LEF1 influences diabetic retinopathy and retinal pigment epithelial cell ferroptosis via the miR-495-3p/GRP78 axis through lnc-MGC

Abstract

BACKGROUND

Diabetic retinopathy (DR) is one of the major eye diseases contributing to blindness worldwide. Endoplasmic reticulum (ER) stress in retinal cells is a key factor leading to retinal inflammation and vascular leakage in DR, but its mechanism is still unclear.

AIM

To investigate the potential mechanism of LEF1 and related RNAs in DR.

METHODS

ARPE-19 cells were exposed to high levels of glucose for 24 hours to simulate a diabetic environment. Intraperitoneally injected streptozotocin was used to induce the rat model of DR. The expression levels of genes and related proteins were measured by RT-qPCR and Western blotting; lnc-MGC and miR-495-3p were detected by fluorescent in situ hybridization; CCK-8 and TUNEL assays were used to detect cell viability and apoptosis; enzyme-linked immunosorbent assay was used to detect inflammatory factors; dual-luciferase gene assays were used to verify the targeting relationship; and the retina was observed by HE staining.

RESULTS

LEF1 and lnc-MGC have binding sites, and lnc-MGC can regulate the miR-495-3p/GRP78 molecular axis. In high glucose-treated cells, inflammation was aggravated, the intracellular reactive oxygen species concentration was increased, cell viability was reduced, apoptosis was increased, the ER response was intensified, and ferroptosis was increased. As an ER molecular chaperone, GRP78 regulates the ER and ferroptosis under the targeting of miR-495-3p, whereas inhibiting LEF1 can further downregulate the expression of lnc-MGC, increase the level of miR-495-3p, and sequentially regulate the level of GRP78 to alleviate the occurrence and development of DR. Animal experiments indicated that the knockdown of LEF1 can affect the lnc-MGC/miR-495-3p/GRP78 signaling axis to restrain the progression of DR.

CONCLUSION

LEF1 knockdown can regulate the miR-495-3p/GRP78 molecular axis through lnc-MGC, which affects ER stress and restrains the progression of DR and ferroptosis in retinal pigment epithelial cells.

Key Words: Diabetic retinopathy; Endoplasmic reticulum stress; miR-495-3p/GRP78; Lnc-MGC; Retinal pigment epithelium cells; Ferroptosis

Core Tip: Inhibiting lnc-MGC protects retinal pigment epithelium cells from high glucose-induced apoptosis, inflammation and oxidative stress. Lnc-MGC targeted miR-495-3p and miR-495-3p targeted GRP78. Lnc-MGC regulated endoplasmic reticulum stress and ferroptosis via miR-495-3p/GRP78. LEF1 regulated diabetic retinopathy as binding protein of lnc-MGC.

INTRODUCTION

Diabetic retinopathy (DR) is one of the most common microvascular complex diseases associated with diabetes[1,2]. It is characterized by progressive changes in the retinal microvasculature, which increase vascular patency, retinal nonperfusion, and pathological intraocular proliferation of retinal vessels[3,4]. DR can cause retinal changes that threaten the vision of one-third of people with diabetes[5]. DR can be clinically divided into proliferative and nonproliferative DR according to whether there are obvious ophthalmic changes and retinal neovascularization[6]. Current treatments for DR include anti-vascular endothelial growth factor (VEGF) drugs, retinal laser photocoagulation, steroid therapy, and surgical remedies[7,8]. Although there have been breakthroughs in DR diagnosis, treating early DR is still challenging.

The endoplasmic reticulum (ER) is an important membranous organelle for protein synthesis, folding and excretion in eukaryotic cells. When stimulated by endogenic or exogenic stimuli, such as a lack of molecular chaperones or cellular energy, Ca²⁺ deficiency, disruption of redox homeostasis, protein variation and reduction in disulfide bonds, the protein folding function of the ER is disrupted. Unfolded or misfolded proteins accumulate in the lumen of the ER and induce ER stress[9,10]. GRP78 separates from ER transmembrane proteins when cells are stimulated by exogenous stimuli, thereby inducing the ER to prevent the amassing of unfolded proteins and maintaining cellular homeostasis[11]. Previous studies have suggested that the ER plays an important role in the body's defense against intracellular bacterial infection by activating apoptosis[12]. Many studies have shown that ER stress is involved in DR progression; for example, studies have shown that TCF7 L2 promotes ER stress signaling in DR[13]. ASK1 promotes the expression of ER stress-related proteins, which are major regulators of DR[14]. Ferroptosis is a form of intracellular iron accumulation and reactive oxygen species (ROS)-dependent programmed cell death, and studies have shown that ER stress plays an important role in the crosstalk between ferroptosis and apoptosis. ER stress-mediated activation of the PERK-eIF2α-ATF4-CHOP pathway induces apoptosis and ferroptosis[15]. However, whether the ER is involved in ferroptosis of retinal pigment epithelium cells in DR remains to be determined. We hypothesized that prevention of capillary degeneration and regulation of the ER stress response in the early stage of DR may be beneficial in delaying the progression of DR and severe irreversible visual loss.

