Ferroptosis: A Novel Mechanism in Diabetic Keratopathy

Abstract

Purpose

To investigate whether ferroptosis, an iron-dependent form of cell death, contributes to diabetic keratopathy (DK) by integrating evidence from human tissues, a diabetic mouse model, and in vitro studies and to explore the potential of ferroptosis inhibitor as a therapeutic strategy.

Methods

Human diabetic and nondiabetic corneal biopsy tissues were examined for iron accumulation using Prussian blue staining. In a streptozotocin-induced diabetic mouse model, corneal iron staining and transferrin receptor 1 (TfR1) expression were assessed. Primary human corneal epithelial cells were exposed to high glucose for 3 days to evaluate iron staining, 4-hydroxynonenal and reactive oxygen species production, and protein expression of TfR1 and glutathione peroxidase 4 (GPX4). To probe the mechanism, human corneal epithelial cells were treated with the ferroptosis inhibitor ferrostatin-1 (Fer-1) and assessed for iron, ROS, proliferation, migration, protein expression, and cell viability.

Results

Diabetic human and mouse corneas showed significantly increased iron staining. In mice, the staining correlated with elevated TfR1 expression. In vitro, high glucose induced iron accumulation, increased 4-hydroxynonenal and ROS, and altered protein expression (increased TfR1 and decreased GPX4). Fer-1 reduced high glucose-induced iron accumulation and ROS, restored GPX4, and improved proliferation, migration, and viability; TfR1 levels remained unchanged.

Conclusions

These findings suggest that ferroptosis plays an important role in DK. High glucose triggers this pathway by increasing iron and disrupting antioxidant defense. Targeting ferroptosis with inhibitors such as Fer-1 may have potential relevance in DK, but further studies are needed to clarify its therapeutic value.

Diabetes mellitus (DM) is a major public health concern, and its long-term complications often lead to severe organ damage.^1–5 Among these, diabetic ocular complications, such as retinopathy and keratopathy, are significant causes of vision impairment and blindness. Diabetic keratopathy (DK) is a common but underdiagnosed complication characterized by delayed corneal wound healing, reduced corneal nerve density, and recurrent epithelial erosions.^6–10 These changes are primarily driven by chronic hyperglycemia and the associated metabolic dysregulation, which collectively impair the function of corneal epithelial cells.

The pathogenesis of DK is complex and multifactorial. Chronic hyperglycemia leads to the accumulation of advanced glycation end-products, increased polyol pathway flux, and an increase in oxidative stress.^11–13 Recent evidence highlights the crucial role of oxidative stress in the progression of diabetic complications, including those affecting the cornea.^14–16 However, the precise mechanisms linking hyperglycemia, oxidative stress, and the specific damage to corneal cells remain to be fully elucidated.

Emerging research has pointed to dysregulated iron metabolism as a key contributor to oxidative stress in various diabetic complications.^17,18 Iron, although essential for cellular processes, can become highly toxic when in excess, because it catalyzes the production of harmful reactive oxygen species (ROS) through the Fenton reaction. This iron-mediated oxidative stress can lead to a specific form of regulated cell death known as ferroptosis.^19–21 Our study aims to investigate the role of iron metabolism and ferroptosis in DK. We hypothesized that hyperglycemia-associated disturbances in iron regulation may increase oxidative stress and impair corneal cell function. Although our findings highlight a potential link between iron dysregulation, ferroptosis, and DK, any therapeutic implications, particularly those involving ferroptosis inhibitors, remain preliminary and require in vivo validation before their translational relevance can be established.

Methods

Donor Clinical Information

The study adhered to the tenets of the Declaration of Helsinki. Donor eyes from both diabetic and nondiabetic subjects were obtained from the Miracles in Sight Eye Bank (Winston-Salem, NC, USA). All procedures were conducted in accordance with federal and institutional guidelines, and all human samples were deidentified before analysis. The nondiabetic donors (NDM group) consisted of three females and two males with an average age of 71.8 years. The diabetic donors (DM group) consisted of one female and four males, with an average age of 66.4 years and an average diabetes duration of 21.2 years.

