In Vivo and In Vitro Impairments in T Helper Cell and Neutrophil Responses against Mucor irregularis in Card9 Knockout Mice

Mucor irregularis is a frequently found fungus in Asia, especially China, and it causes primary cutaneous mucormycosis with a high rate of disfigurement. Caspase recruitment domain-containing protein 9 (Card9) is an essential adaptor molecule downstream of C-type lectin receptors. It mediates the activation of nuclear factor kappa B (NF-κB), regulates T helper 1 (Th1) and Th17 differentiation, and plays an important role in fungal immune surveillance. CARD9 deficiency correlates with the increased susceptibility to many fungal infections, including cutaneous mucormycosis caused by M. irregularis.

ABSTRACT

Mucor irregularis is a frequently found fungus in Asia, especially China, and it causes primary cutaneous mucormycosis with a high rate of disfigurement. Caspase recruitment domain-containing protein 9 (Card9) is an essential adaptor molecule downstream of C-type lectin receptors. It mediates the activation of nuclear factor kappa B (NF-κB), regulates T helper 1 (Th1) and Th17 differentiation, and plays an important role in fungal immune surveillance. CARD9 deficiency correlates with the increased susceptibility to many fungal infections, including cutaneous mucormycosis caused by M. irregularis. However, the underlying immunological mechanisms were not elucidated. Our study established a murine model of subcutaneous M. irregularis infection, and we isolated immune cells, including bone marrow-derived macrophages, bone marrow-derived dendritic cells, naive T cells, and neutrophils, from wild-type (WT) and Card9 knockout (Card9^−/−) mice to examine the antifungal effect of Card9 on M. irregularis in vivo and in vitro. Card9^−/− mice exhibited increased susceptibility to M. irregularis infection. Impaired local cytokine and chemokine production, NF-κB (p65) activation, and Th1/17 cell differentiation and partially impaired neutrophil-dependent antifungal immunity were observed in Card9^−/− mice. This work enriches our knowledge of the relationship between CARD9 deficiency and mucormycosis.

INTRODUCTION

Mucormycosis is characterized by a rapidly progressing angioinvasion with ensuing thrombosis and tissue necrosis, and it is a potentially lethal fungal disease caused by species in the Mucorales (1). Mucormycosis is the third most prevalent invasive fungal disease in immunosuppressed individuals, after candidiasis and aspergillosis. Some cases appear in immunocompetent populations (2). The risk factors of mucormycosis include the use of corticosteroids, neutropenia, trauma, diabetes mellitus, organ transplantations, and chemotherapy (3). There are different clinical forms of infection, including rhinocerebral, cutaneous, pulmonary, gastrointestinal, and disseminated (4). Mucor irregularis (previously Rhizomucor variabilis) is a frequently found fungus in Asia, especially in China, and it generally leads to primary cutaneous mucormycosis in immunocompetent individuals following the disruption of cutaneous barriers (5–7). However, we recently identified the first case of cutaneous mucormycosis caused by M. irregularis in a patient with caspase recruitment domain-containing protein 9 (CARD9) deficiency (8, 9).

Card9 is primarily expressed in myeloid cells, and it is an essential adaptor molecule downstream of C-type lectin receptors (CLRs) that mediates the activation of nuclear factor kappa B (NF-κB), regulates T helper 1 (Th1) and Th17 differentiation, and plays an important role in fungal immune surveillance. CARD9 links innate immunity to adaptive immunity, and CARD9 deficiency is attributed to much more severe immune disorders than the deficiency of a single CLR (10). Human CARD9 genetic mutations are relevant to severe fungal infections: e.g., infections caused by dematiaceous fungal infections (11–13), mucocutaneous or invasive candidiasis (14–16), extrapulmonary Aspergillus infection (17), or deep dermatophytosis (18–20). A previous study reported that a patient with CARD9 mutations and deep dermatophytosis caused by Microsporum ferrugineum exhibited decreased proinflammatory cytokine production and Th cell differentiation (18). Other patients with CARD9 deficiency and phaeohyphomycosis exhibited impaired cytokine and chemokine production and NF-κB activation and decreased Th responses (12). All of the pathogenic fungi identified in CARD9 deficiency patients were from the phylum Ascomycota (21), and M. irregularis is the first fungus belonging to the order Mucorales. Mucorales spores lack a rodlet immunoprotective hydrophobin layer and induce different responses in human mononuclear cells compared to Ascomycota spores (22).