GRP78 can be induced by the ER stress response and is considered a molecular chaperone in the ER[16]. GRP78 performs key functions in maintaining ER homeostasis, assisting in protein folding and assembly, and routing misfolded proteins for degradation[17]. Notably, the expression or elevation of GRP78 is often interrelated with multifarious tumor microenvironmental stresses[18]. In addition, increased expression of GRP78 can regulate the ER stress response and subsequently regulate the occurrence of diabetes[19]. Therefore, GRP78 has the potential to become a more selective therapeutic target for DR.

MicroRNAs (miRNAs) are small RNA molecules of 19 to 25 nucleotides that regulate the transcription of target genes, and a single miRNA can typically affect the expression of many genes[20]. MiRNA-379 has been reported to be a tumor suppressor due to its inhibitory effects on cell proliferation and migration in brain, breast, lung, and liver cancers[21-23]. Previous studies have shown that lncRNA ZFAS1 can actively facilitate endothelial ferroptosis in DR through miR-7-5p[24]. Moreover, studies have confirmed that the upregulation of miR-495-3p can promote the progression of DR, which also shows the specific regulatory effect of miR-495-3p on DR[25]. miR-495-3p restrains cell proliferation and migration by downregulating HMGB1 in colorectal cancer[26]. Previous studies have shown that miR-495-3p is well studied in human tumors and has a high ability to restrain tumor growth and chemoresistance[27]. Furthermore, miR-379-5p has been shown to play a pivotal role in the progression of other tumors[28]. Lnc-MGC and miRNAs regulate ER stress-induced diabetic nephropathy[29]. However, whether lnc-MGC actively participates in the miR-495-3p/GRP78 axis to mediate the possible pathological mechanism of DR remains unclear.

Based on the above studies, this study aimed to verify the mutual effects between lnc-MGC and miR-495-3p and between miR-495-3p and GRP78 to study the role of the miR-495-3p/GRP78 axis in DR cell proliferation, apoptosis and ferroptosis and to ultimately explore its molecular mechanism.

MATERIALS AND METHODS

Cell culture and transfection

ARPE-19 cells were purchased from Otwo Biotech (Shenzhen, China) and cultured in DMEM containing 10% fetal bovine serum (FBS, Thermo, United States) and 1% penicillin/streptomycin. The incubation conditions were 37 °C, 5% CO₂, and 95% relative humidity. A cell model of high glucose (HG) conditions in vitro was established with 30 mmol/L glucose. Tunicamycin (ER stress inducer, 2 μg/mL) or erastin (10 μM, ferroptosis inducer) was added to the medium, which was subsequently incubated for 24 hours. si-/oe-lnc-MGC, the miR-495-3p mimic/inhibitor, oe-GRP78, and si-LEF1 (Shanghai GenePharma, Co., Ltd.) were transfected into cells using Lipofectamine 2000 (Invitrogen) in accordance with the manufacturer's instructions. The cells were transfected 48 hours later for subsequent experiments.

Animal model construction

Eight-week-old male SD rats weighing 180-200 g were purchased from the Animal Experimental Center of Kunming Medical University and fed a 12 hours light/dark cycle at 22 ± 2 °C and 50% ± 5% humidity. The rats were allowed to drink freely. After 1 week of adaptation, the animals were randomly divided into four groups. The normal control group was intraperitoneally injected with the same amount of phosphate buffer. The streptozotocin (STZ)-treated diabetic group, DR + si-LEF1 group and DR + si-LEF1 + oe-lnc-MGC group (after STZ injection, the vector si-LEF1 or si-LEF1 + oe-lnc-MGC was injected into the vitreous body; si indicates knockdown, oe indicates overexpression) were given a single intraperitoneal injection of 65 mg/kg STZ (PBS solution, Sigma-Aldrich, United States) to induce diabetes. Glucose levels in blood obtained from the tail vein were measured to confirm the induction of diabetes, and blood glucose levels > 16.7 mmol/L were considered to indicate diabetes. Nine weeks after STZ injection, the rats were fasted overnight for subsequent studies. All animal experiments were approved by the Medical Ethics Committee of Chuxiong Yi Autonomous Prefecture People's Hospital.