Animals

All animal experiments were approved by the Institutional Animal Care and Use Committee of the Aier Eye Hospital (Permit no. K2024-09-01) and followed the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. To induce diabetes, C57BL/6J mice (male, 8 weeks old) received daily intraperitoneal injections of streptozotocin (STZ) at a dose of 55 mg/kg for 5 consecutive days. Blood glucose levels were measured 2 weeks after the final injection and monitored monthly thereafter. Mice were included in the diabetic group (STZ-DM) only if their nonfasting blood glucose levels exceeded 350 mg/dL. Animals that failed to reach this threshold or that exhibited severe systemic distress unrelated to the study aims were excluded to ensure a consistent and reliable experimental cohort. Age‑matched nondiabetic mice served as controls. The average body weights and blood glucose levels for mice are summarized in Table.

Table.

Body Weight and Blood Glucose of STZ-induced Diabetic Mice

Duration of Diabetes	NDM	STZ-DM
Body weight (g)
2 weeks	36.3 ± 1.8	30.5 ± 1.2
2 months	38.5 ± 1.2	29.2 ± 2.3
3 months	40.3 ± 2.4	28.7 ± 1.5
Blood glucose (mg/dL)
2 weeks	154.8 ± 14.8	528.4 ± 44.2
2 months	168.4 ± 20.4	535.2 ± 52.6
3 months	164.2 ± 21.5	537.0 ± 46.5

NDM, nondiabetic mice.

Values are mean ± SD. n = 10.

Cell Culture

Primary human corneal epithelial cells (HCECs) were purchased from ATCC (PCS-700-010; Manassas, VA, USA). HCEC were cultured in Corneal Epithelial Cell Basal Medium (PCS-700-030; ATCC) with Corneal Epithelial Cell Growth Kit (PCS-700-040; ATCC), following the manufacturer's recommended protocol. The complete culture medium already contained 6.2 mM D-glucose (normal glucose). For high glucose experiments, an extra 18.8 mM of either D-glucose (for high glucose) or mannitol (for osmotic control) was added to the culture medium. The vehicle control for 4-hydroxynonenal (HNE) treatment was ethanol.

Prussian Blue Staining

The samples were immersed in a freshly prepared staining solution, made by mixing equal volumes of 4% aqueous potassium ferrocyanide (#6932; Mallinckrodt, Raleigh, NC, USA) and 4% aqueous hydrochloric acid (A144-212; Fisher Scientific, Hampton, NH, USA), and incubated at room temperature for 30 minutes. They were then rinsed thoroughly in several changes of distilled water, followed by counterstaining with eosin Y for 1 minute.

Semiquantitative Iron Staining Score

To evaluate the iron staining on the corneal epithelial side, two observers, who were blinded to the sample information, independently examined six random fields of view per slide. The average score was calculated based on the subjective evaluation of the intensity and distribution of the blue staining in each field, using the following criteria:

• Score 0: No blue staining, indicating no iron deposition;

• Score 1: A few scattered blue granules, typically in occasional macrophages;

• Score 2: Sparse, fine blue granules with limited, localized distribution;

• Score 3: Moderate, medium-sized blue granules with a broader distribution; and

• Score 4: Strong blue staining in large aggregates or diffusely distributed, covering a significant portion of the tissue.

Immunohistochemistry of Cultured Cells

After fixing the cells with 4% paraformaldehyde for 20 minutes, the sections were incubated with an antibody for HNE (HNE11-S; Alpha Diagnostic, San Antonio, TX, USA) overnight. Then, the sections were incubated with Alexa Fluor 594 AffiniPure Goat Anti-Rabbit IgG (#715-585-150; Jackson ImmunoResearch Laboratories, West Grove, PA, USA) for 2 hours. The samples were mounted with VECTASHIELD Antifade Mounting Medium with DAPI (#H-1200; Vector Laboratories, Burlingame, CA, USA) and imaged using a ZEISS AXIO Observer Z1 microscope (Carl Zeiss Microscopy, Oberkochen, Germany) with standardized exposure settings. The fluorescence intensity was quantified using ImageJ software.

