6-Gingerol Regulates the Exosomal miR-148b-3p/ACSL4/Ferroptosis Axis in an in vitro Model of Diabetic Retinopathy Authors: Shuang Liu, Juan Ye, Long Suo, Tao Xu, Yiyi Chen, Wei Liu Published: 2026 ABSTRACT AbstractBackgroundDiabetic retinopathy (DR), a leading cause of blindness, is driven by redox imbalance and ferroptosis in retinal endothelial cells. While 6-Gingerol exhibits antioxidant properties, its role in modulating exosome-mediated redox signaling in DR remains unexplored. This study investigates whether 6-Gingerol protects against DR by regulating ferroptosis via the exosomal miR-148b-3p/ACSL4 axis.MethodsHuman retinal microvascular endothelial cells (HRMECs) were exposed to high glucose (HG) ± 6-Gingerol. Ferroptosis was assessed via lipid ROS, intracellular iron, and key markers (GPX4, ACSL4, FTH1). Exosomal miRNAs were profiled by sequencing, with miR-148b-3p validated as a regulator of ACSL4 through luciferase assays. Functional rescue experiments confirmed pathway specificity.Results6-Gingerol suppressed HG-induced ferroptosis, reducing lipid ROS by 2.1-fold (P<0.01) and iron accumulation by 1.8-fold (P<0.05), while upregulating GPX4 (3.2-fold, P<0.001). Exosomal miR-148b-3p was downregulated under HG but restored by 6-Gingerol (4.5-fold, P<0.001). miR-148b-3p directly targeted ACSL4, whose knockdown mimicked 6-Gingerol’s effects, whereas ACSL4 overexpression or miR-148b-3p inhibition abolished protection.Conclusions6-Gingerol mitigates DR progression by inhibiting ferroptosis through exosomal miR-148b-3p/ACSL4-dependent redox regulation. These findings unveil a novel crosstalk between exosomal miRNAs and iron metabolism, positioning 6-Gingerol as a potential redox-targeted therapy for DR. Nonetheless, as the data are exclusively obtained from an in vitro model, in vivo validation is essential prior to making any translational inferences.Keywords: 6-gingerol, ferroptosis, miR-148b-3p/ACSL4 axis, diabetic retinopathy, exosome Introduction Diabetic retinopathy (DR), the most prevalent microvascular complication of diabetes, remains a leading cause of blindness in working-age adults.1 While current therapies targeting vascular endothelial growth factor (VEGF) show partial efficacy, high rates of recurrence and treatment resistance underscore the need for alternative strategies addressing DR’s root causes.2 Emerging evidence implicates redox imbalance and iron dysregulation as key drivers of retinal endothelial dysfunction, with ferroptosis—a form of iron-dependent cell death marked by lipid peroxidation—playing a pivotal role in DR progression.3,4 Hyperglycemia-induced oxidative stress disrupts retinal microvascular homeostasis by depleting glutathione (GSH), inactivating glutathione peroxidase 4 (GPX4), and accumulating lipid reactive oxygen species (ROS).5,6 These changes trigger ferroptotic cell death in endothelial cells, exacerbating vascular leakage and ischemia7. Notably, the lipid-metabolizing enzyme acyl-CoA synthetase long-chain family member 4 (ACSL4), a critical ferroptosis regulator, is upregulated in diabetic retinas and promotes lipid peroxidation.3,7 Conversely, microRNAs (miRNAs) transported by exosomes have emerged as redox-sensitive mediators of intercellular communication in diabetic complications.8 For instance, exosomal miR-144-3p exacerbates ferroptosis by targeting GPX4 in osteosarcoma,9 while miR-125b-5p from adipose stem cell exosomes suppresses ferroptosis via the Keap1/Nrf2 pathway.10 Recent research has established a direct connection between exosomal microRNAs (miRNAs) and ferroptosis in retinal diseases. Specifically, exosomal miR-125b-5p, derived from mesenchymal stem cells, has been demonstrated to inhibit ferroptosis in retinal microvascular endothelial cells in the context of diabetic retinopathy by targeting the P53 pathway.11 These findings indicate that exosomal miRNAs may constitute a crucial, albeit underexplored, regulatory mechanism of ferroptosis in diabetic retinopathy. 6-Gingerol (6-Gin), a bioactive phenolic compound from ginger, has demonstrated anti-inflammatory, antioxidant, and anti-diabetic properties.12–14 It activates Nrf2 to counteract oxidative stress in cardiovascular diseases13 and inhibits ferroptosis in cancer via ubiquitin-specific peptidase 14 (USP14) modulation.15 In DR, 6-Gingerol alleviates retinal inflammation and angiogenesis.16 However, whether 6-Gin exerts protective effects against DR by regulating exosome-mediated miRNA transfer and ferroptosis suppression has not been investigated. This study aims to elucidate the mechanism by which 6-Gin modulates the miR-148b-3p/ACSL4 axis via exosomes, thereby inhibiting ferroptosis and protecting retinal endothelial