Fluorescence in situ hybridization

A fluorescence in situ hybridization (FISH) assay was performed with a FISH kit. In brief, ARPE-19 cells were fixed with 4% paraformaldehyde and treated with proteinase K, glycine, and acetylation reagents. ARPE-19 was hybridized with a Cy3-labeled lnc-MGC probe and 647-labeled miR-495-3p for 12 hours at 42 °C after prehybridization for 1 hour at 42 °C. Nuclei were stained with DAPI. Each procedure was performed using a TeFISH kit (GenePharma) following a specific protocol. Images were obtained with a microscope.

Western blot

The proteins of the cells and tissues were extracted with RIPA lysis buffer and BAC kit (Beyotime, product number: P0012) was used to detect protein concentration. A total of 20 μg of protein was loaded into a sodium dodecyl sulfate solution. The samples were then subjected to polyacrylamide gel electrophoresis, and the bands were transferred to membranes. The membranes were incubated in 5% skim milk for blocking for 1 hour, after which they were incubated with anti-Bax (1/1000), anti-Bcl-2 (1/200), anti-cleaved caspase 3 (1/500), anti-GRP78 (1/1000), anti-CHOP (1/2000), anti-XBP-1 (1/1000), anti-GPX4 (1/1000), anti-HO-1 (1/500), anti-FTH1 (1/1000), and anti-xCT (1/2000) antibodies. The membranes were subsequently incubated with a specific secondary antibody (1/1000) at room temperature for 1 hour. The internal reference gene was β-actin. An enhanced chemiluminescence kit was used to develop color, and the protein bands were semiquantitatively analyzed with ImageJ software.

CCK-8

A CCK-8 kit was used to assess cell viability according to the manufacturer’s instructions. A total of 5 × 10³ ARPE-19 cells per well were seeded into 96-well plates and incubated for 24 hours. Then, 100 μL of substrate containing 10 mM CCK-8 was added to each well and incubated at 37 °C for 2 hours. The absorbance at 450 nm was measured with a multimode microplate reader.

RT-qPCR

Total RNA was extracted from cells and tissues with TRIzol (Beyotime, R0016) reagent. A Prime Script RT Kit and a miRNA First Strand cDNA Synthesis Kit were used to reverse transcribe 1 μg of RNA to cDNA according to the manufacturer’s specifications. RT-qPCR was performed with SYBR^® Green Master Mix (Bio-Rad, Part Number: 1725270) or a miRNA Real-time RT-qPCR Kit (Ybscience, Part Number: YB131042-25). The specific sequences of primers used were shown in Table 1.

Table 1 Primer sequences.

Genes	Sequences
Lnc-MGC	F: 5'-ACGAATACCCCACTGGAG-3'
	R: 5'-GTCGAACACAAAGAAGGAA-3'
miR-495-3p	F: 5'-GCGAAACAAACATGGTGCA-3'
	R: 5'-AGTGCAGGGTCCGAGGTATT-3'
GAPDH	F: 5'-ATGGTGAAGGTCGGTGTGAACG-3'
	R: 5'-GGTCTCGCTCCTGGAAGATGGT-3'

F: Forward primer; R: Reversed primer.

Hematoxylin-eosin staining

After dewaxing and hydration, the tissue sections were rinsed with distilled water for 3 minutes, after which a Hematoxylin-eosin staining kit (Solarbio, Beijing, China) was used to stain the sections according to the manufacturer's instructions. Finally, the sections were sealed with neutral gum and observed and imaged under a microscope.

TUNEL staining for apoptosis

A TUNEL assay kit (Roche, Germany) was used to assess apoptosis according to the manufacturer’s specifications. In brief, immobilized ARPE-19 cells were placed on ice with 4% paraformaldehyde solution for 15 minutes. The cells were flushed and permeated on ice with 0.1% Triton X-100 and 0.1% sodium citrate for 2 minutes. Then, the cells were flushed with PBS 3 times and incubated with the TUNEL reaction mixture for 1 hour. Finally, the number of apoptotic cells was assessed by fluorescence microscopy (Nikon, Japan; Cat. No. 400857)[30].

ROS detection

The fluorescent probe DCFH-DA provided in the ROS assay kit (Solarbio, Beijing, China) was added to each sample and incubated in a cell incubator for 30 min without light. The ROS level was measured by flow cytometry.

ELISA

Conditioned medium was harvested from different groups of ARPE-19 cells or collected from rat serum and centrifuged at 4 °C for 5 minutes. The levels of the inflammatory factors TNF-a, IL-6, and IL-1β were subsequently measured using ELISA kits (Solarbio, Beijing, China).