Western Blot Analysis

Western blot analysis was performed on both whole corneal tissue and cultured cells as described previously.^22,23 Briefly, samples were lysed using the RIPA buffer supplemented with protease and phosphatase inhibitors. Corneal samples were first homogenized in the buffer. The lysates were then centrifuged to remove debris, and the total protein in the supernatant was quantified using a bicinchoninic acid assay. Proteins were resolved by SDS-PAGE, transferred to a membrane, and probed with primary antibodies for glutathione peroxidase 4 (GPX4) (sc-166570; Santa Cruz, Dallas, TX, USA) and Transferrin Receptor 1 (TfR1) (sc-65882; Santa Cruz Biotechnology, Dallas, TX, USA). Bands were detected with horseradish peroxidase-conjugated secondary antibodies. The intensity of the bands was measured by densitometry using Image Lab software (Bio-Rad, Hercules, CA, USA) and normalized to the level of β-actin (#A5441; Sigma-Aldrich, St. Louis, MO, USA).

ROS Assay

The production of ROS in HCECs was measured using CM-H₂DCFDA (C6827; Invitrogen, Waltham, MA, USA) following the protocol of the manufacturer. Cells were incubated with a 1:5000 dilution of CM-H2DCFDA for 30 minutes at 37°C. ROS levels were determined by measuring fluorescence intensity at an excitation wavelength of 483 nm and an emission wavelength of 530 nm using CLARIOstar Plus microplate reader. The ROS concentrations were quantified by normalizing the ROS level to the total cell protein concentration.

Cell Migration Assay

To evaluate cell migration, a scratch assay was performed as described previously.^23,24 When cells reached 100% confluence in 12-well plates, a scratch was made using a 200-µL sterile pipette tip. The cells were then washed with PBS to remove any detached cells. The acellular area of the scratch was photographed at 0 and 24 hours using Agilent BioTek Cytation 5 Cell Imaging Multimode Reader (BioTek, Winooski, VT, USA). The size of the acellular area was then measured using ImageJ to quantify the extent of cell migration.

Bromodeoxyuridine (BrdU) Cell Proliferation Assay

Cell proliferation was measured using the BrdU Cell Proliferation ELISA Kit (ab126556; Abcam, Cambridge, UK). Briefly, cells were first incubated with BrdU for 24 hours, then fixed for 30 minutes. An anti-BrdU antibody was added for 1 hour, followed by a peroxidase goat anti-mouse IgG and an horseradish peroxidase substrate for 30 minutes. Absorbance was then measured at 450 nm with CLARIOstar Plus microplate reader (BMG Labtech, Ortenberg, Germany).

Trypan Blue Exclusion Assay

Primary HCEC were seeded into 12‑well plates and treated for 48 hours. Cell viability was assessed using a trypan blue exclusion assay.²⁵ Briefly, cells were washed with PBS and incubated with 100 µL of 0.05 % Trypsin (#25-052-CI; Corning, Glendale, AZ, USA) for 5 minutes. Trypan blue solution (#25-900-CI; Corning) was then added to the cell suspension and mixed thoroughly. A 25-µL aliquot of the mixture was loaded into a Cellometer cell counting chamber (#SD100; Nexcelom, Lawrence, MA, USA) and the percentage of viable cells were quantified using a Nexcelom Cellometer Auto T4.

Statistical Analysis

One-way ANOVA was used for comparisons involving multiple groups, and unpaired Student's t-tests were used for two-group comparisons, all performed using GraphPad Prism 10 (GraphPad Software, Boston, MA, USA). All data are presented as mean ± SD.