To investigate the relationship between CARD9 deficiency and M. irregularis infection and the underlying immune mechanisms, we established an M. irregularis infection murine model and isolated immune cells from wild-type (WT) and Card9 knockout (Card9^−/−) mice to examine the antifungal effects of Card9 on M. irregularis in vivo and in vitro.

RESULTS

Card9^−/− mice exhibited increased susceptibility to M. irregularis infection.

We injected M. irregularis inocula into the hind footpads of WT and Card9^−/− mice to establish a mucormycosis murine model that mimics the clinical onset of cutaneous mucormycosis and investigate the antifungal role of Card9 in M. irregularis infection.

Footpad swelling and lesions gradually healed in WT mice and were completely recovered at 8 weeks postinoculation (wpi). Neither necrosis nor ulceration emerged during the whole observation period (Fig. 1a). Fungal elements and the inflammatory response were absent in histopathological examination of the footpads at 4 wpi (Fig. 1b). Fungal burdens in the footpads and inguinal lymph nodes peaked at 3 days postinoculation (dpi) and were reduced to zero at 7 and 28 dpi (Fig. 1c).

FIG 1.

Card9^−/− mice display increased susceptibility to M. irregularis infection. The hind footpads of WT and Card9^−/− mice were subcutaneously injected with 2.5 × 10⁸ M. irregularis spores. (a) Natural courses of infection in WT and Card9^−/− mice (n = 6/group). Shown are representative images of different observation time points (0 wpi [normal], 4 wpi and 8 wpi). (b) Histopathology of H&E- and PAS-stained footpads from WT and Card9^−/− mice harvested at 8 wpi (n = 3/group). (c) Fungal burdens of the footpads and inguinal lymph nodes of WT and Card9^−/− mice at 3, 7, and 28 dpi (n = 3/group). ns, not significant; *, P < 0.05; **, P < 0.01; and ****, P < 0.0001.

However, persistent swelling was found throughout the infection course in Card9^−/− mice. Fast-growing necrosis, ulcers, crustal lesions, and disfigurement were observed, which resembled the clinical manifestations of M. irregularis infection (Fig. 1a). Inflammatory cells, including lymphocytes, histocytes, and multinuclear giant cells, infiltrated into the footpads of Card9^−/− mice (Fig. 1b). Abundant nonseptate, wide, irregularly branched hyphae were discovered on histopathological examination of the footpads at 4 wpi (Fig. 1b). Although the fungal burdens of the footpads and inguinal lymph nodes gradually decreased over time, these burdens always remained statistically higher than the corresponding sites in WT mice, which illustrates the defects in eliminating fungal infections in Card9^−/− mice (Fig. 1c).

Impaired local cytokine and chemokine production in Card9^−/− mice.

Because of the different phenotypes of WT and Card9^−/− mice, we measured the levels of inflammatory cytokines and chemokines in footpad homogenates. Card9^−/− mice exhibited sharp decreases in the levels of interleukin-6 (IL-6), IL-10, tumor necrosis factor alpha (TNF-α), the Th1-cytokine gamma interferon (IFN-γ), and the Th17 cytokine IL-17A at 3 dpi and a marked increase in the level of the Th2 cytokine IL-4. Card9^−/− mice exhibited significant decreases in the levels of IL-6 and TNF-α at 7 dpi and increases in the levels of IL-4 and IL-10 compared to WT mice (Fig. 2).

FIG 2.

Impaired local cytokine production and Th cell responses in Card9^−/− mice. IFN-γ, IL-17A, IL-6, TNF-α, IL-4, and IL-10 levels were assessed at 0, 3, and 7 dpi (n = 6/group). Data are shown as the mean ± standard deviation (SD) and were analyzed by unpaired t tests. ns, not significant; *, P < 0.05; **, P < 0.01; and ****, P < 0.0001.

Statistically significant reductions in the mRNA levels of the chemokines CXCL1 and CXCL2 and the cytokines TNF-α and IL-6 at 3 dpi were found in the footpads of Card9^−/− mice (Fig. 3a). These results indicated that Card9 participated in the production of cytokines TNF-α and IL-6 and chemokines CXCL1 and CXCL2 in M. irregularis infections.

