Abstract
The aim of this study was to elucidate the expression and functions of IL-24 in C57BL/6 mouse corneas in response to Pseudomonas aeruginosa infection. Among IL-20R cytokines, only IL-24 was induced at both mRNA and protein levels by infection at early time points. The upregulation of IL-24 was dampened by flagellin pretreatment, which protects the corneas from microbial infection. Time course studies revealed bimodal early and later peaks of IL-24 expression, a pattern shared with suppressor of cytokine signaling (SOCS)3 but not IL-1β or IL-6. Silencing of IL-24 enhanced S100A8/A9 expression and suppressed SOCS3, IL-1β, IL-1RN, and matrix metalloproteinase 13 expression at 6 h postinfection. Downregulation of the IL-24 signaling pathway significantly reduced the severity of keratitis, whereas rIL-24 exacerbated P. aeruginosa–mediated tissue destruction. In vitro, rIL-1β induced the expression of SOCS3, IL-24, IL-1β, and IL-6 in primary cultured human corneal epithelial cells. rIL-24, alternatively, stimulated the expression of SOCS3, but not the others. In conclusion, IL-24 promotes P. aeruginosa keratitis through the suppression of early protective mucosal immunity, culminating in increased severity of P. aeruginosa keratitis.
Introduction
Microbial keratitis is a sight-threatening disease that poses a substantial burden to affected individuals. It is reported that corneal infection is a major cause of unilateral blindness around the world (1). Pseudomonas aeruginosa is the predominant bacterium isolated in contact lens–related corneal infections (2). The disease is usually manifested by severe inflammation, formation of corneal ulcers, and hypopyon (3). Both the pathogen and the host immune system contribute to the destruction of the cornea (4). Unrestrained inflammation generated by the host immunity can cause scar tissue formation and vascularization of the cornea. If not properly treated, patients eventually suffer from permanent vision loss.
The corneal epithelial layer forms the first line of defense against pathogens. The discovery of TLRs has elicited keen interest in the innate defensive functions of mucosal epithelial cells (5, 6). Our previous studies showed that mouse corneas pretreated with flagellin, which signals through TLR5, are protective against P. aeruginosa infection (7–9). Flagellin pretreatment has also been demonstrated to be protective in localized infections of the lung (10, 11) and gut (12), radiation pneumonitis (13), and systematically against biological or physical insults (14). We found that this flagellin-induced protection is due to the reprograming of gene expression in epithelial cells to enhance the innate defense against invasive pathogens. To better understand the underlying mechanism, we performed a genome-wide cDNA microarray to assess the differentially expressed genes in corneal epithelial cells (CECs) after P. aeruginosa challenge, with or without flagellin pretreatment (15). Among the genes differentially regulated by flagellin pretreatment, the expression of IL-24 and its downstream negative regulator, suppressor of cytokine signaling 3 (SOCS)3, was notably upregulated in response to infection and significantly suppressed by flagellin pretreatment.
IL-24, along with IL-19, IL-20, IL-22, and IL-26, belongs to the IL-20 subfamily of the IL-10 superfamily of cytokines (16). IL-19, IL-20, and IL-24 (named as IL-20R cytokines here) share the IL-20R1 and IL-20R2 heterodimer receptor. IL-20 and IL-24 can also bind to the IL-22R1 and IL-20R2 receptor complex (17). After binding to their receptors, IL-20R cytokines signal through the JAK/STAT3 pathway to regulate downstream gene expression (18, 19). The SOCS family of proteins, including eight members, plays an important role in the negative regulation of the cytokine–JAK/STAT signaling pathway (20). Whereas SOCS1 inhibits STAT1 activation in the IFN-γ signaling cascade (21), SOCS3 is a major negative regulator of the IL-6–STAT3 signaling pathway (22). SOCS3 binds to both JAK and cytokine receptors and targets them for ubiquitination and proteasomal degradation (23). SOCS3 can also directly inhibit JAK activity (24). Increased expression of IL-24 and SOCS3 has been found within inflamed tissues in various diseases, such as skin inflammation (e.g., psoriasis), inflammatory bowel disease, and rheumatoid arthritis (25–27). IL-24 exerts protective effects against intracellular pathogens, such as Salmonella typhimurium and Mycobacterium tuberculosis (28, 29). Studies in different mouse models have proven the critical importance of SOCS3 in restraining inflammation and promoting optimal levels of protective immunity against the infection (30). Alternatively, some pathogens have evolved to modify host SOCS3 expression to evade the immune response. For example, Epstein–Barr virus and herpes simplex virus can stimulate SOCS3 expression to suppress type I IFN production, thereby subverting the host immune response (31, 32). To date, the functions and mechanism of IL-24 in bacterial keratitis have not been explored.
Given the reported importance of IL-24 in inflammation and host protection against infection, we hypothesized that IL-24 would enhance the host defense against P. aeruginosa keratitis. Surprisingly, we found that IL-24 increased the susceptibility of mouse cornea to P. aeruginosa infection. Our results indicate that P. aeruginosa induces the early expression of IL-24, resulting in a suppression of early protective mucosal immune response.
