Itaconate derivative eye drops deploy anti-inflammatory effect in treating dry eye models

Abstract

Purposes

To investigate the therapeutic potential of 4-Octyl itaconate (4-OI), a derivative of the endogenous immunomodulator itaconate (ITA), as a safe and effective anti-inflammatory agent for the treatment of dry eye disease (DED).

Methods

In vitro cytotoxicity and anti-inflammatory effects of 4-OI were evaluated in hyperosmotic human corneal epithelial cells (HCECs). An in vivo DED model was induced in C57BL/6 mice through environmental controlled chambers. Mice with DED were treated with phosphate buffered saline (PBS), 0.05% cyclosporine A (CsA) eye drops, 4-OI eye drops, or a single 4-OI intraperitoneal injection for 14 days. Corneal epithelial defects and tear secretion were assessed. The safety and underlying molecular mechanisms of topical 4-OI were evaluated using quantitative polymerase chain reaction (qPCR), Western blotting, mRNA sequencing, TdT-mediated dUTP nick-end labeling (TUNEL) assay, and hematoxylin-eosin (H&E) staining. Statistical analyses were performed using independent sample t-tests and one-way ANOVA, with P<0.05 considered statistically significant.

Results

Treatment with 4-OI eye drops significantly reduced corneal epithelial defects in DED mice (P<0.0001) compared to the DED control group. Tear secretion was also elevated in the 4-OI eye drop-treated groups. mRNA sequencing revealed significant downregulation of inflammatory pathways, including interleukin (IL)-1 receptor activity, tumor necrosis factor (TNF) signaling, and IL-17 signaling, in the 2 mM 4-OI eye drop group. Molecular assays confirmed a marked reduction in the mRNA and protein expression of IL-1β, IL-17A, TNF-α, and NFKBIZ (IκBζ) in the corneas of 4-OI-treated mice. No significant corneal epithelial apoptosis or systemic toxicity was observed in the 4-OI-treated groups.

Conclusions

The ITA derivative 4-OI demonstrates potential as a safe and effective therapeutic option for alleviating DED by suppressing ocular inflammation through the downregulation of IL-1, IL-17, and TNF signaling pathways.

Highlights

• •First evidence: 4-OI eye drops reduce dry eye pathology in murine models.
• •Novel mechanism: 4-OI downregulates IL-1/IL-17/TNF pathways in cornea.
• •Topical 4-OI efficacy matches cyclosporine A, suggesting therapeutic potential.
• •4-OI eye drops show excellent corneal safety and tolerability profile.

1. Introduction

Dry eye disease (DED) is a chronic ocular surface disorder influenced by a variety of factors, often accompanied by symptoms such as eye dryness, foreign body sensation, photophobia, and blurred vision.¹ If left untreated, it can lead to severe complications, including blindness. Previous studies have shown that patients with dry eye account for 21%–30% of outpatient visits in China.² As the most common ocular surface disease, DED significantly reduces patients’ quality of life and contributes to a substantial socio-economic burden.

Emerging evidence indicates that the primary drivers of DED are tear film instability and hyperosmolarity, which initiates a self-perpetuating vicious cycle of inflammation on the ocular surface.3, 4, 5 Disruption of ocular surface homeostasis leads to the activation of stress signaling pathways, such as mitogen-activated protein kinase (MAPK) and nuclear factor κ-light chain enhancers (NF-κB) pathways, triggering the production of an array of pro-inflammatory cytokines interleukin (IL)-1 (IL-1α; IL-1β), tumor necorsis factor (TNF)-α and matrix metalloproteinases (MMPs).6, 7, 8, 9 These mediators promote the activation, maturation, and migration of antigen-presenting cells, which in turn prime effector T cells, including T helper (Th)1 and Th17 subsets, to migrate to the ocular surface.¹⁰ Of particular pathophysiological relevance, the IL-17 signaling pathway has been identified as one of the central drivers in this process. Elevated IL-17 expression in the ocular surface and tears of DED patients and model mice significantly exacerbates disease severity, and its signaling is critically modulated by IκBζ, a key regulatory protein encoded by the NFKBIZ gene that influences Th17 cell differentiation and IL-17A-induced cytotoxicity.11, 12, 13, 14, 15, 16, 17 The subsequent release of cytokines such as interferon (IFN)-γ and IL-17 further amplifies the inflammatory response, leading to epithelial damage, goblet cell loss, and further aggravation of tear film instability, thereby entrenching the pathological cycle.18, 19, 20, 21 Consequently, therapeutic strategies aimed at inhibiting this immune-inflammatory response and disrupting the cycle at key points are considered fundamental to addressing the underlying cause of DED.²²

