The Wnt/StarD7 axis protects retinal ganglion cells from glutamate excitotoxicity by inhibiting ferroptosis

Abstract

Background

Glutamate (Glu) accumulation-induced excitotoxicity is a major cause of retinal ganglion cell (RGC) death in glaucoma, and the role of ferroptosis, a novel form of cell death, is critical in this process. The aim of this study was to investigate the function and regulatory mechanisms of the lipid transport protein StarD7 in RGC ferroptosis.

Methods

An N-methyl-D-aspartate (NMDA)-induced retinal excitotoxicity mouse model and a Glu-induced RGC cell model were constructed for experimental investigation. RT‒qPCR and Western blotting were used to assess the expression of related genes and proteins, HE staining was used to assess pathological retinal damage, and kits were used to evaluate ferroptosis-related indicators.

Results

Ferroptosis was involved in NMDA-induced RGC damage in glaucoma mice. StarD7 expression was upregulated in glaucoma, and overexpression of StarD7 decreased the levels of total iron, Fe²⁺, ROS, and MDA in vitro and in vivo while increasing the expression levels of GSH, GPX4, and xCT, thereby suppressing RGC ferroptosis. Mechanistically, Glu treatment significantly reduced the expression of the Wnt signaling pathway proteins Wnt1 and β-catenin. Activating the Wnt/β-catenin pathway promoted StarD7 expression, which in turn inhibited Glu-induced ferroptosis in mRGCs.

Conclusion

The Wnt/β-catenin signaling pathway inhibits Glu-induced RGC ferroptosis by upregulating StarD7 expression, revealing the potential neuroprotective role of StarD7 in glaucoma treatment and providing a scientific basis for the development of new therapeutic strategies.

Clinical trial number

Not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13062-026-00835-y.

Introduction

Glaucoma is the leading cause of irreversible blindness and is characterized primarily by the loss of retinal ganglion cells (RGCs) [1]. The global prevalence of glaucoma among individuals aged 40 to 80 years is estimated to be 3.5% [2]. With the increasing aging population, approximately 111.8 million people are projected to be affected by glaucoma by 2040 [3]. Elevated intraocular pressure (IOP) remains the most significant risk factor and therapeutic target for glaucoma, but other factors also play crucial roles in disease onset and progression [4]. Therefore, exploring new molecular mechanisms beyond IOP control is essential for understanding the neuroprotective effects of RGCs.

Glutamate (Glu), the primary excitatory neurotransmitter in the central nervous system, plays a central role in neuronal communication [5]. In the progression of glaucoma, the excitotoxicity of Glu caused by excessive activation of N-methyl-D-aspartate (NMDA) receptors due to an overabundance of Glu has been demonstrated to play a critical role in RGC death [6]. Additionally, Glu is recognized as a typical inducer of ferroptosis [7], which is an iron-dependent programmed cell death process that has been shown to be involved in the pathophysiological processes of glaucoma [8]. For instance, proteomic analysis by Suo et al. [9] reported that ferroptosis may play a significant role in RGC loss in NMDA-induced glaucoma models. Other studies have shown that iron chelators can mitigate NMDA-induced neuronal damage by reducing iron levels and oxidative stress in vivo [10]. These findings suggest that inhibiting ferroptosis could serve as a potential therapeutic approach in NMDA-induced glaucoma models.

StarD7, a lipid transport protein, was initially discovered in a chorionic carcinoma cell line (JEG-3) and subsequently detected in a series of tumor cell lines and in normal cells/tissues [11]. It is involved in various physiological processes, such as lipid transfer, metabolism, and signaling pathway regulation [12]. Studies have shown that mutations in the StarD7 gene are associated with mitochondrial dysfunction and neurodegenerative diseases [13], but no research to date has indicated that StarD7 is involved in the pathological progression of glaucoma. Notably, one study identified StarD7 as a critical factor in intracellular coenzyme Q transport and an inhibitor of ferroptosis [14]. Therefore, the aim of this study was to further investigate whether StarD7 is involved in RGC ferroptosis.

Wnt pathway proteins belong to the family of secreted lipid-modified glycoproteins involved in various physiological processes, such as cell proliferation, apoptosis, organ development, and embryogenesis [15]. They have been shown to significantly influence the pathological progression of neuronal diseases such as glaucoma [16]. For example, activation of Wnt/β-catenin in Müller cells protects against NMDA-mediated excitotoxic damage to RGCs [17]. Downregulation of expression of factors associated with the Wnt/β-catenin pathway can increase IOP in the trabecular meshwork (TM) of patients with primary open-angle glaucoma (POAG) [18]. Additionally, the Wnt/β-catenin signaling pathway has been demonstrated to regulate ferroptosis in other diseases [19], and Rena et al. [20] reported that the Wnt/β-catenin pathway can also promote the transcription of StarD7 by modulating its gene promoter. However, whether the Wnt signaling pathway alleviates NMDA-induced RGC injury by regulating StarD7-mediated ferroptosis remains to be further explored.

