The adaptor CARD9 is required for adaptive but not innate immunity to oral mucosal Candida albicans infections

Abstract

Oropharyngeal candidiasis (OPC [thrush]) is an opportunistic infection caused by the commensal fungus Candida albicans. OPC is common in individuals with HIV/AIDS, infants, patients on chemotherapy, and individuals with congenital immune defects. Immunity to OPC is strongly dependent on the interleukin-23 (IL-23)/IL-17R axis, as mice and humans with defects in IL-17R signaling (IL17F, ACT1, IL-17RA) or in genes that direct Th17 differentiation (STAT3, STAT1, CARD9) are prone to mucocutaneous candidiasis. Conventional Th17 cells are induced in response to C. albicans infection via signals from C-type lectin receptors, which signal through the adaptor CARD9, leading to production of Th17-inducing cytokines such as IL-6, IL-1β, and IL-23. Recent data indicate that IL-17 can also be made by numerous innate cell subsets. These innate "type 17" cells resemble conventional Th17 cells, but they can be activated without need for prior antigen exposure. Because C. albicans is not a commensal organism in rodents and mice are thus naive to this fungus, we had the opportunity to assess the role of CARD9 in innate versus adaptive responses using an OPC infection model. As expected, CARD9(-/-) mice failed to mount an adaptive Th17 response following oral Candida infection. Surprisingly, however, CARD9(-/-) mice had preserved innate IL-17-dependent responses to Candida and were almost fully resistant to OPC. Thus, CARD9 is important primarily for adaptive immunity to C. albicans, whereas alternate recognition systems appear to be needed for effective innate responses.

FIG 1

CARD9 is required for adaptive Th17 responses to OPC. (A) Schematic diagram of OPC model. (B) CARD9^−/− mice fail to show enhanced clearance of C. albicans from the oral mucosa following rechallenge. CARD9^−/− or WT mice were subjected to a 1° or 1°+2° oral inoculation with C. albicans, and fungal load in tongue was assessed at days 1 and 2 postinfection. The percentage of mice with a detectable fungal count at each time point is shown. **, P < 0.01 (by chi-square analysis). (C) CARD9^−/− mice show increased fungal loads in the oral cavity following rechallenge. Data represent fungal loads in WT and CARD9^−/− mice at days 1 to 3 postinfection during 1° infection or 1°+2° infection. Data are expressed as CFU/g tongue tissue. Error bars show geometric mean with 95% confidence interval (CI). **, P < 0.01 (by t test with Mann-Whitney correction). (D) CARD9^−/− mice exhibit impaired Th17 cell frequencies in cLN. CD3⁺ CD4⁺ cells from cLN of infected WT or CARD9^−/− mice were analyzed for intracellular expression of IL-17A, and the percentage of total LN cells is indicated. **, P < 0.01 (by t test with Mann-Whitney correction). (E) CARD9^−/− mice have reduced Candida-specific Th17 responses following rechallenge. Total cLN cells from CARD9^−/− or WT mice subjected to the indicated conditions were cultured in vitro for 5 days with HK Candida extract or pALS peptide. Supernatants were analyzed in triplicate for IL-17A. **, P < 0.01 (by t test with Mann-Whitney correction). n.s., not significant.

FIG 2

CARD9 is dispensable for immunity to acute OPC. (A) CARD9^−/− mice show only mild susceptibility to acute C. albicans oral infection. The indicated mice were subjected to OPC for 5 days, and fungal load in tongue was assessed. Data are expressed as CFU/g tongue tissue. **, P < 0.01; ***, P < 0.005 (by t test with Mann-Whitney correction). (B) CARD9^−/− mice regain weight following oral infection with C. albicans. The indicated mice were subjected to OPC and weighed daily for 5 days. Data are expressed as the average percentage of starting weight on day 1. (C) IL-17RA signaling but not CARD9 is required for clearance of C. albicans from the oral cavity. The indicated mice were subjected to infection with C. albicans, and fungal load in tongue was assessed at days 1 to 5 postinfection. Data are expressed as CFU/g tongue tissue. Error bars show geometric mean with 95% CI. (D) CARD9^−/− mice exhibit delayed clearance of fungi from the oral mucosa. (Left) WT or CARD9^−/− mice were subjected to oral infection with C. albicans, and fungal burden in tongue was assessed at the indicated days postinfection. Data are expressed as CFU/g tongue tissue. Error bars show geometric mean with 95% CI. (Right) At 6 weeks postinfection, an oral swab was plated onto YPD-AMP agar, and the percentage of mice with evidence for C. albicans colonization is indicated. ***, P < 0.01 (by t test with Mann-Whitney correction).

FIG 3

Reduced IL-17 signature gene expression in CARD9^−/− mice following OPC. Tongue tissue from WT or CARD9^−/− mice at day 1 or day 3 postinfection was analyzed for expression of il17a (A), IL-17 target genes il6 and lcn2 (encoding lipocalin 2, 24p3) (B), and Th17 inductive cytokines il23 and il1b (C). Expression was normalized to GAPDH. n.s., not significant. ‡, P < 0.05 (compared to WT mice at day 1); *, P < 0.05; ***, P < 0.001 (compared to WT mice at the matched time points by analysis of variance [ANOVA] with post hoc Tukey's test).

FIG 4

CARD9^−/− mice do not exhibit fungal dissemination from the GI tract following OPC. The indicated mice were subjected to OPC for 5 days. In the experiments whose results are shown in panel B, cortisone acetate (Cort.) was administered to WT mice at days −1, +1, and + 3 postinoculation at 225 mg/kg of body weight. The indicated organs were analyzed for C. albicans colonization. Bars show geometric means.