Materials and Methods
Animals
Wild-type C57BL/6 (B6) mice (8 wk old; female) were purchased from The Jackson Laboratory (Bar Harbor, ME). All animal procedures were performed in compliance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Visual Research and were approved by the Institutional Animal Care and Use Committee of Wayne State University.
Mouse model of P. aeruginosa keratitis
Mice were anesthetized with an i.p. injection of ketamine (90 mg/kg) and xylazine (10 mg/kg) before surgical procedures. Mouse corneas were scratched gently with a sterile 26-gauge needle to create three 1-mm incisions to break the epithelial barrier. Purified flagellin (500 ng in 5 μl of PBS) was applied topically to the injured cornea as an eye drop. PBS was used as control. Twenty-four hours later, the corneas were scratched again and inoculated with 1.0 × 104 CFU of ATCC 19660 in 5 μl of PBS.
Administration of small interfering RNA, recombinant protein, or neutralizing Ab
All of the small interfering RNAs (siRNAs) used in this study were SMARTpool (a mixture of four siRNAs) ON-TARGETplus siRNAs designed by GE Dharmacon (Lafayette, CO). Mice were subconjunctivally injected twice with siRNA targeting to a specific gene (50 pmol in 5 μl of RNase-free water) during 2 d. Six hours after the second siRNA injection, mouse corneas were inoculated with P. aeruginosaP. aeruginosa on the corneas.
Isolation of mouse CECs
A razor blade was tailored to ∼5 mm wide in the edge and placed in a Castroviejo razor blade breaker and holder. Mice were euthanized by cervical dislocation. Under the microscope, CECs were surgically scraped off from the basement membrane. Cells were collected to the razor blade from the basement membrane. Liquid nitrogen was used to snap freeze the cells and cool off the tip of a sharp surgical scalper at the same time. Cells were immediately transferred into precooled 1.5-ml Eppendorf tubes placed on dry ice by scraping the razor blade with the scalper. Cells were processed immediately for RNA isolation or protein extraction, or they were stored at −80°C for later use.
Clinical examination, quantification of P. aeruginosa CFU, and determination of myeloperoxidase units
Corneas were photographed daily up to 3 d postinfection (dpi) for the assessment of infection severity. Clinical scores were assigned to the infected corneas in a blinded fashion according to the scale previously reported (33). Whole corneas were excised and placed in 200 μl of sterile PBS. Tissue was homogenized with a TissueLyser II (Qiagen, Valencia, CA). The homogenates were divided into two parts. The first fraction (50 μl) was subjected to serial log dilutions for the assessment of viable bacterial number. The remaining homogenates were further lysed for myeloperoxidase (MPO) measurement. MPO units were determined according to a previously reported method (34). One MPO unit corresponds to 2.0 × 105 polymorphonuclear leukocytes (PMNs).
Cell culture
Primary human CECs were obtained from diseased human corneal samples. CECs were dislodged from underlying basement membrane by dispase and then digested by trypsin. Cells were pelleted, resuspended, and grown in defined keratinocyte serum-free medium (Life Technologies, Waltham, MA) with supplements and antibiotics.
Semiquantitative and quantitative PCR
The primers used in this study are listed in Table I. Total RNA was extracted with an RNeasy mini kit (Qiagen) following the manufacturer’s instructions. For semiquantitative PCR, cDNA was amplified with TaqMan technology (Promega, Madison, WI). PCR products were subjected to electrophoresis on 2% agarose gels containing ethidium bromide. For quantitative PCR (qPCR), cDNA was amplified using a StepOnePlus real-time PCR system (Applied Biosystems, University Park, IL) with SYBR Green PCR master mix (Applied Biosystems). Data were analyzed by using the ΔΔ cycle threshold method with β-actin or GAPDH as the internal control.
ELISA
Western blot
705) (9145; Cell Signaling Technology), and anti–β-actin (A1978; Sigma-Aldrich).
Statistical analysis
A nonparametric Mann–Whitney U test was used to compare the clinical scores. A Student t test or one-way ANOVA was used to compare quantitative means. A p value <0.05 was considered to be significant.
Results
P. aeruginosa infection induces IL-24 and SOCS3 expression in mouse CECs
To verify our previously reported cDNA microarray results (15), we infected mouse corneas with 1.0 × 104 CFU of P. aeruginosa strain ATCC 19660. CECs were collected at 6 h postinfection (hpi) and subjected to PCR analysis. As shown in Fig. 1A, flagellin challenge alone minimally induced IL-24 expression relative to the naive cornea. P. aeruginosa infection increased the expression of IL-24, and this elevation of IL-24 expression was dampened by flagellin pretreatment in B6 mouse CECs. As IL-24 shares receptors with IL-19 and IL-20, we explored the induction of these cytokines in response to infection. Whereas low levels of IL-19 were detected in all samples, no IL-20 mRNA was detected. The infection-induced expression of SOCS3 was also dampened by the flagellin pretreatment (Fig. 1B). The expression of IL-1β, an innate immune and major inflammatory mediator, significantly increased after P. aeruginosa infection, and it was attenuated by pretreatment with flagellin (Fig. 1B, Table I).