Results

Increased Prussian Blue Iron Staining in Human Diabetic Donor Corneal Biopsy Tissue

Prussian blue staining revealed dense, granular deep blue deposits throughout the corneal epithelium in biopsy punches from diabetic human donors, with staining distributed broadly across the epithelial layer (Figs. 1A, 1B). These iron deposits were confined to the epithelium, whereas the underlying stroma showed minimal accumulation. In contrast, nondiabetic corneal biopsies exhibited only sparse, minute blue punctate granules in a small number of epithelial cells. The semiquantitative iron staining score suggested a significant increase in iron accumulation in corneal biopsy tissues from diabetic donors compared with nondiabetic controls, which is consistent with ferroptotic processes (Fig. 1C).

Figure 1.

Increased Prussian blue iron staining in human diabetic donor corneal biopsy tissue. (A) Representative images of whole mount preparations of human corneal biopsy punches from 69-year-old nondiabetic (NDM) and 64-year-old diabetic (DM) human donors. These punches were stained with Prussian blue and eosin. Scale bar, 200 µm. (B) Representative corneal epithelial images used for the semiquantitative iron staining score analysis. Scale bar, 50 µm. (C) Increased Prussian blue iron staining in human diabetic donor corneal epithelium (n = 5). All values are mean ± SD.

Increased Ferroptosis in STZ-induced Diabetic Mouse Corneas

We have previously reported that the STZ-induced diabetic mouse model exhibits significantly delayed corneal wound healing and mitochondrial dysfunction.^26,27 Using this same model, we observed increased iron accumulation in the corneas after 3 months of diabetes. Paraffin sections showed that iron staining was predominantly localized to the corneal epithelium in STZ-DM mice (Fig. 2A), accompanied by a decrease in epithelial thickness (Fig. 2B). Fresh whole mount corneas further confirmed enhanced iron staining in diabetic mice (Figs. 2C, 2D). Western blot analysis revealed an elevated expression of TfR1 in the corneas of the diabetic mice (Figs. 2E, 2F). TfR1 is a protein that plays a key role in cellular iron uptake.^28,29 This finding suggests that the corneas of diabetic mice exhibit increased ferroptosis, characterized by increased iron and elevated expression of the iron uptake protein TfR1.

Figure 2.

Iron accumulation and TfR1 upregulation in STZ-diabetic (STZ-DM) mouse corneas. (A) Representative Prussian blue staining images showing iron staining in paraffin sections from STZ-DM mouse corneas. The arrow indicates the location of the iron staining in the cornea. (B) STZ-DM mice exhibit a significant reduction in corneal epithelial thickness compared with nondiabetic (NDM) controls (n = 5). (C) Representative corneal flat mounts image (with epithelium side up) showing Prussian blue staining for iron in the corneas of NDM and STZ-DM mice. The arrows indicate the location of the iron staining in the cornea. (D) Increased iron staining scores in corneal flat mounts from STZ-DM mice (n = 5). (E, F) Western blot showing increased TfR1 expression in STZ-DM mouse corneas (n = 5). All values are mean ± SD.

High Glucose Exposure Induced Ferroptosis in Primary HCEC In Vitro

To investigate the cellular mechanism, primary HCECs were treated with 25 mM total glucose for 3 days. The high glucose treatment resulted in increased iron staining (Figs. 3A, 3B) and a notable increase in the production of both 4-hydroxynonenal (HNE) (Figs. 3C, 3D) and ROS (Fig. 3E). HNE is a toxic byproduct produced when oxidative stress induces lipid peroxidation of fatty acids in the cell membranes.³⁰ Our results indicate that high glucose conditions increase oxidative stress, causing HNE to accumulate and further damage the cells, which can lead to ferroptosis. Western blot analysis showed that high glucose treatment increased the expression of TfR1 while decreasing the expression of GPX4 (Figs. 3F–H). GPX4 is a critical antioxidant enzyme that protects cells from oxidative stress and ferroptosis by neutralizing lipid peroxides.^31–33 This collective evidence strongly suggests that high glucose exposure induces ferroptosis in corneal epithelial cells.