FIG 3.

Impaired local chemokine production in Card9^−/− mice and impaired NF-κB (p65) activation in Card9^−/− bone marrow-derived macrophages. (a) CXCL1, CXCL2, TNF-α, and IL-6 levels were assessed at 3 dpi (n = 4/group). Data are shown as the mean ± SD and were analyzed by unpaired t tests. ns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001. (b) NF-κB (p65) activation was impaired at 60 min in Card9^−/− bone marrow-derived macrophages compared with WT bone marrow-derived macrophages.

Impaired NF-κB (p65) activation in Card9^−/− bone marrow-derived macrophages.

To further clarify the signaling pathways participating in the immune response to M. irregularis, we detected NF-κB pathway activity in bone marrow-derived macrophages (BMDMs) from WT or Card9^−/− mice stimulated with inactivated M. irregularis swollen spores.

NF-κB (p65) translocates into the cell nucleus when Card9 is activated via the Card9-dependent NF-κB pathway. We measured the specific NF-κB (p65) DNA binding activity in BMDMs and found a marked reduction in NF-κB (p65) activation in Card9^−/− BMDMs (Fig. 3b). These results indicated the importance of Card9 in the NF-κB (p65) activation in M. irregularis infections.

Differences in cytokine secretion by dendritic cells and Th1/17 cell differentiation induced by resting or swollen spores of M. irregularis.

Innate immunity and adaptive immunity are essential in antifungal immunity. As the most important professional antigen-presenting cells (APCs), immature bone marrow-derived dendritic cells (BMDCs) process and present antigens to prime naive T cells. To identify the cytokines secreted by DCs that shaped Th1/Th17 cell differentiation, we cocultured immature DCs with paraformaldehyde-inactivated resting spores of M. irregularis for 24 h and examined the levels of IL-1β, IL-23, and IL-12p70 in the culture supernatant. We did not observe obvious IL-1β, IL-23, or IL-12p70 secretion (Fig. 4a). To clarify whether cytokines secreted by DCs against M. irregularis were stage specific, we prepared swollen spores of M. irregularis and found statistically significant increases in the secretion of IL-1β, IL-23, and IL-12p70 with the swollen spores compared to resting spores (Fig. 4a).

FIG 4.

Differences in cytokine secretion by BMDCs and Th1/17 cell differentiation induced by resting or swollen spores of M. irregularis. (a) Swollen spores of M. irregularis induced more secretion of IL-1β, IL-23, and IL-12p70 and differentiation of Th1/17 cells than resting spores. (b) Swollen spores of M. irregularis induced higher expression of T-bet and RORγt than did resting spores. (c) Swollen spores of M. irregularis induced higher secretion of IL-17A and IFN-γ than did resting spores. (d) Swollen spores of M. irregularis induced more Th1/17 cell differentiation than resting spores. ns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

We investigated whether the adaptive immune responses induced by different stages of spores were disparate and noticed that the transcriptional levels of T-bet and ROR-γt and the protein levels of secreted IFN-γ and IL-17A were also significantly higher in the swollen spore-stimulated group than the resting spore-stimulated group (Fig. 4b and c). Swollen spores also induced a statistically significant improvement in Th1/Th17 cell differentiation (Fig. 4d). All the results mentioned above indicated that resting or swollen spores of M. irregularis could induce different cytokine secretion by dendritic cells and Th1/17 cell differentiation.

Card9 is indispensable in cytokine secretion by DCs and Th1/17 cell differentiation induced by swollen spores of M. irregularis.

Based on the in vivo findings in Card9^−/− mice, we further determined the in vitro antifungal effect of Card9 on cytokine secretion by DCs and Th1/17 cell differentiation induced by swollen spores of M. irregularis. We found that the secretion of IL-1β, IL-23, and IL-12p70 by DCs from Card9^−/− mice stimulated by swollen spores was significantly lower than that by the DCs from WT mice (Fig. 5a). We further observed that the expression of T-bet and ROR-γt and the secretion of IL-17A and IFN-γ were also lower in Card9^−/− mice than WT mice (Fig. 5b and c). The swollen spore-induced Th1/Th17 cell differentiation was lower in Card9^−/− mice than WT mice (Fig. 5d). Our results indicated that Card9 is indispensable in cytokine secretion by DCs and Th1/17 cell differentiation induced by swollen spores of M. irregularis.