cells. Our findings may reveal a novel therapeutic approach for mitigating diabetic retinopathy progression. Using human retinal microvascular endothelial cells (HRMECs), we (1) Identify miR-148b-3p as a glucose-sensitive exosomal miRNA that targets ACSL4; (2) Demonstrate that 6-Gin upregulates miR-148b-3p to inhibit ACSL4-driven ferroptosis; (3) Establish the miR-148b-3p/ACSL4 axis as a novel redox regulatory pathway in DR. Our findings reveal a previously unrecognized mechanism by which 6-Gin restores endothelial redox homeostasis, provides a natural compound-based strategy for targeting exosome-mediated signaling in DR and addresses an unmet need in DR treatment by focusing on ferroptosis, distinct from current VEGF-centric therapies. We recognize that the HRMEC monoculture model doesn’t fully mimic the complex in vivo retinal environment of diabetic retinopathy, which involves interactions among neurons, glia, and pericytes. Thus, our findings indicate a potential mechanistic pathway rather than a comprehensive disease model, requiring validation in co-culture or animal models. Methods Cell Culture and Transfection Human retinal microvascular endothelial cells (HRMECs) were obtained from IMMOCELL (Xiamen, China) and cultured in a complete ECM medium supplemented with 100 U/mL penicillin and 100 mg/L streptomycin (Cat 15140122, Sigma, USA). The cells were maintained in a humidified incubator at 37°C with 5% CO2. Overexpression plasmids for ACSL4 (pcDNA-ACSL4) and empty control vectors were provided by Geenseed Biotech (China). The miR-148b-3p mimics, inhibitors, and their respective scramble negative controls (NC) were synthesized by RiboBio (Guangzhou, China). Cells were seeded in 6-well plates at 60% confluence and transfected with 50 nM of miRNA mimics or 1 µg of overexpression plasmids using Lipofectamine 3000 (L3000015, Life Technologies, USA) according to the manufacturer’s instructions. Following a 48-hour transfection period, the cells were harvested for further analysis. For 6-gingerol treatment, cells were exposed to a medium containing 25 µM of 6-Gin for 24 hours before collection. HRMECs cultured under normal glucose (NG, 5 mM D-glucose) served as the control group. High glucose (HG, 30 mM D-glucose) was applied to induce diabetic-like oxidative stress and ferroptosis, with or without 6-Gingerol (25 µM) treatment. In preliminary experiments, mannitol (25 mM) was used as an osmotic control to confirm that the observed effects were attributable to glucose-induced metabolic stress rather than osmotic pressure. Cell Viability Cells were seeded in 96-well plates at a density of 5000 cells per well and cultured for 24 hours. Afterward, 5 μL of CCK-8 reagent (C0038, Beyotime, China) was added to each well and incubated for an additional 4 hours. Following incubation, the supernatants were gently removed, and 100 μL of DMSO was added to dissolve the formazan crystals. Cell proliferation was assessed by measuring the absorbance at 570 nm using a microplate reader (Bio-Rad, USA), and the relative values were used to generate proliferation curves. Isolation and Identification of Exosomes HRMECs were seeded and exposed to low glucose (5.5mM) or high glucose (30mM) conditions for 48 h, respectively, followed by exosome extraction from the culture supernatants using ultracentrifugation. Briefly, the collected supernatants were first centrifuged at 2000 × g for 20 minutes at 4°C to remove dead cells, followed by centrifugation at 10,000 × g for 30 minutes at 4°C to eliminate cell debris. The resulting supernatants were transferred to clean ultracentrifuge tubes and subjected to ultracentrifugation at 100,000 × g for 70 minutes at 4°C. The supernatants were discarded, and the pelleted exosomes were resuspended in PBS, then ultracentrifuged again at 100,000 × g for another 70 minutes. The final exosome pellets were resuspended in PBS for immediate use or stored at −80°C. To assess batch-to-batch reproducibility, three independent exosome isolations were performed. Western blot analysis was performed to confirm the presence of exosome-specific markers, including CD9 and TSG101. The size distribution and concentration of exosomes were determined using nanoparticle tracking analysis (NTA) with a Nanosight NS300 system (Malvern, UK), and the morphology of exosomes was visualized using a transmission electron microscope (TEM, JEM-3010, JEOL, Japan). To assess exosome internalization, exosomes were labeled with the Cyanine5.5 fluorescent dye (HY-D0924, MCE, USA) following the manufacturer’s instructions. The labeled exosomes were then added to cells cultured in 6-well plates and incubated for 12 hours at 37°C. After incubation, the cells were washed with