P. aeruginosa infection induces IL-24 and SOCS3 expression in B6 mouse CECs; flagellin pretreatment attenuates their expression. Mouse corneas were scratched with a needle and topically pretreated with 500 ng of flagellin for 24 h. PBS was used as control. Then the corneas were scratched again and inoculated with P. aeruginosa (ATCC 19660, 1.0 × 104 CFU). Corneal epithelia were collected at 6 hpi for semiquantitative RT-PCR (A) or quantitative real-time PCR (B) analysis of IL-24, IL-19, and IL-20 (A) and SOCS3 and IL-1β (B). Results are presented relative to those of naive corneas, set as 1. Data are representative of three independent experiments (B, mean ± SEM). The p values were generated by one-way ANOVA. ***p < 0.001. Flag, flagellin; PA, P. aeruginosa.
IL-24, SOCS3, and IL-1β are early responsive genes during P. aeruginosa infection
Having shown the effects of infection and flagellin pretreatment on cytokine expression, we then performed a time course study to compare the expression of IL-24 with IL-1β, SOCS3, and IL-6 in response to P. aeruginosa infection. Two peaks of IL-24 upregulation were observed: a rapid and marked increase at 3 hpi, decreasing by 6 hpi, and then a gradual increase to a higher level at 18 hpi. In contrast, IL-1β expression was induced at 3 hpi and continued to increase as the infection progressed (Fig. 2A). The expression of SOCS3 followed the observed bimodal expression kinetics of IL-24. IL-6 has been shown to be critical to the host defense of the cornea, including neutrophil infiltration and bacterial clearance, during P. aeruginosa infection (35) as well as in other systems (36). IL-6 expression was first observed at 9 hpi and reached a significantly higher level at 18 hpi, as compared with the naive cornea (Fig. 2A). Hence, the infection-induced expression of SOCS3 at 3 hpi is correlated with that of IL-24, but not IL-6, which is also known to activate the JAK/STAT3 pathway.
IL-24, SOCS3, and IL-1β are early responsive genes after P. aeruginosa infection in B6 mouse cornea. Mouse corneas were scratched with a needle and inoculated with P. aeruginosa. Whole corneas were collected at 3, 6, 9, and 18 hpi for qPCR analysis of IL-24, IL-1β, SOCS3, and IL-6 (A). Results are presented relative to those of naive corneas, set as 1. (B) Immunoblot analysis of IL-24, SOCS3, p-STAT3, and STAT3 in cell lysates of whole corneas infected with P. aeruginosa for 3, 6, 9, and 18 h. β-Actin serves as the loading control. (C) Quantification of protein levels based on the densitometry of the Western blots in (B). (D) ELISA assays of IL-1β and IL-6 in cell lysates of whole corneas infected with P. aeruginosa for 9 or 18 h. Data are representative of three independent experiments (A, C, and D, mean ± SEM). *p < 0.05, **p < 0.01, ***p < 0.001 (ANOVA), compared with the control. ND, not detected; PA, P. aeruginosa.
The expression of some of these cytokines and the activation of STAT3 were also assessed by using Western blotting or ELISA in whole corneal extracts (Fig. 2B–D). IL-24 expression at the protein level was markedly increased at 3 hpi and reduced at 6 hpi and thereafter. The activation of STAT3 detected by the phospho-STAT3 level followed a similar expression pattern with that of IL-24. SOCS3 protein expression increased at 3 and 6 hpi, down by 9 hpi, followed by an increase at 18 hpi (Fig. 2B, 2C). The protein levels of IL-1β and IL-6 increased in a time-dependent fashion, with peak levels observed at 18 hpi (Fig. 2D).
Silencing of IL-24 improves the outcome of P. aeruginosa keratitis in mouse cornea
To determine whether IL-24 influences the host defense against P. aeruginosa keratitis, we subconjunctivally injected B6 mice with IL-24 siRNA or control siRNA. Western blotting analysis of the whole corneas at 1 dpi revealed an effective knockdown of IL-24 by siRNA, as well as the downregulation of STAT3 activation (Fig. 3A, 3B).
