CARD9 is involved in the microbiota balance, playing a critical role in the fungal landscape in the gut (Fig. 2). Malik et al. [21] co-housed CARD9-deficient mice and littermate controls during the AOM/DSS treatment protocol. Subsequently, 18S internal transcribed spacer and 16S rRNA sequencing was used to assay the fungal diversity between the wild-type (WT) and CARD9-deficient mice [21]. Under these conditions, WT mice displayed a mycobiota predominantly consisting of Tremellomycetes (Basidiomycota) and Leotiomycetes (Ascomycota). CARD9-deficient mice revealed a distinct mycobial landscape, including an increased number of Saccharomycetes, particularly Diutinacatenulata and Cladosporium of Dothideomycetes, along with the decreased presence of several members of Ascomycota, including Pseudocercospora cordiana of Dothideomycetes and Agaricomycetes and Claussenomyces of Leotiomycetes [21]. As compared to the control group, the anti-fungal treatment group had a similar shift in the mycobial landscape leading to a significant decrease in several members of Ascomycota, particularly Leotiomycetes, Helotiales, and Pseudocercospora (Dothideomycetes) [21]. In addition, Lamas et al. [24] explored the composition of the fungal microbiota in CARD9${}^{-/-}$ and WT mice. CARD9${}^{-/-}$ mice developed a higher fungal load at the colonic tissues, and reached a peak at day 7. The study further analyzed the genus composition of fecal fungal microbiota via high-throughput internal transcribed spacer 2 sequencing revealing a major difference in Monographella nivalis, Microchium gen, Agaricomycetes class, Microbotryomycetes class, Ascomycota phy, and Hypocrealesord, and Sporobolomyces gen. Of note, diversity measurements in CARD9${}^{-/-}$ mice exhibited a fungal microbiota dominated by members of the phyla Zygomycota, Basidiomycota, and Ascomycota. In a recent study, Wang et al. [10] reported the same results (i.e., that the total fungal burden in feces from tumor-bearing CARD9${}^{-/-}$ mice were significantly higher than that from tumor-bearing WT mice) [24]. Importantly, a significant difference regarding fungal biodiversity was observed between tumor-bearing WT and CARD9${}^{-/-}$ mice, showing the fungal microbiota in feces dominated by Candida (C. glabrata, C. tropicalis, and C. albicans) in tumor-bearing CARD9${}^{-/-}$ mice [24]. In accordance with previous work, CARD9 null mice were found have a higher susceptibility with an increased C. rodentium fecal load compared with WT mice [25]. Although it was difficult to distinguish the normal and diseased fungi due to the diversity of the gut fungi [26], there was increasing evidence that C. tropicalis acted as an inducer of CRC [10]. Thus, CARD9 may correlate with CRC by controlling the gut fungal landscape.

Fig. 2.

CARD9 controls the microbiome composition in the gut, consisting of commensal bacteria and fungi.

In parallel, additional work also demonstrated that CARD9 was associated with the composition of the gut bacteria (Fig. 2) [10, 21, 24]. Ankit et al. [21] revealed a distinct bacterial landscape in WT and CARD9 null mice, showing a relative increase in members of Mycoplasmataceae, Bacteroidaceae, Anaeroplasmataceae, Rikenellaceae, and Odoribacteraecae, and a relative decrease in Alphaproteobacteria RF32, Helicobacteraceae, Prevotellaceae, Ruminococcaceae, Paraprevotellaceae, Deferribacteraceae, Acetobacteraceae, and Phormidiaceae in the CARD9 null mice. Under the condition of cohousing mice, the abundance of Mycoplasmataceae was higher while Phormidiaceae and Alphaproteobacteria RF32 was lower in the CARD9 null mice housed with the WT mice [21]. Similarly, Lamas et al. [24] identified the composition of the bacterial microbiota in feces by sequencing the 16S rDNA and revealed similar biodiversity in WT and CARD9${}^{-/-}$ mice, but a decreased microbiota including Lactobacillus reuteri, Adlercreutzia (genus), and Actinobacteria (phylum) in the CARD9${}^{-/-}$ mice. Recently, a study demonstrated that the baseline faecal bacterial composition was different between CARD9${}^{-/-}$$\to$germ-free (GF) mice and WT$\to$GF mice, especially in an increase in Ruminococcus genera in the CARD9${}^{-/-}$$\to$GF mice [25]. GF mice are devoid of all microorganisms including bacteria, fungi and viruses. CARD9${}^{-/-}$ GF mice refer to the CARD9${}^{-/-}$ mice without microorganisms. CARD9${}^{-/-}$ mice refer to the CARD9${}^{-/-}$ mice with microorganisms. Different from the above reports, this study demonstrated no difference in the bacterial burden or composition between tumor-bearing WT and CARD9${}^{-/-}$ mice [10].

To determine the contribution of the microbiome in the gut, mice were depleted of anaerobic bacteria with oral administration of metronidazole before AOM-DSS administration [21, 27]. Fecal samples of these mice demonstrated a significant increase in Aerococcaceae, Enterobacteriaceae, Porphyromonadaceae, Lactobacillaceae, and Mycoplasmataceae, and a decrease in Rikenellaceae and S24_7 of Bacteroidales, Alphaproteobacteria, Deferribacteraceae. Of note, metronidazole treatment could depletecommensal anaerobes, led to distinct changes in the bacterial landscape, and eventually promoted CRC development [21]. Thus, CARD9 may correlate with CRC progression by controlling the bacterial landscape in the gut; however, further confirmation is needed.