PBS to remove any unbound exosomes, and the uptake of exosomes was visualized using a confocal laser scanning microscope (Carl Zeiss, Germany). Transwell Assay The transfected cells were collected and resuspended in serum-free medium. A total of 2 × 104 cells were seeded into the upper chamber of a Transwell insert. The lower chamber was filled with medium containing 10% FBS as a chemoattractant. After 24 hours of incubation at 37°C, the cells that had migrated through the membrane were fixed with 4% paraformaldehyde and stained with crystal violet. The migrated cells were then counted and photographed under a microscope, with five random fields analyzed per well. The experiment was conducted in triplicate to ensure reproducibility. TUNEL Assay The TUNEL assay was carried out on HRMECs using the One Step TUNEL Apoptosis Assay Kit (C1086, Beyotime, China) in accordance with the manufacturer’s instructions. The sections were incubated with digoxigenin-dNTP and terminal deoxynucleotidyl transferase to label fragmented DNA, followed by treatment with anti-digoxigenin antibodies conjugated with fluorescein to detect apoptotic cells. Nuclei were counterstained with DAPI to facilitate visualization. Apoptotic cells were identified and quantified by counting five randomly selected fields under 400× magnification in a blinded manner to minimize bias. Detection of Ferroptosis Cells were treated with 5 µM C11-BODIPY (S0043M, Beyotime, China) to assess lipid ROS levels by incubating in the dark at 37 °C for 30 minutes. Following incubation, the cells were harvested and resuspended in phenol-red-free medium. The fluorescence signal was subsequently measured using a BD FACsCalibur system to evaluate oxidative stress. Additionally, intracellular iron levels were quantified using the Iron Assay Kit (ab83366, Abcam, USA) according to the manufacturer’s protocol, providing a measure of iron accumulation in cells. Western Blot Assay Cells were lysed using ice-cold radioimmunoprecipitation assay (RIPA) buffer containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 0.1% SDS, 1% Triton X-100, and 1% sodium deoxycholate, supplemented with protease and phosphatase inhibitors (Cat 78440, Thermo, USA). The total protein concentration was determined using the BCA Protein Assay Kit (Thermo, USA). Equal amounts of protein samples were loaded and separated by SDS-PAGE, followed by transfer onto nitrocellulose membranes. The membranes were blocked with 5% skim milk and incubated with primary antibodies specific to CD9 (1:2000 dilution, ab223055, Abcam, USA), TSG101 (1:2000 dilution, ab223055, Abcam, USA), ACSL4 (1:2000 dilution, ab155282, Abcam, USA), GPX4 (1:2000 dilution, ab252833, Abcam, USA), FTH1 (1:2000 dilution, ab183781, Abcam, USA), and β-actin (1:2000 dilution, ab8226, Abcam, USA). After incubation with HRP-conjugated secondary antibodies (Abcam, USA), the protein bands were visualized using an enhanced chemiluminescence (ECL) system (A38554, Thermo, USA). Quantitative PCR Total RNA was isolated from cultured cells using Trizol reagent (Takara, Japan) following the manufacturer’s instructions. Subsequently, 1 μg of RNA was converted to cDNA using the PrimeScript RT reagent kit with gDNA Eraser (RR037Q, Takara, Japan). Quantitative PCR (qPCR) was conducted with a SYBR-based real-time PCR system using SmartChip® Real-Time PCR System (640022, Takara, Japan). The cycling conditions were set as follows: initial denaturation at 94 °C for 30 seconds, followed by 40 cycles of denaturation at 94 °C for 5 seconds and annealing at 60 °C for 30 seconds. Gene expression levels were quantified using the comparative Ct method, with β-actin serving as the internal reference. The relative expression of target genes was calculated using the 2ΔΔCt method.17 Luciferase Reporter Gene Assay The ACSL4 promoter region spanning positions −1113 to +195 was amplified from genomic DNA. Both the wild-type (WT) and mutant ACSL4 promoter fragments were cloned into the pGL3-Basic vector and confirmed by sequencing. HEK293T cells were co-transfected with the ACSL4 promoter constructs and corresponding miRNA using Lipofectamine 3000. After 48 hours, luciferase activity was measured using a dual-luciferase reporter assay system (E1910, Promega, Madison, WI, USA). Firefly luciferase activity was normalized to Renilla luciferase activity, and the results were presented as the firefly/Renilla luciferase ratio. Statistics All statistical analyses were conducted using GraphPad Prism 8.0 and SPSS 22.0 software. Data are expressed as mean ± SEM from multiple independent experiments, each performed in triplicate, and each experiment was repeated at least three times. Group differences were assessed