Silencing of IL-24 attenuates the severity of P. aeruginosa infection in mouse cornea. Mice were subconjunctivally injected with IL-24 siRNA (50 pmol in 5 μl of RNase-free water) twice in 2 d. Corneas were inoculated with P. aeruginosa 6 h after the second injection. Nonspecific siRNA serves as control. (A) Immunoblot analysis of IL-24, p-STAT3, and STAT3 in cell lysates of IL-24 siRNA– or control siRNA–treated corneas at 1 dpi. β-Actin serves as the loading control. (B) Quantification of protein levels based on the densitometry of the Western blots in (A). (C) Mice were monitored and photographed daily up to 3 dpi. (D) Clinical scores were assigned to each cornea daily and plotted as median plus interquartile range. At 3 dpi, the corneas were excised and subjected to bacterial plate counting (E) and MPO assay (F). Data are representative of three independent experiments with five mice per group (B, E, and F, mean ± SEM). The p values were generated by a Mann–Whitney test (D) or an unpaired Student t test (B, E, and F). *p < 0.05, **p < 0.01, ***p < 0.001. PA, P. aeruginosa.
The progression of P. aeruginosa keratitis in the control siRNA– and IL-24 siRNA–treated corneas was monitored for up to 3 dpi. Micrographs taken daily showed that knockdown of IL-24 revealed reduction in the severity of keratitis, as compared with the control group (Fig. 3C). At 3 dpi, the control corneas developed severe keratitis, thick opacity covering the entire corneal surface, and surface irregularity, whereas the corneas injected with IL-24 siRNA had less severe keratitis with lighter opacity located at the center of the cornea. The severity of keratitis was quantitated using a 12-scale clinical scoring system by evaluating the area and density of opacity and surface irregularity (33). The clinical scores assigned to IL-24–silenced mice were significantly lower than those of their control counterparts at all 3 d examined (Fig. 3D). Mice were euthanized at 3 dpi, and whole corneas were subjected to bacterial counting and MPO measurement to evaluate neutrophil infiltration. Silencing of IL-24 resulted in lower bacterial burden and reduced influx of neutrophils compared with the control corneas (Fig. 3E, 3F).
We next treated the corneas with IL-24 siRNA and assessed its effects on the expression of several innate immune-responsive genes at 6 hpi in mouse CECs in vivo. S100A8/A9 and CXCL10 are antimicrobial peptides and chemotactic proteins. The heterodimer of S100A8 and S100A9 (also referred to as calprotectin) has been shown to prevent the growth of bacteria, fungi, and protozoa by chelating Mn2+ and Zn2+ ions (37–40). CXCL10, also known as IFN-γ–inducible protein 10, can act as a chemoattractant for NK cells and an important antimicrobial peptide capable of directly killing P. aeruginosa and Candida albicans (8, 41, 42). As shown in Fig. 4A, the expression of the antimicrobial peptide S100A8 increased 3-fold (p < 0.05, ANOVA) after IL-24 knockdown in uninfected CECs. P. aeruginosa infection induced the expression of both S100A8 and S100A9, and silencing of IL-24 further elevated their expression in the infected corneas. The expression of CXCL10 was also induced by P. aeruginosa infection and further increased after silencing of IL-24. The expression of SOCS3 was upregulated after P. aeruginosa infection. IL-24 knockdown resulted in the downregulation of SOCS3 expression. IL-1 receptor antagonist (gene IL1rn), the natural inhibitor of IL-1–type cytokines, has been demonstrated to ameliorate symptoms in several autoimmune diseases (43, 44). Matrix metalloproteinase (MMP)13 is reported to contribute to basement membrane damage and facilitate P. aeruginosa invasion into the stroma (45). The expression of IL-1β, IL-1RN, and MMP13 was also induced by P. aeruginosa infection. IL-24 knockdown dampened their expression to a level similar to the control siRNA without infection in mouse CECs. The expression of calprotectin and CXCL10 was further confirmed in protein level (Fig. 4B). Basal expression of both calprotectin and CXCL10 was detected in the uninfected corneas. Consistent with the mRNA expression, the protein expression of calprotectin and CXCL10 was induced by P. aeruginosa infection, and further enhanced when knocking down IL-24.
Silencing of IL-24 upregulates the expression of calprotectin and CXCL10 in mouse CECs in response to P. aeruginosa infection. Mouse corneas were treated with IL-24 siRNA or control siRNA and inoculated with P. aeruginosa as in Fig. 3. Corneal epithelial samples were collected at 6 hpi for qPCR (A) or ELISA (B) analysis. For qPCR, results are presented relative to those of control siRNA–treated, uninfected corneas, set as 1. Data are representative of three independent experiments with at least four corneas per group (mean ± SEM). *p < 0.05, **p < 0.01, ***p < 0.001 (ANOVA). PA, P. aeruginosa.