Therefore, in this study, we aimed to investigate the effects of StarD7 on RGC ferroptosis and the regulatory role of the Wnt signaling pathway in this process by constructing in vitro and in vivo models of Glu excitotoxicity to provide new insights and methods to mitigate the progression of glaucoma.

Materials and methods

Bioinformatics analysis

The glaucoma RGC gene expression profile dataset GSE12596 was downloaded from the GEO public database. This dataset was constructed on the basis of RNA sequencing data from RGCs from three rats with experimental glaucoma (unilateral elevation of IOP). An equal number of RGCs were collected from normal eyes and eyes with elevated IOP. Gene expression in glaucomatous RGCs was compared with that in fellow-eye RGCs. The fellow eye served as the control. The raw data were obtained in the form of a gene count matrix. Data processing and statistical analysis were completed in the R language environment (v4.3.0). The edgeR package (v3.42.0) was used to normalize the raw counts, after which the normalized counts were converted to log2 count values.

The identification of differentially expressed genes (DEGs) was performed using the limma package (v3.56.0). The screening thresholds for differentially expressed genes were set as |log2FC| ≥ 1 and P value < 0.05. For visualization, the pheatmap package was used to generate heatmaps of the differentially expressed genes. Ten highly expressed genes and ten genes with low expression levels were selected for heatmap plotting. The R package ggplot2 was used to construct volcano plots of the differentially expressed genes, which clearly illustrated the changes in gene expression levels and statistical significance.

Retinal excitotoxicity mouse model construction and experimental design

Sixty 6–8-week-old male C57BL/6 mice (weight: 18–22 g) were purchased from Hunan Slack Jinda Experimental Animal Co., Ltd. After one week of acclimatization under specific pathogen-free conditions with a 12 h/12 h light‒dark cycle, the mice were randomly divided into 5 groups: control, NMDA, NMDA + Fer-1, NMDA + AAV-NC, and NMDA + AAV-StarD7. NMDA induces RGC damage through excitatory neurotoxicity [21, 22]. In this study, a mouse model of retinal excitotoxicity was established via intravitreal injection of NMDA. In accordance with the methods of Liu et al. [6], mice were first anesthetized with 3% isoflurane inhalation, followed by the topical application of oxybuprocaine hydrochloride eye drops (HY-B1288; MedChemExpress, USA) and tropicamide eye drops (HY-B0321; MedChemExpress, USA) for ocular anesthesia and pupil dilation. Subsequently, 1 µL of 20 mM NMDA (M3262; Sigma‒Aldrich, USA) was injected into the bilateral vitreous cavities using a 33-gauge needle connected to a 10-µL syringe (Hamilton Company, USA). The control group received an equivalent volume of saline injected into the vitreous cavities. The entire procedure was performed under a stereomicroscope. Post-surgery, 0.3% tobramycin ointment (HY-B0441; MedChemExpress, USA) was applied to maintain eye moisture and prevent bacterial infections. To investigate the role of ferroptosis, on the basis of previous methods [23], 2 µg/kg ferrostatin-1 (Fer-1; HY-100579; MedChemExpress, USA) was administered via intravitreal injection 1 h prior to model induction in mice.

Adenoviral vectors targeting StarD7, AAV-NC and AAV-StarD7 were purchased from OBiO Technology Company (Shanghai, China). To investigate the role of StarD7, AAV-NC and AAV-StarD7 were administered via intravitreal injection as previously described [24]. Briefly, 1.5 µL of AAV-StarD7 (2.3 × 10¹² vg/mL) was injected into the vitreous body of mice using a 10 µL microsyringe equipped with a 33-gauge needle. The injection was performed slowly over 10 s, and the needle was left in place for 1 min to allow diffusion of the viral solution. The NMDA + AAV-NC group received intravitreal injections of AAV-NC at the same concentration. The mice were housed for 1 week before they underwent NMDA-induced ocular procedures. All the mice were euthanized 24 h after NMDA induction for subsequent analyses (5 mice per group for retinal immunofluorescence staining and HE staining; another 5 mice were used for ferroptosis-related indicators, Western blot detection, and RT‒qPCR analysis).

Cell culture, transfection, and model construction

Mouse retinal ganglion cells (mRGCs) were purchased from Procell Life Science & Technology Co., Ltd. (CP-M122; Wuhan, China) and cultured in the provided medium (CM-M122; Procell, China) in a standard cell culture incubator at 37 °C with 5% CO₂. mRGCs isolated by the Procell Laboratory were identified by immunofluorescence staining for CD90 and were free of mycoplasma contamination. A cell model was established by treating the cells with 10 mM Glu [25, 26] (G1626; Sigma‒Aldrich, USA) for 24 h at 37 °C and 5% CO₂.

