miR-26a-5p Attenuates Oxidative Stress and Inflammation in Diabetic Retinopathy through the USP14/NF-κB Signaling Pathway

Purpose Diabetic retinopathy (DR) is an ocular disease caused by diabetes and may lead to vision impairment and even blindness. Oxidative stress and inflammation are two key pathogenic factors of DR. Recently, regulatory roles of different microRNAs (miRNAs) in DR have been widely verified. miR-26a-5p has been confirmed to be a potential biomarker of DR. Nevertheless, the specific functions of miR-26a-5p in DR are still unclear. Methods Primary cultured mouse retinal Müller cells in exposure to high glucose (HG) were used to establish an in vitro DR model. Müller cells were identified via morphology observation under phase contrast microscope and fluorescence staining for glutamine synthetase. The in vivo animal models for DR were constructed using streptozotocin-induced diabetic C57BL/6 mice. Western blotting was performed to quantify cytochrome c protein level in the cytoplasm and mitochondria of Müller cells and to measure protein levels of glial fibrillary acidic protein (GFAP), ubiquitin-specific peptidase 14 (USP14), as well as factors associated with NF-κB signaling (p-IκBα, IκBα, p-p65, and p65) in Müller cells or murine retinal tissues. ROS production was detected by CM-H2DCFDA staining, and the concentration of oxidative stress markers (MDA, SOD, and CAT) was estimated by using corresponding commercial kits. Quantification of mRNA expression was conducted by RT-qPCR analysis. The concentration of proinflammatory factors (TNF-α, IL-1β, and IL-6) was evaluated by ELISA. Hematoxylin-eosin staining for murine retinal tissues was performed for histopathological analysis. Immunofluorescence staining was conducted to determine NF-κB p65 nuclear translocation in Müller cells. Furthermore, the interaction between miR-26a-5p and USP14 was verified via the luciferase reporter assays. Results HG stimulation contributed to Müller cell dysfunction by inducing inflammation, oxidative injury, and mitochondrial damage to Müller cells. miR-26a-5p was downregulated in Müller cells under HG condition, and overexpression of miR-26a-5p relieved HG-induced Müller cell dysfunction. Moreover, miR-26a-5p targeted USP14 and inversely regulated USP14 expression. Additionally, HG-evoked activation of NF-κB signaling was suppressed by USP14 knockdown or miR-26a-5p upregulation. Rescue assays showed that the protective impact of miR-26a-5p upregulation against HG-induced Müller cell dysfunction was reversed by USP14 overexpression. Furthermore, USP14 upregulation and activation of NF-κB signaling in the retinas of DR mice were detected in animal experiments. Injection with miR-26a-5p agomir improved retinal histopathological injury and weakened the concentration of proinflammatory cytokines and oxidative stress markers in the retinas of DR mice. Conclusion miR-26a-5p inhibits oxidative stress and inflammation in DR progression by targeting USP14 and inactivating the NF-κB signaling pathway.