Dual-luciferase gene assay

The wild-type (WT) or mutant 3' untranslated region of lnc-MGC or GRP78 incorporating miR-495-3p binding sites was cloned and inserted into the pmirGLO luciferase reporter vector (MAIBio, LM-1439). lnc-MGC-WT, lnc-MGC-mutation (MUT), GRP78-WT, and GRP78-MUT were transfected into 293T cells together with miR-495-3p mimics or normal control (NC) mimics using Lipofectamine 2000 (Invitrogen, Catalog No. 11668019). After 48 h of transfection, luciferase activity was determined.

Statistical analysis

GraphPad Prism 8.0 was used to analyze the experimental data and plot the graphs. The Shapiro-Wilk test was employed to ensure normality. If the data passed the normality test, we utilized the following statistical tests: Student’s t-test for comparing two groups; one- or two-factor analysis of variance (ANOVA) for multigroup comparisons; and Tukey’s post hoc test. P < 0.05 was considered statistically significant. The cell-based experiments included 3 parallel experiments with 3 technical replicates for each experimental group. The animal experiments were conducted in triplicate, with 3 technical replicates in each experimental group.

RESULTS

Differential expression of lnc-MGC and miR-495-3p in DR rat retinas and cell models

In this study, we first investigated the differential expression of lnc-MGC and miR-495-3p in DR rat retinas and cell models. The RT-qPCR results of the levels of lnc-MGC and miR-495-3p demonstrated that the level of lnc-MGC was markedly elevated whereas that of miR-495-3p was significantly decreased in both the DR rat retina and the HG cell model (Figure 1A and B). FISH was used to detect the localization of lnc-MGC and miR-495-3p in ARPE-19 cells, and the results showed that both were localized in the nucleus (Figure 1C and D). These results demonstrated that lnc-MGC was observably upregulated in the DR rat retina and HG cell model, whereas miR-495-3p was observably decreased, both of which were differentially expressed.

Figure 1 Differential expression of lnc-MGC and miR-495-3p in the diabetic retinopathy rat retina and cell model. A: Detection of lnc-MGC and miR-495-3p expression in diabetic retinopathy rat retina tissue by RT-qPCR; B: Expression levels of lnc-MGC and miR-495-3p in the cell model were detected by RT-qPCR; C: Localization of lnc-MGC in ARPE-19 cells was detected by fluorescence in situ hybridization (FISH). The scale bar represents 10 μm; D: Localization of miR-495-3p in ARPE-19 cells was detected by FISH. The scale bar represents 10 μm. ^aP < 0.001. NC: Normal control; DR: Diabetic retinopathy; HG: High glucose.

Inhibition of lnc-MGC protects retinal pigment epithelial cells

We then verified the role of lnc-MGC in the regulation of DR progression. The level of lnc-MGC was markedly lower in the HG + si-lnc-MGC group, indicating successful transfection of si-lnc-MGC (Figure 2A). The cell viability and apoptosis results demonstrated that the transfection of si-lnc-MGC increased the vigor of ARPE-19 cells induced by HG and decreased the rate of apoptosis (Figure 2B and C). Compared with the NC group, the HG group presented increased expression levels of inflammatory factors and inflammation, whereas the expression levels of TNF-α, IL-6 and IL-1β in the HG group were markedly decreased after the cells were transfected with si-lnc-MGC (Figure 2D). Flow cytometry demonstrated that the transfection of si-lnc-MGC reduced ROS levels in HG-treated cells (Figure 2E). Western blotting indicated that the levels of Bax and cleaved-caspase 3 were prominently elevated by HG, which was reversed by the presence of si-lnc-MGC. The opposite was true for Bcl-2 (Figure 2F). These results indicate that the inhibition of lnc-MGC has a protective effect on RPE cells.

Figure 2 Inhibition of lnc-MGC protects retinal pigment epithelial cells. A: RT-qPCR was used to detect the transfection efficiency; B: CCK-8 was used to detect cell proliferation activity; C: TUNEL was used to detect cell apoptosis. The scale bar represents 100 μm; D: ELISA was used to detect the expression of the inflammatory factors TNF-α, IL-6 and IL-1β; E: Flow cytometry was used to detect reactive oxygen species; F: Western blotting was used to detect the expression of the apoptosis-related proteins Bax, Bcl-2 and cleaved-caspase 3. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. NC: Normal control; HG: High glucose; ROS: Reactive oxygen species.