StarD7 knockdown (si-StarD7), StarD7 overexpression (oe-StarD7), and control (si-NC or oe-NC) plasmids were synthesized by GenePharma (Shanghai, China). Cells were grown in 6-well plates, and when the cell density reached 60–70%, the plasmids were transfected into mRGCs using Lipofectamine 3000 reagent (L3000150; Invitrogen, USA). Transfection efficiency was assessed via Western blotting or RT‒qPCR after 48 h.

To investigate the effect of StarD7 on Glu-induced ferroptosis in mRGCs, oe-StarD7-transfected cells were treated with or without 10 µM of the ferroptosis activator erastin [27] (HY-15763; MedChemExpress, USA) for 24 h after 48 h of transfection, followed by induction with Glu. To explore the role of the Wnt signaling pathway in Glu-induced mRGCs, mRGCs were treated with 1 mM of the Wnt signaling pathway activator lithium chloride (LiCl; V900067; Sigma‒Aldrich, USA) for 30 min 24 h prior to Glu induction. Additionally, to investigate whether the Wnt signaling pathway affects Glu-induced ferroptosis in mRGCs through StarD7, cells were transfected with si-StarD7 48 h before LiCl treatment.

HE staining

Mouse eyeballs were fixed in 4% formaldehyde, dehydrated through a graded alcohol series, and embedded in paraffin. The samples were subsequently cut into 3 μm vertical slices using a cryostat microtome (Leica, Germany). The sections were stained using an HE staining kit (G1120; Solarbio, China), and pathological changes in the retinal tissues were observed under an optical microscope (Olympus, Japan).

Immunofluorescence staining

Multiple splicing RNA-binding protein (RBPMS) is a specific marker of RGCs. Immunofluorescence detection of RBPMS was performed via retinal flat mounts. After the mice were euthanized, the eyeballs were removed and fixed in 4% paraformaldehyde (G1101; Servicebio, China) for 1 h. The retinas were then isolated, blocked in PBS containing 0.5% Triton X-100 for 1 h, and incubated overnight at 4 °C with a rabbit anti-RBPMS antibody (1:200; ab152101; Abcam, UK). The retinas were subsequently washed three times with PBS containing 0.5% Triton X-100 (5 min each), followed by incubation with fluorescently labeled secondary antibodies at room temperature in the dark for 2 h. Finally, observation and imaging were conducted using a fluorescence microscope (Leica, Germany).

Immunofluorescence staining for StarD7 was performed on retinal vertical paraffin sections and mRGCs. Vertical paraffin sections were dewaxed, and endogenous peroxidase activity was blocked with 3% hydrogen peroxide. After 5 min, the sections were immersed in citrate sodium antigen retrieval solution (C1032; Solarbio, China) and subjected to antigen retrieval in a microwave oven. Moreover, the treated mRGCs were fixed with 4% paraformaldehyde at room temperature for 20 min, washed three times with PBS, permeabilized with 0.1% Triton X-100 for 10 min, and blocked with 4% BSA for 1 h. The processed sections or cells were subsequently incubated overnight at 4 °C with anti-StarD7 antibody (1:200; PA5-115966; Thermo Fisher Scientific, USA). The next day, the samples were incubated with fluorescently labeled secondary antibodies in the dark for 2 h, followed by nuclear staining with DAPI (C1106; Beyotime, China), after which they were observed and imaged under a fluorescence microscope (Leica, Germany).

Fe²⁺ assessment

A Tissue Iron Content Assay Kit (BC4355; Solarbio, China), Elabscience^® Total Iron/Ferrous Iron Colorimetric Assay Kit (E-BC-K880-M/E-BC-K881-M; Elabscience, China), and Ferrous Iron Assay Kit (BC5415; Solarbio, China) were used to measure total iron and Fe²⁺ levels in retinal tissues and mRGCs. After mouse retinal tissues or mRGCs were homogenized or lysed, the supernatant was collected by centrifugation. Working solutions were added to determine the total iron and Fe²⁺ contents in the tissues or cells.

ROS assessment

Total ROS levels were measured using CellROX™ Green (C10444; Thermo Fisher Scientific, USA). Briefly, treated mouse retinal tissues or cells were washed with prewarmed PBS, incubated with 5 µM CellROX Green at room temperature for 30 min, and imaged using a fluorescence microscope (Leica, Germany).

MDA and GSH assessments

An MDA assay kit (A003-1-2) and a GSH assay kit (A006-2-1) from Nanjing Jiancheng Bioengineering Institute (China) were used to determine the levels of MDA and GSH in retinal tissue homogenates and cells. A microplate reader was used to measure the OD values at 532 nm and 420 nm.

