Fusarium solani Activates PANoptosis and Modulates Immune Response in Fungal Keratitis

Abstract

Purpose

This study aimed to investigate how Fusarium solani triggers PANoptosis in fungal keratitis and to examine the toxic impacts and underlying mechanisms of deoxynivalenol (DON) on human corneal epithelial (HCE-T) cells.

Methods

A F. solani keratitis mouse model was established through intrastromal injection of fungal spores. Clinical severity was evaluated using a standardized scoring system. PANoptosis markers in corneal tissues were analyzed via western blotting, immunofluorescence, and TUNEL assay. Real-time PCR was used to quantify inflammatory cytokines. RNA sequencing was conducted to select differentially expressed genes (DEGs) and analyze immune cell infiltration profiles. HCE-T cells were exposed to DON, and PANoptosis-related proteins were detected via western blot and immunofluorescence. Gene expression alterations were analyzed using RNA sequencing.

Results

PANoptosis activation was validated through upregulated expression of cleaved caspases-1, -3, -7, and -8; gasdermin D (GSDMD); and phosphorylated mixed lineage kinase domain-like protein (p-MLKL) in infected corneas, accompanied by augmented TUNEL-positive cells. Inflammatory cytokines (IL-6, IL-1β, TNF-α) were significantly upregulated. RNA sequencing revealed significant changes in PANoptosis, immune response, and extracellular matrix organization. Immune infiltration profiling indicated a marked increase in M1 macrophages in infected corneas. In vitro experiments demonstrated that DON exposure increased PANoptosis markers (cleaved caspases-1, -3, -7, and -8; BAX; GSDMD; and p-MLKL) in HCE-T cells, along with upregulation of inflammatory cytokines (IL-6, IL-1β, IL-18, TNF-α). RNA sequencing further revealed alterations in ribosomal RNA (rRNA) processing and extracellular matrix organization pathways in DON-treated cells.

Conclusions

F. solani infection and DON exposure induce PANoptosis in fungal keratitis, leading to significant inflammatory responses. These results suggest that targeting PANoptosis signaling pathways may represent a novel therapeutic approach for treating fungal keratitis.

Fungal keratitis is a severe infectious corneal disease that poses a significant threat to vision, being one of the major contributors to global blindness and visual impairment.^1,2 Remarkably, its incidence has increased in tropical and subtropical regions over the past decade.³ Fungal keratitis is primarily caused by pathogenic species such as Fusarium, Aspergillus, and Candida. Among these, Fusarium solani is the most commonly isolated pathogen, accounting for 25% to 73.3% of global cases.^4–6

Producing various mycotoxins, including trichothecenes, zearalenone, fumonisin B1, and T-2 toxins, is a notable characteristic of Fusarium species.⁷ These secondary metabolites act as virulence factors, impairing host tissues and immune responses while promoting fungal invasion.⁸ Deoxynivalenol (DON), a sesquiterpenoid mycotoxin, is primarily synthesized by filamentous fungi within the Fusarium genus. Emerging evidence has demonstrated that F. solani strains contain high levels of DON.^9,10 This mycotoxin induces a spectrum of toxicological effects in both humans and animals, including damage to the gastrointestinal tract, liver, kidneys, reproductive system, and nervous system.^11–13 However, the pathogenic mechanisms of F. solani and DON in fungal keratitis remain unclear.

PANoptosis, a newly characterized integrated cell death modality, is defined by crosstalk among apoptotic, pyroptotic, and necroptotic signaling cascades.¹⁴ Research has shown that effector molecules from these pathways can be co-activated in tissues or cells to modulate host immune responses.^15,16 Studies have demonstrated that infections by diverse pathogens, including fungi such as Candida albicans and Aspergillus fumigatus, can cause extensive cell death associated with PANoptosis.¹⁷ These studies indicate that PANoptosis is pivotal in the host response to various infectious agents. Despite these insights, the role of PANoptosis in F. solani–induced fungal keratitis has not been previously explored. Therefore, our study sought to bridge this gap by examining the potential for F. solani and DON to trigger PANoptosis in fungal keratitis. Elucidating these mechanisms may uncover novel insights into pathogenesis and facilitate the improvement of novel therapeutic strategies.

Materials and Methods

Establishment of a Mouse Corneal F. solani Infection Model

All animal experiments adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Male C57BL/6 mice (18–20 g, 6–8 weeks old) from Daoke Pharmaceutical Technology Co. (Guangdong, China) were housed under specific pathogen-free conditions. Experimental eyes were topically treated with levofloxacin eye drops (Santen, Osaka, Japan) twice daily for 3 days. Povidone iodine solution (Heng Jian, Guangdong, China) was applied for preoperative disinfection. Corneal infection was induced via intrastromal injection of 2 µL of a 4.5 × 10⁶ colony-forming units (CFU)/mL F. solani spore suspension into the right eye with a 33-gauge Hamilton syringe. The left eye remained uninjected and served as a control.