Downregulation of lnc-MGC can mediate the protective effect of retinal pigment epithelial cells through ER stress and ferroptosis

After confirming that lnc-MGC can promote the progression of DR, we investigated the specific regulatory mechanism involved. Kusaka et al[31] reported that ER stress was related to retinal nerve cell damage, and we speculated that the progression of DR was also related to ER stress. Then, we treated the cells with tunicamycin or erastin and measured cell viability and apoptosis. The addition of tunicamycin or erastin significantly decreased cell viability and increased apoptosis in the HG + si-lnc-MGC group (Figure 3A and B). Compared with those in the HG + si-lnc-MGC group, after the addition of tunicamycin or erastin, the levels of intracellular inflammatory factors were significantly increased, and the degree of cellular inflammation was aggravated (Figure 3C); the addition of tunicamycin or erastin increased the levels of ROS in the HG cell group after the addition of si-lnc-MGC (Figure 3D). Finally, the levels of interrelated proteins were measured by Western blotting after tunicamycin or erastin was added to the HG + si-lnc-MGC group. The expression levels of the Bax, cleaved-caspase 3, GRP78, CHOP, XBP-1, and HO-1 proteins were markedly increased, whereas the expression levels of the GPX4, FTH1, xCT, and Bcl-2 proteins were significantly decreased, indicating increased ER stress and the promotion of cellular ferroptosis (Figure 3E-G). These results confirmed that downregulation of lnc-MGC can mediate the protective effect on retinal pigment epithelial cells through ER stress and ferroptosis.

Figure 3 Downregulation of lnc-MGC can mediate the protective effect of retinal pigment epithelial cells through ER stress and ferroptosis. A: CCK-8 detection of cell proliferation activity; B: TUNEL detection of cell apoptosis. The scale bar represents 100 μm; C: ELISA detection of the expression of the inflammatory cytokines TNF-α, IL-1β and IL-6; D: Flow cytometry detection of reactive oxygen species; E: Western blotting was used to detect the expression of the apoptosis-related proteins Bax, Bcl-2 and cleaved-caspase 3; F: Western blotting was used to detect the expression of the ER-related proteins GRP78, CHOP and XBP-1; G: Western blotting was used to detect the expression of the ferroptosis-associated proteins GPX4, HO-1, FTH1, and xCT. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. NC: Normal control; HG: High glucose; ROS: Reactive oxygen species.

Correlation between lnc-MGC and miR-495-3p

Based on previous studies, we validated the targeting relationship between lnc-MGC and miR-495-3p. The potential targeting relationship between lnc-MGC and miR-495-3p was predicted using the StarBase. The predicted results showed that there was a targeted binding site between lnc-MGC and miR-495-3p (Figure 4A). The targeting relationship was further confirmed by double luciferase gene reporting assay. The results revealed that the luciferase activity in the combined region of miR-495-3p and lnc-MGC-WT was significantly reduced, whereas the luciferase activity in lnc-MGC-MUT was not observably changed (Figure 4B). Our results show that lnc-MGC can specifically bind to miR-495-3p. RT-qPCR revealed that si-lnc-MGC prominently upregulated miR-495-3p compared to the HG group (Figure 4C). These results indicate that lnc-MGC targets miR-495-3p.

Figure 4 Correlation between lnc-MGC and miR-495-3p. A: The StarBase was used to predict the binding sites of lnc-MGC and miR-495-3p; B: Dual-luciferase gene analysis was used to verify the targeting binding relationship between lnc-MGC and miR-495-3p; C: The expression of miR-495-3p was detected by RT-qPCR. ^aP < 0.001, ^bP < 0.01. WT: Wild type; MUT: Mutation; NC: Normal control; HG: High glucose.

lnc-MGC affects DR progression through miR-495-3p

After verifying the targeting relationship between lnc-MGC and miR-495-3p, we further explored the effect of lnc-MGC on the progression of DR through miR-495-3p. The miR-495-3p inhibitor was transfected into the cells, and the level of miR-495-3p decreased upon inhibition of its expression (Figure 5A). CCK-8 and TUNEL assays indicated that, compared with the HG + si-lnc-MGC group, transfection with the miR-495-3p inhibitor inhibited cell viability and promoted cell apoptosis (Figure 5B and C). Restraint of miR-495-3p increased the levels of TNF-α, IL-6, and IL-1β in the HG + si-lnc-MGC group (Figure 5D). Flow cytometry indicated that the miR-495-3p inhibitor promoted the content of ROS in the HG + si-lnc-MGC group of cells (Figure 5E). Finally, when the levels of interrelated proteins were measured, transfection with the miR-495-3p inhibitor increased the levels of Bax and cleaved-caspase 3 proteins and decreased the level of the Bcl-2 protein compared with those in the HG + si-lnc-MGC group (Figure 5F). In addition, the protein expression levels of GRP78, CHOP, XBP-1, and HO-1 were also increased, whereas the protein expression levels of GPX4, FTH1, and xCT were decreased in the HG + si-lnc-MGC + miR-495-3p inhibitor group (Figure 5G and H). These results indicate that knockdown of lnc-MGC can upregulate miR-495-3p, inhibit ER stress, inhibit ferroptosis, and alleviate DR progression.