Western blot analysis

Proteins were extracted from mouse retinal tissues and mRGCs using RIPA lysis buffer (P0013B; Beyotime, China) supplemented with protease phosphatase inhibitors (A32959; Thermo Fisher Scientific, USA). Protein concentrations were quantified using a BCA protein assay kit (23225; Thermo Fisher Scientific, USA). The protein samples were separated by SDS‒PAGE and then transferred to PVDF membranes (Millipore, USA). The membranes were blocked with QuickBlock™ blocking buffer (P0231; Beyotime, China) for 10 min, after which they were incubated overnight with primary antibodies against GPX4 (1:1000; ab125066; Abcam, UK), xCT (1:1000; ab307601; Abcam, UK), StarD7 (1:1000; PA5-115966; Thermo Fisher Scientific, USA), Wnt1 (1:25; ab15251; Abcam, UK), β-catenin (1:1000; ab32572; Abcam, UK), and β-actin (ab8227; Abcam, UK). The membranes were then incubated with HRP-conjugated secondary antibodies at room temperature for 2 h. The protein bands were visualized using an enhanced chemiluminescence kit (Millipore, USA) and a chemiluminescence imager (Tanon Science & Technology, China) and analyzed with ImageJ software.

RT‒qPCR

Total RNA was extracted from retinal tissues and mRGCs using TRIzol reagent (15596026CN; Invitrogen, USA), and the RNA concentration and purity were measured with a Nanodrop2000 microultraviolet spectrophotometer (Nanodrop, USA). RNA was subsequently reverse transcribed into cDNA using a PrimeScript RT kit (RR037A; TaKaRa, China). qPCR was performed on a 7500 Real-Time System (Applied Biosystems, USA) using a TB Green Premix Ex Taq II (Tli RNase H Plus) kit (RR820A; TaKaRa, China). β-actin was used as the internal reference, and relative quantification was performed using the 2^−ΔΔCt method. The primer (5′-3′) sequences used were as follows: StarD7 forward: 5’-CTCCCGAACCAAAGTCCCC-3’, reverse: 5’-GGTGGGTGCCTGTAATTGGG-3’; β-actin forward: 5’-GTGACGTTGACATCCGTAAAGA-3’, reverse: 5’-GCCGGACTCATCGTACTCC-3’.

Statistical analysis

The data were analyzed using GraphPad Prism 8.0 (GraphPad Software, USA) and are expressed as the mean ± standard deviation (SD). The Shapiro–Wilk test and Levene test were used to evaluate the normality of the data and the homogeneity of variance, respectively. All the data are suitable for parameter tests. Independent t tests were used for comparisons between two groups. One-way analysis of variance (ANOVA) was used for comparisons among multiple groups, followed by Tukey’s test to assess differences between groups. P < 0.05 was considered to indicate statistical significance.

Results

Ferroptosis is involved in NMDA-induced RGC damage in mice with glaucoma

We first investigated the effect of ferroptosis on NMDA-induced RGC damage in mice with glaucoma. Retinal staining using RBPMS, a marker of RGCs, revealed clearly visible RBPMS-labeled RGCs in the control group but a significant reduction in the number of RBPMS-labeled RGCs in the NMDA group. Fer-1 injection resulted in a notable recovery of the number of RGCs (Fig. 1A). HE staining revealed that the retinal structure in the control group was clear and orderly, whereas compared with that in the control group, the total retinal thickness in the NMDA group was significantly reduced. Fer-1 injection partially restored the total retinal thickness (Fig. 1B). The detection of ferroptosis-related indicators revealed that compared with the control group, the NMDA group exhibited significantly elevated levels of total iron, Fe²⁺, ROS, and MDA and significantly decreased GSH levels. Fer-1 injection reduced the total iron, Fe²⁺, ROS, and MDA levels but increased the GSH content (Fig. 1C–G). Additionally, the Western blot results demonstrated that compared with the control group, the NMDA group had reduced expression of GPX4 and xCT, whereas Fer-1 injection increased the expression levels of GPX4 and xCT (Fig. 1H). These results indicate that ferroptosis promotes NMDA-induced RGC damage in mice with glaucoma.

Fig. 1.