Cyclosporine A (CsA) is a first-line ocular immunosuppressant for moderate-to-severe DED in China.²³ It exerts its therapeutic effect by inhibiting upstream calcineurin-Nuclear Factor of Activated T-cells (NFAT) signaling, thereby suppressing T-cell activation and proliferation to achieve broad immunosuppression.²⁴ This action ultimately alleviates ocular surface inflammation and increases goblet cell density, contributing to tear film homeostasis. In DED mouse models, CsA treatment has been shown to significantly reduce the expression of pro-inflammatory cytokines and chemokines, including IL-1β, TNF-α, IL-6, IL-17, and intercellular adhesion molecules.^25,26 However, the clinical utility of CsA is limited by its delayed onset of action (often requiring 3–6 months), variable efficacy in large-scale trials, and frequent local adverse effects such as burning and stinging that compromise patient adherence.^27,28 These limitations of CsA might be multifactorial: the delayed response may be inherent to its upstream immunomodulatory mechanism, while the irritative symptoms are largely attributable to the oleaginous, preservative-laden emulsion required for its clinical formulation due to CsA's insoluble macromolecular properties.

To address these limitations, targeting downstream inflammatory effectors represents a promising alternative strategy. Recently, itaconate (ITA), an endogenous immunometabolite with a small molecular structure, has emerged as a promising candidate for such a strategy.²⁹ Produced in abundance by activated mammalian macrophages, ITA has demonstrated potent anti-inflammatory effects in models of sepsis, psoriasis, gout, viral infections, and pulmonary fibrosis.^11,30, 31, 32, 33 Mechanistically, ITA competitively inhibits succinate dehydrogenase (SDH) to suppresses IL-1β transcription,^34,35 and, more notably, modulates the activating transcription factor 3 (ATF3)/IκBζ axis to counteract IL-17A-induced cytotoxicity.11, 12, 13, 14 By disrupting these key downstream mediators, ITA may offer a faster therapeutic response compared to upstream agents like CsA. To harness this potential, we focused on 4-Octyl itaconate (4-OI), a cell-permeable prodrug efficiently converted to ITA in human cells, which has shown efficacy across various inflammatory disease models.^31,36

Here, we explored the possibility of the therapeutic effect of 4-OI as a safe and effective anti-inflammatory treatment for treating DED model, as well as its underlying molecular mechanism.

2. Materials and methods

2.1. Preparation of 4-OI

4-OI in this study was purchased from MedChemExpress (HY-112675, NJ, USA). The 4-OI powder was added in phosphate buffered saline (PBS) solution containing 20% (2-Hydroxypropyl)-β-cyclodextrin (HY-101103, NJ, USA). Ultrasound was then used until the drug solution became clear. The formation was adjusted with a pH of 7 and kept at 4 °C for short-term storage (≤2 weeks). It was recovered to room temperature 30 min before application. As there is currently no literature regarding the appropriate concentration of 4-OI formation for DED, we conducted preliminary experiments to explore the optimal concentration of 4-OI used for this study (supplemental information). The concentration of 4-OI used in vitro was determined based on Cell Counting Kit-8 (CCK-8) assay results to ensure cell viability and minimize cytotoxicity. For in vivo experiments, the optimal concentration of 4-OI eye drops was established through a dose-escalation study. Briefly, a gradient of 4-OI concentrations was administered twice daily for seven consecutive days in healthy mice. The highest concentration that did not induce corneal epithelial damage was selected as the high-dose formulation, while the low-dose formulation was set at half of this concentration.

2.2. DED mouse model

Sixty SPF C57BL/6J mice (120 eyes, Shanghai Jiesijie Laboratory Animal Co., Ltd., Shanghai, China), 6–8 weeks old, were randomized into six groups: control group, DED control group, CsA eye drop group, 4-OI eye drop (high concentration) group, 4-OI eye drop (low concentration) group, and 4-OI intraperitoneal (IP) injection group, each group n=10. The mice were housed in controlled environment (25 °C, 12h/12h light-dark cycle) with adequate rodent chow and water available. The humidity for control group mice was 50%, while other groups was 18.5% ± 5.1% for 14 days of DED modeling.³⁷ All animal experiments were approved by the Animal Experimentation Ethics Committee of Eye & ENT Hospital of Fudan University with the approval number of IRB-EENT-150301, and conformed to the Association for Research in Vision and Ophthalmology statement on the use of animals. DED control mice received topical PBS bilaterally. The CsA-treated group was administered 0.05% CsA eye drops (Sinqi Pharmaceutical, Shenyang, China).³⁸ For the 4-OI eye drop groups, low- and high-concentration formulations (1 mM and 2 mM, respectively) were applied. All topical treatments (PBS, CsA, and both 4-OI concentrations) were administered to both eyes twice daily for 14 consecutive days. Mice in the IP injection group received a single dose of 50 mg/kg 4-OI (∼13 mM); this regimen was designed to sustain systemic drug exposure throughout the chronic DED phase while minimizing handling-related stress.³⁰ Schirmer's I-test and the corneal fluorescent staining (CFS) were recorded after DED modeling and 14 days of drug administration.