Introduction
Diabetic retinopathy (DR) is a chronic diabetic complication characterized by an ocular disorder that can lead to impaired vision and even blindness.DR is seriously threatening the eye health of people worldwide [1] and partly results from unhealthy eating habits and lifestyles nowadays [2].Clinically, fundus lesions in DR are manifested as retinal detachment, macular edema, vitreous proliferation, neovascularization, and cotton patch.At present, the frstline therapy for diabetic macular edema, a common cause of vision loss in patients with DR, is anti-VEGF administration.
However, some patients with DR show tachyphylaxis or refractoriness to the repeated injection of anti-VEGF drugs [3].Terefore, fnding preventive and therapeutic methods to control DR is of great urgency.
Oxidative stress and infammatory response induced by chronic hyperglycemia are key factors for the occurrence and progression of DR [4].Oxidative stress is the consequence of the imbalance between the release of endogenous antioxidant factors and the release of reactive oxygen species (ROS) [5].Te excessive release of ROS can result in the damage of tissues around and in retinal vessels, thereby leading to DR [6].Moreover, hyperglycemia-induced oxidative stress can dramatically enhance infammation and angiogenesis [6].Emerging evidence has manifested the crucial role of ROS in the activation of a typical proinfammatory signaling, i.e., nuclear factor-kappa B (NF-κB) signaling [7].Once NF-κB is activated, nuclear translocation of NF-κB occurs and the transcription of proinfammatory factors happens.Interleukin-1β (IL-1β), interleukin 6 (IL-6), and tumor necrosis factor-α (TNF-α) are common proinfammatory factors [8].As the main glial cells in the retinas, Müller cells undergo oxidative injury and show reactive phenotypes under hyperglycemic conditions [9].Activation of Müller cells is manifested as the upregulation of glial fbrillary acidic protein (GFAP) expression and the subsequent release of proinfammatory cytokines [10].Tus, inhibition of the GFAP, oxidative stress, and infammatory response in Müller cells under hyperglycemic conditions may prevent the pathological process of DR [11].
MicroRNAs (miRNAs) are small, endogenous, noncoding RNAs of ∼22 to 26 nucleotides in length.Te posttranscriptional regulation is a key function of miR-NAs [12].miRNAs interact with the 3′ untranslated regions (UTRs) of target genes, thereby repressing the translation or promoting the degradation of mRNAs [13].Many studies have demonstrated that miRNAs can modulate various biological processes, including cell proliferation, metabolism, and organ development [14,15].miRNAs have been considered as promising candidates for targeted therapeutic approaches to DR [16,17].Emerging evidence has identifed the close association of miRNAs with the occurrence or development of DR.For example, depleted miR-29a/b leads to Müller cell dysfunction to exacerbate DR by targeting mRNA forkhead box protein O4 [18].Overexpressed miR-18a-3p alleviates blood-retinal barrier disruption in DR mice and mitigates infammatory injury in Müller cells under high glucose conditions by targeting the gene bone morphogenetic protein 4 [19].miR-138-5p displays low expression in DR and its upregulation represses the proliferative potential of retinal capillary endothelial cells by interacting with mRNA neuro-oncological ventral antigen 1 [20].miR-26a-5p is a well-known miRNA and its regulatory role has been demonstrated in diferent human diseases, including diabetes and diabetic complications.For example, miR-26a-5p attenuates myocardial injury in diabetic rats by downregulating phosphatase and tensin homolog [21].Long noncoding RNA and small nucleolar RNA host gene 5 target miR-26a-5p to promote renal tubular cell damage by accelerating cell apoptosis, oxidative stress, and infammation in diabetic nephropathy [22].Upregulation of miR-26a-5p prevents infammatory response in tubular epithelial cells by binding with target gene CHAC1 and inhibiting NF-κB signaling, thereby attenuating diabetic kidney disease [23].Importantly, a previous study has identifed that miR-26a-5p suppresses retinal neuronal cell death to prevent retinal impairment by interacting with mRNA phosphatase and tensin homolog in diabetic mice [24].Another study has pointed out that circulating miR-26a-5p is downregulated in nonproliferative DR, suggesting its potential involvement in retinal neurodegeneration at an early stage [25].Nevertheless, the molecular mechanism mediated by miR-26a-5p in DR and its functions in Müller cells need further investigation.
Ubiquitin-specifc peptidase 14 (USP14), a ubiquitinspecifc protease associated with the proteasome, exerts essential functions in infammation, cellular functions, neurodegenerative diseases, and tumor development [26].It is reported that USP14 is highly expressed in DR patients and can alter the functions of Müller cells treated with high glucose by enhancing IκBɑ deubiquitination and degradation, thereby promoting IκBɑ phosphorylation and NF-κB activation [27].However, the miR-26a-5p/USP14 regulatory network in DR has not been explored.
In the current work, the in vitro and in vivo models for DR are established using high glucose-(HG-) stimulated Müller cells and streptozotocin-induced diabetic mice to investigate the biological role and molecular mechanism of miR-26a-5p in DR.Our study might provide promising insight into potential therapeutical strategies for DR treatment.

Cell Identifcation.
Müller cells were identifed through morphology observation under a phase contrast microscope.Additionally, fuorescent staining for glutamine synthetase (GS), a specifc enzyme of Müller cells in the retina, was also performed for cell identifcation.A fuorescence microscope (Leica, Germany) was utilized for visualization of the fuorescence of Müller cells.

Cell Treatment.
When cell confuence reached 80%, Müller cells were subjected to high glucose (HG) or normal glucose (NG) treatment.To mimic the diabetic environment, 33.3 mM of glucose (HG) was utilized to stimulate Müller cells for 24 h.Müller cells treated with 5 mM of glucose (normal glucose) for 24 h served as the control group.