Ferroptosis is involved in NMDA-induced RGC damage in mice with glaucoma. A: Immunofluorescence staining assessment of retinal tissue RBPMS expression; scale bar: 100 μm; B: HE staining to assess histopathological damage in mouse retinal tissue; scale bar: 25 μm (GCL: ganglion cell layer; INL: inner nuclear layer; ONL: outer nuclear layer); C: total iron content assessment in retinal tissue; D: Fe²⁺ content assessment in retinal tissue; E: ROS content assessment in retinal tissue; scale bar: 100 μm; F: kit assessment of retinal tissue MDA content; G: kit assessment of GSH content in retinal tissue; H: Western blot assessment of GPX4 and xCT protein expression in retinal tissue. n = 5 mice per group for the pathological staining experiments; n = 5 mice per group for the molecular experiments. ^***P < 0.001 vs. control; ^##P < 0.01, ^###P < 0.001 vs. NMDA

StarD7 is differentially expressed in mice with glaucoma

Previous studies have shown that StarD7 is involved in the pathological progression of neurodegenerative diseases such as epilepsy and leukoencephalopathy [28], but its role in glaucoma remains unknown. First, through bioinformatics analysis, gene expression in glaucomatous RGCs was compared with that in fellow-eye RGCs, which revealed that StarD7 was upregulated in glaucomatous RGCs (Fig. 2A–B). Moreover, RT‒qPCR and Western blot results demonstrated that compared with the control treatment, NMDA treatment significantly increased the mRNA and protein expression levels of StarD7 (Fig. 2C–D). Immunofluorescence staining further revealed an increase in StarD7 expression in the NMDA-treated model group (Fig. 2E). These findings indicate that StarD7 expression is upregulated in the NMDA-induced glaucoma model, which may be closely related to the pathological mechanisms of glaucoma.

Fig. 2.

Differential expression of StarD7 in glaucoma. A: Volcano plot; B: clustering heatmap; OD represents the normal control group; OS represents the glaucoma model group; C: RT‒qPCR assessment of StarD7 mRNA expression in the NMDA mouse model; D: Western blot assessment of StarD7 protein expression in the NMDA mouse model; E: immunofluorescence staining assessment of StarD7 expression in the NMDA mouse model; scale bar: 100 μm. n = 5 mice per group for pathological staining experiments; n = 5 mice per group for molecular experiments. ^**P < 0.01, ^***P < 0.001 vs. control

StarD7 inhibits ferroptosis to alleviate NMDA-induced RGC injury in mice with glaucoma

Previous studies have shown that StarD7 is a key inhibitor of ferroptosis [14]. To investigate whether StarD7 participates in the development of RGC damage in glaucoma mice through the regulation of ferroptosis, we injected a StarD7-overexpressing adenovirus into the vitreous body of mice before the NMDA model was established. RT‒qPCR and Western blot results demonstrated that StarD7 overexpression led to the upregulation of StarD7 expression (Fig. 3A–B). Immunofluorescence staining revealed that compared with that in the NMDA + AAV-NC group, the number of RGCs labeled with RBPMS was significantly greater in the StarD7-overexpressing group (Fig. 3C). HE staining results indicated that StarD7 overexpression significantly inhibited the reduction in total retinal thickness caused by NMDA (Fig. 3D). Subsequent detection of ferroptosis-related markers revealed that StarD7 overexpression reduced total iron, Fe²⁺, ROS, and MDA levels and increased the GSH content (Fig. 3E–I). Additionally, the Western blot results revealed that StarD7 overexpression increased the expression levels of GPX4 and xCT (Fig. 3J). These results suggest that StarD7 alleviates NMDA-induced RGC damage in mice with glaucoma by inhibiting ferroptosis.

Fig. 3.

StarD7 inhibits ferroptosis to alleviate NMDA-induced RGC damage in mice with glaucoma. A: RT‒qPCR assessment of StarD7 mRNA expression in the NMDA mouse model; B: Western blot assessment of StarD7 protein expression in the NMDA mouse model; C: immunofluorescence staining assessment of RBPMS expression in retinal tissue; scale bar: 100 μm; D: HE staining assessing of retinal tissue pathological damage in mice; scale bar: 25 μm (GCL: ganglion cell layer; INL: inner nuclear layer; ONL: outer nuclear layer); E: total iron content assessment in retinal tissue; F: Fe²⁺ content assessment in retinal tissue; G: ROS content assessment in retinal tissue; scale bar: 100 μm; H: kit assessment of the MDA content in retinal tissue; I: kit assessment of the GSH content in retinal tissue; J: Western blot assessment of GPX4 and xCT protein expression in retinal tissue. n = 5 mice per group for pathological staining experiments; n = 5 mice per group for molecular experiments. ^***P < 0.001 vs. control; ^###P < 0.001 vs. NMDA + AAV-NC