Clinical Scoring

Corneal opacity, surface regularity, and opacity extent were evaluated and photographed at 1, 3, 5, 7, and 10 days after modeling. Clinical scores were graded using a quantitative visual scoring system previously established by Wu et al.,¹⁸ with the specific scoring criteria detailed in Supplementary Table S1. On day 3 post-infection, corneas were paraffin–embedded, sectioned, and stained by hematoxylin and eosin (H&E) for microscopic imaging.

Cell Culture

The human corneal epithelial (HCE-T) cell line was obtained from Procell Life Science & Technology (CL-0743; Wuhan, China). HCE-T cells were cultured in Gibco Dulbecco's Modified Eagle's Medium/Nutrient Mixture F-12 (DMEM/F-12) medium containing 10 ng/mL human epidermal growth factor (hEGF, E9644; Sigma-Aldrich, St. Louis, MO, USA), 10% Gibco fetal bovine serum (A5669701; Thermo Fisher Scientific, Waltham, MA, USA), 0.1 mg/mL penicillin/streptomycin (15140122; Thermo Fisher Scientific), and 5 µg/mL insulin (I2643; Sigma-Aldrich). Cells were seeded into 25-cm² flasks and incubated at 37°C, 5% CO₂, and 80% to 90% humidity, with medium changes every 2 days. Cells were subcultured at a 1:3 ratio when cell density reached 80% to 90%.

Cell Counting Kit-8 Assay

HCE-T cells were collected and inoculated into 96-well plates (1 × 10⁵ cells/mL), with five replicates per treatment group. Peripheral wells were filled with PBS, and background and blank controls were included. After 24-hour incubation for cell adhesion, media containing DON (D0156; Sigma-Aldrich) at concentrations of 1.25, 2.5, 5, 10, or 20 µM were added. After 24-hour treatment, 100 µL of fresh medium with 10 µL Cell Counting Kit-8 (CCK-8) solution (GK10001; GlpBio, Shanghai, China) was added per well. Optical density (OD) values were measured at 450 nm, and cell viability was calculated as cell viability (%) = [(A1A)/(A0A)] × 100%, where A1 is the sample OD, A is the background OD, and A0 is the blank control OD.

Western Blotting

Corneal tissue/HCE-T cells were mixed with radioimmunoprecipitation assay (RIPA) buffer (K1125; APExBio Technology, Houston, TX, USA), followed by ultrasonic grinding and centrifuging. Protein concentration was determined using a bicinchoninic acid (BCA) assay kit (PT0001; Leagene, Shanghai, China). Then, 5× loading buffer (LT101; Epizyme, Shanghai, China) was added to samples, which were then heated at 100°C for 10 minutes; 10% and 15% SurePAGE gels (M00665/M00719; GenScript, Jiangsu, China) were used to transfer them. Membranes probed with primary antibodies against cleaved caspase-1 (1:2000, PA5-105049; Thermo Fisher Scientific), cleaved gasdermin D (GSDMD; 1:800, NBP2-33422; Novus Biologicals, Centennial, CO, USA), cleaved caspase-3 (1:1000, 9661T; Cell Signaling Technology, Danvers, MA, USA), cleaved caspase-7 (1:1000, 9491T; Cell Signaling Technology), cleaved caspase-8 (1:1000, 8592T; 1:1000, 9496T; Cell Signaling Technology), phosphorylated mixed lineage kinase domain-like protein (p-MLKL) (1:1000, 37333S, Cell Signaling Technology; 1:1000, ab187091, Abcam, Cambridge, UK), and GAPDH (1:10.000, 10494-1-AP; Proteintech, Rosemont, IL, USA) overnight at 4°C. Horseradish peroxidase (HRP)-conjugated secondary antibodies (1:10,000, bs-0295G; Bioss Antibodies, Woburn, MA, USA) were used, and blots were visualized by enhanced chemiluminescence (Tanon Life Science, Shanghai, China).

Real-Time Quantitative PCR

Total RNA was extracted from mouse corneas or HCE-T cells using the EZ-Press RNA Purification Kit (B0004D, EZBioscience, Roseville, MN, USA). RNA concentration was detected using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Synthesis of cDNA was performed using a reverse transcription kit (RR036A, Takara, Shiga, Japan), and PCR amplification was performed using a SYBR Green qPCR kit (A0012; EZBioscience). Primer sequences are listed in the Table.

Table.