As an early responsive cytokine, IL-24 may induce the expression of other cytokines and/or antimicrobial peptides in infected corneas. To that end, we used real-time PCR to assess the expression of eight different genes at 1 dpi in the whole corneas (Fig. 5A). Whereas most genes were unaffected by the downregulation of IL-24, the expression of IL-1RN (7.2-fold increase, p < 0.05), S100A9 (3.3-fold increase, p < 0.05), and IL-17A (0.6-fold decrease, p > 0.05) were altered by IL-24 silencing in noninfected corneas. IL-17A is reported to be secreted by innate γδ T cells to recruit neutrophils (46). Activated neutrophils further produce IL-17A in an autocrine fashion, contributing to bacterial and fungal clearance (47, 48). Infection induced marked upregulation of SOCS3, IL-6, MMP13, IL-17A, CXCL10, IL-1β/IL-1RN, antimicrobial peptide S100A8, and S100A9. Knocking down IL-24 resulted in the downregulation of the SOCS3 (1.47-fold), IL-6 (4-fold), MMP13 (1.73-fold), IL-17A (3.3-fold), CXCL10 (5.27-fold), IL-1β (3.0-fold), IL-1RN (5.72-fold), S100A8 (1.98-fold), and S100A9 (1.45-fold) in infected corneas (Fig. 5A). The expression of IL-6, IL-1β, IL-1Ra (protein encoded by gene IL-1RN), and calprotectin was confirmed at the protein level by ELISA analyses (Fig. 5B). Whereas the basal levels of IL-6 and IL-1β were undetected, basal expression of IL-1RA and S100A8/A9 was observed in the control siRNA– or IL-24 siRNA–treated, noninfected corneas. Similar to the mRNA level, IL-6, IL-1β, IL-1RA, and S100A8/9 were elevated at the protein level by P. aeruginosa infection and significantly attenuated by IL-24 silencing. The downregulation of the cytokines and antimicrobial peptides after IL-24 silencing at 1 dpi in the cornea may be related to the diminished host response to much reduced bacterial burden in the silencing group.
Silencing of IL-24 dampens the expression of cytokines and antimicrobial peptides in mouse cornea in response to P. aeruginosa infection. Mouse corneas were treated with IL-24 siRNA or control siRNA and inoculated with P. aeruginosa as in Fig. 3. (A) Whole cornea samples were collected at 1 dpi for qPCR analysis. Results are presented relative to those of control siRNA–treated, uninfected corneas, set as 1. (B) Cell lysates from whole corneal samples at 1 dpi were subjected to ELISA analysis. Data are representative of three independent experiments with at least four corneas per group (mean ± SEM). *p < 0.05, **p < 0.01, ***p < 0.001 (ANOVA or unpaired Student t test). ND, not detected; PA, P. aeruginosa.
Blockage of IL-20R2 decreases the severity of P. aeruginosa keratitis in mouse cornea
Previous reports based on PCR studies demonstrated that the receptors for IL-20R cytokines are mainly expressed in epithelial cells, but not in immune cells (49, 50). We showed that whereas equal levels of IL-22R1 were detected in both the naive and infected CECs at ∼95 kDa, the abundance of IL-20R2 proteins was decreased in the infected CECs, compared with the naive control (Supplemental Fig. 1). We next assessed the effect of blocking IL-20R2, the common receptor chain for IL-19, IL-20, and IL-24, to further verify the effect of the IL-24 signaling pathway on P. aeruginosa keratitis (Fig. 6). Whereas the IgG control corneas developed keratitis gradually within 3 d, the anti–IL-20R2 Ab-treated corneas had slight corneal opacification at 1 dpi, which progressed to cloudy corneas at day 2, and resolved to cloudiness with the outline of the iris and pupil discernable at day 3 (Fig. 6A). Clinical scoring revealed that the anti–IL-20R2 group started with a lower score at day 1 and a slope similar to that observed in the IgG control group at day 2, with scores slightly declining at day 3 (Fig. 6B). The blockage of IL-20R2 reduced the bacterial burden and MPO units in the corneas at 3 dpi as well (Fig. 6C, 6D).
IL-20R2 neutralizing Ab decreases the severity of P. aeruginosa keratitis in mouse cornea. Mice were subconjunctivally injected with IL-20R2 neutralizing Ab (200 ng/5 μl) 4 h before the inoculation with P. aeruginosa. Mouse IgG serves as the control. (A) Mice were monitored and photographed daily up to 3 dpi. (B) Clinical scores were assigned to each cornea daily and plotted as median plus interquartile range. At 3 dpi, the corneas were excised and subjected to bacterial plate counting (C) and MPO assay (D). Data are representative of three independent experiments with five mice per group (C and D, mean ± SEM). The p values were generated by a Mann–Whitney test (B) or unpaired Student t test (C and D). *p < 0.05, **p < 0.01, ***p < 0.001. PA, P. aeruginosa.
rIL-24 increases the susceptibility of mouse cornea to P. aeruginosa infection
As a complementary approach, we assessed the effects of rIL-24 on the severity of P. aeruginosa keratitis. Subconjunctival injection of recombinant mouse IL-24 at 4 h prior to P. aeruginosa inoculation resulted in more severe keratitis than that of the control with 0.1% BSA injection (Fig. 7A). Recombinant mouse IL-24 was reconstituted with 0.1% BSA in PBS according to the manufacturer’s recommendation. BSA, 0.1% in PBS, was used as a vehicle control. The control corneas were covered with opacity by ∼30% at day 1, ∼80% at day 2, and fully covered at day 3. Surface irregularity was also apparent at day 3. In rIL-24–treated corneas, there was a sign of central melting at day 1, central ring with heavy opacity at day 2, and heavy opacity covered the entire cornea with corneal ulceration and neovascularization present at day 3, pathologies usually seen in the control corneas at day 5 (Fig. 7A). The clinical scores revealed significantly higher disease severity in the rIL-24–treated group than that observed in the BSA control group (Fig. 7B). A single dose of rIL-24 resulted in significantly higher bacterial burden and neutrophil infiltration as compared with the BSA control group at 3 dpi (Fig. 7C, 7D).