Figure 3.

Oxidative stress and iron dysregulation in HCEC under high-glucose conditions. (A) Representative images showing Prussian blue iron staining in HCEC after high-glucose treatment (HG, 25 mM in total). Normal glucose treated cells (NG, 6.2 mM D-glucose and 18.8 mM mannitol) were used as control. (B) Increased iron staining scores in HCEC treated with high glucose for 3 days (n = 5). (C, D) Immunostaining showing that high-glucose treatment increased HNE production in HCEC (n = 5). Secondary-only negative control was performed to confirm minimal background relative to primary HNE signals. (E) High glucose exposure increased ROS production in HCEC (n = 5). (F–H) Western blot shows that high glucose treatment leads to decreased GPX4 expression and increased TfR1 expression (n = 3). All values are mean ± SD.

Treatment With HNE Induces Ferroptosis in HCECs

Previously, we showed that exposure of HNE induced mitochondrial damage and inhibits the proliferation and migration of HCECs.²⁷ To further explore the role of HNE in ferroptosis, HCECs were treated with this compound for 2 days. HNE treatment resulted in an increase in iron staining (Figs. 4A, 4B). Western blot analysis (Figs. 4C–E) showed that HNE treatment mirrored the effects of high glucose, causing an increase in TfR1 expression and a decrease in GPX4 expression. This finding suggests that HNE acts as a key indicator of the cellular damage, triggering the iron accumulation and altered antioxidant defense pathways central to ferroptosis.

Figure 4.

HNE-induced iron accumulation and ferroptosis-related protein changes in HCEC. (A) Representative images showing Prussian blue iron staining in HCEC after HNE treatment. (B) Increased iron staining scores in HCEC treated with HNE (20 µM) for 2 days (n = 5). Ethanol served as vehicle control (Ctrl) at the same volume. (C–E) Western blot analysis shows that HNE treatment leads to decreased GPX4 expression and increased TfR1 expression (n = 3). All values are mean ± SD.

Ferrostatin-1 Treatment Ameliorated the Ferroptosis Induced by High Glucose in HCECs

To confirm that high glucose exposure induces ferroptosis in HCECs, we treated the cells with Ferrostatin-1 (Fer-1), a compound that inhibits ferroptosis by scavenging lipid peroxidation free radicals.^34–36 Prussian blue staining showed that iron accumulation was significantly reduced, indicating that Fer-1 successfully prevented iron-dependent cellular damage under high-glucose conditions (Figs. 5A, 5B). Western blot analysis showed that Fer-1 restored GPX4 expression to normal levels, but the TfR1 level remained unchanged (Figs. 5C–E). This may be because Fer-1 primarily acts by inhibiting lipid peroxidation, a process that occurs downstream of iron uptake, and therefore may not directly regulate the expression of the iron uptake protein TfR1. Furthermore, Fer-1 significantly reduced ROS production in high-glucose cells (Fig. 5F), improved cell viability (Fig. 5G), and ameliorated the inhibition of cell proliferation (Figs. 5H, 5I) and migration (Figs. 5J, 5K) by high-glucose exposure. Taken together, these results suggest that Fer-1 protects the corneal cells from the damage caused by high-glucose–induced ferroptosis.

Figure 5.