C-type lectin receptors (CLRs), known as membrane-associated receptors, are critical for the detection of pathogenic and commensal fungal signals [31]. CARD9 functions as an adaptor molecule downstream of the CLRs. First, CLRs induce the activation of NF-$\kappa$B in a CARD9 dependent manner [32]. After CLR activation, Protein Kinase C $\delta$ (PKC$\delta$) is recruited to the immunological synapse. PKC$\delta$ enables CARD9 phosphorylation at T231 in the coiled-coil domain, forming the assembly of a Bcl10 and Malt1 proteins, termed the CARD9-Bcl10-Malt1 (CBM) complex. The CBM complex serves as a critical regulator to control the canonical NF-$\kappa$B pathway, which regulates the expression of a variety of immune-regulatory factors, including TNF-$\alpha$, IL-6, IL-1$\beta$, etc. Notably, chronic inflammation is linked to tumor carcinogenesis, which leads to inflammation-associated tumors [33]. NF-$\kappa$B modulates inflammatory signal in the tumor microenvironment indicating potential crosstalk between inflammation and tumor [34]. Second, CARD9 signal stimulates inflammasome-mediated cytokine IL-18 secretion and regulated T cell immunity in the fungal-sensing pathways [21, 35]. CLRs are activated by commensal gut fungi, which contributed to inflammasome production via the CARD9 signal. The inflammasome forms a platform to trigger the proteolytic processing of IL-18 [36]. IL-18 has been demonstrated to be critical to maintaining intestinal epithelial integrity and stimulating interferon gamma (IFN-$\gamma$) expression from intestinal CD8${}^{+}$ T cells [37, 38]. As expected, CARD9-deficient mice demonstrate greatly reduced inflammasome activation and IL-18 maturation and increased susceptibility to CRC [21]. Conversely, exogenous supplementation of IL-18 in CARD9-deficient mice has been shown to restore antitumor CD8${}^{+}$ T cell function, increase IFN-$\gamma$ production, ameliorate inflammation and reduce intestinal tumor development [21]. Third, in a study of the intestinal mycobiota, CARD9 deficiency increased the accumulation of myeloid-derived suppressor cells (MDSCs) in response to fungal dysbiosis [10]. CARD9${}^{-/-}$ macrophages exhibited impaired fungicidal functions, which led to an obvious increase in the gut fungal burden, particularly in C. tropicalis. Specific fungi could decrease the number of CD8${}^{+}$ and CD4${}^{+}$ T cells and promote the accumulation of T regulatory (Treg) cells and MDSCs in the colonic lamina propria, particularly with a notable increase in MDSCs [10]. It is well known that MDSCs are a group of immature myeloid cells that suppress the functions of effector T cells, negatively interfere with T cell-mediated antitumor immunity and are positively associated with the development of CRC [39, 40]. CARD9 is a central adaptor protein in anti-fungal immune systems via CLRs signal, which is involved in CRC carcinogenesis and progression.

CARD9 is associated with cellular immunity in patients with systemic fungal infections. Little is known about its antifungal antibody response. Recently, a study revealed that commensal C. albicans boosted the production of systemic antifungal IgG antibodies in a process dependent on CARD9 macrophages (Fig. 4) [41, 42]. C. albicans was identified as a strong inducer to elicit germinal centre B cell expansion in the spleen, which shapes the human antibody repertoire and induces the generation of high affinity antibodies. CARD9 deficiency is an inherited immune disorder that has been specifically linked to systemic fungal infections. Despite the normal function of germinal centre B cells, there is a failure to generate systemic antifungal IgG antibodies in the presence of a high systemic Candida burden in CARD9 deficiency. In contrast, increased expression of CARD9 was able to induce germinal centre B cells to produce systemic IgG antibodies against C. albicans [41, 42]. Thus, these results suggest that CARD9 signal in macrophages elicits germinal centre B cell expansion for the production of antifungal antibodies.

Fig. 4.

Commensal C. albicans modulates antifungal immunity by systemic antifungal IgG antibodies produced germinal centre B cells dependent on CARD9 expressing macrophages.

The receptor nucleotide-binding oligomerization domain containing protein 2 (NOD2) is an intracellular biosensor that can recognize a myriad of types of infectious bacteria. In this pathway, CARD9 signal plays a critical role in NOD2-mediated recognition of the microbiota [43]. Muramyl dipeptide (MDP), a bacterial cell wall component peptidoglycan produced Gram-positive bacteria, and whole Listeria monocytogenes, specifically activates NOD2, promotes the interaction of CARD9 with NOD2, leads to the activation of the JNK and MAP kinases p38 [43].

Mincle, a pattern recognition receptor, contributes to the detection of commensal bacteria and inflammation resolution in macrophages. Mincle has been shown to sense the Surface (S)-layer of the probiotic bacteria Lactobacillus brevis in a CARD9-dependent manner [44]. Lactobacillus is regarded as one of the major bacteria in the intestinal tract of mammals [45]. The macrophage inducible Mincle interacts with the S-layer of commensal bacteria that contributed to forming the Mincle/Syk/CARD9 axis. Importantly, this is accompanied by an altered release of both pro- and anti-inflammatory cytokines and impaired immune priming capacity of CD4${}^{+}$ T cells [44].

Toll-like receptors (TLRs) such as TLR3 and TLR7 on the cell surface detect viral infection [46]. In a study CARD9${}^{-/-}$ macrophages failed to produce IL-6 and TNF-$\alpha$ after treatment with poly (I:C) (a TLR3 ligand), loxoribine (a TLR7 ligand), but not PGN (a TLR2 ligand), LPS (a TLR4 ligand). This suggested that CARD9 was involved in the antivirus immune via TLR3 and TLR7 [43]. In addition, CARD9 was required for the recognition of viral nucleic acids via the nucleic acid receptors including RIG-I [47] and RAD50 [48, 49], which triggered the NF-$\kappa$B pathway.