using Student’s t-test or one-way ANOVA for normally distributed data; for nonparametric data, the Mann–Whitney test was used. All tests were two-tailed, and a P-value of < 0.05 was considered statistically significant. Results 6-Gingerol Protects HRMECs from Dysfunction and Ferroptosis in a High-Glucose Environment To investigate the therapeutic potential of 6-Gin in DR, we first assessed its effects on HRMECs under high-glucose (HG, 30 mM) conditions. CCK-8 assays demonstrated that 6-Gin treatment (10–50 μM) significantly improved HRMECs viability in a dose-dependent manner, with optimal protection observed at 25 μM (Figure 1A). Furthermore, Transwell assay revealed that 6-Gin enhanced HRMECs migration capacity by approximately 1.8-fold compared to HG-treated controls (Figure 1B). TUNEL staining demonstrated that 6-Gin (25 μM) reduced HG-induced apoptosis by 42% (Figure 1C). Importantly, we observed that 6-Gin profoundly inhibited HG-associated ferroptosis, as evidenced by a 2.1-fold decrease in intracellular labile iron pool (LIP) (Figure 1D) and a 65% reduction in lipid peroxidation compared to HG-treated cells (Figure 1E). At the molecular level, 6-Gin treatment significantly modulated key ferroptosis regulators (Figure 1F and G). qRT-PCR and Western blot analyses showed that 6-Gin suppressed ACSL4 expression by 60%, while upregulating the antioxidant enzyme GPX4 (1.7-fold) and the iron storage protein ferritin heavy chain 1 (FTH1, 2.3-fold). These findings demonstrate that 6-Gin protects HRMECs from HG-induced damage by suppressing aberrant ferroptosis through coordinated regulation of both iron metabolism and lipid peroxidation pathways. 6-Gingerol protects HRMECs from dysfunction and ferroptosis in a high-glucose environment. (A) CCK-8 analysis of cellular viability. (B) Transwell analysis of cell migration (C) TUNEL staining using to analyze apoptosis. (D) The total iron level was analyzed by flow cytometry. (E) Lipid ROS was analyzed by flow cytometry. (F) qPCR analysis of the mRNA expression of ACSL4, GPX4 and FTH1. (G) Western blotting analysis of the protein expression of ACSL4, GPX4 and FTH1. (bar: mean ± SE, n = 3, ***P<0.001 vs cont group, ##P<0.01 HG+6G group vs HG group, ###P<0.001 HG+6G group vs HG group). Identification of Glucose-Regulated Exosomes and Dysregulated miR-148b-3p in HRMECs To elucidate the role of exosomal signaling in diabetic retinopathy, we first established a robust exosome isolation and characterization protocol. Transmission electron microscopy (TEM) revealed typical cup-shaped vesicles with diameters of 30–150 nm from both normal glucose (NG, 5.5 mM) and high glucose (HG, 30 mM)-treated HRMECs (Figure 2A). Nanoparticle tracking analysis (NTA) demonstrated that HG conditions increased exosome secretion by 2.8-fold compared to NG controls (P<0.01), with peak particle concentrations at 110 nm (Figure 2A). Three independent exosome preparations yielded consistent results across all characterization assays. The exosomal nature of these vesicles was confirmed through (1) Western blot showed strong expression of exosome-specific proteins CD9 and TSG101, while excluding the negative control calnexin (Figure 2B); (2) Confocal microscopy using PKH67-labeled exosomes demonstrated rapid internalization (>80% incorporation within 4h) by recipient HRMECs (Figure 2C). To identify key regulatory miRNAs in diabetic conditions, we performed small RNA sequencing of HG vs NG exosomes (Figure 2D). Bioinformatic analysis revealed 73 significantly upregulated miRNAs (eg, miR-21-5p, miR-155-5p) and 73 significantly downregulated miRNAs (including miR-126-3p, miR-148b-3p). Notably, miR-148b-3p showed the most dramatic downregulation (−7.2 fold, FDR<0.001) under HG conditions (Figure 2E). This miRNA was selected for further investigation due to its predicted targeting of ACSL4 (a key ferroptosis regulator) and previously reported roles in metabolic diseases.7 Identification of glucose-regulated exosomes and dysregulated miR-148b-3p in HRMECs. (A) The exosomes were examined using nanoparticle tracking analysis (NTA) and transmission electron microscopy (TEM). (B) Western blotting was employed to confirm the presence of the exosomal marker proteins CD9 and TSG101. (C) Confocal laser scanning immunofluorescence was utilized to assess the uptake of exosomes by HRMECs. (D) High-throughput miRNA sequencing identified 73 miRnas that were upregulated and 73 miRnas that were downregulated. (E) qPCR was used to validate the expression levels of these miRnas. (F) HRMECs were treated with 6-gingerol, and the expression of miR-148b-3p was subsequently verified using qPCR (bar: mean ± SE, n = 3, **P<0.01 vs NC group, ***P<0.001 vs NC