Fer-1 mitigates high-glucose–induced ferroptosis and functional impairment in HCEC. (A, B) HCEC were pretreated with Fer-1 (2 µM) for 8 hours and then exposed to high glucose for 3 days. Prussian blue iron staining showed that Fer-1 reduced iron staining scores in HCEC (n = 5). (C–E) Western blot analysis shows that Fer-1 treatment ameliorated the high-glucose–induced GPX4 downregulation but TfR1 expression remained unchanged in HCEC (n = 3). (F) Fer-1 treatment attenuated high-glucose–induced ROS generation in HCEC (n = 5). (G–K) Cell viability (n = 5), cellular migration (n = 5) and proliferation (n = 5) were measured after the cells were exposed to normal glucose (6.2 mM), high glucose (25 mM) and Fer-1 (2 µM) for 3 days. Cell proliferation was quantified using a BrdU cell proliferation ELISA kit, which measures BrdU incorporation as absorbance at 450 nm; therefore, results are presented as fold‑change in absorbance relative to the control group. Values are mean ± SD.

Discussion

Ferroptosis is an iron-dependent form of regulated cell death that is characterized by the accumulation of lipid peroxides and is distinct from other forms of cell death, such as apoptosis.^37–39 It has been implicated in the pathology of numerous conditions, including various systemic diseases and injuries. For instance, studies have shown that ferroptosis contributes to the pathogenesis of cancer, neurodegenerative diseases, and cardiovascular conditions.^40–42 In the context of the eye, research has begun to show that ferroptosis may also play a role in several ocular diseases. For example, it has been linked to the development of dry eye disease, glaucoma, and AMD.^43–45

Notably, a growing body of evidence has established a strong link between ferroptosis and diabetic retinopathy, one of the most common and severe complications of diabetes. Studies have shown that hyperglycemia leads to iron overload and increased lipid peroxidation in retinal cells, contributing to microvascular dysfunction, retinal cell death, and the breakdown of the blood–retinal barrier. Research in both cell cultures and mouse models of diabetic retinopathy has demonstrated that high glucose levels induce ferroptosis in retinal endothelial cells and photoreceptors, which can be mitigated by the specific ferroptosis inhibitor, Fer-1.⁴⁶ This finding indicates that targeting ferroptosis may be a promising therapeutic strategy for treating diabetic retinopathy.

Although the role of ferroptosis in the diabetic retina is becoming well-established, its involvement in the diabetic cornea remains largely unexplored. To our knowledge, our study provides the first comprehensive evidence that diabetes induces ferroptosis in the cornea, identifying it as a key mechanism contributing to DK. Through a multipronged approach encompassing human, mouse, and in vitro human corneal cell studies, we have demonstrated a clear link between high-glucose conditions and the induction of this unique form of cell death.

The clinical relevance of our findings is underscored by the direct observation of a significant increase in iron accumulation within the corneal epithelium of diabetic human donors compared with nondiabetic controls. This finding, derived from human donor biopsy tissue, provides a powerful initial piece of evidence supporting our central hypothesis. We further validated this finding in an STZ-induced diabetic mouse model, which consistently showed enhanced iron staining and elevated expression of TfR1. This protein is crucial for cellular iron uptake, and its upregulation provides a molecular explanation for the observed iron accumulation.^28,29 The consistent findings across both human and animal models strongly suggest that altered iron metabolism is a fundamental hallmark of the diabetic cornea.

Although Prussian blue staining revealed focal iron accumulation, multiple lines of evidence indicate tissue-wide proferroptotic conditions rather than propagation from isolated foci. Widespread HNE accumulation (Fig. 3C) demonstrates extensive lipid peroxidation throughout the epithelium, indicating that oxidative stress affects the entire tissue rather than emanating from focal iron-overloaded cells. Western blot analysis confirms global upregulation of TfR1 and downregulation of GPX4 throughout the corneal epithelium. These molecular changes indicate that the diabetic corneal epithelium exists in a state predisposed to ferroptosis, with the focal Prussian blue–positive cells representing those that have progressed to overt iron accumulation and active ferroptotic cell death. Recent studies show that ferroptosis can propagate between adjacent cells via direct cell–cell contact,^47,48 suggesting that these focal ferroptotic cells may amplify local damage through propagation to immediate neighbors. However, the diffuse molecular and biochemical changes we observe indicate that tissue-wide vulnerability is the primary driver of ferroptotic dysfunction in the diabetic corneal epithelium, with local cell-to-cell propagation potentially contributing to regional amplification of damage.

