Our results suggest that T-cell deficiency might underlie the lack of CaK induction in nude mice. To test this, we investigated the role of CD4+ T cells during CaK initiation in BALB/c mice. BALB/c mice treated with anti-CD4 antibodies, to deplete CD4+ T cells, were more resistant to CaK formation (Fig. 2A). However, depleting Treg cells (with anti-CD25 antibodies) or γδ T cells (with anti-TCRγδ antibodies) had no effect on the development of CaK in BALB/c mice (Supporting Information Fig. Y-27632 concentration 2A). Furthermore, immunodepletion of IL-23 or IL-17, but not IFN-γ, ablated CaK induction in BALB/c mice (Fig. 2B and Supporting Information Fig. 2B). These data indicate that at least two immune components are

responsible for CaK initiation: a group of T cells that do not belong to either the Treg or γδ T-cell groups, and

cytokines involving the IL-23-IL-17 axis that excludes IFN-γ. In line with these results, IL-17A−/– mice on C57BL/6J background did not develop CaK, unlike WT mice, in response to C. albicans inoculation (Supporting Information Fig. 3A and B). Lastly, depleting neutrophils completely blocked the initiation of CaK (Fig. 2A and Supporting Selleckchem PLX4032 Information Fig. 4), demonstrating for the first time a critical role for neutrophils in the pathogenesis of CaK. Indeed, neutrophils are the predominant immune cells in corneas with infectious keratitis caused by other pathogens [17-20]. To determine whether CaK onset affects IL-17 activity, expression levels of IL-17 in cornea and

serum were measured at various times of the CaK induction course. In immunocompetent BALB/c mice, serum levels of IL-17 were high at day 4 postinfection and then decreased over the 1-month experimental period (Fig. 3A). In contrast, neither IL-23-neutralized BALB/c mice nor nude mice induced IL-17 expression after infection (Fig. 3A), correlating with their inability to develop CaK. Expression analysis at earlier times revealed a peak of IL-17 expression at 24 h postinoculation, while expression was undetectable in inoculated nude mice (Fig. 3B and C). These data indicate that IL-17 is induced acutely after inoculation and is correlated with the development of CaK. To identify the source of IL-17 at early phase of infection, especially before activation of antigen-driven Th17, immunofluorescence labeling was performed on corneal tissues. ID-8 This analysis revealed that IL-17-producing cells were generally positive for Ly-6G, CD4, or Gr-1 (Fig. 4). Quantitative measurement using flow cytometry showed that, among all infiltrating cells, Ly-6G+ neutrophils outnumbered CD4+ lymphocytes by about 40-fold in BALB/c mice at day 1 postinoculation (Fig. 5A). Additionally, anti-IL-17 antibody treatment greatly decreased CD4+ cell and neutrophil infiltration in the corneas (Fig. 5A). Neutrophil infiltration was also greatly inhibited in corneas of IL-17A−/− mice (Supporting Information Fig. 3C and D).