FIG 5.

Card9 is indispensable in cytokine secretion by dendritic cells and Th1/17 cell differentiation induced by swollen spores of M. irregularis. (a) The secretion of IL-1β, IL-23, and IL-12p70 by BMDCs from Card9^−/− mice stimulated by swollen spores was significantly lower than that by BMDCs from WT mice. (b) The expression of T-bet and RORγt induced by swollen spores of M. irregularis was decreased in Card9^−/− mice. (c) The secretion of IL-17A and IFN-γ induced by swollen spores of M. irregularis was reduced in Card9^−/− mice. (d) Th1/Th17 cell differentiation induced by swollen spores of M. irregularis was lower in Card9^−/− mice than in WT mice. ns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001.

The binding of β-glucan to dectin-1 might be the key factor for the different CARD9-dependent host responses to M. irregularis resting spores and swollen spores.

To figure out why Card9-dependent host responses to resting spores and swollen spores are different, we examined β-glucan on the surface of the different forms of M. irregularis spores by confocal technology. We found that β-glucan was absent in the cell walls of resting spores, while its level dramatically increased in the cell walls of swollen spores (Fig. 6a).

FIG 6.

The binding of β-glucan to dectin-1 might be the key factor for the different CARD9-dependent host response to M. irregularis resting spores and swollen spores. (a) Confocal images of β-glucan staining on the surface of resting spores and swollen spores. β-Glucan was absent in the cell walls of resting spores, while dramatically increased in the cell walls of swollen spores. (b) The binding of dectin-1 soluble proteins to swollen spores was much greater than to resting spores.

β-Glucan is mainly recognized by dectin-1; therefore, we tested the binding ability of different forms of M. irregularis with soluble dectin-1 proteins. The binding of dectin-1 soluble proteins to swollen spores was much greater than binding to resting spores (Fig. 6b).

From the results mentioned above, we speculate that the binding of β-glucan to dectin-1 might be the key factor for the different CARD9-dependent host responses to M. irregularis resting spores and swollen spores.

Card9 deficiency partially impairs neutrophil-dependent antifungal immunity.

Because the expression of the neutrophil-targeting chemokines CXCL1 and CXCL2 was significantly lower in Card9^−/− mice than WT mice, we investigated the responses of neutrophils in combating M. irregularis infections.

We used flow cytometry to determine the ability of neutrophils from WT mice or Card9^−/− mice to phagocytose fluorescein isothiocyanate (FITC)-labeled M. irregularis spores, and no significant difference was found between the mice (data not shown). Second, we detected the production of reactive oxygen species (ROS), and there was no difference (data not shown). Third, we assessed the release of neutrophil extracellular traps (NETs) following neutrophil stimulation with M. irregularis hyphae. We demonstrated that neutrophils from Card9^−/− mice released dramatically fewer NETs than WT mice (Fig. 7a). Finally, we cocultured neutrophils with inactivated swollen spores of M. irregularis overnight and examined the expression of IL-6 and TNF-α in neutrophils. We found significantly lower expression in the neutrophils from Card9^−/− mice than WT mice (Fig. 7b). Our results indicated that Card9 deficiency partially impairs neutrophil-dependent antifungal immunity.

FIG 7.

Card9 deficiency partially impairs neutrophil-dependent antifungal immunity. (a) Neutrophils from Card9^−/− mice released dramatically fewer NETs than those from WT mice. (b) The expression of IL-6 and TNF-α was significantly lower in Card9^−/− neutrophils than in WT neutrophils. *, P < 0.05; ***, P < 0.001.

DISCUSSION

The present study used WT and Card9^−/− mice and established a murine model of subcutaneous M. irregularis infection to mimic the route of human cutaneous mucormycosis. We isolated immune cells, including BMDCs, naive T cells, BMDMs, and neutrophils, to investigate the antifungal effects of Card9 on M. irregularis in vivo and in vitro. Our results demonstrated that Card9^−/− mice had higher susceptibility to M. irregularis infection than WT mice, which may be associated with impaired local cytokine and chemokine production, impaired NF-κB pathway activation, and impaired Th cell and neutrophil responses in Card9^−/− mice.