2.3. Schirmer's I-Test

Schirmer's I-test was performed with Zone-Quick Phenol-Red cotton threads (Jingming New Technology Development Co., Ltd., Tianjin, China). After general anesthesia with 1.25% avertin (0.2 mL/10 g) intraperitoneally, 1 mm of the phenol red cotton line was inserted into the lower lateral conjunctival fornix for 30 s. The color of phenol cotton changes from yellow to red due to the secretion of tears. The wetted length (mm) of the strip indicated by red dye was recorded. The eyelids were closed after examination to prevent corneal damage.

2.4. Corneal fluorescein staining

CFS was scored as follows: 2 μL of 0.5% (w/v) sodium fluorescein (Sigma-Aldrich, St. Louis, MO, USA) was instilled onto the ocular surface, after which the eyelids were gently closed 3–4 times to distribute the dye, and excess fluid was carefully blotted. Staining was visualized under a slit lamp (Zeiss, Germany) with cobalt blue light and graded by the same experienced ophthalmologist in a masked manner. The cornea was divided into five quadrants (optic zone, upper, lower, left, right), and each quadrant was scored according to the following criteria: 0 points for no staining; 1 point for fewer than 30 stained dots; 2 points for more than 30 non-diffuse stained dots; 3 points for severe diffuse staining or plaque staining. The sum of all quadrants constitutes the final score with a maximum total of 15 points.¹⁸ All eyes in each group were included in the analysis.

2.5. Real-time qPCR

The corneas of mice were carefully removed for subsequent real-time quantitative polymerase chain reaction (qPCR), Western blot and whole transcriptome RNA sequencing. For real-time qPCR, the total RNA of each cornea was extracted and purified using the RNA extraction kit (EZB-RN4, EZBioscience, China). The first strand cDNA was reverse transcribed with the Reverse Transcription kit (A0010CGQ, EZBioscience, China). Quantitative real-time PCR was examined using Quanti Nova SYBR Green kit (A0012-R2, EZBioscience, China) following the manufacturer's instructions. Primer design was showen in the supplementary information (Table S1). Data were analyzed using the comparative Ct method. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control.

2.6. Western blot

Western blots were performed as described in a previous study.³⁹ Following antibodies were used: NFKBIZ (14014-1-AP, 1:500; Proteintech, Rosemont, USA), IL-17A (ab218013, 1:500; Abcam, Cambridge, MA), IL-1β (ab234437, 1:500; Abcam, Cambridge, MA), TNF-α (60291-1-Ig, 1:500; Proteintech, Rosemont, USA) and β-actin (3700, 1:25000; Cell Signaling Technology, Danvers, MA).

2.7. RNA-sequencing

The transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). In order to obtain sufficient RNA for transcriptome sequencing, the two corneas from each mouse were pooled as a single biological replicate. Corneal tissues from three mice in the high-concentration 4-OI eye drop group (2 mM) and three mice in the PBS control group were processed this way, thus enabling a 3 vs. 3 comparative analysis. Total RNA for the construction of gene libraries was extracted using the VAHTS Universal V6 RNA-seq Library Prep Kit (Vazyme Biotech Co., Ltd., Nanjing, China) following the manufacturer's instructions. Then, the libraries were sequenced on a llumina Novaseq 6000 platform and 150 bp paired-end reads were generated. Differential mRNA expression analysis was performed using the DESeq2. q value<0.05 and log₂FC>0.585 or log₂FC<-0.585 was set as the threshold for significantly differential expression gene (DEGs). The bioinformatic analyses of membrane proteins were performed by the online software Database for Annotation, Visualization and Integrated Discovery (DAVID, https://david.ncifcrf.gov/home.jsp). The expression of proteins in both groups and the analysis result were visualized by R software.