group). Remarkably, 6-Gin (25 μM) treatment reversed HG-induced miR-148b-3p suppression, restoring its expression to 85% of NG control levels (P<0.001, Figure 2F). This suggests 6-Gin may exert its therapeutic effects through specific modulation of exosomal miRNA profiles. miR-148b-3p Mediates HRMEC Protection and Ferroptosis Suppression To functionally characterize miR-148b-3p’s role in diabetic retinopathy, we performed gain-of-function experiments in HRMECs under high glucose stress. Cell viability (CCK-8 assay) and migratory capacity (Transwell assay) were significantly enhanced by miR-148b-3p mimic transfection (50 nM), showing 1.9-fold increased viability compared to negative control (NC) mimic (P<0.001, Figure 3A) and 2.4-fold greater migration through 8 μm pores (P<0.01, Figure 3B). TUNEL staining confirmed miR-148b-3p’s anti-apoptotic effects, reducing HG-induced DNA fragmentation by 63% vs NC controls (Figure 3C). More critically, miR-148b-3p exhibited potent anti-ferroptotic properties. Additionally, miR-148b-3p exhibited potent anti-ferroptotic properties that reduce intracellular Fe2+ by 55% (P<0.01, Figure 3D) and increased Fe3+ storage capacity (+2.1-fold ferritin aggregation, Prussian blue staining). Meanwhile, it lowered lipid ROS by 48% (C11-BODIPY assay, P<0.01, Figure 3E) and increased total GSH/GSSG ratio (1.7-fold vs NC). Mechanistically, qPCR and immunoblotting revealed miR-148b-3p’s dual regulatory action on ferroptosis pathways (Figure 3F and G). Both mRNA and protein level of ACSL4 are decreased while the protein level of FTH1 and GPX4 are increased. miR-148b-3p mediates HRMEC protection and ferroptosis suppression. After transfection with miR-148b-3p mimics, CCK-8, Transwell and TUNEL assay were used to detect cell viability (A), migration (B) and apoptosis (C), respectively; flow cytometry detected the levels of Fe2+ and lipid ROS (D and E); ACSL4, GPX4 and FTH1 protein and mRNA expressions were detected by qPCR and Western blot, respectively (F and G). (bar: mean ± SE, n = 3, **P<0.01 vs miR mimic group). miR-148b-3p Directly Targets ACSL4 to Mediate 6-Gingerol’s Protective Effects We systematically validated the targeting relationship between miR-148b-3p and ACSL4 and its protective mechanism in diabetic retinopathy through multi-dimensional experiments. The miRNA–ACSL4 seed match was shown (Figure 4A). First, the overexpression efficiency of miR-148b-3p in HRMECs was confirmed, with mimic transfection increasing expression approximately 400-fold (P<0.001), while transfection of its inhibitor reduced endogenous levels by 85%, indicating a robust experimental system (Figure 4B). Bioinformatics analysis further identified ACSL4 as a potential target of miR-148b-3p, with three prediction algorithms (TargetScan, miRDB, miRWalk) recognizing a conserved binding site in its 3’UTR region (1123–1129 nt). Moreover, ACSL4 was found to be significantly associated with the ferroptosis pathway (KEGG: hsa04216) and diabetic complications (Reactome: R-HSA-382551), reinforcing the biological rationality of this regulatory interaction. miR-148b-3p directly targets ACSL4 to mediate 6-Gingerol’s protective effects. (A–C) miR-148b-3p treated HRMEC, (n=3, **P<0.01 vs miR-148b-3p group). (A) The expression of miR-148b-3p was analyzed by qPCR. (B) The activity of ACSL4 3’UTR was analyzed by double luciferase reporter gene. (C) qPCR was used to analyze the expression of ACSL4. (D–F) Treating HRMECs with 6-gingerol or 6-gingerol combined with miR-148b-3p inhibitor, (bar: mean ± SE, n = 3, **P<0.01 vs 6G group, ##P<0.01 6G+inhibitor group vs 6G group). (D) qPCR was used to analyze the expression of miR-148b-3p. (E) The activity of ACSL4 3’UTR was analyzed by double luciferase reporter gene. (F) qPCR was used to analyze the expression of ACSL4. To verify this interaction, a dual-luciferase reporter assay was performed using constructs containing either wild-type (WT) or mutant (MUT) ACSL4 3’UTR. The results demonstrated that miR-148b-3p mimic significantly suppressed luciferase activity in WT constructs (reduction by 62%, P<0.001), whereas no effect was observed in MUT constructs (Figure 4C). This confirms that miR-148b-3p exerts its regulatory effect through specific binding to the ACSL4 3’UTR. Additionally, overexpression of miR-148b-3p in HRMECs reduced mRNA levels of ACSL4 (by 68% and 54%), whereas inhibition of miR-148b-3p reversed this effect, indicating that the suppression of ACSL4 by miR-148b-3p is both direct and reversible (Figure 4D). To explore the physiological significance of this regulatory relationship, the study investigated the role of the natural compound 6-Gin. Treatment with 6-Gin significantly upregulated miR-148b-3p