Sequence of Primers for qPCR

IL-1β	Mouse	Forward	TGGACCTTCCAGGATGAGGACA
		Reverse	GTTCATCTCGGAGCCTGTAGTG
TNF-α	Mouse	Forward	GGTGCCTATGTCTCAGCCTCTT
		Reverse	GCCATAGAACTGATGAGAGGGAG
IL-6	Mouse	Forward	TTCACAGAGGATACCACTCC
		Reverse	AAGTGCATCATCGTTGTTCA
GAPDH	Mouse	Forward	GTGGCAAAGTGGAGATTGTTG
		Reverse	CGTTGAATTTGCCGTGAGTG
IL-6	Human	Forward	CACACAGACAGCCACTCACC
		Reverse	TTTTCTGCCAGTGCCTCTTT
IL-18	Human	Forward	GCTTGAATCTAAATTATCAGTC
		Reverse	GAAGATTCAAATTGCATCTTAT
IL-1β	Human	Forward	TGTACCTGTCCTGCGTG
		Reverse	ACTGGGCAGACTCAAATTC
TNF-α	Human	Forward	CCCTCACACTCAGATCATCTTCT
		Reverse	GCTACGACGTGGGCTACAG
GAPDH	Human	Forward	TGGCACCCAGCACAATGAA
		Reverse	CTAAGTCATAGTCCGCCTAGAAGCA

Immunofluorescence Assay

Antigen was restored by a citrate-based buffer (pH 6.0) for paraffin-embedded sections. Cells were fixed and permeabilized with 4% paraformaldehyde and 0.2% Triton X-100. Sections were incubated with primary antibodies against cleaved caspase-1 (1:500, PA5-105049; Thermo Fisher Scientific), cleaved GSDMD (1:160, NBP2-33422; Novus Biologicals), cleaved caspase-3 (1:400, 9661T; Cell Signaling Technology), cleaved caspase-8 (1:400, 8592T; Cell Signaling Technology), p-MLKL (1:1600, 37333S; Cell Signaling Technology), EGF-like module-containing mucin-like hormone receptor-like 1 (EMR1, F4/80; 1:500, GB113373; Servicebio, Wuhan, China), and α-smooth muscle actin (α-SMA; 1:500, GB111364; Servicebio). Positive cells were quantified under high magnification in three randomly selected fields, and positive cell ratio was calculated based on 4′,6-diamidino-2-phenylindole (DAPI) colocalization.

TUNEL Staining

Corneal paraffin sections were deparaffinized by formaldehyde and a graded ethanol series, followed by PBS washing. Sections were treated with 20 µg/mL proteinase K (ST533; Beyotime, Shanghai, China) and washed with PBS, followed by incubation at 37°C after applying TUNEL reagent (C1088; Beyotime). Fluorescence images were captured, and positive cells were quantified as in immunofluorescence staining.

Live/Dead Staining

HCE-T cells were cultured with 5 µM DON for 24 hours, with untreated cells serving as controls. After treatment, cells were stained by a live/dead cell staining kit (C2015; Beyotime). Following a final PBS wash, cells were imaged using a fluorescence microscope. Positive cells were quantified as described for immunofluorescence staining.

RNA Sequencing and Analysis

RNA sequencing (RNA-seq) of RNA from mouse corneas (noninfected vs. F. solani–infected) and HCE-T cells (untreated vs. 5-µM DON-treated) was done at BGI Genomics (Shenzhen, China). Sequencing data were quantified as FPKM, log₂ transformed, and analyzed bioinformatically. DEGs were identified using the Bayes test and limma Bioconductor package (P < 0.05), ranked by P value. Gene annotation and visualization were performed with Database for Annotation, Visualization, and Integrated Discovery (DAVID) (P < 0.05). Reactome pathway enrichment analysis was used for pathway analysis, focusing on Gene Ontology (GO) Biological Process (BP). Protein–protein interaction (PPI) networks were built with STRING l1.5 (confidence score > 0.7), and hub genes were identified by topological analysis in Cytoscape 3.9.1 with MCODE and cytoHubba plugins based on Matthews correlation coefficient (MCC) scores.

Immune Cell Infiltration Analysis

Immune cell infiltration profiles in F. solani–infected mouse corneas were characterized using ImmuCell-AI-mouse 1.0 (ImmuCellAI_mouse; wchscu.cn), a machine-learning platform. It quantifies 36 immune subsets from bulk RNA-seq data across three hierarchical layers. The tool uses support vector machine classifiers trained on mouse single-cell datasets to calculate cell-type enrichment scores, classifying immune cells into seven major lineages, 15 intermediate subtypes, and 14 terminal subsets. Corneal sample raw counts (noninfected vs. infected) were normalized by DESeq2 and processed to generate relative abundance matrices. Wilcoxon rank-sum test with Bonferroni correction was applied to the differential analysis to identify significantly changed immune subsets (q < 0.05) to better understand infection-induced immune responses.

