TLR-mediated induction of pro-allergic cytokine IL-33 in ocular mucosal epithelium

Abstract

Interleukin (IL) 33 has been recently identified as a ligand to the ST2 receptor that mediates Th2-dominant allergic inflammation. The purpose of this study was to explore the role of toll-like receptor (TLR)-mediated innate immunity in IL-33 induction by mucosal epithelium. Human corneal tissues and cultured primary human corneal epithelial cells (HCECs) were treated with a variety of viral or bacterial components without or with different inhibitors to evaluate the IL-33 regulation and signaling pathways. The level of mRNA expression was determined by reverse transcription and real time PCR, and protein was measured by ELISA, immunostaining and Western blotting. IL-33 mRNA and protein were largely induced by various microbial components, mainly by polyI:C and flagellin, the ligands to TLR3 and TLR5, respectively in human corneal epithelium ex vivo and in vitro cultures. Pro-IL-33 protein was normally restricted inside cells, and could be secreted outside when activated by ATP. The PolyI:C induced IL-33 production was blocked by TLR3 antibody or TRIF Inhibitory peptide, while flagellin stimulated IL-33 was blocked by TLR5 antibody or MyD88 Inhibitory peptide. Interestingly, IκB-α inhibitor (BAY11-7082) or NF-κB inhibitor (quinazoline) blocked NF-κB p65 protein nuclear translocation, and suppressed IL-33 production induced by PolyI:C and flagellin. These findings demonstrate that IL-33, an epithelium-derived pro-allergic cytokine, is induced by microbial ligands through TLR-mediated innate signaling pathways, suggesting a possible role of mucosal epithelium in Th2-dominant allergic inflammation.

Figure 1

TLR-dependent induction of IL-33 by microbial ligands in primary HCECs. A. Levels of IL-33 mRNA at 8 hours after stimulation by 50μg/ml polyI:C or 10μg/ml of Pam3CSK4, PGN, polyI:C, LPS, flagellin, FSL-1, R-837, ssRNA40 or C-CpG-ODN. B. The concentrations of IL-33 protein detected by ELISA in cell lysates treated with various TLR ligands (X-axis) for 48 hours. C. Levels of IL-33 mRNA at different time periods (4-24 h) in cells treated with 50μg/ml polyI:C (left), or by increasing concentration (10, 25, or 50μg/ml) of polyI:C at 8 hours (right), evaluated by quantitative RT and real-time PCR. Results shown are the mean ± SD of three to five independent experiments. *P< 0.05; **P < 0.01.

Figure 2

Stimulated secretion of IL-33 cellular protein to culture medium by ATP in primary HCECs. The HCECs were exposed to microbial ligand 50μg/ml polyI:C (A, C, E, G) or 10μg/ml flagellin (B, D, F, H) for 48 hours without or with addition of 5mM ATP to media 30min prior to measurement. The IL-33 protein concentrations in culture media (A & B) and in cell lysates were detected by ELISA (C & D), and the level of cellular IL-33 protein was also evaluated by Western blot analysis using β-actin as control (E & F) with quantitative ratio of IL-33/β-actin (G & H). Results shown are the mean ± SD of three to five independent experiments. *P< 0.05; **P < 0.01.

Figure 3

IL-33 induction in an ex vivo human corneal tissues evaluated by immunohistochemical staining. A fresh corneoscleral tissue was cut into four equal-sized pieces. Each quarter was placed into a well of eight-chamber slides, epithelial side up in 150μL of serum-free SHEM medium, without or with polyI:C (50μg/ml) or flagellin (10μg/ml) for 24 hours in a 37°C incubator. Frozen sections of corneal tissues were used for IL-33 immunohistochemical staining with isotype IgG as the negative control. Magnification 400x.

Figure 4

TLR and NF-κB signaling pathways are involved in IL-33 induction by polyI:C or flagellin. The HCECs exposed to polyI:C (50μg/mL) or flagellin (10μg/mL) were preincubated in the absence or presence of rabbit TLR3Ab (10μg/mL), TLR5Ab (10μg/mL), BAY11-7082 (10μM) or NF-kB activation inhibitor quinazoline (NF-kB-I, 10μM) for 1 hour, and Pepinh-MYD (40μM) or Pepinh-TRIF (40μM) for 6 hours. The cultures treated by ligands for 8 hours were subjected to RT and real-time PCR to measure IL-33 mRNA (A & B), the cultures treated for 48 hours were used to evaluate IL-33 protein in cell lysates by ELISA (C & D) and Western blot using β-actin as control (E & F) with quantitative ratio of IL-33/β-actin (G & H).. Results shown are the mean ± SD of three to five independent experiments. *P< 0.05; **P < 0.01.

Figure 5

Activation of IkBα signaling induced by polyI:C and flagellin in HCECs. The HCECs were exposed to polyI:C (50μg/mL) or flagellin (10μg/mL) in the absence or presence of preincubated Pepinh-MYD (40μM) or Pepinh-TRIF (40μM) for 6 hours, and BAY11-7082 (10μM) or NF-kB activation inhibitor quinazoline (NF-kB-I, 10μM) for 1 hour. The cells treated for 1 hour were used for IκB-α phosphorylation (A & C, B & D) and degradation (A & E, B & F) with β-actin as control by Western blotting and quantitative ratio analysis of p-IkB-α/β-actin (C & D) or IkB-α/β-actin (E & F). Results shown are the mean ± SD of three to five independent experiments. *P< 0.05; **P < 0.01.

Figure 6

Activation of NF-κB signaling induced by polyI:C and flagellin in HCECs. A. The HCECs were exposed to polyI:C (50μg/mL) or flagellin (10μg/mL) in the absence or presence of preincubated Pepinh-MYD (40μM) or Pepinh-TRIF (40μM) for 6 hours, and BAY11-7082 (10μM) or NF-kB activation inhibitor quinazoline (NF-kB-I, 10μM) for 1 hour. The cells treated for 4 hours were subjected to cytoplasm and nuclear protein extraction for NF-κB p65 nuclear translocation analysis by Western blotting with quantitative ratio analysis of N-p65/C-p65. Results shown are the mean ± SD of three to five independent experiments. **P < 0.01. C-p65, cytoplasm p65; N-p65, nuclear p65. B. The cells treated for 4 hours in 8-chamber slides were fixed in acetone for immunofluorescent staining with rabbit antibody against p65 followed by AlexaFluor 488-conjugated second antibody. The images are representatives from three independent experiments.