rIL-24 increases the susceptibility of the mouse cornea to P. aeruginosa infection. Mice were subconjunctivally injected with either recombinant mouse IL-24 (200 ng in 5 μl of 0.1% BSA) or 0.1% BSA as control. Mouse corneas were inoculated with P. aeruginosa 4 h after the injection of rIL-24. (A) The infected corneas were photographed daily. (B) The severity of keratitis was assessed with clinical scores. At 3 dpi, the corneas were excised and subjected to bacterial load (C) and MPO unit determination (D). Data are representative of three independent experiments with five mice per group (B, median plus interquartile range; C and D, mean ± SEM). The p values were generated by a Mann–Whitney test (B) or unpaired Student t test (C and D). *p < 0.05, **p < 0.01. PA, P. aeruginosa.
The expression of innate immune response genes was also assessed in rIL-24–treated corneas (Fig. 8). In noninfected corneas, rIL-24 stimulated significant upregulation of IL-17A (4.8-fold increase, p < 0.01), whereas the expression of IL-1RN (2.5-fold decrease, p < 0.01) and S100A8 (0.5-fold decrease, p > 0.05) was downregulated, contrasting to those of IL-24 siRNA–treated corneas. The presence of rIL-24 significantly augmented the infection-induced expression of these genes.
rIL-24 enhances cytokine expression in mouse cornea in response to P. aeruginosa infection. Mouse corneas were treated with recombinant mouse IL-24 or 0.1% BSA and inoculated with P. aeruginosa as in Fig. 7. Whole cornea samples were collected at 1 dpi for qPCR analysis. Results are presented relative to those of 0.1% BSA-treated, uninfected corneas, set as 1. Data are representative of three independent experiments with at least four corneas per group (mean ± SEM). *p < 0.05, **p < 0.01, ***p < 0.001 (ANOVA). PA, P. aeruginosa.
IL-24 specifically induces SOCS3 expression in human CECs
Our in vivo data imply that both IL-1β and IL-24 are the downstream products of TLR signaling, and that these two cytokines may regulate the expression of each other, as well as other genes involved in the innate defense of the corneas. To that end, we first treated primary human CECs with recombinant human IL-1β and found enhanced expression of SOCS3, IL-24, IL-1β, and IL-6 mRNA (Fig. 9A). The level of SOCS3 mRNA was increased 2.2-fold at hour 1 and further elevated (6.1-fold) at 2 h after IL-1β stimulation. IL-1β significantly induced the mRNA expression of IL-24, IL-1β, and IL-6 at both hour 1 and hour 2. Conversely, rIL-24 stimulated robust upregulation of SOCS3, peaking at 1 hpi, but it had no effect on the expression of IL-1β, IL-24, or IL-6 (Fig. 9B).
Effects of recombinant human IL-1β or IL-24 on primary human CECs. (A) Human CECs were treated with recombinant human IL-1β (50 ng/ml). Cells were collected at 1 and 2 h for qPCR analysis of SOCS3, IL-24, IL-1β, and IL-6. (B) Human CECs were treated with recombinant human IL-24 (100 ng/ml). Cells were collected at 1 and 2 h for qPCR analysis of SOCS3, IL-1β, IL-24, and IL-6. Results are presented relative to those of untreated control cells, set as 1 (mean ± SEM). Data are representative of three independent experiments. **p < 0.01, ***p < 0.001 (ANOVA).
Discussion
The present study demonstrated that among the three mouse IL-20R cytokines, only IL-24 was markedly induced by P. aeruginosa infection. Interestingly, SOCS3 shared a similar expression pattern with IL-24, whereas the expression of another STAT3 activator, IL-6, was not increased until 9 hpi. Functional analysis revealed that whereas exogenous IL-24 increased the severity of P. aeruginosa keratitis, downregulation of the IL-24 signaling pathway resulted in reduced clinical scores, less opacity, lower bacterial burden, and reduced infiltration of neutrophils during P. aeruginosa infection. During the initial hours of infection (6 hpi) when most invading pathogens remain at the epithelial layer, the inhibition of IL-24 markedly increased the infection-induced expression of antimicrobial peptides S100A8, S100A9, and CXCL10. At 1 dpi, the expression of the infection response genes (SOCS3, IL-1β, IL-6, IL-17A, MMP13, CXCL10, and S100A8/A9) was all markedly elevated in infected corneas. Induction of these mediators was dampened to different extents by IL-24 knockdown at 1 dpi, correlating with the reduced number of pathogens. Targeting IL-20R2, the common receptor chain in the IL-24 signaling pathway, improved the outcome of P. aeruginosa keratitis. Application of rIL-24, alternatively, markedly increased the severity of P. aeruginosa keratitis and reversed the expression pattern of infection response genes seen in IL-24 downregulated corneas at 1 dpi. In vitro, whereas IL-1β induced the expression of IL-1β (self-amplification), IL-24, IL-6, and SOCS3, treatment with IL-24 induced only the expression of SOCS3. Taken together, our data suggest that the IL-20R cytokines have immunosuppressive properties and contribute to corneal susceptibility to P. aeruginosa infection.