Statistical Analysis

Data are presented as mean ± SD. Statistical analyses and visualization were performed using R 4.2.1 (R Foundation for Statistical Computing, Vienna, Austria) and Prism 9.0.0 (GraphPad, Boston, MA, USA). Group comparisons made using unpaired two-tailed Student's t-tests or one-way ANOVA. Statistical significance was defined as *P < 0.05, **P < 0.01, and ***P < 0.001 (see details in figure legends).

Results

Establishment of a F. solani Keratitis Mouse Model

Clinical symptoms of fungal keratitis, including corneal opacity and neovascularization, were monitored at 1, 3, 5, 7, and 10 days post-infection (dpi) (Fig. 1A). On the first day, corneal edema was noted, with significant cloudiness in the inoculation site and congestion of the corneal limbal vascular network. By day 3, corneal opacity and protrusion were exacerbated, with neovascularization extending into the corneal stroma. On the fifth day, partial corneal transparency was restored, but notable protrusion persisted. Fluorescein staining confirmed that the corneal ulcer area reached its maximum size at 3 dpi. Ocular surface inflammation was evaluated using clinical parameters, which collectively indicated peak inflammation at 3 dpi (Fig. 1B). Based on these observations, subsequent experiments utilized samples collected at this time point. Calcofluor white (CFW) staining was used to confirm the fungal infection (Supplementary Fig. S1). H&E staining demonstrated marked histological differences between the normal control cornea and the F. solani–infected cornea. Infected corneas showed extensive fibroblast proliferation, disorganized collagen fibers, and dense inflammatory cell infiltration, in contrast to the intact structure of control corneas (Figs. 1C, 1D).

Figure 1.

Establishment of the F. solani keratitis mouse model. (A) Clinical manifestations of F. solani keratitis observed on days 1, 3, 5, 7, and 10 post-infection (dpi). (B) Clinical scores show the severity of F. solani keratitis. (C, D) H&E staining of a normal control cornea (C) and cornea infected with F. solani (D).

Activation of PANoptosis and Inflammatory Response in F. solani–Infected Mouse Corneas

To characterize PANoptosis activation, western blot analysis revealed significantly upregulated expression of key markers (including cleaved caspases-1, -3, -7, and -8; cleaved GSDMD; and p-MLKL) in F. solani–infected corneas relative to uninfected controls (Figs. 2A–F). Immunofluorescence staining was used to localize and quantify PANoptosis markers in corneal tissues, corroborating western blot findings. The results showed an absence of PANoptosis-related proteins in healthy corneas, but high expression levels were detected in corneas infected with F. solani. Immunofluorescence staining showed that apoptotic and pyroptotic cells were primarily localized to the stroma (Figs. 2G–J), but necroptotic cells accumulated in epithelial and stromal layers of ulcerated regions (Fig. 2K). TUNEL staining confirmed a significant increase in apoptotic cell density in infected corneas relative to controls (Fig. 2L). Additionally, F. solani infection induced a significant upregulation of inflammatory cytokines in the corneal tissue (Fig. 2M).

Figure 2.

F. solani infection induces upregulation of PANoptosis-related protein expression in corneal tissue. (A–F) Western blot analysis was performed to assess the expression levels of PANoptosis markers in in corneal tissue from F. solani–infected mice. GAPDH was used as a loading control. (G–L) Representative immunofluorescence images of corneal sections stained for cleaved caspase-3 (green), cleaved caspase-8 (green), cleaved caspase-1 (green), cleaved GSDMD (green), p-MLKL (red), and TUNEL (red), with DAPI (blue) marking cell nuclei. (M) qPCR analysis of inflammation-associated factors IL-1β, TNF-α, and IL-6. Quantification of positive cell counts in the corneal tissue is shown. Scale bar: 50 µm. **P < 0.01; ***P < 0.001; ****P < 0.0001.