2.8. Cell culture and the hyperosmolar stress model

Human Corneal Epithelial Cell (HCEC) Line was kindly provided by Prof. Weiyun Shi 's team at Shandong Eye Institute.⁴⁰ Cells were cultured as per the standardized protocol.⁴¹ In detail, HCECs were cultured in Dulbecco's Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F12) medium (Gibco, Carlsbad, CA, USA) containing 100 U/mL penicillin, 100 U/mL streptomycin (Gibco, Carlsbad, CA, USA), and 10% fetal bovine serum (Gibco, Carlsbad, CA, USA). Cells were incubated in a humidified 37 °C incubator containing 5% CO₂. Hyperosmolar stress cell model was performed by treating HCECs for 24 h in hyperosmolar (500 mOsM) medium. Hyperos molarity was achieved by adding 94 mM sodium chloride.⁴² Cell experiments included in this study were conducted using HCECs that were within 20 generations.

2.9. CCK-8 assay

An enhanced CCK-8 (Beyotime, Shanghai, China) was used for cell viability assays according to the manufacturer's protocol. Briefly, HCECs were cultured in 96-well plates at the density of 1 × 10⁴ cells per well for 24 h. After the cells were adhered, remove the supernatant, add 100 μL of regular culture medium to the control wells, and add 100 μL of different concentrations of 4-OI (0, 2, 5, 10, 15, 20, 30, 50, 100 μM) prepared using hypertonic medium to the experimental wells and incubated in the 37 °C-incubator containing 5% CO₂ for 24 h. After washed off the culture medium with PBS, 100 μL of fresh medium (consisted of 90 μL DMEM/F12 medium and 10 μL of CCK-8 solution) was added per well. Cells were cultivated at 37 °C in 5% CO₂ for 30 min, and the absorbance was measured at wavelengths of 450 nm using a microplate reader (Tecan, Mannedorf, Switzerland).

2.10. Statistical analysis

Data analysis was performed using SPSS (version 22.0, Chicago, USA) and Graphad P rism software (version 9.0, California, USA). Independent sample t-test and one-way ANOVA were used for data analysis. Results were considered significantly different at P<0.05. For mRNA-seq results, genes with expression levels | log₂FC |>1 and Bonferroni corrected P≤0.05 were considered significantly different. Hierarchical clustering analysis of differentially expressed genes was performed using R (v 3.2.0), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway enrichment analysis of differentially expressed genes based on a hypergeometric distribution algorithm, and bar graphs were drawn for significantly enriched functional entries, etc.

3. Results

3.1. 4-OI reduces inflammation associated factor gene/protein expression in a dry eye cell model

To assess the safety and determine the optimal concentration of 4-OI in an in vitro DED cell model, hyperosmotic HCECs were treated with a gradient of 4-OI concentrations.⁴² Cytotoxicity was evaluated using the CCK-8 assay, allowing for the identification of concentrations that preserved cell viability without inducing adverse effects. Results showed that 4-OI significantly restored cell activity in the dry eye cell model at concentrations ranging from 2 to 30 μM, compared to the DED control group (P<0.001) (Fig. S1). To investigate whether 4-OI modulates the expression of IκBζ or IL-1β, as seen in other inflammatory conditions, western blotting was performed on normal and hypertonic cells treated with varying concentrations of 4-OI (Fig. 1A-C). Additionally, qPCR analysis assessed inflammation-related gene expression in control HCECs, hyperosmolar-treated HCECs (DED control group), and hyperosmolar HCECs treated with 5 μM 4-OI (4-OI group). Results demonstrated that the protein levels of IL-1β and IκBζ (encoded by NFKBIZ) were reduced in the 4-OI-treated group, with a concentration-dependent correlation. Furthermore, qPCR revealed significant downregulation of mRNA levels for IL-1β (P=0.043), IL-17A (P=0.014), TNF-α (P=0.003), and NFKBIZ (P=0.004) in 4-OI-treated hyperosmolar HCECs compared to the untreated group, confirming the anti-inflammatory effects of 4-OI in alleviating hyperosmolar stress in the dry eye cell model (Fig. 1D–G).

Fig. 1.

In vitro anti-inflammatary results of 4-OI. (A) Representative Western blot images demonstrate the protein expression levels of IL-1β and IκBζ from different treatment groups (n=3 independent experiments). Control group: normal HCECs without treatment. DED control group: hyperosmotic HCEC group. 4-OI treatment groups: hyperosmotic HCECs treated with the indicated concentrations (5, 10, 15, 20 μM) of 4-OI for 24 h. (B,C) Quantitative analysis of IL-1β (B) and IκBζ (C) protein levels normalized to GAPDH. (D-G) Relative mRNA expression level of IL-1β (D), IL-17A (E), TNF-α (F) and NFKBIZ (G) measured by q-PCR (n=5 biological replicates). Data are presented as mean ± SEM. ∗P<0.05, ∗∗P<0.01, ∗∗∗P<0.001, ∗∗∗∗P<0.0001 vs. DED control group (one-way ANOVA with Dunnett's post-hoc test).