StarD7 inhibits Glu-induced ferroptosis in mRGCs

The effect of StarD7 on Glu-induced ferroptosis in mRGCs was examined in vitro. First, the RT‒qPCR and Western blot results revealed that compared with the control treatment, the Glu treatment significantly increased the mRNA and protein expression levels of StarD7 (Fig. 4A–B). Next, StarD7 was overexpressed, and the expression level of StarD7 in the overexpression group was significantly upregulated (Fig. 4C). To further investigate the role of StarD7 in the Glu-induced ferroptosis of mRGCs, experiments were conducted in vitro by adding erastin to induce ferroptosis. Ferroptosis-related indicator assays revealed that compared with the control treatment, the Glu treatment significantly increased total iron, Fe²⁺, ROS, and MDA levels but significantly reduced GSH levels. However, overexpression of StarD7 reduced the total iron, Fe²⁺, ROS, and MDA contents and increased the GSH content, and the addition of erastin partially attenuated the effects of StarD7 overexpression (Fig. 4D–H). Additionally, the Western blot results revealed that compared with the control treatment, the Glu treatment significantly decreased the expression levels of GPX4 and xCT. Further overexpression of StarD7 effectively increased the expression of these two proteins, whereas treatment with erastin partially reversed the ability of StarD7 overexpression to promote their expression (Fig. 4I). These results indicate that StarD7 alleviates Glu-induced mRGC damage by inhibiting ferroptosis.

Fig. 4.

StarD7 inhibits Glu-induced ferroptosis in mRGCs. A: RT‒qPCR assessment of StarD7 mRNA expression in mRGCs; B: Western blot assessment of StarD7 protein expression in mRGCs; C: Western blot assessment of oe-StarD7 transfection efficiency in mRGCs; D: total iron content in mRGCs; E: Fe^²+ content in mRGCs; F: ROS content in mRGCs; scale bar: 100 μm; G: MDA content in mRGCs assessed by a kit; H: GSH content in mRGCs assessed by a kit; I: Western blot assessment of GPX4 and xCT protein expression in mRGCs. n = 3 independent experiments. ^**P < 0.01, ^***P < 0.001 vs. control or oe-NC; ^###P < 0.001 vs. Glu; ^{^}P < 0.05, ^{^^}P < 0.01, ^{^^^}P < 0.001 vs. Glu + oe-StarD7

The Wnt signaling pathway promotes StarD7 expression

It has been reported that the Wnt signaling pathway plays a critical role in the normal development and maintenance of the retina [29], and the transcription of the StarD7 gene may be regulated by the Wnt/β-catenin signaling pathway [20]. Therefore, we investigated the regulatory relationship between the Wnt signaling pathway and StarD7 in disease. Compared with that in the control group, the expression of the Wnt signaling pathway proteins Wnt1 and β-catenin was reduced in the Glu group. LiCl treatment increased the expression of Wnt1 and β-catenin. (Fig. 5A). The RT‒qPCR and Western blot results revealed that compared with the control group, the Glu group had increased StarD7 expression, and LiCl treatment further increased StarD7 expression (Fig. 5B–C). Immunofluorescence staining further confirmed the above results (Fig. 5D). Thus, these findings indicate that the Wnt signaling pathway can activate StarD7 expression.

Fig. 5.

The Wnt signaling pathway promotes StarD7 expression. A: Western blot assessment of the expression of Wnt1 and β-catenin in mRGCs; B: RT‒qPCR assessment of StarD7 mRNA expression in mRGCs; C: Western blot assessment of StarD7 protein expression in mRGCs; D: Immunofluorescence staining assessment of StarD7 expression in mRGCs; scale bar: 10 μm. n = 3 independent experiments. ^**P < 0.01, ^***P < 0.001 vs. control; ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 vs. Glu

The Wnt signaling pathway activates StarD7 to inhibit Glu-induced mRGC ferroptosis

Finally, we verified whether the Wnt signaling pathway could influence Glu-induced ferroptosis in mRGCs through the regulation of StarD7. After the expression of StarD7 was knocked down, the Western blot results revealed that the expression of StarD7 in the si-StarD7 group was significantly downregulated, indicating successful transfection (Fig. 6A). The detection of ferroptosis-related indicators revealed that compared with those in the Glu group, the total iron, Fe²⁺, ROS, and MDA levels were significantly lower and the GSH content was greater in the LiCl-treated group. Further knockdown of StarD7 partially attenuated the effects of LiCl (Fig. 6B–F). Additionally, the expression levels of GPX4 and xCT significantly increased after LiCl treatment but markedly decreased following further StarD7 knockdown (Fig. 6G). These results demonstrate that the Wnt signaling pathway inhibits Glu-induced ferroptosis in mRGCs through the activation of StarD7 expression.

Fig. 6.