Differential Gene Expression Analysis in Fungal Keratitis-Infected Corneas

To explore transcriptional alterations in F. solani–infected corneas, RNA-seq was conducted on four infected and four normal corneal samples (Fig. 3A). RNA-seq identified 3364 upregulated and 2591 downregulated genes in infected corneas relative to controls (Fig. 3B). The top 10 upregulated and downregulated DEGs are shown in Figure 3C. Reactome pathway analysis of upregulated DEGs revealed enrichment in immune system processes, neutrophil degranulation, and extracellular matrix (ECM) organization (Fig. 3D). GO analysis indicated that the upregulated DEGs were significantly enriched in biological processes such as inflammatory response, immune system processes, chemotaxis, immune response, and innate immune response. These genes were also enriched in cellular components, including the outer membrane, extracellular space, cytoplasmic membrane, extracellular region, and biofilm. Additionally, molecular functions such as chemokine activity, cytokine activity, signal receptor activity, transmembrane signal receptor binding, and chemokine receptor binding were significantly represented (Fig. 3E). In contrast, Reactome pathway analysis of downregulated DEGs identified enrichment in neuronal system, metabolism, and visual phototransduction pathways (Supplementary Fig. S2A). GO analysis indicated that downregulated DEGs were enriched in biological processes including visual perception, response to stimuli, and light signal transduction. These genes were also enriched in cellular components such as synapses, myosin II complex, axon terminals, neuronal cell bodies, and plasma membrane. Additionally, molecular functions, including calmodulin binding, microfilament motility activity, 3-deoxyglucosamine dehydrogenase activity, benzaldehyde dehydrogenase (NAD⁺) activity, and carboxylesterase activity, were significantly represented (Supplementary Fig. S2B). PPI analysis of the top 1000 upregulated and downregulated DEGs identified the top 10 hub genes. The top 10 hub upregulated and downregulated genes are shown in Supplementary Figures S2C and S2D, respectively. To further validate PANoptosis activation, the expression of PANoptosis-related genes in F. solani–infected corneas of mice were detected (Fig. 3F). The DEG analysis showed that the transcriptional expression levels of caspase family (caspases-3, -6, -7, and -8), BCL2 family (Bax, Bid), death receptor family (Fas), gasdermin family (GSDMA, GSDMC, GSDMD), PANoptosome components (ZBP1, RIPK1, NLRP3, AIM2), and others (MLKL, HSP90, HMGB1, NLRC4, TRADD, IL-18, TNF) were significantly upregulated but the transcriptional expression levels of Bcl2 were downregulated. Based on the above sequencing results, further confirmation of PANoptosis activation (cleaved GSDMD, cleaved caspase-3, p-MLKL), macrophage infiltration (F4/80), and fibroblast activation (α-SMA) was obtained through immunofluorescence staining (Fig. 3G).

Figure 3.

Differential gene expression analysis in fungal keratitis-infected corneas. (A) Heatmap of DEGs from RNA-seq of four infected corneas and four normal corneas. (B) Volcano plot illustrating 3364 upregulated and 2591 downregulated genes in F. solani–infected corneas compared to normal corneas. (C) List of the top 10 upregulated and downregulated DEGs. (D) Reactome pathway enrichment analysis of upregulated DEGs indicating significant pathways related to the immune system, neutrophil degranulation, and ECM organization. (E) GO analysis of upregulated DEGs. (F) PANoptosis-related gene expression in F. solani–infected corneas and normal corneas. (G) Immunofluorescence staining of α-SMA, F4/80, cleaved GSDMD/cleaved caspase-3/p-MLKL in F. solani–infected and normal corneas.

Immune Cell Infiltration Analysis of Corneas Infected with F. solani

Excessive activation of the host innate immune response is a key pathological driver of corneal injury in fungal keratitis. Pathogenic fungi activate pattern recognition receptors, leading to induction of chemokines and cytokines that recruit immune cells to the cornea for fungal clearance. To characterize immune cell infiltration in F. solani–infected corneas, we analyzed immune cell profiles using bulk RNA-seq data. A heatmap illustrates the composition of immune cells in corneal tissue following F. solani infection; macrophages were the predominant immune cell type in infected corneas, contrasting with T cells in normal corneas (Fig. 4A). Within infected corneas, M1 macrophages were the dominant macrophage subset, whereas Treg cells were the most abundant T-cell subtype in normal corneas (Fig. 4B). Further analysis of the differential expression of immune cells between the corneal tissue following F. solani infection and normal cornea revealed significant differences (Fig. 4C). Activated CD8⁺ T cells were enriched in infected corneas, whereas naïve CD8⁺ T cells predominated in normal tissues. M0 and M1 macrophages were enriched in infected corneas, but M2 macrophage levels remained unchanged. Notably, Treg cells were significantly enriched in normal corneas.

Figure 4.

Immune cell infiltration analysis of corneas infected with F. solani. (A) Heatmap depicting the changes in immune cell composition in corneal tissue following F. solani infection. (B) Proportional comparison of immune cell subtypes in infected and normal corneas. (C) Differential expression analysis of specific immune cell subtypes between F. solani–infected and normal corneas.