3.2. 4-OI eye drops ameliorate corneal epithelial damage and tear secretion in dry eye mice

To investigate the efficacy of 4-OI in a mouse model of dry eye, mice were exposed to an environmental chamber set at 18.5% ± 5.1% humidity, 24 °C ± 2 °C temperature, and 15 L/min airflow for 14 days.³⁷ Corneal epithelial defects were evaluated by CFS, revealing uniform and consistent scores (Fig. S2). Tear film stability was maintained at 1–2 mm (Figs. S3 and S4), confirming the successful establishment of the dry eye model in mice.

To determine the maximum tolerated concentration of 4-OI in the mouse cornea, gradient concentrations of 4-OI were topically administered to healthy mice twice daily for 7 days. Based on CFS results (Fig. S5), 2 mM was selected as the high concentration, and 1 mM as the low concentration for subsequent experiments. DED model mice were then assigned to five treatment groups: (i) pathological DED control (treated with PBS), (ii) 0.05% CsA eye drops, (iii) 2 mM 4-OI eye drops, (iv) 1 mM 4-OI eye drops, and (v) 4-OI administered via intraperitoneal injection (50 mg/kg, as previously reported).³¹ Corneal epithelial defects were significantly reduced in the 2 mM 4-OI eye drop group (P<0.0001), the 1 mM 4-OI eye drop group (P=0.016), and the CsA eye drop group (P=0.007) compared to the DED control group (Fig. 2A and B). Notably, the 2 mM 4-OI eye drop group showed a more pronounced improvement than the intraperitoneal injection group (P=0.016). Tear secretion also tended to increase in the 4-OI eye drop groups, with the higher concentration showing a more substantial effect (Fig. 2C).

Fig. 2.

Evaluation of clinical signs after 14 days of treatment in DED mice. (A) Representative images of corneal fluorescein staining. (B) Corneal fluorescein staining scores (n=10 eyes per group). (C) Tear secretion volume measured (n=10 eyes per group). Data are presented as mean ± SEM. ∗P<0.05, ∗∗P<0.01, ∗∗∗∗P<0.0001 vs. DED control group (one-way ANOVA with Tukey's post-hoc test).

3.3. 4-OI instillation decreased IL-1β, IL-17A, TNF-α and NFKBIZ gene and protein expression in mouse corneas

To validate the in vivo anti-inflammatory effects, we assessed the expression of key inflammation-related factors in the corneas of dry eye model mice via qPCR and Western blot across different treatment groups. qPCR results showed significant reductions in mRNA levels of IL-1β (P=0.002), IL-17A (P=0.024), TNF-α (P=0.011), and NFKBIZ (P=0.022) in the 2 mM 4-OI eye drop group compared to the DED control group (Fig. 3A–D). CsA eye drops significantly reduced IL-1β expression (P=0.031) and tended to decrease TNF-α. Moreover, mRNA levels of these inflammatory factors were significantly lower in the 2 mM 4-OI eye drop group than in the intraperitoneal injection group (P<0.05), suggesting that intraperitoneal 4-OI administration has a limited effect on corneal inflammation. Western blot analysis and subsequent densitometry quantification corroborated these findings, revealing markedly reduced protein expression levels of IκBζ, IL-17A, IL-1β, and TNF-α in the corneas of the 2 mM 4-OI-treated mice (Fig. 3E–I). Notably, after 14 days of application, 2 mM 4-OI eye drops elicited a significantly greater reduction in IL-17A mRNA levels compared to the 0.05% CsA eye drops treatment (P=0.0195). This superior inhibitory effect of 4-OI was more pronounced at the protein level. Western blot analysis revealed that 4-OI administration led to a dramatically stronger suppression of key pro-inflammatory proteins relative to CsA, as evidenced by the significantly lower expression levels of IL-17A (P=0.0001), its key transcriptional regulator NFKBIZ (P=0.0028), and TNF-α (P=0.0009). Together, these results demonstrate that topical 4-OI eye drops effectively reduce both mRNA and protein expression of key inflammatory factors, highlighting its potential as a therapeutic agent for ocular surface inflammation.

Fig. 3.

4-OI suppresses the expression of key inflammatory factors in the corneas of DED mice. (A-D) Relative mRNA expression levels of IL-1β (A), IL-17A (B), TNF-α (C) and NFKBIZ (D) measured by qPCR (n=5 corneas per group). (E) Representative Western blot images of IL-1β, IL-17A, TNF-α and IκBζ protein levels. (F-I) Quantitative analysis of IL-1β (F), IL-17A (G), TNF-α (H) and IκBζ (I) protein levels normalized to GAPDH (n=3 independent experiments). Data are presented as mean ± SEM. ∗P<0.05, ∗∗P<0.01, ∗∗∗P<0.001, ∗∗∗∗P<0.0001 vs. DED control group (one-way ANOVA with Dunnett's test).