Figure 5 lnc-MGC affects the progression of diabetic retinopathy through miR-495-3p. A: Transfection efficiency was detected by RT-qPCR; B: Cell proliferation was detected by CCK-8 assay; C: Cell apoptosis was detected by TUNEL assay. The scale bar represents 100 μm; D: The expression of the inflammatory cytokines TNF-α, IL-1β, IL-6 was detected by ELISA; E: reactive oxygen species were detected by flow cytometry; F-H: Western blotting was used to detect the expression of related proteins. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. Inhibitor: MiR-495-3p inhibitor; NC: Normal control; HG: High glucose; ROS: Reactive oxygen species.

miR-495-3p targets GRP78

We further verified the targeted regulation of GRP78 by miR-495-3p. TargetScan predicted that miR-495-3p could target GRP78 (Figure 6A), and we then confirmed the targeting relationship between them. The dual-luciferase reporter assay results revealed that the relative luciferase activity of the GRP78-WT ⁺miR-495-3p mimic group was clearly decreased (Figure 6B). These results show that miR-495-3p targets GRP78.

Figure 6 miR-495-3p targets GRP78. A: The TargetScan was used to predict binding sites for miR-495-3p and GRP78; B: Dual-luciferase verification that miR-495-3p targets GRP78. ^aP < 0.01. WT: Wild type; MUT: Mutation; NC mimic: Negative control mimic.

miR-495-3p regulates ER stress and ferroptosis through GRP78

It has been confirmed that miR-495-3p has a targeting relationship with GRP78. We verified whether miR-495-3p is involved in GRP78-mediated regulation of ER stress and ferroptosis. After confirming that the miR-495-3p mimic and oe-GRP78 were successfully transfected into HG-induced cells (Figure 7A and B), we performed verification experiments. Cell viability and apoptosis assays showed that the miR-495-3p mimic increased cell viability and decreased apoptosis in the HG group, whereas oe-GRP78 reversed these effects (Figure 7C and D). Transfection of the miR-495-3p mimic reduced the levels of inflammatory cytokines and ROS in the HG group, and oe-GRP78 attenuated this downregulation (Figure 7E and F). Western blot analysis revealed that the miR-495-3p mimic downregulated the expression levels of the Bax, cleaved-caspase 3, GRP78, CHOP, XBP-1 and HO-1 proteins in HG-treated cells. Bcl-2, GPX4, FTH1 and xCT protein expression levels were significantly increased, and this regulatory effect was reversed after transfection with oe-GRP78 (Figure 7G-I). These results indicate that miR-495-3p regulates ER stress and ferroptosis through GRP78.

Figure 7 miR-495-3p regulates endoplasmic reticulum stress and ferroptosis through GRP78. A: RT-qPCR was used to detect transfection efficiency; B: Western blot was used to detect transfection efficiency; C: CCK-8 was used to detect cell proliferation activity; D: A TUNEL assay was used to detect apoptosis. The scale bar represents 100 μm; E: The expression of TNF-α, IL-1β and IL-6 was detected by ELISA; F: Reactive oxygen species were detected by flow cytometry; G-I: Western blotting was used to detect the expression of related proteins. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. NC: Normal control; HG: High glucose; ROS: Reactive oxygen species; mimic: miR-495-3p mimic.

LEF1, as a binding protein of lnc-MGC, regulates the progression of DR

LEF1 has been identified as an important transcription factor for the Wnt signaling pathway, which has been implicated in cell proliferation in tumors[32]. We hypothesized that LEF1 regulation may underlie the role of lnc-MGC-binding proteins in the regulation of DR progression. Utilizing the database on the JASPAR website (https://jaspar.genereg.net/), we found that the transcription factor LEF1 can be established with lnc-MGC promoter DNA sequences and has three possible binding sites (Figure 8A and B). si-LEF1 and oe-lnc-MGC were then transfected into the cells, and Western blotting and RT-qPCR confirmed that the transfection was successful (Figure 8C and D). Compared with HG, si-LEF1 increased cell viability and decreased apoptosis, whereas oe-lnc-MGC reversed this regulatory influence (Figure 8E and F). Transfection of si-LEF1 reduced the expression of inflammatory cytokines and intracellular ROS in HG-treated cells, and oe-lnc-MGC alleviated this downregulation (Figure 8G and H). Finally, the expression levels of relevant proteins were measured, and the results revealed that the transfection of si-LEF1 could reduce the degree of apoptosis of the cells in the HG group, attenuate the ER stress response, and inhibit the ferroptosis response of the cells in the HG group, which was retrogressed by oe-lnc-MGC (Figure 8I-K). These results demonstrate that LEF1 can be used as a binding protein of lnc-MGC to regulate DR progression.