The skin lesions and histopathological findings of the footpads of Card9^−/− mice were consistent with M. irregularis-infected patients with CARD9 deficiency (8). Based on the similar phenotypes between the mouse model and human patients, we hypothesized that a close connection existed between CARD9 deficiency and increased susceptibility to M. irregularis infection.

Autosomal-recessive CARD9 deficiency is a primary immunodeficiency, and it leads to life-threatening, invasive fungal infections in otherwise healthy people (21). The spectrum of fungal susceptibility and clinical presentation varies based on CARD9 deficiency and fungal identity (23). Neutrophils are the most numerous white blood cells and the first-line defensive effector cells that are rapidly recruited into infectious sites after infection occurs (24). Published data demonstrated that CARD9 deficiency was associated with an impaired accumulation of neutrophils and induction of neutrophil-recruiting CXC chemokines against black yeasts, Candida and Aspergillus (12, 16, 25). A recent study by R. A. Drummond et al. showed that CARD9 mediated the accumulation of neutrophils into the central nervous system (CNS) via IL-1β and CXCL1 (26). Significantly decreased expression of neutrophil-targeting chemokines CXCL1 and CXCL2 was found in Card9^−/− mice in our study, which highlighted the role of Card9 in recruiting neutrophils into infection sites to mediates immunity against M. irregularis. Our results are consistent with previous studies.

Based on the reduced numbers of neutrophils recruited to the infection sites, we wondered whether there were deficiencies in the functions of neutrophils against M. irregularis. Neutrophils are one of the most essential immune cells in innate immunity and use killing mechanisms, including phagocytosis, to eliminate pathogens, and they kill the pathogens in phagosomes using NADPH oxygenase-dependent mechanisms (ROS) or antibacterial proteins (cathepsins, defensins, lactoferrin, and lysozyme) (27). P. Liang et al. indicated that Card9^−/− polymorphonuclear leukocytes (PMNs) had defects in Phialophora verrucosa killing and proinflammatory cytokine production but normal ROS generation and phagocytotic ability (28). The abilities of Card9^−/− neutrophils in ROS generation and phagocytotic ability toward M. irregularis were normal in our study, which indicates the unnecessary role of Card9 in ROS generation and phagocytotic ability of neutrophils, as published previously (29, 30; data not shown). Neutrophils selectively release NETs in response to pathogens that are too large to be phagocytosed (31). NETs are extracellular weblike structures comprised of DNA and neutrophil antimicrobial proteins released after microbial or sterile inflammatory stimuli. J. T. Loh et al. demonstrated that Dok3 adaptor negatively regulated antifungal immunity in neutrophils via suppression of Card9 signaling (32). Card9^−/− neutrophils produced fewer NETs against M. irregularis hyphae in our study and expressed decreased levels of IL-6 and TNF-α, which indicated that Card9 influenced the production of NETs and proinflammatory cytokines.

Impaired IL-17-dependent immunity was also associated with CARD9 deficiency (21). Th cells in adaptive immunity play essential roles in the clearing of pathogenic fungi: Th1 cells and Th17 cells render protective immunity against fungi, but Th2 cells increase susceptibility to fungal infections (33). Chronic fungal infections arise once Th1/Th2 cellular immunity becomes unbalanced, in which case, Th1 responses turn into Th2 immune responses (34). Previous studies demonstrated the significant roles of Th17 cells in fungal infections caused by Candida spp. and dematiaceous fungi (21), and Th1 cells played essential roles in Aspergillus fumigatus or Coccidioides posadasii infections (35, 36). Mucorales-specific T cells produced IFN-γ, IL-4, and IL-10, but less IL-17, and caused damage to fungal hyphae (37). Rhizopus arrhizus hyphae induced a strong release of IL-23 and TNF-α by human DCs, which drives Th1 and Th17 responses (38). Reduced production of the Th1 cytokine IFN-γ and the Th17 cytokine IL-17A and increased production of IL-10 and the Th2 cytokine IL-4 were found during chronic M. irregularis infection in Card9^−/− mice in our study. Card9^−/− BMDCs induced less secretion of proinflammatory cytokines and lower Th1/17 differentiation than WT BMDCs. Collectively, these findings demonstrated the importance of Th cells in the pathogenesis of mucormycosis caused by CARD9 deficiency.

