Effect of pro-inflammatory mediators on membrane-associated mucins expressed by human ocular surface epithelial cells

Abstract

Membrane-associated mucins are altered on the ocular surface in non-Sjögren's dry eye. This study sought to determine if inflammatory mediators, present in tears of dry eye patients, regulate membrane-associated mucins MUC1 and -16 at the level of gene expression, protein biosynthesis and/or ectodomain release. A human corneal limbal epithelial cell line (HCLE), which produces membrane-associated mucins, was used. Cells were treated with interleukin (IL)-6, -8, or -17, tumor necrosis factor-alpha (TNF-alpha), and Interferon-gamma (IFN-gamma), or a combination of TNF-alpha and IFN-gamma, or IFN-gamma and IL-17, for 1, 6, 24, or 48 h. Presence of receptors for these mediators was verified by RT-PCR. Effects of the cytokines on expression levels of MUC1 and -16 were determined by real-time PCR, and on mucin protein biosynthesis and ectodomain release in cell lysates and culture media, respectively, by immunoblot analysis. TNF-alpha and IFN-gamma each significantly induced MUC1 expression, cellular protein content and ectodomain release over time. Combined treatment with the two cytokines was not additive. By comparison, one of the inflammatory mediators, IFN-gamma, affected all three parameters-gene expression, cellular protein, and ectodomain release-for MUC16. Combined treatment with TNF-alpha and IFN-gamma showed effects similar to IFN-gamma alone, except that ectodomain release followed that of TNF-alpha, which induced MUC16 ectodomain release. In conclusion, inflammatory mediators present in tears of dry eye patients can affect MUC1 and -16 on corneal epithelial cells and may be responsible for alterations of surface mucins in dry eye.

Fig. 1

A. HCLE cells express the receptors for the inflammatory mediators used in this study, indicating their ability to be responsive to the treatments. RT-PCR performed on RNA from untreated control cultures demonstrates the presence of message for the receptors for IL-6, -8, -17, TNF-α, and IFN-γ. The predicted base pair size for each of the receptors is indicated at the bottom of the figure. B. Representative immunoblots used to assay effects of cytokines on MUC1 and MUC16 cellular protein and ectodomain release. These blots demonstrate effect of 24 hour exposure to IL-17, IFN-γ, or IL-17 plus IFN-γ. GAPDH is the loading control for cellular protein quantitation.

Fig. 2

MUC1 expression, cellular protein amount and/or ectodomain release was affected by four of the five inflammatory mediators tested. TNF-α and IFN-γ significantly increased expression and cellular protein by 6 and 24 hours, respectively, followed by ectodomain release at 24 (TNF-α) or 48 hours (IFN-γ). IL-6 decreased cellular MUC1 amount at 1 hour, followed by a decrease in ectodomain release at 6 and 24 hours. IL-17 also decreased the cellular amount of MUC1 at 1 hour, but by 48 hours, there was a significant increase. No significant effects were seen after treatment with IL-8. Combined treatment with TNF-α and IFN-γ or with IL-17 and IFN-γ had no effect over treatment with the cytokines alone.

Fig. 3

MUC16 cellular protein amount and/or ectodomain release was significantly affected by all five of the inflammatory mediators tested. There was a decrease in ectodomain release after 1 hour of treatment with IL-17. IL-6 increased cellular protein and decreased ectodomain release at 1 hour and decreased both at 24 hours. IL-8 decreased ectodomain release at 24 hours. TNF-α decreased cellular protein at 1 hour and increased ectodomain release at 24 hours. IFN-γ induced the most significant effects, increasing cellular protein at 24 and 48 hours, decreasing ectodomain release at 24 and 48 hours without changes in expression level at 24 hours. By 48 hours, expression levels were significantly upregulated. Combined treatment with TNF-α and IFN-γ or IL-17 and IFN-γ had no effect over treatment with the cytokines alone.