Detection of Oxidative Stress Markers.
Te homogenized retinal tissues or Müller cells were subjected to centrifugation to obtain the supernatants.Te levels of SOD (superoxide dismutase), CAT (catalase), and MDA (malondialdehyde) in the supernatants were estimated by a SOD assay kit (#S0101M, Beyotime, China), a CAT assay kit (#CAT100, Sigma-Aldrich, USA), and a MDA assay kit (#S0131M, Beyotime) according to the suppliers' instructions, respectively.

Enzyme-Linked Immunosorbent Assay (ELISA).
Measurement of the IL-1β, IL-6, VEGF, and TNF-α in the supernatants of homogenized retinal tissues or Müller cells was performed in the collected supernatants by utilizing corresponding kits based on the suppliers' suggestions, respectively.
2.14.Statistical Analysis.SPSS 18.0 software (SPSS Inc., USA) was used for data analyses.Experimental data are shown as the mean ± standard error of the mean (SEM).Diferences between the two groups were tested by using an independent sample t-test, and multigroup comparisons were evaluated by using a one-way analysis of variance, followed by Tukey's post hoc test.A P value of less than 0.05 was set as the threshold for statistical signifcance.

Identifcation of Mouse Primary Retinal Müller Cells.
A phase contrast microscope showed that Müller cells were oval-shaped and in a pale color with a big nucleus and abundant cytoplasm (Figure 1(a)).Additionally, there are long pyramidal projections at two ends of the cell body (Figure 1(a)).Te results of GS staining revealed that nearly all the cells were stained with GS which is a Müller cellspecifc enzyme in the retina (Figure 1(b)).Te fndings confrmed that the cells used in the study are Müller cells.b-d) denoted that the activities of antioxidant enzymes (SOD and CAT) were decreased and the concentration of prooxidant MDA was increased after HG treatment, while miR-26a-5p overexpression reversed these changes.CM-H 2 DCFDA staining assay recorded that miR-26a-5p mimics inhibit HGstimulated ROS production (Figure 2(e)).Te abovementioned results indicated that miR-26a-5p upregulation mitigates HG-evoked oxidative stress in Müller cells.Moreover, cytochrome c release from the mitochondria into the cytosol is a key induction of cell apoptosis [32].Te outcome of western blotting demonstrated HG-induced cytochrome c release from the mitochondria to cytosol, which was counteracted by miR-26a-5p enhancement (Figure 2(f )), implying that upregulated miR-26a-5p suppresses HG-triggered Müller cell apoptosis.Furthermore, retinal gliosis characterized by upregulation of GFAP and VEGF in Müller cells is an important pathological characteristic of DR [33].As western blotting and ELISA denoted,

HG-Evoked Activation of NF-κB Signaling Was Suppressed by USP14 Knockdown or miR-26a-5p Elevation.
Many reports have mentioned that USP14 can activate NF-κB signaling.Additionally, it is well documented that NF-κB activation promotes the progression of DR.Hence, USP14 protein expression in the in vitro cell model was quantifed.
It is shown that the USP14 protein level was increased after HG stimulation (Figure 3(a)), suggesting the potential involvement of USP14 in DR.Subsequently, the relationship of miR-26a-5p, USP14, and NF-κB signaling in HG-stimulated Müller cells was fgured out.USP14 was downregulated via transfecting sh-USP14 into Müller cells under HG condition (Figure 3(a)).Quantifcation of the protein levels of p-IκBα, IκBα, NF-κB (nuclear), and NF-κB (cytoplasm) was also conducted via western blotting.HG exposure triggered the increase in p-IκBα and NF-κB (nuclear) levels and the decrease in IκBα and NF-κB (cytoplasm) protein levels in Müller cells.However, these changes were reversed by the interference of USP14.Te abovementioned fndings confrm that silenced USP14 represses HG-mediated activation of NF-κB signaling in Müller cells.A binding site between miR-26a-5p and USP14 3′UTR is provided by a bioinformatics tool, starBase (Figure 3(b)).To verify the interaction of the two RNAs, the luciferase reporter plasmid USP14-Wt/Mut was constructed and transfected into HEK293T cells together with miR-26a-5p mimics/NC mimics.Data showed that miR-26a-5p mimics caused a reduction of luciferase activity in the USP14-Wt group, but miR-26a-5p mimics had little impact on the activity in the USP14-Mut group (Figure 3