Toxic Effects of DON on HCE-T Cells and Induction of PANoptosis

Our previous research has confirmed the presence of high levels of DON in the culture medium for F. solani and in rabbit corneas infected with F. solani. Also, DON treatment prolonged the corneal epithelial regeneration cycle.⁹ To explore the role of DON toxicity in fungal keratitis, we examined the specific pathogenic mechanism of DON on HCE-T cells. HCE-T cells were treated with gradient concentrations of DON for 24 hours. DON exposure reduced HCE-T cell viability in a concentration-dependent manner, decreasing viability to 77.4%, 62.6%, 53.6%, 51.3%, and 42.3% at 1.25 to 20 µM DON (Fig. 5A). After 24 hours of DON treatment of HCE-T cells, the expression levels of cleaved caspase-3, cleaved caspase-7, cleaved caspase-8, Bax, cleaved caspase-1, and cleaved GSDMD significantly increased, but there was no significant change in p-MLKL protein expression (Figs. 5B–G). When the DON treatment time was extended to 48 hours, the expression level of p-MLKL significantly increased (Fig. 5H). Immunofluorescence staining confirmed DON-induced PANoptosis by detecting cleaved caspase-3 and cleaved GSDMD expression and localization. The group treated with 5-µM DON exhibited strong cytoplasmic expression of cleaved caspase-3 and GSDMD (Figs. 5I–J). Live/dead cell staining showed that the number of dead cells in the group treated with 5-µM DON increased compared to the control group (Fig. 5L). Additionally, messenger RNA (mRNA) fold changes of IL-6, IL-1β, IL-18, and TNF-α were elevated in the HCE-T cells treated with 5-µM DON compared to the control group (Fig. 5M).

Figure 5.

Toxic effects of DON on HCE-T cells and induction of PANoptosis. (A) Cell viability of HCE-T cells after treatment with gradient concentrations of DON (1.25, 2.5, 5, 10, and 20 µM) for 24 hours, assessed by CCK-8 assay. Results show a concentration-dependent reduction in cell viability. (B–H) Western blot analysis of PANoptosis marker expression in HCE-T cells after DON treatment. (I, J) Immunofluorescence staining of cleaved caspase-3 (green) and cleaved GSDMD (green) in HCE-T cells following 5-µM DON treatment for 24 hours. (K) Live/dead cell staining showing a significant increase in cell mortality in the 5-µM DON treatment group compared to the control group. (L) Quantitative PCR analysis of inflammation-associated factors IL-6, IL-1β, IL-18, and TNF-α in HCE-T cells. Data were analyzed using the Student's t-test. *P < 0.05, **P < 0.01, ***P < 0.001.

Molecular Mechanisms of HCE-T Cell Damage Induced by DON Based on RNA-Seq Analysis

To uncover the molecular mechanisms of DON-induced HCE-T cell damage, RNA-seq analysis was performed. The analysis identified a total of 6747 DEGs in the DON-treated group compared to the control group, with 3238 genes upregulated and 3509 genes downregulated (Figs. 6A, 6B). The top 10 upregulated and downregulated DEGs are presented in Figure 6C. Upregulated DEGs were primarily enriched in pathways related to ribosomal RNA (rRNA) processing (nuclear and cytoplasmic modifications), transfer RNA (tRNA) processing, and ribosomal biogenesis (Supplementary Fig. S3A). In contrast, the downregulated DEGs were enriched in pathways linked to ECM organization, collagen formation, non-integrin membrane–ECM interactions, steroid metabolism, and elastin fiber formation (Fig. 6D). GO analysis showed that upregulated genes were enriched in mRNA processing, non-coding RNA (ncRNA) metabolism, and ribosome biogenesis. Enriched cellular components included RNA polymerase complexes, nuclear membranes, and pre-ribosomal complexes. In terms of molecular functions, the upregulated genes were enriched in DNA transcriptional activation/inhibition factor binding, RNA/rRNA catalytic activity, and methyltransferase activity (Supplementary Fig. S3B). Downregulated DEGs were enriched in cell–matrix adhesion, glycoprotein metabolism, and lipid transport pathways. Enriched cellular components included collagen-rich ECM and endoplasmic reticulum lumen. Molecular functions such as transmembrane transporter activity, amide binding, ECM structural components, glycosyltransferase activity, and integrin binding were also significantly enriched (Fig. 6E). To further confirm PANoptosis activation, the expression of PANoptosis-related genes in HCE-T cells treated with DON were detected (Fig. 6F). The DEG analysis showed that the transcriptional expression levels of the caspase family (caspases-2, -3, and -7), BCL2 family (Bak1, Bid), gasdermin family (GSDMB), PANoptosome (RIPK1, FADD), and others (CYLD, IL-18, TNFRSF10A, TNFRSF10B) were significantly upregulated.