At the signaling pathway level, Card9 lies downstream of C-type lectin receptors (CLRs), and it associates with BclX and Malt1 to activate the canonical NF-κB pathway and induce Th1 and Th17 cell responses (39). Card9-dependent NF-κB activation is essential in Candida albicans infection, Exophiala spinifera infection and Rhizopus arrhizus infection (12, 40). Our study showed a statistically significant reduction in NF-κB (p65) activation in Card9^−/− BMDMs. This result, in combination with the findings from Th cells mentioned above, suggests that CARD9 mediates downstream Th1 and Th17 responses toward M. irregularis via the NF-κB signaling pathway.

A published report showed that the developmental stage of spores of Rhizopus oryzae corresponded to various macrophage activities, and swollen spores induced stronger macrophage activity than resting spores (41). The ability of M. irregularis spores to induce host immune responses was also stage specific in our study, which was likely due to differences in pathogen-associated molecular pattern (PAMP) exposure. However, this hypothesis needs further investigation.

Our study found that the increased susceptibility to M. irregularis in Card9^−/− mice may be related to impaired neutrophil and Th cell responses. Therefore, interventions targeting neutrophils and Th cells may effectively improve the antifungal abilities of CARD9 deficiency patients. Researchers showed that granulocyte-macrophage colony-stimulating factor (GM-CSF) or granulocyte colony-stimulating factor (G-CSF) increased the number of PMNs in the blood circulation and decreased the morbidity of infections (42). Exogenous GM-CSF and G-CSF as adjunctive immunotherapy successfully controlled Candida CNS infections in three patients with CARD9 deficiency, but no significant effect was observed in a fourth patient with same infection (43–45). Two patients with CARD9 deficiency were successfully treated with hematopoietic stem cell transplantation (HSCT) (46). Neutropenia is an important risk factor of mucormycosis, and several patients suffering from mucormycosis were successfully cured with combined applications of exogenous GM-CSF or G-CSF and antifungal therapy (47, 48). Published studies demonstrated the positive antifungal potentials of IFN-γ in several patients with candidiasis or Aspergillus terreus infection (49, 50). All of these studies support the importance of leukocytes in CARD9-related antifungal immunity.

The immune system is a complex network that is composed of multiple immune cells, cytokines, and complement. Our study performed in vivo and in vitro experiments, which provided some results on the antifungal effects of Card9 on M. irregularis. These results enriched our knowledge of the relationship between CARD9 deficiency and mucormycosis, but they may not fully reflect the host immune response against Mucorales. Due to the heterogeneity between mice and humans, our results require further verification in humans. More attention must be paid to the links between primary immunodeficiency, especially CARD9 deficiency, and M. irregularis infection in clinical practice, and deeper fundamental studies elucidating the host immune responses against M. irregularis are urgently needed to exploit novel immunotherapeutic strategies.

MATERIALS AND METHODS

Mice.

Card9-deficient (Card9^−/−) mice on the C57BL/6 background were generous gifts from Xin Lin (Tsinghua University School of Medicine, Beijing, China). The C57BL/6 WT mice (Vital River Laboratory, Beijing, China) and Card9^−/− mice used in this study were male (6 to 10 weeks old) and maintained under specific-pathogen-free conditions. Animal experiments were performed following protocols approved by the Ethics Committee of Peking University First Hospital.

Fungus.

A clinical isolate of M. irregularis (BMU09468) kept in the Research Center for Medical Mycology at Peking University was used in this study. This isolate was from the first Card9-deficient patient with cutaneous mucormycosis caused by M. irregularis.

The isolate was grown on potato dextrose agar (PDA) for 4 to 6 days at 28°C to obtain abundant resting spores. A fungal suspension was prepared at 2.5 × 10⁹ CFU/ml. Resting spores were inactivated with 4% paraformaldehyde. Swollen spores were prepared via the shaking of resting spores in RPMI 1640 medium at 37°C for 5 h at a speed of 200 rpm/min and subsequently inactivated according to the protocol used for resting spores.