To investigate the underlying cellular and molecular pathways, we conducted in vitro experiments using primary HCECs treated with high glucose. Our results revealed a cascade of events characteristic of ferroptosis. The treatment led to increased cellular iron accumulation, an increase in ROS, and a notable increase in HNE, a commonly accepted marker of lipid peroxidation.^47–49 Western blot analysis confirmed that high glucose exposure increased the expression of the iron-uptake protein TfR1 while simultaneously decreasing the expression of GPX4. GPX4 is a key antioxidant enzyme, and its downregulation leaves the cell vulnerable to the oxidative damage that drives ferroptosis.^31–33 These findings collectively demonstrate that high glucose directly triggers ferroptosis at the cellular level by disrupting the delicate balance of iron homeostasis and antioxidant defense.

Our study further confirmed the role of HNE and GPX4 in this process. When HCECs were treated with HNE alone, the effects on the cells mirrored those of high glucose exposure: increased iron accumulation and altered expression of both TfR1 and GPX4. This finding suggests that the HNE generated in the high-glucose environment is not merely a byproduct, but an active signaling molecule that contributes to the ferroptotic cascade.

To confirm that the observed cell death was ferroptosis, we used Fer-1, a specific inhibitor that scavenges lipid peroxidation–derived free radicals. Fer-1 treatment effectively prevented high-glucose–induced cellular damage: it markedly reduced iron staining and ROS production and mitigated the delay in cell proliferation and migration. Western blot analysis further showed that Fer-1 restored GPX4 expression under high-glucose conditions. These findings indicate that Fer-1 acts downstream of iron uptake by neutralizing toxic lipid peroxidation products, the terminal effectors of ferroptosis. As a radical-trapping antioxidant, Fer-1 inhibits lipid peroxidation without directly altering upstream iron-handling pathways such as TfR1. The persistent upregulation of TfR1 likely reflects a direct response to high-glucose–induced metabolic stress. Fer-1 prevents the lethal consequences of this stress, but does not remove the hyperglycemic stimulus driving TfR1 expression. Together, our data suggest that, in DK, TfR1-mediated iron overload initiates ferroptosis, whereas GPX4 loss and lipid peroxidation execute it. Fer-1 specifically targets the execution phase, leaving the initiation phase (TfR1 levels) unchanged.

Our study, although significant, has several limitations. First, the human corneal tissue samples were from a limited number of donors and were not collected in a prospective manner. A larger sample size and a prospective cohort study would be necessary to further confirm our findings. Second, although we have established that high glucose and HNE induce ferroptosis, the specific upstream signaling molecules that initiate this process have yet to be identified. Additionally, our current study focused on a limited number of ferroptosis markers, such as TfR1, GPX4, and HNE. Although these markers provide strong evidence, a more comprehensive analysis of other key components in the pathway, such as acyl-coenzyme A synthetase long-chain family member 4 and ferroptosis suppressor protein 1, is warranted. We plan to address this issue in future research to more fully characterize the ferroptotic cascade in DK. Further research is needed to explore the specific upstream pathways that mediate altered iron metabolism and GPX4 downregulation in the context of DK. Last, future animal studies should also be expanded to include therapeutic interventions. Administering Fer-1 or other ferroptosis inhibitors to diabetic mice could directly test their potential as a treatment for DK. These studies would be essential for translating our findings into clinical applications.

Conclusions

Our study is the first to systematically investigate and confirm the role of ferroptosis in DK, providing a new perspective on the pathogenesis of this condition. Although our findings are compelling, future research is needed to explore the precise signaling pathways linking high glucose to ferroptosis initiation. Additionally, these results open the door for investigating Fer-1 or similar compounds as a novel therapeutic strategy to prevent or treat DK by targeting this newly identified cell death mechanism.