Figure 8 LEF1, as the binding protein of lnc-MGC, regulates the progression of diabetic retinopathy. A and B: LEF1 can bind to the lnc-MGC promoter; C: Western blotting was used to detect the inhibition efficiency of LEF1; D: RT-qPCR was used to detect the transfection efficiency; E: CCK-8 was used to measure the cell proliferation activity; F: TUNEL was used to detect cell apoptosis. The scale bar represents 100 μm; G: ELISA was used to determine the expression of the inflammatory cytokines TNF-α, IL-1β and IL-6; H: Flow cytometry was used to detect reactive oxygen species; I-K: Western blotting was used to detect the expression of related proteins. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. NC: Normal control; HG: High glucose; ROS: Reactive oxygen species.

In vivo, LEF1 combined with lnc-MGC regulates DR progression

To further elucidate the mechanism by which LEF1 combined with lnc-MGC regulates DR, we conducted animal experiments to verify that after si-LEF1 treatment, the body weight of DR rats was elevated, the blood glucose content was decreased, and oe-lnc-MGC reversed the effect of si-LEF1 (Figure 9A and B). The levels of apoptosis, inflammatory factors, and ROS in DR rats markedly decreased after the injection of si-LEF1, whereas the coinjection of oe-lnc-MGC alleviated this downregulation effect (Figure 9C-E). The results showed that the injection of si-LEF1 could reduce the degree of apoptosis in DR rats, weaken the ER stress response, and inhibit the ferroptosis response, whereas the injection of oe-lnc-MGC alleviated this regulatory effect (Figure 9F-H). Changes in retinal thickness and structure were observed by HE staining. Compared with those of the rats in the NC group, the thickness of each layer of retinal tissue in the DR group and the thickness of the photoreceptor cell layer were observably greater, and cellular edema and edema of the retinal ganglion cell layer and plexiform outer layer occurred, whereas the knockdown of LEF1 reduced the severity of DR, and the overexpression of lnc-MGC restrained this reduction effect (Figure 9I).

Figure 9 Animal experiments to verify that LEF1 combined with lnc-MGC regulates diabetic retinopathy progression. A and B: Diabetic retinopathy rat body weight and blood glucose detection; C: A TUNEL assay was used to detect apoptosis. The scale bar represents 100 μm; D: ELISA was used to detect the expression of the inflammatory factors TNF-α, IL-1β and IL-6; E: Reactive oxygen species were detected by flow cytometry; F-H: Western blotting was used to detect the expression of related proteins; I: HE staining was used to observe the retinal structure. The scale bar represents 20 μm. ^aP < 0.05, ^bP < 0.01, and ^cP < 0.001. ONL: Outer nuclear layer; INL: Inner core layer; RGC: retinal ganglion cells; NC: Normal control; DR: Diabetic retinopathy; ROS: Reactive oxygen species.

DISCUSSION

In recent decades, studies have shown that DR is largely the product of microvascular complications, inflammation and retinal neurodegeneration that develop in the later stages of diabetes[33]. Many studies have focused on the influences of a single genetic aberration in animal models of DR, all of which have provided deeper insights into how this inflammatory, neurodegenerative, and microvascular process occurs in DR[34]. For example, normal cell‒cell mutual effects are varied by diabetes, resulting in severe vascular abnormalities, loss of the blood-retinal barrier, and impaired neuronal function[35]. Previous studies have identified the differential expression of different miRNAs as latent biomarkers for the diagnosis and treatment of DR[36]. LncRNAs also participate in DR progression[37]. Our study revealed that, in both the DR rat model and the HG-induced cell model, the expression of lnc-MGC was high, whereas the expression of miR-495-3p was decreased. Moreover, downregulation of lnc-MGC promoted the viability of HG-induced ARPE-19 cells and inhibited apoptosis and the production of inflammatory factors and ROS, indicating that inhibiting the expression of lnc-MGC could alleviate HG-induced damage to ARPE-19 cells and thus alleviate DR progression.