In vivo infection with M. irregularis.

The fungal suspension (100 μl) was subcutaneously injected into the hind footpads of WT and Card9^−/− mice. Normal saline was used as a negative control. All animals were monitored daily to assess survival rates. Mice were sacrificed at 3 and 7 dpi for cytokine profiling, at 3 dpi for chemokine profiling, at 3, 7, and 28 dpi for fungal burden analysis, and at 28 dpi for histopathological examination.

In vivo analysis—histopathological examination, fungal burden analysis, cytokine profiling.

For histopathological examination, formalin‐fixed footpads from WT and Card9^−/− mice were processed and embedded in paraffin. Sections (5 μm) were stained with hematoxylin and eosin (H&E) and periodic acid‐Schiff (PAS) stain.

For fungal burden analysis, footpads were weighed up and homogenized manually in 1 ml normal saline by using sterile glass homogenizers, then the homogenates were collected, 10‐fold dilutions were performed, and then 100-μl samples of each of the diluted homogenates were added into PDA culture dishes. CFU were counted after incubation on PDA supplemented with chloramphenicol at 28°C for 12 h. Values were marked as log₁₀ spore equivalents per gram of tissue.

For cytokine profiling, Cytometric Bead Array (CBA) kits (Becton Dickinson) were used to measure the levels of inflammatory cytokines in the supernatant of footpad homogenates from WT and Card9^−/− mice according to the manufacturer's instructions. Data were analyzed by FCAP Array software according to the “Guide to Analyzing Data from BD CBA Kits Using FCAP Array Software” at bdbiosciences.com/cbasetup (https://www.bdbiosciences.com/en-us/applications/research-applications/bead-based-immunoassays).

Chemokine profiles.

RNA was extracted from 3-dpi footpad homogenates using the E.Z.N.A Total RNA kit (Omega Bio-tek, Norcross, GA, USA), and cDNA was reverse transcribed using Prime-Script RT master mix (TaKaRa, Shiga, Japan). The gene expression of cxcl1, cxcl2, Il6, tnfa, T-bet, and ROR-γt was quantified by quantitative PCR (qPCR). The primers used are listed in Table 1.

TABLE 1.

Primers used for qPCR amplification

Gene	Sequence of:
	Forward primers	Reverse primers
Tnfa	5′-GCCTCTTCTCATTCCTGCTTG-3′	5′‐CTGATGAGAGGGAGGCCATT‐3′
Il6	5′‐ACGGCCTTCCCTACTTCACA‐3′	5′‐CATTTCCACGATTTCCCAGA‐3′
Cxcl1	5′‐GGCGCCTATCGCCAATG‐3′	5′‐CTGGATGTTCTTGAGGTGAATCC‐3′
Cxcl2	5′‐AAGTTTGCCTTGACCCTGAA‐3′	5′‐AGGCACATCAGGTACGATCC‐3′
T-bet	5′‐GCCAGGGAACCGCTTATATG‐3′	5′‐GGACGATCATCTGGGTCACAT‐3′
ROR-γt	5′‐TCTACACGGCCCTGGTTCTC‐3′	5′‐GCCTTGTCGATGAGTCTTGCA‐3′
gapdh	5′‐TCGTCCCGTAGACAAAATGGT‐3′	5′‐TCTCCACTTTGCCACTGCAA‐3′

NF-κB (p65) transcription factor assay.

BMDMs and their nuclear extracts were obtained as previously described (40). An NF-κB (p65) transcription factor assay kit (Cayman Chemical, USA) was used to detect specific NF-κB (p65) DNA-binding activity in the nuclear extracts following the manufacturer’s instructions.

Cell culture medium and reagents.

The medium for cell culture was RPMI 1640 complete medium (Gibco, Invitrogen, San Diego, CA, USA) containing 10% fetal bovine serum (FBS), 100 U/ml penicillin, and 100 g/ml streptomycin. Mouse recombinant GM-CSF and IL-4 were obtained from PeproTech (Rocky Hill, NJ, USA). LPS, curdlan, trypan blue and β-1,3-d-glucanase were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ionomycin was obtained from Enzo (Farmingdale, NY, USA). β-(1-3)-Glucan-specific monoclonal antibody and goat anti-mouse secondary antibody were from BioSupplies (Australia) and Life Technologies (USA). Dectin-1 soluble protein and anti-mouse dectin antibody were purchased from R&D Systems (USA).