3.4. 4-OI significantly down-regulated IL-1β, IL-17, TNF-α and other pathways in the corneas of dry eye mice and reduced inflammatory responses

To elucidate the molecular mechanisms underlying the therapeutic effects of 4-OI in dry eye models, we performed mRNA sequencing to identify DEGs in the corneas of mice treated with 2 mM 4-OI and PBS for 14 consecutive days. Comparative analysis of the transcriptomic data revealed significant changes in gene expression, with 1756 DEGs identified between the 4-OI-treated and DED control groups (Fig. 4A and B). Specifically, 876 genes were downregulated, while 880 genes were upregulated in response to 4-OI treatment. GO and KEGG enrichment analyses highlighted several inflammatory pathways, including IL-1 signaling, IL-17 signaling, TNF pathways, and cytokine-receptor interactions, as the most downregulated (Fig. 4C and D). These findings corroborate previous literature on the molecular mechanisms of DED and are consistent with our qPCR and Western blot results, which demonstrated a reduction in pro-inflammatory mediators following 4-OI treatment. Beyond inflammation suppression, 4-OI also regulated genes involved in critical biological processes such as apoptosis, keratinization, and epithelial barrier function, all of which were downregulated, suggesting a protective effect against corneal epithelial damage. Conversely, genes associated with cell cycle progression and DNA repair mechanisms were upregulated (Fig. 4E and F), indicating that 4-OI may promote epithelial regeneration and maintain corneal homeostasis. These findings suggest that 4-OI alleviates DED through a dual mechanism of action, mitigating inflammation while promoting corneal repair, warranting further investigation.

Fig. 4.

Transcriptomic analysis reveals the anti-inflammatory mechanism of 4-OI in the corneas of DED mice. (A) Volcano plot of DEGs between the 4-OI-treated and DED control groups (n=3 biological replicates per group, with two corneas from one mice pooled per replicate). (B) Heatmap of clustered DEGs. (C-D) Significantly enriched KEGG pathways (C) and GO terms (D) for the downregulated DEGs. (E-F) Significantly enriched KEGG pathways (E) and GO terms (F) for the upregulated DEGs.

3.5. High concentration 4-OI eye drops do not induce corneal epithelial apoptosis or morphological abnormalities in mice

To evaluate the safety of 4-OI ophthalmic solution in a dry eye mouse model, we conducted TdT-mediated dUTP nick-end labeling (TUNEL) assays to assess corneal keratocyte apoptosis and performed hematoxylin-eosin (H&E) staining to examine corneal morphology across different treatment groups. After 14 days of treatment, the TUNEL assay showed no significant increase in apoptotic cells in the corneal epithelium of mice treated with high-dose 4-OI eye drops compared to the DED control and CsA groups (Fig. 5A). Quantitative analysis confirmed that the proportion of apoptotic cells in the high-dose 4-OI group was significantly lower than in the DED control group and exhibited the lowest trend among all groups (Fig. 5B). No significant difference was observed compared to the CsA group. Histological evaluation further revealed that high-dose 4-OI eye drops did not induce any noticeable epithelial damage (Fig. 5C). Corneal epithelial thickness was significantly reduced in the DED control group, while the high-dose 4-OI group showed near-complete restoration of normal thickness without signs of epithelial erosion (Fig. 5D). To assess potential systemic effects, we examined liver, spleen, and kidney tissues from both the eye drop and intraperitoneal injection groups, with no evidence of necrosis, deformation, or structural abnormalities compared to untreated mice (Fig. 5E). These findings collectively suggest that high-dose 4-OI does not induce corneal epithelial apoptosis, disrupt corneal morphology, or cause systemic organ damage in the mouse model.

Fig. 5.

Safetyassessment of 4-OI application in a DED mouse model. (A) Representative images of TUNEL (red) and DAPI (blue) staining of mouse corneas. Scale bar: 100 μm. Red: TUNEL-positive cells, Blue: DAPI. (B) Quantification of TUNEL-positive cells per field (n=6 fields per group from 3 mice). (C) Representative H&E-stained sections of the cornea. The yellow double-headed arrows mark the corneal epithelial layer. (D) Quantification of corneal epithelial thickness (n=6 measurements per group from 3 mice). (E) Representative H&E-stained sections of liver, kidney, and spleen tissues. Scale bar: 100 μm. Data in B and D are presented as mean ± SEM. ∗P<0.05, ∗∗P<0.01, ∗∗∗P<0.001 vs. DED control group (one-way ANOVA with Tukey's post-hoc test).