Interestingly, emerging evidence suggests a clear association between ER stress-related physiological functions and the pathogenesis of DR[38]. Therefore, we explored the regulatory role of ER stress in DR. The ER plays important roles in protein and lipid synthesis, ion homeostasis, and organelle communication[39]. An imbalance in ER homeostasis in beta cells can lead to loss of function and cell death and further induce diabetes in humans[40]. In addition, Alzheimer's disease is caused by the irregular accumulation of proteins due to changes in the protein balance of ER stress[41]. In previous studies, blueberry anthocyanin extract was found to restrain oxidative stress and ER stress to reduce HG-induced apoptosis and prevent the development of DR[42]. ER stress and ferroptosis are closely related, and some studies have shown that ferroptosis is related to ulcerative colitis intestinal epithelial cell death mediated by ER stress signals[43]. In addition, ferroptosis was found to be associated with RPE cell death induced by oxidative stress in vitro[44]. Moreover, previous studies have shown that multiple mediators are involved in ferroptosis, including ROS accumulation, GSH depletion, and GPX4 activity inhibition. In addition, HO-1, FTH1 and xCT also mediate the occurrence of ferroptosis[45]. Our study revealed that downregulation of lnc-MGC could reduce HG-induced ER stress and inhibit ferroptosis in ARPE-19 cells and that the chaperone GRP78 is essential for the regulation of multiple processes, including ER stress and inflammation[46].

ER stress cannot be separated from the participation of its molecular chaperone GRP78, which can prevent protein misfolding and aggregation by increasing ER activity to assemble new peptides[47]. In addition, GRP78 targets misfolded proteins for degradation through the proteasome degradation pathway, thereby restoring the balance of ER stress to normal[48]. Studies have shown that GRP78 is able to bind to unfolded proteins and subsequently fold or degrade them through ER-associated protein degradation pathways[49]. The increased expression of GRP78 is considered to be an important marker of ER stress[50]. Previous studies have shown that GRP78 can promote mitochondrial stability and regulate autophagy after ER stress, which indicates that its translocation dependence plays an important role in regulating autophagy[51]. Previously, Blackwood et al[52] reported that ER stressors can promote the secretion of GRP78 in cardiomyocytes and protect cardiomyocytes. Consistent with these studies, our study revealed that the overexpression of GRP78, a molecular chaperone in the ER, significantly inhibited cell viability, promoted cell apoptosis and the production of inflammatory factors, increased intracellular ROS accumulation, and promoted ER stress and ferroptosis.

MiRNAs usually repress gene expression at the posttranscriptional level[53]. However, other means of regulation have been identified, such as transcriptional regulation through interactions with promoters or cofactors of transcription factors in the nucleus[54,55]. MiR-495-3p is considered a therapeutic target for the assessment of sepsis-related acute kidney injury because its expression is reduced in the tissues and cells assessed for sepsis-associated acute kidney injury, and it interacts with other genes to inhibit cell proliferation and promote apoptosis[56]. Inhibiting miR-495-3p can improve angiogenesis after ischemia[57]. A study revealed that miR-495-3p participates in ER stress to regulate kidney injury induced by sepsis[58]. Therefore, we examined whether miR-495-3p affects DR progression by influencing ER stress. In this study, we found that lnc-MGC can target and regulate miR-495-3p, while miR-495-3p can target and bind to GRP78, and the inhibition of lnc-MGC can upregulate the expression of miR-495-3p, thus affecting the expression of GRP78 and thereby reducing ER stress and alleviating ferroptosis in cells. Inhibition of lnc-MGC has a protective effect on HG-induced ARPE-19 cells and alleviates DR progression.

LEF1, a transcription factor located on chromosomes, is involved in the occurrence of a variety of diseases and regulates gene expression by inducing changes in the helical structure of DNA[59].Our study revealed that LEF1 can bind to the promoter region of lnc-MGC, affecting the activity of the lnc-MGC promoter, and that downregulating LEF1 can increase cell viability, inhibit cell apoptosis, reduce the production of inflammatory factors and ROS, and inhibit ER stress and ferroptosis, whereas overexpressing lnc-MGC can reverse the downregulatory effect of LEF1. These results indicated that LEF1 could be used as a lnc-MGC-binding protein to regulate DR progression, a conclusion that was also verified in animal experiments.

CONCLUSION

Our study demonstrated that LEF1 regulates the miR-495-3p/GRP78 axis via lnc-MGC to influence ER stress-mediated DR and ferroptosis in RPE cells. The results of this study provide a theoretical basis for therapeutic strategies centered on ER stress-mediated DR. However, this study still has certain limitations. This study relied on in vitro cell models and animal models. Although these models stimulate the DR environment to a certain extent, further verification in the clinical setting will be required. Moreover, this study focused mainly on the LEF1/lnc-MGC/miR-495-3p/GRP78 axis; other potential regulatory mechanisms have not yet been explored. Future research directions include further validating these molecular mechanisms in clinical samples, exploring the role of this signaling axis in other diabetic complications, and expanding to other related molecules and pathways to comprehensively understand the pathogenesis of DR.