β-(1-3)-Glucan staining in the fungal cell wall.

Inactivated resting spores and swollen spores were stained with β-(1-3)-glucan antibody (1 μg/ml) for 1 h, washed twice with PBS containing 1% goat serum and then stained with Alexa Fluor 488 goat anti-mouse secondary antibody for another 30 min, and washed twice with phosphate-buffered saline (PBS). Images were acquired in a Leisa TCS SP5 fluorescence confocal microscope (Leica, Micro-systems, GmbH).

Dectin-1 binding assay.

Inactivated resting spores and swollen spores were incubated with dectin-1 soluble protein (10 μg/ml) for 1 h, washed twice with PBS, and stained with phycoerythrin (PE) anti-mouse dectin-1 antibody for 30 min, and washed twice with PBS, and fluorescence was detected with a BD FACSCalibur flow cytometer.

Measurement of cytokine secretion by DCs.

Bone marrow cells were isolated by flushing the femur and tibia of mice with RPMI 1640 medium under aseptic conditions and cultured in RPMI 1640 complete medium containing GM-CSF and IL-4. The cells were harvested and purified using CD11c⁺ microbeads (Miltenyi Biotec, Auburn, CA, USA) on the seventh day. Ninety-six-well plates were seeded with DCs (2 × 10⁵ cells/well), and resting spores or swollen spores were cocultured with the DCs at a multiplicity of infection (MOI) of 2 or 1, respectively. The mock group was the negative control, and curdlan (100 μg/ml) and LPS (10 μg/ml) were prepared as positive controls. The supernatants were collected after 24 h, and the levels of the cytokines IL-1β, IL-12p70, and IL-23 were detected using mouse enzyme linked immunosorbent assay (ELISA) Ready-Set-GO! kits (eBioscience, San Diego, CA, USA).

Measurement of Th1/17 cell differentiation.

DCs and spores were added to 48-well plates as described above and cultured for 24 h. Naïve mouse CD4⁺ T cells were sorted using a naive CD4⁺ T cell isolation kit (Miltenyi Biotec, Auburn, CA, USA). The T cells were added into the coculture system containing immature DCs and spores (at a ratio of DCs to T cells of 1:10). Four days later, the cells were restimulated with phorbol myristate acetate (PMA; 50 ng/ml), ionomycin (500 ng/ml), and GolgiStop (10 μg/ml) for 5 h. The cells were stained with FITC-conjugated anti-mouse CD3, PE- and Cy5-conjugated anti-mouse CD4, PE-conjugated anti-mouse IL-17A, allophycocyanin-conjugated anti-mouse IFN-γ and isotype control antibodies (BD Pharmingen, USA). The percentage of Th1/Th17 cells was analyzed using flow cytometry. Supernatants from the DC and T cell coculture system were collected before restimulation, and the levels of IFN-γ and IL-17A were measured using a mouse IFN-γ ELISA kit and mouse IL-17A ELISA kit (Dakewei, Shenzhen, China), respectively.

Preparation of neutrophils, the NETosis assay, and qPCR.

Mature neutrophils were prepared as described previously (51). A 24-well plate was seeded with purified neutrophils (5 × 10⁴ cells/well) before stimulation with inactivated hyphae (MOI of 10). Propidium iodide (PI) was added 4 h later, and the cells were analyzed for NET release using microscopy. A 24-well plate was also seeded with purified neutrophils (1 × 10⁶ cells/well) before stimulation with inactivated swollen spores at an MOI of 2 overnight. RNA was extracted from the neutrophils, cDNA was reversed transcribed, and the gene expression of Il6 and tnfa was quantified by qPCR as described above.

Statistical analysis.

All the results are shown as the means ± standard errors of the means (SEM) and were calculated using GraphPad Prism (version 6.0). A two-tailed Student's t test was used to compare differences (significant at P < 0.05) between two groups, and one-way analysis of variance (ANOVA) was used to compare three or more groups. P values are represented as follows: ns, not significant; *, P < 0.05; **, P < 0.01; ***, P < 0.001; and ****, P < 0.0001.