miR-26a-5p Overexpression Alleviates Retinal Injury of Mice with DR and Alleviates Infammation and Oxidative
Stress In Vivo.Whether miR-26a-5p exerts a retinal protective efect in vivo was validated.H&E staining of retinal tissues in indicated groups was conducted to observe histopathological changes.Te control group displayed normal retinal structure and smooth retinal surfaces.In the retinas of the control group mice, each layer showed a complete and clear structure, and the cells in each layer were densely distributed and neatly arranged.However, the retinas of DR mice showed disrupted retinal structure, unclear inner and outer layers, disordered retinal cells in each layer, and angiogenesis.After injection with miR-26a-5p agomir, all these pathological changes in the retinas of DR mice were alleviated (Figure 5(a)).Quantifcation of retinal thickness revealed that DR induced a signifcant decrease in retinal thickness, and the change in DR mice was ofset by miR-26a-5p overexpression (Figure 5(a)).Based on the results of the concentrations of VEGF, TNF-α, IL-1β, and IL-6 in retinal tissues were elevated in the DR group compared to those in the normal group, and the accumulations of VEGF and proinfammatory cytokines were downregulated in the DR + miR-26a-5p agomir group compared with those in the DR + NC-agomir group (Figures 5(b-e)).Te fnding indicated that miR-26a-5p prevents retinal infammation in DR mice.Consistently, the decreased activities of SOD and CAT as well as the increased MDA level were shown in the retinal tissues of DR mice when compared to those in normal mice, while the injection of miR-26a-5p agomir reversed these changes in DR mice (Figures 5(f-h)).Te results demonstrated the repressive impact of miR-26a-5p on retinal oxidative stress in DR mice.Additionally, USP14 protein expression was surged in the retinal tissues of DR mice compared to that in the normal group (Figure 5(i)).Additionally, the protein level of USP14 was reduced upon the injection of miR-26a-5p agomir into DR mice (Figure 5(i)), which implied that miR-26a-5p inversely regulates USP14 in vivo.Te regulatory role of miR-26a-5p in modulating NF-κB signaling in DR mice was also verifed.High p-IκBα protein expression and low IκBα protein expression were found in retinal tissues of DR mice compared with those in the control group, which was counteracted by miR-26a-5p agomir injection, indicating that miR-26a-5p induces the inactivation of NF-κB pathway in vivo.To sum up, overexpressed miR-26a-5p alleviates retinal injury of DR mice and suppresses infammation and oxidative stress in vivo.

A Schematic Diagram Revealing the Role of miR-26a-5p/ USP14/NF-κB Pathway in HG-Stimulated Müller Cells.
Tis study revealed the protective efects of miR-26a-5p against oxidative stress and infammation in DR in vitro.In Müller cells, HG exposure induces infammation by upregulating levels of VEGF, TNF-α, IL-1β, and IL-6, and HG aggravates oxidative stress by reducing antioxidant enzymes (SOD and CAT) and increasing MDA and ROS levels.HG treatment also decreases miR-26a-5p expression in Müller cells.Moreover, miR-26a-5p directly targets USP14 and inversely regulates USP14 expression.Since USP14 can activate the NF-κB signaling by promoting NF-κB phosphorylation, downregulation of USP14 induced by miR-26a-5p leads to the inactivation of NF-κB signaling.Activated NF-κB signaling contributes to infammation and oxidative stress in Müller cells, while the inactivation of NF-κB signaling exerts the opposite efect.In conclusion, miR-26a-5p suppresses HG-triggered oxidative stress and infammation in Müller cells by targeting USP14 and thereby inactivating the NF-κB signaling (Figure 6).