while downregulating ACSL4. However, when combined with the miR-148b-3p inhibitor, these effects were completely abolished (Figure 4E and F). These results not only confirm that 6-Gin’s mechanism depends on the miR-148b-3p/ACSL4 axis but also further strengthen the notion that miR-148b-3p exerts a negative regulatory effect on ACSL4. ACSL4 Regulates HRMEC Function and Ferroptosis Knockdown of ACSL4 significantly enhanced cell viability and migration while reducing apoptosis in HRMECs (Figure 5A–C). Conversely, ACSL4 overexpression exerted the opposite effects, impairing cellular function (Figure 6A–C). At the molecular level, ACSL4 silencing markedly decreased ferroptosis-associated indicators, including intracellular iron levels and reactive oxygen species (ROS) accumulation (Figure 5D and E), while upregulating key ferroptosis-suppressive proteins, FTH1 and GPX4 (Figure 5F and G). In contrast, ACSL4 overexpression exacerbated ferroptosis, further confirming its critical role in regulating ferroptotic cell death in HRMECs (Figure 6D–G). Effects of knocking down the expression of ACSL4 on the function and ferroptosis of HRMECs. After knocking down the expression of ACSL4, CCK-8, Transwell and TUNEL assay were used to detect cell viability (A), migration (B) and apoptosis (C), respectively; flow cytometry detected the levels of Fe2+ and lipid ROS (D and E); ACSL4, GPX4 and FTH1 protein and mRNA expressions were detected by qPCR and Western blot, respectively (F and G). (bar: mean ± SE, n = 3, **P<0.01 vs siACSL4 group). Effects of overexpression of ACSL4 on the function and ferroptosis of HRMECs. After overexpression of ACSL4, CCK-8, Transwell and TUNEL assay were used to detect cell viability (A), migration (B) and apoptosis (C), respectively; flow cytometry detected the levels of Fe2+ and lipid ROS (D and E); ACSL4, GPX4 and FTH1 protein and mRNA expressions were detected by qPCR and Western blot, respectively (F and G). (bar: mean ± SE, n = 3, **P<0.01 vs ceACSL4 group). 6-Gingerol Modulates Ferroptosis via miR-148b-3p/ACSL4 Axis Rescue experiments demonstrated that 6-Gin effectively downregulated ACSL4 while upregulating miR-148b-3p in high glucose (HG)-treated HRMECs. However, these protective effects were reversed by either miR-148b-3p inhibition, or ACSL4 overexpression. Functional assays further confirmed that 6-Gin’s beneficial effects—including enhanced cell viability, reduced apoptosis, and suppression of ferroptosis (Figure 7A–E)—are dependent on the miR-148b-3p/ACSL4 regulatory axis. These findings conclusively establish that 6-Gin exerts its protective role against ferroptosis in diabetic retinopathy by modulating the miR-148b-3p-mediated suppression of ACSL4. 6-Gingerol modulates ferroptosis via miR-148b-3p/ACSL4 axis. HRMECs cells induced by HG were treated with 6-gingerol, or 6-gingerol combined with miR-148b-3p inhibitor or overexpressed ACSL4. (A) CCK-8 detected cell activity. (B) Transwell detected cell migration. (C and D) Flow cytometry detected the levels of Fe2+ and lipid ROS. (E) qPCR analyzed ACSL4, GPX4 and FTH1 expression. (bar: mean ± SE, n = 3, ***P<0.001 vs cont group, ###P<0.001 HG+6G group vs HG group, ^^P<0.01 HG+6G+ACSL4 group vs HG+6G+inhibitor group, ^^^P<0.001 HG+6G+inhibitor or HG+6G+ACSL4 group vs HG+6G group). Discussion Diabetic retinopathy (DR) persists as a leading cause of vision loss worldwide, with retinal endothelial cell dysfunction representing a critical pathogenic driver.1,18 The present study uncovers a novel redox-sensitive mechanism by which the bioactive compound 6-Gin protects against DR progression. We demonstrate that 6-Gin inhibits HG-induced ferroptosis in HRMECs through modulation of the exosomal miR-148b-3p/ACSL4 axis, a proposed working model is summarized in Figure 8. This finding aligns with the growing recognition of ferroptosis as a key mediator of endothelial injury in diabetic microangiopathy.4,19 Schematic showing the mechanism of 6-gingerol attenuates diabetic retinopathy by suppressing ferroptosis via exosomal miR-148b-3p/ACSL4 axis. High glucose suppresses miR-148b-3p in exosomes from HRMECs and increases ACSL4 expression, leading to ferroptosis. This is indicated by higher ROS levels and reduced GPX4 and FTH1 expression. 