The Wnt signaling pathway inhibits Glu-induced ferroptosis in mRGCs via StarD7. A: Western blot assessment of StarD7 expression in mRGCs; B: total iron content in mRGCs; C: Fe^²+ content in mRGCs; D: ROS content in mRGCs; scale bar: 100 μm; E: MDA content in mRGCs assessed by an assay kit; F: GSH content in mRGCs assessed by an assay kit; G: Western blot assessment of GPX4 and xCT protein expression in mRGCs. n = 3 independent experiments. ^***P < 0.001 vs. si-NC or control; ^###P < 0.001 vs. Glu; ^{^^^}P < 0.001 vs. Glu+LiCl

Discussion

Glaucoma is the leading cause of irreversible vision impairment worldwide, with its pathological characteristics manifesting as RGC loss [30]. Studies have shown that Glu excitotoxicity is among the important mechanisms leading to RGC loss in glaucoma [2]. The accumulation of Glu in the retina, which acts as a classic inducer of ferroptosis, can induce ferroptosis in various cell types, and ferroptosis has been demonstrated to participate in the pathological progression of glaucoma [7]. Consistent with the findings of previous studies, the results of this study further confirmed that ferroptosis is involved in NMDA-induced RGC damage in mice with glaucoma. Additionally, StarD7 expression was found to increase in disease, and overexpression of StarD7 inhibited NMDA-induced ferroptosis in the RGCs of mice with glaucoma and Glu-induced ferroptosis in mRGCs, whereas activation of the Wnt signaling pathway also suppressed Glu-induced ferroptosis in mRGCs by promoting StarD7 expression. Our research highlights the potential neuroprotective roles of the Wnt signaling pathway and StarD7 in glaucoma treatment.

Ferroptosis is a novel form of cell death triggered by the inactivation of cellular GSH-dependent antioxidant defenses, which leads to iron-dependent accumulation of toxic lipid ROS, thereby affecting the pathological processes of various diseases [23, 31]. Epidemiological studies have shown that high-iron diets and elevated serum ferritin levels are positively correlated with the incidence of glaucoma [32, 33], suggesting that abnormal iron metabolism and ferroptosis are closely related to the pathogenesis of glaucoma. Ferroptosis has been validated in subacute intraocular hypertension models, optic nerve crush models, and microbead injection-induced glaucoma models [8, 34]. Additionally, in NMDA-induced excitotoxicity models, the iron chelator deferoxamine (DFO) has been shown to have significant neuroprotective effects on RGC-5 cells [35]. Other studies have indicated that inhibiting METTL3-mediated HMGCS1 m6A RNA modification protects rats from Glu excitatory toxicity-induced ferroptosis [7]. In this study, C57BL/6 mice and mRGCs were selected to construct in vivo and in vitro models of Glu excitotoxicity, which confirmed that ferroptosis is involved in NMDA-induced RGC damage in mice with glaucoma. Fer-1 effectively alleviated pathological damage to retinal tissues and promoted RGC survival. Fer-1 treatment also reduced the levels of total iron, Fe²⁺, ROS, and MDA in retinal tissues while increasing the levels of GSH and the expression of GPX4 and xCT. These results indicate that inhibiting ferroptosis can mitigate NMDA-induced RGC damage in mice with glaucoma.

To further explore the molecular targets that regulate ferroptosis in RGCs, we conducted bioinformatics analysis and found that StarD7 expression is increased in glaucoma, which was also validated in an NMDA-induced glaucoma mouse model. These findings suggest that the upregulation of StarD7 expression may be an early, endogenous compensatory protective response initiated by RGCs when they face excitotoxic injury. However, this endogenous upregulation did not suppress the ferroptosis phenotype, indicating that the magnitude of the endogenous StarD7 upregulation was insufficient to completely counteract NMDA/Glu-induced ferroptosis pressure. Therefore, in both in vivo and in vitro experiments, the StarD7 gene was overexpressed to enhance its protective function. Previous studies have shown that StarD7, a member of the START protein family, is involved in various biological processes, such as lipid transport, tumorigenesis, and skeletal muscle myogenic differentiation [36–38]. However, its function in glaucoma RGCs remains unclear. Additionally, StarD7 has been reported to act as a ferroptosis inhibitor [14], and its inhibition can induce ROS production [39]. This study further investigated the effect of StarD7 on excitotoxicity-induced ferroptosis in in vivo and in vitro models. The results revealed that StarD7 overexpression ameliorated retinal histopathological damage, reduced RGC death, and decreased total iron, Fe²⁺, ROS, and MDA levels in retinal tissue while promoting the expression of GSH, GPX4, and xCT, thereby inhibiting ferroptosis. Furthermore, in a Glu-induced mRGC cell model, the addition of erastin weakened the inhibitory effect of StarD7 overexpression on ferroptosis. These findings reveal that further increasing StarD7 expression can inhibit ferroptosis and mitigate RGC injury.