Figure 6.

Molecular mechanisms of HCE-T cell damage induced by DON based on RNA-seq analysis. (A, B) Heatmap and volcano plot showing the DEGs between DON-treated and control groups. (C) The top 10 upregulated and downregulated DEGs in response to DON treatment. (D) Reactome pathway enrichment analysis of downregulated DEGs. (E) GO functional enrichment analysis of downregulated DEGs. (F) PANoptosis-related gene expression in DON-treated and control groups.

Discussion

Fungal keratitis is a sight-threatening ophthalmic disease, often resulting from ocular trauma that disrupts the corneal epithelial barrier, allowing conidia to infiltrate the corneal stroma.¹⁹ Upon corneal invasion, pathogens secrete toxins, hydrolytic enzymes, and proteases to facilitate adhesion, invasion, and disease progression. In response, the innate immune system is activated to recognize and combat the infection, thereby influencing the severity of the disease. PANoptosis, a distinct form of inflammatory programmed cell death (PCD), integrates apoptosis, pyroptosis, and necroptosis to enhance host defense against pathogens, representing a critical component of innate immunity. Previous studies have demonstrated that PANoptosis can be triggered by pathogens such as Candida albicans and Aspergillus fumigatus. However, it remains unclear whether F. solani induces PANoptosis in fungal keratitis and how this mechanism contributes to the pathogenesis of the disease. In this study, our purpose was to investigate the pathological mechanisms of F. solani–induced fungal keratitis and explore the toxic effects of DON on HCE-T cells. Our results demonstrate that F. solani infection triggers PANoptosis, marked by concurrent activation of apoptotic, pyroptotic, and necroptotic pathways. Furthermore, we identified substantial changes in gene expression, immune cell infiltration, and inflammatory responses in both infected corneas and DON-treated HCE-T cells, offering new perspective on the molecular mechanisms of fungal keratitis and its potential exacerbation by DON.

Pyroptosis, a form of cellular inflammatory necrosis, is critical for regulating inflammation, immune responses, and pathogen infections.²⁰ In its canonical pathway, activated inflammasomes process pro-caspase-1, which cleaves GSDMD into N-terminal and C-terminal fragments. The N-terminal domain induces plasma membrane pore formation, leading to cell lysis. Concurrently, caspase-1 processes pro-IL-1β and pro-IL-18 into their active forms, amplifying inflammation.²¹ In F. solani–infected corneas, we observed elevated levels of pyroptosis markers (cleaved caspase-1 and GSDMD), localized primarily to the stroma via immunofluorescence. These results highlight pyroptosis as a key driver of inflammation and tissue injury in fungal keratitis.

Apoptosis, a fundamental PCD mechanism, involves caspase cascades where initiator caspases (e.g., 8, 9, 10) activate effector caspases (e.g., 3, 6, 7), leading to cellular proteolysis.^22,23 The extrinsic pathway is initiated by death ligands (e.g., FasL, TNF-α) binding to receptors, activating caspases-8 and -10.²⁴ In contrast, the intrinsic pathway responds to cellular stress or DNA damage, with BCL2 family members Bax/Bak permeabilizing mitochondria to release apoptogenic factors, activating caspase-9.^25–27 In F. solani–infected corneas, we detected elevated cleaved caspases-3, -7, and -8 and TUNEL-positive cells, localized to the stroma. These data implicate the extrinsic apoptotic pathway in fungal keratitis pathogenesis.

Necroptosis, also known as programmed necrosis, is a pathway activated by ligand-induced aggregation of death receptors, leading to RIPK1/RIPK3 activation and MLKL phosphorylation. Phosphorylated MLKL migrates to plasma membrane, forming pore-like complexes that compromise the integrity of cell membrane.²⁸ In F. solani–infected corneas, we observed elevated p-MLKL expression, primarily localized to epithelial and stromal layers of ulcerated regions. These findings indicate that necroptosis contributes to fungal keratitis pathogenesis.

The observed elevation of pro-inflammatory cytokines in F. solani–infected corneas underscores a robust host immune response. These cytokines mediate leukocyte recruitment and initiate inflammatory cascades essential for fungal clearance. However, persistent inflammation contributes to corneal damage—a hallmark of fungal keratitis—and exacerbates disease progression. RNA-seq analysis indicated that ZBP1, AIM2, RIPK1, and caspase-8 are important components of the panoptosome in F. solani–infected corneas. Reactome pathway analysis identified neutrophil degranulation and ECM remodeling as critical processes in the infected microenvironment. Neutrophils, although vital for pathogen elimination, release proteolytic enzymes and reactive oxygen species that may induce collateral tissue injury. Dysregulated ECM remodeling, characterized by collagen disorganization and fibroblast proliferation, not only delays wound healing but also perpetuates scarring. Immune profiling revealed a macrophage subset shift in infected corneas, with M1 macrophages dominating over M2 counterparts. This pro-inflammatory bias corresponds to their role in phagocytosis and cytokine secretion. Conversely, healthy corneas maintained higher Treg cell frequencies, which likely suppress excessive inflammation to preserve tissue homeostasis. The augmented activated CD8⁺ T-cell infiltration in infected tissues suggests a heightened adaptive immune response, although their precise contribution to pathogenesis remains unclear. Collectively, these findings highlight the dual role of immunity in fungal keratitis.

PANoptosis is a multi-mechanistic coordinated defense of the body in response to complex stress. Pyroptosis and necroptosis destroy the membrane rapidly and release potent inflammatory mediators. These signals recruit natural immune cells such as neutrophils and macrophages to eliminate pathogens or damaged tissues, thereby initiating defense in the early stages and avoiding the spread of pathogens.²⁹ In contrast, apoptosis is a highly regulated and non-inflammatory process in which the cell membrane retains its integrity. The initiation of the apoptotic pathway can further clear damaged cells in infection or stress to avoid sustained inflammatory stimulation and ensures normal development and cellular homeostasis.^22,30 This integration not only ensures the efficient defense of the body against complex stress but also avoids the limitations such as excessive inflammation or insufficient defense of a single mode of death.

Although our study focused on F. solani, PANoptosis mechanisms vary across fungal pathogens. C. albicans and A. fumigatus induce PANoptosis (integrating pyroptosis, apoptosis, and necroptosis) via ZBP1, with the ZBP1 Zα2 domain being critical for this activation.³¹ However, their triggers differ, as C. albicans relies on secreted toxins and cell wall components, whereas the A. fumigatus ZBP1 ligand remains unknown, possibly linked to its unique cell wall polysaccharides.^32,33 These differences in virulence factors and sensor engagement may underlie their divergent pathogenicity in keratitis.

DON, a Fusarium-derived mycotoxin, has been shown to cause oxidative stress, DNA damage, and cellular signaling perturbations. In the current study, DON treatment of corneal tissues induced upregulation of PANoptosis markers. These findings suggest that DON not only contributes to cellular injury through activation of multiple PCD pathways but also exacerbates inflammation, which may further promote tissue damage in fungal keratitis. Accumulating evidence indicates that corneal pathologies are influenced by epigenetic modifications.³⁴ These reversible, heritable changes in gene expression—independent of DNA sequence—regulate cellular differentiation and stress responses.³⁵ In the current study, RNA-seq analysis of DON-treated HCE-T cells identified significant transcriptional changes, with upregulated genes enriched in RNA metabolism pathways (rRNA modification, tRNA processing, ribosomal biogenesis). These results suggest that DON disrupts epigenetic regulation of RNA biosynthesis, thereby impairing protein translation and augmenting its cytotoxic effects.

Following corneal injury, disruption of the epithelial basement membrane initiates phenotypic reprogramming of corneal and stromal cells. Early apoptotic responses in stromal cells are followed by proliferative repair and fibroblast differentiation, processes tightly regulated by ECM remodeling. However, DON exposure downregulated genes involved in ECM organization, collagen biosynthesis, and integrin signaling in HCE-T cells. This impairment of matrix remodeling likely disrupts epithelial–stromal communication, thereby delaying wound healing and promoting pathological remodeling. The resulting inability of stromal cells to adapt to the altered microenvironment may contribute to delayed healing, corneal melting, or perforation—a critical complication of fungal keratitis.

This study has several limitations. First, although it identified the presence of PANoptosis in F. solani keratitis and used transcriptome sequencing to explore potential mechanisms, the sequencing results were not further validated, limiting the confirmation of key genes involved. Second, findings from in vitro HCE-T cell models require validation in primary human corneal cells or clinical samples. Future studies should integrate patient-derived samples to corroborate these observations.

Conclusions

This research provides valuable perspective to the pathogenesis of F. solani keratitis and highlights the role of PANoptosis in the host's inflammatory response. We identified significant activation of apoptosis, pyroptosis, and necroptosis pathways in response to the infection, contributing to tissue damage and inflammation. Additionally, the study explored the toxic effects of DON, revealing its potential to exacerbate corneal injury by disrupting RNA biosynthesis, impairing matrix remodeling, and altering cell communication between corneal epithelial and stromal cells. These findings underscore the complicated interactions among fungal infection, PCD, and inflammation in fungal keratitis.