3.6. Discussion

The pathogenesis of DED is characterized by a self-perpetuating inflammatory vicious cycle on the ocular surface, driven by a complex network of stress signaling pathways, activation of adaptive immunity (e.g., T-cell activation), and the release of effector cytokines such as IL-1β, TNF-α, and IL-17, which are directly responsible for epithelial damage.3, 4, 5, 6, 7, 8, 9 These mediators act synergistically to amplify inflammatory response, disrupt epithelial homeostasis, and perpetuate disease chronicity. Therapeutic strategies capable of simultaneously targeting multiple nodes within this inflammatory network may offer superior efficacy in breaking the cycle.

In this study, topical administration of 4-OI demonstrated significant anti-inflammatory effects in DED models. Specifically, in hyperosmolar HCECs and the corneas of DED mice, 4-OI reduced the levels of pro-inflammatory cytokines IL-1β, IL-17A, TNF-α, and the key autoimmune regulator IκBζ protein. Following 4-OI instillation, the disease phenotype, including corneal epithelial defects and tear secretion in dry eye model mice, was alleviated without significant local or systemic adverse effects. To our knowledge, this is the first study to identify and confirm that the ITA derivative 4-OI can be used safely and effectively in the treatment of DED in animal models. Our results demonstrate that the therapeutic effect of 4-OI is achieved through its action as a comprehensive immunometabolite modulator. The downregulation of IκBζ and IL-17A expression supports the notion that 4-OI functions by modulating the ATF3/IκBζ axis and directly countering IL-17A-induced cytotoxicity.11, 12, 13 Beyond this targeted effect, our transcriptomic data revealed that 4-OI induced a broad anti-inflammatory landscape, significantly suppressing MAPK and NF-κB signaling pathways, IL-1 receptor activity and binding, TNF and IL-17 signaling pathway, thereby neutralizing the toxic effects of cytokines themselves and interrupting the amplification process in the vicious cycle of inflammatory cascade. This multi-faceted suppression is consistent with ITA's known mechanisms through inhibition of the SDH/succinate/hypoxia-inducible factor (HIF)-1α axis, thereby downregulating IL-1β,^34,35 and activation of the nuclear factor erythroid 2-related factor 2 (Nrf2) pathway to antagonize NF-κB-mediated TNF signaling.³⁰ The downregulation of MAPK and NF-κB signaling also suggests that 4-OI may function by suppressing key stress-response pathways in the ocular surface. Furthermore, the downregulation of genes involved in apoptosis and keratinization, coupled with the upregulation of genes associated with cell cycle and DNA repair, suggests a dual mechanism of action: mitigating inflammatory damage while actively promoting epithelial regeneration. For instance, the downregulation of keratinization processes might be partially attributed to the direct influence of 4-OI on IL-1β and IL-17A. IL-1β is associated with the squamous metaplasia of corneal epithelial cells,⁴³ while IL-17A —often acting through the regulator NFKBIZ to drive keratinocyte proliferation⁴⁴—is a critical driver of hyperkeratinization in pathologies like psoriasis.¹³ Our observation of upregulated DNA repair and cell cycle genes aligns with the well-established role of 4-OI as a potent Nrf2 activator.³⁰ The Nrf2 pathway is a master regulator of cellular defense and has been mechanistically linked to enhanced DNA repair capacity and cytoprotection.⁴⁵ Also, a recent study demonstrated that 4-OI directly reduced DNA damage and promoted repair in endothelial cells by engaging a defined E2F transcription factor 1-aconitate decarboxylase 1-itaconate axis.⁴⁶ This provides a mechanistic precedent for our observation that 4-OI may promote corneal epithelial condition not only by suppressing inflammation but also by actively enhancing DNA repair and cellular resilience. This suggests that 4-OI might exert a multifaceted, systems-level effect on ocular surface homeostasis, effectively disrupting the vicious cycle of DED at multiple nodes. However, the superior efficacy of topical 4-OI over intraperitoneal injection underscores the importance of local delivery, achieving higher drug concentrations in the corneal epithelium for enhanced anti-inflammatory effect, as supported by the concentration-dependent response between 1 mM and 2 mM doses.