6-Gingerol can counteract this process, suggesting its potential in dietary interventions to prevent diabetic retinopathy. Our data show that 6-Gin significantly attenuated HG-triggered lipid reactive oxygen species (ROS) accumulation, intracellular iron overload, and downregulation of the antioxidant enzyme GPX4–hallmarks of the ferroptotic process. This protection was accompanied by upregulation of miR-148b-3p in HRMECs-derived exosomes, which directly targeted the ferroptosis regulator ACSL4. Importantly, both miR-148b-3p overexpression and ACSL4 knockdown recapitulated the beneficial effects of 6-Gin, whereas miR-148b-3p inhibition or ACSL4 overexpression abolished this protection. These results establish the miR-148b-3p/ACSL4 pathway as a critical redox-sensitive mechanism through which 6-Gin confers anti-ferroptotic effects in retinal endothelial cells. Our findings complement recent studies demonstrating the role of exosomal miRNAs in regulating ferroptosis. For instance, exosomal miR-125b-5p from adipose stem cells was shown to alleviate pulmonary microvascular endothelial cell ferroptosis by modulating the Keap1/Nrf2/GPX4 axis.10 Similarly, exosomal miR-144-3p was found to promote ferroptosis in osteosarcoma by targeting GPX4.9 By identifying miR-148b-3p as a novel redox-sensitive miRNA that suppresses ACSL4 and ferroptosis in the diabetic retina, our study expands the repertoire of exosome-mediated signaling pathways that govern endothelial cell fate in vascular complications. However, analogous miRNA-ferroptosis pathways are present in other contexts, and our findings contribute to this body of knowledge by applying it to DR, rather than proposing a completely novel mechanism. Although our in-vitro system uses HRMECs as both donor and recipient, this setting is biologically plausible because extracellular vesicles (EVs)/exosomes operate not only in paracrine but also in autocrine modes, whereby vesicles released to the pericellular space are rapidly re-captured by the same cell population via clathrin-mediated endocytosis and macropinocytosis, establishing short-range feedback loops that modulate stress responses and phenotype stability.20,21 Under hyperglycemic oxidative stress, exosome biogenesis and cargo sorting are dynamically regulated through ESCRT/ceramide (nSMase2) and Rab27 pathways and by RNA-binding adaptors (eg, YBX1/hnRNPA2B1), enabling shifts in miRNA payloads that tune lipid redox homeostasis.22,23 In this context, 6-Gingerol’s activation of redox-protective programs (eg, Nrf2 axis reported across cardiovascular/metabolic models) provides a mechanistic basis for biasing exosomal miRNA loading toward miR-148b-3p and for enhancing an autocrine, self-reinforcing suppression of ACSL4-dependent ferroptosis in HRMECs, even within a monoculture.21,24,25 Thus, the “same-cell” source–recipient configuration reflects an autocrine EV circuit that is well-documented in endothelial and other systems and is sufficient to explain how 6-Gingerol can exert therapeutic-like effects through exosomal miRNA modulation in our model. Nonetheless, this autocrine hypothesis remains conjectural in the absence of direct experimental validation. Our data highlight the miR-148b-3p/ACSL4 axis as a key player in 6-Gin’s anti-ferroptotic effects, but other pathways might also be involved. The Nrf2 pathway, known for regulating antioxidant responses, could influence ferroptosis via GPX4 and HO-1, though its role compared to the miR-148b-3p/ACSL4 axis is unclear.26 Additionally, Nrf2-independent mechanisms like iron chelation, modulation of ALOX lipoxygenases, ferritinophagy, or the mevalonate pathway might play a role.27 Other exosomal miRNAs could also affect ferroptosis.11 Comprehensive studies using specific inhibitors and genetic tools are needed to fully understand 6-Gin’s protective mechanisms. The ability of 6-Gin to upregulate miR-148b-3p and inhibit ACSL4-dependent ferroptosis represents a promising therapeutic strategy for DR. ACSL4, a lipid-metabolizing enzyme that catalyzes the formation of long-chain fatty acyl-CoAs, is a critical determinant of ferroptosis susceptibility.28 Its upregulation in diabetic retinas and association with myocardial dysfunction underscore the importance of targeting this pathway.6,29–31 By restoring the miR-148b-3p/ACSL4 axis, 6-Gin may offer a multifaceted approach to preserving endothelial redox homeostasis and delaying DR progression. Notably, we found that GPX4 expression increased when ACSL4 was inhibited, reflecting an adaptive feedback within the lipid peroxidation network of ferroptosis. As ACSL4 promotes the incorporation of polyunsaturated fatty acids into phospholipids,32,33 its inhibition reduces the pool of oxidizable substrates and alleviates lipid peroxide accumulation, thereby relieving the oxidative pressure on GPX4 and facilitating its stabilization.34 In addition, activation of redox-regulatory pathways, such as Nrf2- and SREBP1-mediated transcriptional programs, may further enhance GPX4 expression as part of a compensatory antioxidant response.35,36 Consistent with previous findings that pharmacologic or genetic suppression of ACSL4 often elevates GPX4 levels, our data suggest that the observed GPX4 upregulation represents an intrinsic self-limiting mechanism by which cells restore redox homeostasis and counteract excessive ferroptotic stress. Clinically, these findings provide mechanistic insight with potential translational value for diabetic retinopathy (DR). Endothelial ferroptosis has been increasingly recognized as an early and targetable driver of microvascular degeneration in diabetes,19,29 preceding irreversible neurovascular damage. Consistent with recent clinical evidence demonstrating that systemic inflammatory burden correlates with the severity of diabetic retinopathy in patients with type 2 diabetes mellitus, our findings suggest that redox dysregulation and ferroptosis in retinal endothelial cells may contribute to the pathophysiology underlying DR progression.37 By identifying the exosomal miR-148b-3p/ACSL4 axis as a regulatory node of ferroptosis, our study highlights a novel biomarker and therapeutic target for monitoring and modulating retinal endothelial injury.8 The ability of 6-Gingerol to restore this axis suggests that dietary or pharmacologic supplementation of this compound could complement current anti-VEGF or antioxidant regimens, offering a multitarget strategy that addresses the redox and metabolic roots of DR rather than its late angiogenic manifestations.25 Moreover, because ferroptosis contributes to broader diabetic microangiopathies—including nephropathy and cardiomyopathy,30 our results may extend beyond ocular disease, pointing to ferroptosis modulation as a shared therapeutic avenue for systemic vascular protection in diabetes. However, these clinical implications should be considered preliminary, given that all data have been obtained from in vitro experiments. Notably, 6-Gin’s antioxidant and anti-inflammatory properties have been previously documented in diabetic complications, including its ability to activate the Nrf2 pathway and suppress angiogenesis.13,25 Our findings extend this knowledge by revealing a previously unrecognized mechanism involving exosome-mediated miRNA regulation of ferroptosis. This suggests that 6-Gin’s therapeutic potential for DR may involve the integration of complementary redox-modulatory pathways. Conclusion In conclusion, our study demonstrates that 6-Gin inhibits HG-induced ferroptosis in retinal endothelial cells by upregulating exosomal miR-148b-3p, which in turn suppresses the ferroptosis regulator ACSL4. These findings reveal a redox signaling axis that governs endothelial dysfunction in DR, although similar miRNA-ferroptosis mechanisms have been reported in other systems, and suggest a potential role for 6-Gin that warrants further in vivo investigation as a multitarget therapy for this blinding disease. Future investigations should focus on validating these mechanistic insights in vivo and exploring potential synergies with existing DR treatments. This study has several limitations: (1) In vitro HRMECs model may not fully mimic diabetic retinopathy’s complex in vivo milieu; (2) The broader functional impacts of miR-148b-3p beyond ACSL4 regulation remain unexplored; (3) Ferroptosis-independent effects of 6-Gin were not assessed; (4) Translational concerns regarding drug delivery and pharmacokinetics require further study. Future work should incorporate diabetic animal models and human tissue validation. Funding Statement This work was supported by Jiangsu Traditional Medicine Science and Technology Development Project (MS2022136); Natural Science Foundation of Suqian (K202312); and Scientific Research Project of Suqian First People’s Hospital (KY202317). Data Sharing Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. Author Contributions Shuang Liu: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data Curation, Writing–Original Draft, Visualization, Funding acquisition. Juan Ye: Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data Curation, Writing – Review & Editing. Wei Liu: Conceptualization, Methodology, Resources, Writing–Review & Editing, Supervision, Project administration, Funding acquisition. Tao Xu: Formal analysis, Visualization, Software, Investigation, Writing – review & editing. Long Suo: Formal analysis, Visualization, Project administration, Funding acquisition, Writing – review & editing. Yiyi Chen: Formal analysis, Visualization, Software, Investigation, Writing – review & editing. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. Disclosure The authors declare no competing interests. A version of this paper was made available as a pre-print: https://doi.org/10.21203/rs.3.rs-7357034/v1. References Associated Data Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.