The Wnt signaling pathway is associated with multiple processes and sites [40] and has been proven to regulate various biological and pathological processes [41]. In glaucoma, the Wnt/β-catenin signaling pathway is closely related to pathological progression [42]. Wang et al. [43] reported increased expression of Wnt/β-catenin signaling pathway inhibitors in the eyes of glaucoma patients, suggesting that impaired Wnt/β-catenin signaling may be associated with retinal degeneration. Other studies have shown that Wnt/β-catenin signaling can protect RGCs from excitotoxic damage [17]. Additionally, studies have reported that the Wnt signaling pathway can regulate ferroptosis [44, 45]. Consistent with previous findings that activated Wnt/β-catenin signaling exerts protective effects on RGCs [46], our results demonstrated that the Wnt signaling pathway proteins Wnt1 and β-catenin were underexpressed in a Glu-induced cell model, revealing that Wnt signaling is inhibited under injury conditions. However, the addition of LiCl reduced the total iron, Fe²⁺, ROS, and MDA levels in Glu-induced mRGCs but increased the GSH levels and the expression of GPX4 and xCT, indicating that activating Wnt signaling can inhibit Glu-induced RGC ferroptosis. Furthermore, previous studies have shown that in the hexosamine biosynthetic pathway (HBP), the Wnt/β-catenin signaling pathway can exert regulatory effects by activating the StarD7 promoter [47]. Our experiments further demonstrated that exogenous activation of the Wnt signaling pathway significantly increased StarD7 expression and that StarD7 knockdown weakened the inhibitory effect of Wnt signaling activation on Glu-induced mRGC ferroptosis. Our data indicated that Wnt signaling activation further increased StarD7 expression, thereby inhibiting Glu-induced mRGC ferroptosis. Furthermore, our study revealed that Glu treatment decreased Wnt1 and β-catenin protein levels, whereas endogenous StarD7 expression increased. Therefore, the Glu-induced upregulation of StarD7 expression may not be solely Wnt dependent. Research has revealed that steroidogenic factor 1 (SF-1) can activate the StarD7 promoter and promote its transcription [12, 48]. However, whether Glu can promote StarD7 expression by affecting SF-1 requires further investigation.

Overall, our study demonstrated that activating the Wnt signaling pathway could inhibit RGC ferroptosis by promoting StarD7 expression. However, this study had several limitations. First, our research focused solely on the Glu excitotoxicity model. Future studies are needed to further validate the protective role of this regulatory pathway in glaucoma across various models, such as acute hypertension models, optic nerve crush models, and microbead injection models. Second, the results of this study revealed that StarD7 affected RGC ferroptosis levels by regulating ferroptosis-related indicators. However, the specific mechanism through which StarD7 regulates ferroptosis remains unclear. Studies have shown that StarD7 may regulate ferroptosis resistance by affecting the transport of mitochondrial coenzyme Q [14]. Whether this mechanism is applicable to glaucoma requires further exploration in the future. Third, further validation of the protective effects of Wnt signaling pathway activators on RGCs in animal models is still needed. Finally, this study focused on retinal ganglion cells, which are distributed mainly in the ganglion cell layer. However, the sample preparation was based on the entire retinal tissue of the mice. Although the results are consistent with our expectations, future studies still require the acquisition of a single ganglion cell layer for analysis to strengthen the reliability of the conclusions (e.g., by collecting specific retinal regions via laser microdissection).

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: Liwei Zhang and Wen Chu; methodology: Liwei Zhang, Wen Chu and Xiaoxiao Feng; software: Wen Chu and Hua Peng; validation: Xiaoxiao Feng and Yabin Yang; formal analysis: Liwei Zhang, Wen Chu and Yang Yabing; investigation: Liwei Zhang, Hua Peng and Lei Kong; resources: Lei Kong; data curation: Wen Chu, Xiaoxiao Feng, and Liyun Guo; writing—original draft preparation: Liwei Zhang and Wen Chu; writing—review and editing: Liwei Zhang, Yabin Yang and Lei Kong; visualization: Hua Peng and Liyun Guo; supervision: Lei Kong; funding acquisition: Lei Kong. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the National Natural Science Foundation of China (81860171, 82560205, 82560316), Medical Reserve Talents of Yunnan Province (H-2018020), Yunnan Provincial Science and Technology Plan Project (no. 202105AF150044 and no. 202105AF150067, no. 202605AF350044), the China Scholarship Council (202208535051), Yunnan Provincial Special Funding for Training High-level Health Technology Talents (D2025043), and Tianjin Natural Science Foundation (24JCQNJC02050), the Science and Technology Program of Yunnan Province: 202401AY070001-287, the Yunnan University Medical Research Fund (Grant No.YDYXJJ2025-0012), the Project of Yunnan Provincial Clinical Research Center (Grant No.2024YNLCYXZX0331).

Data availability

The data are available from the corresponding author upon reasonable request.

Declarations

Ethical approval and consent to participate

All animal experiments were approved by the Experimental Animal Ethics Committee of Yunnan Zequn Biotechnology Co., Ltd., and all animal experimental protocols adhered to the ARRIVE guidelines 2.0.

Consent to publish

Not applicable.

Competing interests

The authors declare no competing interests.