Discussion
Diabetic retinopathy (DR) is a common microvascular complication of diabetes mellitus, leading to vision defcits and even irreversible blindness [1].Müller cells are predominant glial cells in the retinal tissues, and reactive gliosis can happen in response to hyperglycemia featured by an upregulated level of GFAP (a glial activation marker) [34].Oxidative injury and chronic infammation are two unignorable pathogenic factors in the pathogenesis of DR [35].During DR progression, hyperglycemic condition contributes to excessive ROS release and induces the production of infammatory cytokines (e.g., TNF-α, IL-1β, and IL-6), which activates NF-κB by promoting IκBα and p65 phosphorylation, and p65 nuclear translocation, in turn provoking infammatory response [36,37].Müller cells under HG conditions initiate subsequent oxidative stress and induce retinal infammation by secreting various proinfammatory cytokines [27].Tus, strategies aimed to prevent early neuroglial dysfunction and maintaining the homeostasis of Müller cells may be benefcial to treat DR or deter DR progression.
In this study, Müller cells were exposed to HG for the establishment of in vitro DR models.It was found that HG stimulation induced ROS accumulation, decreased antioxidative CAT and SOD concentration, and increased MDA level in Müller cells, implying the enhanced oxidative stress provoked by HG in DR in vitro.Cytochrome c release from the mitochondria into the cytosol is a key event in cell apoptosis [32].HG induced cytochrome c release from the mitochondria to cytosol in Müller cells, implying that HG triggers Müller cell apoptosis.Moreover, Müller cells were activated under HG condition, as manifested by the upregulated GFAP expression.HG challenge also contributed to infammation in Müller cells by inducing NF-κB signaling activation and proinfammatory factor secretion.Te abovementioned fndings illustrated that HG treatment aggravates oxidative stress, infammation, and Müller cell gliosis, which are in line with the results of previous reports [29].Additionally, we also established the in vivo DR models by using streptozotocin-injected mice as previously described [38].As a result, DR mice exhibited obvious retinal pathological injury compared to normal mice.Moreover, upregulated SOD and CAT levels, elevated proinfammatory cytokine concentration, and increased GFAP expression, as well as the activated NF-κB signaling were observed in the retinal tissues of DR mice.All the data confrmed the successful establishment of in vivo models of DR.
Accumulating evidence has revealed that aberrantly expressed miRNAs are closely related to the initiation or development of DR since miRNAs can directly regulate genes associated with the disease [39][40][41].Despite the lack of proteincoding capacity, miRNAs can modulate the expression of target mRNAs by binding to the 3′UTR of mRNAs [13].Te miRNA-mRNA regulatory mechanism is frequently found in DR.For example, miR-365 facilitates Müller gliosis, retinal injury, and oxidative stress in DR by targeting TIMP3 [42].Downregulated miR-200b hinders Müller cell apoptosis and oxidative injury in DR via interaction with OXR1 [43].miR-26a-5p is a miRNA that has been reported to participate in various diseases, 10 Journal of Ophthalmology including diabetic nephropathy [44], diabetic myocardial injury [21], diabetic cataract [45], and DR [24].In this study, we found that miR-26a-5p was downregulated in HG-stimulated Müller cells, further confrming the potential role of miR-26a-5p in DR pathogenesis.Moreover, overexpressed miR-26a-5p reversed the impact of HG on Müller cell gliosis and cytochrome c release from the mitochondria into the cytosol.Upregulation of miR-26a-5p attenuated oxidative stress and infammation in Müller cells under HG conditions by reducing ROS levels, enhancing SOD and CAT activities, and inhibiting the release of proinfammatory cytokines.Importantly, injection with miR-26a-5p agomir alleviated retinal histopathological alternations, infammation, and oxidative injury in DR mice.Furthermore, the elevation of miR-26a-5p inactivated the NF-κB signaling in cell models and animal models for DR.Terefore, miR-26a-5p may mitigate DR through the inactivation of the NF-κB pathway.USP14 belongs to the ubiquitin-specifc processing family and can regulate various cellular activities, such as infammation, apoptosis, oxidative stress, and proliferation [46,47].Moreover, emerging studies have validated its involvement in diabetes mellitus [48], DR [27], and diabetic nephropathy [49].In this study, we found that USP14 was highly expressed in HG-treated Müller cells.Importantly, through bioinformatics prediction and experimental validation, it was confrmed that miR-26a-5p could bind to USP14 and negatively regulate USP14 expression in HG-exposed Müller cells and retinas of DR mice.Furthermore, rescue assays illustrated that USP14 upregulation reversed the mitigative efects of miR-26a-5p overexpression on oxidative stress and infammation in Müller cells under HG conditions.Te inhibitory efect of overexpressed miR-26a-5p on Müller cell gliosis and cytochrome c release from the mitochondria into the cytosol was also countervailed by upregulating USP14.Terefore, a high USP14 level aggravates DR development by enhancing HG-induced Müller cell dysfunction.-d) Measurement of SOD, CAT, and MDA levels in Müller cells of the NC mimics, miR-26a-5p mimics, miR-26a-5p mimics + pcDNA, and miR-26a-5p mimics + pcDNA/USP14 groups was performed using specifc commercial kits.(e) CM-H2DCFDA staining was performed to measure ROS levels in Müller cells of the NC mimics, miR-26a-5p mimics, miR-26a-5p mimics + pcDNA, and miR-26a-5p mimics + pcDNA/USP14 groups.(f ) Western blotting was performed to quantify cytochrome c protein levels in the cytoplasm and mitochondria of Müller cells of the NC mimics, miR-26a-5p mimics, miR-26a-5p mimics + pcDNA, and miR-26a-5p mimics + pcDNA/USP14 groups.(g) Western blotting for evaluating GFAP protein expression in Müller cells of the NC mimics, miR-26a-5p mimics, miR-26a-5p mimics + pcDNA, and miR-26a-5p mimics + pcDNA/USP14 groups.(h-k) ELISA was performed to measure (h) VEGF level and the accumulation of (i-k) proinfammatory cytokines (TNF-α, IL-1β, and IL-6) in Müller cells of the NC mimics, miR-26a-5p mimics, miR-26a-5p mimics + pcDNA, and miR-26a-5p mimics + pcDNA/USP14 groups.* * p < 0.01 and * * * p < 0.001.molecules or other signaling pathways implicated with the miR-26a-5p/USP14 axis in DR also deserve more exploration in future studies.
Relieves HG-Induced Müller Cell Dysfunction.An in vitro cell model for DR was established by stimulating Müller cells with a high concentration of glucose.Results from RT-qPCR illustrated that miR-26a-5p was downregulated in HG-treated Müller cells compared to that in Müller cells without any treatment (Figure 2(a)), implying that miR-26a-5p might substantially participate in DR.To examine the functions of miR-26a-5p in HG-induced Müller cell injury models, miR-26a-5p was overexpressed in HG-treated Müller cells via transfection of miR-26a-5p mimics, as demonstrated by RT-qPCR (Figure 2(a)).Te outcome of Figures 2( miR-26a-5p augmentation ofset the promoting efect of HG on GFAP protein expression and VEGF release in Müller cells (Figures2(g and h)), which implied that HG-induced Müller cell gliosis was mitigated by upregulating miR-26a-5p.Additionally, HG stimulation caused Müller cell infammation as evidenced by the increase in the concentration of proinfammatory cytokines (TNF-α, IL-1β, and IL-6) in the HG group, but miR-26a-5p overexpression reversed the alterations in the release of infammatory cytokines (Figures2(i-k)).Overall, miR-26a-5p overexpression alleviated HG-induced Müller cell dysfunction.