CsA is an established first-line immunomodulatory therapy for DED, serving as the clinically relevant benchmark to evaluate the therapeutic potential of 4-OI in this study. Results showed that while topical 0.05% CsA reduced the CFS score and IL-1β content in DED mice after 14 days, it exhibited no significant effect on the expression of key downstream cytokines such as TNF-α, IL-17A, and NFKBIZ/IκBζ. In contrast, 4-OI elicited a more pronounced and consistent suppression of these critical inflammatory mediators. This differential efficacy might stem from their fundamental mechanisms of action: CsA acts as an upstream immunosuppressant by inhibiting calcineurin-NFAT signaling in T-cells, achieving broad but indirect immunosuppression.²⁴ Upon entering T-cells, CsA forms a complex with cyclophilin that inhibits calcineurin. This prevents the dephosphorylation and nuclear translocation of NFAT, a master regulator of IL-2 transcription. The consequent reduction in IL-2 production halts T-cell clonal expansion and differentiation. This T-cell suppression consequently reduces the production of various DED-related cytokines. Specifically, the diminished differentiation of Th1 and Th17 cells lowers levels of IFN-γ and IL-17, which contribute to epithelial barrier damage and goblet cell loss. Furthermore, by modulating T-cell activity, CsA indirectly dampens broader inflammatory networks, reducing levels of other mediators like IL-1β and TNF-α.^25,26 This broad but indirect immunosuppressive action, coupled with its relatively large and hydrophobic nature that results in poor cell membrane penetration and limited bioavailability, may account for its delayed clinical onset, often requiring prolonged treatment (e.g., 60 days with a higher 0.09% concentration) to achieve significant downstream effects.^27,47 Conversely, 4-OI functions as a comprehensive immunomodulator, directly targeting pro-inflammatory signals (e.g., IL-17 via the ATF3/IκBζ axis and IL-1β via SDH/succinate/HIF-1α axis) within the effector phase of established inflammation.11, 12, 13, 14^,34,35 This direct mechanism may facilitate a more rapid therapeutic response. The superior efficacy of the topical 4-OI formulation over intraperitoneal injection, evidenced by the concentration-dependent response between the 1 mM and 2 mM doses, underscores the critical importance of achieving high local drug concentrations in the corneal epithelium for effective anti-inflammatory action. Additionally, as an endogenous small molecule with favorable water solubility, 4-OI presents a potential formulation advantage over the irritative potential associated with some conventional CsA emulsions,^31,36 which often require oleaginous vehicles and preservatives that can cause burning and stinging.^27,28

Our study provides the first preclinical evidence that topical 4-OI eye drops effectively alleviates clinical signs and suppresses ocular surface inflammation in a DED mouse model, with a favorable safety profile. The therapeutic efficacy of 4-OI was associated with the significant downregulation of key inflammatory pathways as confirmed by transcriptomic and molecular analyses. These findings validate our initial hypothesis that 4-OI, by targeting comprehensive immunometabolic checkpoints, represents a novel and promising therapeutic strategy for DED. However, the promising findings of this initial investigation must be considered in the context of its limitations, including the 14-day treatment period that precludes assessment of long-term durability, the preliminary dose-ranging falling short of a full dose-response analysis, and the use of a simple aqueous solution without comparison to advanced formulations. Consequently, further research is essential to define the long-term pharmacological profile, optimize formulations for improved ocular bioavailability, and fully elucidate the molecular mechanisms to unlock the complete therapeutic potential of 4-OI on DED.

Study approval

The authors confirm that any aspect of the work covered in this manuscript that involved human patients or animals was conducted with the ethical approval of all relevant bodies and the study was performed in accordance with the Declaration of Helsinki，and the protocol was approved by the Animal Experimentation Ethics Committee of Eye & ENT Hospital of Fudan University (approval number: IRB-EENT-150301) and conformed to the Association for Research in Vision and Ophthalmology statement on the use of animals.

Author contributions

The authors confirm contribution to the paper as follows: ZZ: Data curation, Writing-Original draft preparation; YL: Investigation; XJZ: Software, Visualization; XTZ: Conceptualization, Methodology; FZ: Writing- Reviewing and Editing; JH: Methodology, Supervision. All authors reviewed the results and approved the final version of the manuscript.

Funding

This work was supported by the National Science Fund for Distinguished Young Scholars (82425015), National Natural Science Foundation of China (82171102 and 32171432), National Key Research and Development Program of China (2023YFA0915003 and 2022YFC3400202), the "Dawn" Program of Shanghai Municipal Education Commission (24SG11), Shanghai Science and Technology Innovation Action Plan for Cell and Gene Therapy (24J22800500), Shanghai Science and Technology Innovation Action Plan for Advanced Materials (24CL2900802), Shanghai Municipal Commission of Health (20254Z0019), and Shanghai High-Quality Development Plan for Sciencee and Technology Industry (25SF1907000).

Declaration of competing interests

The authors declare that they have no financial or personal re lationships that could inappropriately influence, or be perceived to in f luence, the research presented in this manuscript.

Appendix A. Supplementary data

The following is the Supplementary data to this article: