Proteomic profiling reveals immunomodulatory role of IL-33 in ocular bacterial and fungal infections

Abstract

Bacterial and fungal pathogens are major causes of infectious endophthalmitis following eye surgery or trauma, often leading to vision impairment or blindness. The distinct clinical outcomes observed in bacterial and fungal endophthalmitis suggest differences in host immune responses. To investigate these differences, we utilized cytokine arrays and murine models of bacterial (Staphylococcus aureus) and fungal (Aspergillus fumigatus) endophthalmitis. Our analysis revealed that cytokine responses peaked in bacterial infections at 12-24 h, whereas fungal infections exhibited a delayed peak at 48 h. Several inflammatory mediators, including MMP9, MMP3, CD14, LIX, LCN2, retinol-binding protein 4, ICAM1, and VCAM1, were differentially elevated. Notably, interleukin-33 (IL-33) levels peaked early in bacterial infections but continued to rise throughout all time points in fungal endophthalmitis. Analysis of patient vitreous samples further confirmed higher levels of IL-33 in bacterial (n=40) and fungal (n=20) endophthalmitis cases. Functional studies in IL-33-deficient mice revealed an increased fungal burden and elevated TNF-α and IL-6 levels, but bacterial endophthalmitis severity remains largely unaffected. Additionally, bone marrow-derived macrophages from IL-33^-/- mice exhibited increased cell death in response to fungal and bacterial infection. Our findings reveal divergent innate immune responses between bacterial and fungal endophthalmitis and emphasize the immunomodulatory function of IL-33 in ocular infections.

Keywords: IL-33; cell death; endophthalmitis; inflammation; proteomic profiler; retina.

Fig 1

Temporal proteome profiling of whole-eye lysate during S. aureus endophthalmitis. Endophthalmitis was induced in the eyes (n = 4) of B6 mice by intravitreal inoculation of S. aureus (SA) RN6390 (5,000 CFUs/eye). After 12, 24, and 48 h post-infection, whole eyes were enucleated for proteome profiler analysis, and PBS-injected mice were used as a control (C). (A) Schematic representation of the experimental design. (B) Venn diagram illustrating the overlap and distinct sets of cytokines differentially expressed at the indicated time points. (C) Heatmap depicting the expression patterns of proteins across the time points. (D) Representative temporal profiles of key inflammatory mediators. The data represented are the culmination of two independent experiments and are shown as means ± SD. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison tests by comparing infected samples: (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 2

Temporal proteome profiling of retinal lysate during S. aureus endophthalmitis. Eyes (n = 4). Endophthalmitis was induced in the eyes of B6 mice by intravitreal inoculation of S. aureus (SA) RN6390 (5,000 CFUs/eye). After 12, 24, and 48 h post-infection, retinal tissues were harvested for proteome profiler analysis, and PBS-injected mice were used as a control (C). (A) Schematic representation of the experimental design. (B) Venn diagram illustrating the overlap and distinct sets of cytokines differentially expressed at the indicated time points. (C) Heatmap depicting the expression patterns of proteins across the time points. (D) Representative temporal profiles of key inflammatory mediators. The data represented are the culmination of two independent experiments and are shown as means ± SD. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison tests by comparing infected samples: (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 3

Temporal proteome profiling of retinal lysate during A. fumigatus endophthalmitis. Eyes (n = 4). Endophthalmitis was induced in the eyes of B6 mice by intravitreal inoculation of A. fumigatus (15,000 CFUs/eye). After 12, 24 h, and 48 h post-infection, retinal tissues were harvested for proteome profiler analysis, and PBS-injected mice were used as a control (C). (A) Schematic representation of the experimental design. (B) Venn diagram illustrating the overlap and distinct sets of cytokines differentially expressed at the indicated time points. (C) Heatmap depicting the expression patterns of proteins across the time points. (D) Representative temporal profiles of key inflammatory mediators. The data represented are the culmination of two independent experiments and are shown as means ± SD. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison tests by comparing infected samples: (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 4

Comparative analysis of proteins in bacterial vs fungal endophthalmitis. The protein profiles of S. aureus (red line) and A. fumigatus (blue line)-infected retinal lysates at 12, 24, and 48 h post-infection were compared head-to-head to visualize the relative expression. The data represented are the culmination of two independent experiments and are shown as means.

Fig 5

Validation of protein profiles during experimental bacterial and fungal endophthalmitis. Eyes (n = 4) of C57BL/6 mice were intravitreally injected with S. aureus (5,000 CFUs/eye) or A. fumigatus (15,000 CFUs/eye). Retinas were harvested at indicated time points post-infection, lysed with sonication in RIPA buffer with protease inhibitor, and subjected to western blot analysis using β-actin as the endogenous control. Western blot detection of MMP-9, CHI3L1, VCAM-1, and ICAM-1 during S. aureus (A) and A. fumigatus (C) infections. Densitometric analysis (B and D) was performed using ImageJ software, and the results are expressed as relative fold changes normalized to the respective loading control, β-actin. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison tests by comparing infected samples: ns, not significant; (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 6

Assessment of IL-33 levels in patients with bacterial and fungal endophthalmitis. (A–C) Representative clinical images from patients with endophthalmitis with confirmed bacterial (gram-positive and gram-negative) or fungal endophthalmitis. (D) Vitreous samples (25 µL) from patients with culture-positive bacterial (n = 40, [gram positive and gram negative, 20 each]) and fungal (n = 20) endophthalmitis, along with HCs (n = 20), were used for the detection of IL-33 using ELISA. Statistical analysis was performed using the one-way ANOVA by comparing control with endophthalmitis samples: (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001

Fig 7

Effect of IL-33 deficiency on the pathogenesis of bacterial and fungal endophthalmitis. Endophthalmitis was induced in B6 WT and IL-33^−/− mice (B6 background) by intravitreal inoculation of S. aureus (SA) or A. fumigatus (AF) (n = 4 per condition). (A and D) Representative slit-lamp micrograph showing corneal haze/opacity at the indicated time point post-infection. Quantitation of intraocular bacterial (B) and fungal (E) burden in whole-eye lysates by serial dilution and plate counting method. (C and F) ELISA of indicated inflammatory cytokines. The data represented are the culmination of two independent experiments and are shown as means ± SD. Statistical analysis was performed using one-way ANOVA with Tukey’s multiple comparison tests by comparing infected samples: (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 8

String analysis and the role of IL-33 on cell survival during fungal infection. (A) PPI network of IL-33 was constructed using the STRING database. (B) GO enrichment analysis of the STRING-derived PPI network, demonstrating IL-33’s role in regulating cell survival. Real-time analysis of cell death in BMDMs was performed using SYTOX Green dye and the Incucyte SX5 live-cell imaging system following infection with A. fumigatus (C) and S. aureus (E). LDH release was estimated from infected BMDMs following A. fumigatus (D) and S. aureus (F) infection. Data are represented as mean ± SD. Statistical analysis was performed using one-way ANOVA: (∗) P < 0.05; (∗∗) P < 0.01; (∗∗∗) P < 0.001; (∗∗∗∗) P < 0.0001.

Fig 9

Schematic of study design. (A) Proteomic analysis was performed in experimental models of S. aureus or A. fumigatus endophthalmitis using whole-eye or retinal tissue lysates. (B) Biological and functional categorization of altered proteins, including IL-33, during endophthalmitis. (C) Validation of IL-33 levels in vitreous samples from bacterial or fungal endophthalmitis patients. (D) Functional role of IL-33 in KO mice.