The host responds to tissue infection by eliciting acute inflammation in an attempt to control invading pathogens (51). These early events are coordinated by several families of proinflammatory mediators, including cytokines, chemokines, and lipid mediators. The major proinflammatory cytokines/chemokines elicited by microbial infection include IL-1β, IL-6, IL-8, and IL-17A, which orchestrate leukocyte recruitment to the sites of infection (52, 53). It is now clear that during the acute inflammatory response, an array of molecules, such as IL-10 and resolvins (51) that function in the resolution of inflammation, are also induced. Among the IL-10 superfamily of cytokines, IL-20 subfamily cytokines, including IL-19, IL-20, and IL-24 (referred to as IL-20R cytokines), are known to be produced mainly by myeloid cells and epithelial cells, and to target epithelial cells (54). In this study, we reported that the expression of IL-24, but not IL-19 or IL-20, was induced in mouse corneas by P. aeruginosa infection, a pattern similar to that described for such induction by inflammatory cytokines such as IL-1β and CXCL2 (mouse homolog of human IL-8), but not IL-6. It has been reported that the expression of IL-19 and IL-24, but not of IL-20, was induced by infection with methicillin-resistant Staphylococcus aureus (MRSA) in mouse skin (48). Thus, we conclude that IL-24, similar to IL-1β, is an early response gene in response to P. aeruginosa infection in the cornea.
We observed that IL-24 knockdown decreased the expression of IL-1β, IL-6, and MMP13, but increased the expression of S100A8/A9 and CXCL10 during the initial hours of infection (6 hpi). Additionally, IL-24 alone increased the expression of IL-17A and decreased the expression of IL-1RN, a natural IL-1R antagonist in noninfected corneas (Fig. 8), suggesting a proinflammatory role of IL-24 in B6 mouse corneas. It is unlikely that IL-24 directly induces the expression of these cytokines from epithelial cells, because the treatment of rIL-24 alone to B6 mouse corneas (Fig. 8) or cultured human CECs (Fig. 9) did not alter the expression of IL-1β and MMP13. Importantly, IL-24 was found to inhibit the expression of the antimicrobial peptides S100A8/A9 and CXCL10 in mouse CECs (Fig. 4), resulting in impaired mucosal immunity and reduced bacterial clearance at the corneal surface. We postulate that it is the corresponding increase in bacterial burden that drives the enhanced inflammatory responses at the mucosal surface, including increased in vivo expression of inflammatory cytokines (IL-1β, IL-6, and MMP13). Hence, the effects of IL-24 on the corneal inflammatory response are likely through an indirect action, such as the induction of SOCS3.
Given well-accepted anti-inflammatory and protective properties of the IL-10 family cytokines (28, 55, 56), we thought that IL-24 would enhance the host innate defense against P. aeruginosa in the corneas. Our data, however, showed that whereas downregulation of IL-24 or the Ab blockade of the IL-20R2 protected the corneas from P. aeruginosa infection, exogenous IL-24 worsened P. aeruginosa keratitis, suggesting a detrimental role of IL-24 during P. aeruginosa infection of the corneas. Although most studies to date in the literature suggest a protective role of IL-20 subfamily cytokines, particularly IL-22 (28, 55–57), a more recent study of MRSA infection of mouse skin revealed that IL-19 and IL-24 play an immunosuppressive role by downregulating IL-1β– and IL-17A–dependent pathways, which mediate host innate immune defense against the infectious agent (48). Our results of IL-24 as a negative mediator in P. aeruginosa keratitis are in line with the MRSA skin infection study, yet by entirely different mechanisms. In the MRSA skin infection model, IL-19/IL-24 target IL-1β and IL-17A expression; in our P. aeruginosa keratitis model, IL-24 hampers the expression of innate immune defense molecules, resulting in an increase in bacterial burden, which in turn stimulated stronger inflammatory response. Hence, the early induced IL-24 acts as an innate immune suppressor in the corneas in response to P. aeruginosa infection.