Figure 1 :Figure 2
Figure 1: Identifcation of mouse primary retinal Müller cells.(a) Müller cell morphology under a phase contrast microscope.(b) Fluorescence staining for glutamine synthetase (GS) was performed for cell identifcation.

Figure 2 :Figure 3 :
Figure2: miR-26a-5p overexpression relieves HG-induced Müller cell dysfunction.(a) RT-qPCR was performed to measure miR-26a-5p expression in retinal cells with or without HG treatment and to evaluate the transfection efciency of miR-26a-5p mimics in HG-exposed Müller cells.(b-d) SOD, CAT, and MDA levels in Müller cells of four groups (control, HG, HG + NC mimics, and HG + miR-26a-5p mimics groups) were measured using specifc commercial kits.(e) CM-H 2 DCFDA staining was conducted to evaluate ROS levels in Müller cells of the control, HG, HG + NC mimics, and HG + miR-26a-5p mimics groups.(f ) Western blotting was performed to quantify cytochrome c protein levels in the cytoplasm and mitochondria of Müller cells in the control, HG, HG + NC mimics, and HG + miR-26a-5p mimics groups.(g) Western blotting was performed to quantify GFAP protein expression in Müller cells of the control, HG, HG + NC mimics, and HG + miR-26a-5p mimics groups.(h-k) ELISA was conducted to measure (h) VEGF level and (i-k) the concentration of proinfammatory cytokines (TNF-α, IL-1β, and IL-6) in Müller cells of the control, HG, HG + NC mimics, and HG + miR-26a-5p mimics groups.* * p < 0.01 and * * * p < 0.001.

Table 2 :
Primer sequences used for RT-qPCR.