). In addition, when compared with the model group, ROCK2 and caspase-3 expression markedly reduced in the miR-335 mimic group (P=0.0107; P=0.0272), whereas TIA1 and Bcl-2 expression increased (P<0.0001; P=0.0020). The immunohistochemical staining of TIA1 was utilized to detect the production of SG. According to the results of immunohistochemistry, the miR-335 mimic significantly promoted TIA1 expression when compared with the model group (P<0.0001; Fig. 3E). The apoptosis levels were detected by TUNEL staining in frozen brain sections. In contrast to the model group, the miR-335 mimic significantly reduced the apoptotic levels of the ischemic brain (P=0.0236; Fig. 3F). These results indicated that miR-335 increased SG formation and reduced the ischemia-induced injury, ROCK2 protein expression and apoptosis levels in MCAO rats. By analyzing the data, the present study revealed that there was a negative correlation between ROCK2 and TIA1 expression in vivo.

To study the potential mechanism of miR-335 promoting SG assembly, the present study conducted immunoprecipitation to detect the phosphorylation level of TIA1 in the ischemic cortex of MCAO rats. As presented in Fig. 4A, TIA1 phosphorylation was increased in the model and inhibitor groups when compared with control group. In addition, TIA1 phosphorylation was significantly inhibited in the miR-335 mimic group compared with the model (P=0.0011). The mechanism underlying how miR-335 promotes SG formation and the regulatory relationship between ROCK2 and TIA1 phosphorylation is presented in Fig. 4B. These results indicated that miR-335 inhibits the phosphorylation level of TIA1 by decreasing the kinase activity of ROCK2 in MCAO rats.

Using bioinformatics software (TargetScan, www.targetscan.org; miRBase, www.mirbase.org), the present study demonstrated that there were predicted binding sites of miR-335 in the 3′-UTR of ROCK2 mRNA (Fig. 5A). To verify this result, a dual luciferase reporter system assay was performed in PC12 cells. Cells transfected with miR-335 NC and pLUC-ROCK2 WT plasmid were treated as the NC. The plasmid profile is presented in Fig. 5B. Compared with the NC group, the relative luciferase activity of cells co-transfected with pLUC-ROCK2 plasmid and miR-335 mimic was significantly decreased (P=0.0022; Fig. 5C). These results confirm that the 3′-UTR of ROCK2 mRNA may be the target of miR-335 in PC12 cells.

To investigate the protective effect of miR-335 in vitro, a serum-free cell model was used in the present study. The stimulatory time experiments indicated that 24 h stimulation generated the most obvious SG formation (Fig. 6A). Then the PC12 cells were cultured in 6/24-well plates and divided into 6 groups: Control, model, miR-335 mimic, miR-335 NC, miR-335 inhibitor and miR-335 inhibitor NC groups. The relative expression of miR-335 was measured by RT-qPCR (Fig. S1B). Following transfection with an miR-335 mimic, mimic NC, inhibitor and inhibitor NC respectively, the cells were cultured with serum-free stimulation for 24 h in RPMI-1640 medium. As presented in Fig. 6B, compared with the model group, ROCK2 expression significantly decreased and TIA1 expression increased significantly in the miR-335 mimic group (P<0.0001; P<0.0001). According to the results of immunofluorescence staining (Fig. 6C), the miR-335 mimic significantly promoted SG production (P<0.0001). As presented in Fig. 6D, compared with the model group, apoptosis levels were significantly reduced in the miR-335 mimic group (P<0.0001). These results indicated that miR-335 promoted SG formation, decreased ROCK2 expression and prevented cell apoptosis in the serum-free cell model.

To analyze the effect of ROCK2 in vitro, the present study performed RNA interference (RNAi) to silence the expression of ROCK2 in serum-free cells. PC12 cells were transfected with siRNA-ROCK2-1, siRNA-ROCK2-2 and siRNA-ROCK2-3 (100 nM) for 6 and 12 h in 6-well plates (1×10⁵ cells per well) to select the most effective segment and transfection time point. The relative expression of ROCK mRNA was measured by RT-qPCR (Fig. S1C). Western blotting results demonstrated that siRNA-ROCK2-1 transfected for 6 h was the most effective method to silence ROCK2 expression (Fig. 7A). Furthermore, compared with the model group, ROCK2 expression significantly decreased and TIA1 expression significantly increased in the siRNA-ROCK2 group (P=0.0013; P=0.0046; Fig. 7B). As presented in Fig. 7C, in contrast to the model group, SG formation was increased in the siRNA-ROCK2 group (P=0.0002). Flow cytometry analysis was used to evaluate apoptosis levels in the cell model. Compared with the model group, the apoptosis level significantly reduced in the siRNA-ROCK2 group (P<0.0001; Fig. 7D). These results indicated that ROCK2 silencing promoted SG formation, elevated TIA1 protein expression and prevented cell apoptosis in serum-free cells.

According to the aforementioned results, the present study revealed that miR-335 promoted SG formation and suppressed cell apoptosis in PC12 cells with serum-free stimulation. Previous studies have demonstrated that the target gene can be regulated by multiple miRNAs (31,32). Therefore, to investigate the association between the protective function of miR-335 and ROCK2, co-transfection of miR-335 and siRNA-ROCK2 was performed in the serum-free cell model. As presented in Fig. 8A, compared with the model group, ROCK2 expression was significantly inhibited in the co-transfection and siRNA-ROCK2 groups (P=0.0024; P=0.0035), and TIA1 expression was significantly increased in these two groups (P=0.0009; P=0.0030). Furthermore, there was no significant difference in ROCK2 and TIA1 expression between the siRNA-ROCK2 and co-transfection groups (P=0.9885; P=0.7090). In contrast to the model group, SG formation was markedly increased in the co-transfection and siRNA-ROCK2 groups (P=0.0019; P=0.0034), however, no significant difference was identified between these two groups (P=0.9591; Fig. 8B). Flow cytometry analysis was used to detect the level of apoptosis. Compared with the model group, the apoptotic level was significantly down-regulated in the co-transfection and siRNA-ROCK2 groups (P=0.0003; P=0.0008), however, no significant difference was identified when comparing these two groups (P=0.7818; Fig. 8C). These results indicated that miR-335 specifically targeted the ROCK2 3′-UTR to promote SG formation, reduce ROCK2 expression and suppress cell apoptosis.