In IL-24–manipulated B6 mouse corneas, the inflammatory response, measured by clinical scoring, the levels of proinflammatory cytokines, and the intensity of PMN infiltration, is directly correlated to the numbers of recoverable bacteria. This seems paradoxical, given that acute inflammation is known to be required for controlling invading pathogens. This paradox can perhaps be resolved by the locations where bacterial clearance may occur at different times of infection. In our mouse model of microbial keratitis, epithelium-scratched wounds are made with minimal injury of the basement membrane, which serves as a barrier for P. aeruginosa invasion of the stroma (45). During the initial hours of infection (6 hpi), fewer than 1000 CFU of P. aeruginosa were detected (1 × 104 inoculated), and most of them were detected within the epithelium. By 24 hpi, the basement membrane was dissolved, and the epithelium partially disappeared with a large number of P. aeruginosa in the stroma (45). We observed robust expression of potent antimicrobial peptides and bactericidal chemokines, including calprotectin and CXCL10, by P. aeruginosa infection, which was further enhanced by downregulation of IL-24 signaling in corneal epithelium at 6 hpi. These potent bactericidal peptides participate in innate killing of pathogens within the epithelium (58), and greatly reduced pathogen invasion to the stroma at the initial hours of infection, 3–6 hpi when most invading pathogens remain in the epithelial layer (8, 15, 41). Consistent with the notion that epithelium is the major site of innate defense, we observed that IL-24 receptors were mostly expressed by epithelium in B6 mouse corneas (Supplemental Fig. 1), suggesting that the effects of IL-20R cytokines on innate defense mechanisms are likely limited to the epithelial layer. It is plausible that the reduction of pathogen burden results in a less innate inflammatory response, yet the activated innate immune defense apparatus further eliminates remaining pathogens. Downregulation of IL-24, similar to flagellin administration (9), also greatly reduced the intensity of inflammatory response in the corneas, including a marked decrease in the number of PMNs at 1 dpi, giving a marked decrease in the corneal content of calprotectin (60% of total cytosolic proteins in PMN) (59) seen in Fig. 5B.
The presence of rIL-24, similar to downregulation of CXCL10 (8), calprotectin (15), or cathelicidin (9), alternatively, significantly increased bacterial burden due to a decrease in pathogen killing, resulting in much more severe keratitis with greatly increased infiltration of PMNs. Hence, the pathogen burden appears to be a determining factor of the inflammatory response during microbial infection in the cornea. This phenomenon may not be unique for the cornea, as the inflammatory response in the mouse ceca was shown to be related to the numbers of Salmonella with or without flagellin administration (60). Moreover, our study using a lethal murine P. aeruginosa model also showed a decrease in P. aeruginosa burden as well as in PMN infiltration in flagellin-pretreated B6 mouse lung at 24 hpi (10).
How might IL-24 suppress innate immune response during tissue infection? We observed that in cultured human primary CECs, whereas IL-1β stimulated the expression of IL-24, IL-1β (self-amplification), IL-6, as well as SOCS3, IL-24 was only able to induce SOCS3 expression in a robust fashion at 1 h after treatment, with SOCS3 rapidly declining to a lower level at 2 h after treatment (Fig. 9). This early and robust expression of SOCS3 during the initial hour of infection may be one of underlying mechanisms for IL-24 to possess immunosuppressive activity in the cornea. Consistent with this notion, cell-specific loss of SOCS3 in keratinocytes has been shown to upregulate S100A8/A9 expression and to cause skin inflammation and epidermal hyperplasia in an IL-20R cytokine–related manner (27).
In summary, our data demonstrate that IL-24 expression is enhanced at an early time point in mouse corneas after P. aeruginosa infection. IL-24 dampens inflammatory response necessary for mounting an effective defense against microbial infection during the initial hours of infection. This observation highlights the concern for the use of IL-24 as an anticancer agent, as it may render patients more susceptible to opportunistic infection at a time when their immune systems are already comprised. Additonally, the IL-20R2 signaling pathway could be therapeutically targeted to alter the susceptibility of the corneas to P. aeruginosa infection.
Disclosures
The authors have no financial conflicts of interest.
Acknowledgments
We thank Sharowynn Smith for proofreading the manuscript, Haijing Sun for technical support, and Paul D. Walker, Ashok Kumar, and Jena Steinle for serving as committee members.
Footnotes
This study was supported by National Institutes of Health/National Eye Institute Grants R01-EY010869, R01-EY017960, and P30-EY004068 for research to prevent blindness.
The online version of this article contains supplemental material.
Abbreviations used in this article:
- B6
- C57BL/6
- CEC
- corneal epithelial cell
- dpi
- day postinfection
- hpi
- hour postinfection. MMP, matrix metalloproteinase
- MPO
- myeloperoxidase
- MRSA
- methicillin-resistant Staphylococcus aureus
- PMN
- polymorphonuclear leukocyte
- qPCR
- quantitative PCR
- siRNA
- small interfering RNA
- SOCS
- suppressor of cytokine signaling.
- Received December 13, 2016.
- Accepted February 27, 2017.
- Copyright © 2017 by The American Association of Immunologists, Inc.
References
In this issue
