1. Introduction
Cryptococcus neoformans is a globally distributed soil-borne fungus that causes pneumonia following inhalation of infectious propagules. Pulmonary cryptococcosis is typically mild or even asymptomatic; however, in some cases it may progress to meningoencephalitis, a morbid condition that is universally fatal if left untreated [
1].
C. neoformans is estimated to cause approximately 152,000 cases of meningitis annually with an estimated death toll of 112,000 [
2]. Among adults living with HIV in sub-Saharan Africa,
C. neoformans is the most common cause of meningitis and accounts for 15–20% of AIDS-related deaths.
C. neoformans is also increasingly recognized to cause disease in patients without HIV, including solid organ transplant recipients, patients receiving immunosuppressive therapy, and individuals who are otherwise considered immunocompetent [
3]. Cryptococcal disease is characterized by complex host–pathogen interactions that result in damage to both structural and immune cells and an effective crosstalk between lymphocytes and mononuclear phagocytes is required for fungal clearance [
4,
5].
Recognition of fungal pathogens by various families of host pattern recognition receptors (PRRs) including Toll-like receptors (TLR), C-type lectin receptors (CLRs), and intracellular Nucleotide Oligomerization Domain-like receptors (NLR) is a critical first step in activation of innate immunity against infection [
6]. Upon activation, transmembrane PRRs initiate intracellular signaling cascades to trigger cellular responses such as cytokine production and phagocytosis that result in microbial killing. Members of the CLR family including Dectin-1, Dectin-2, Dectin-3, Mincle, and the Mannose Receptor have been shown to recognize various cell wall components of pathogenic fungi [
7]. Following ligand recognition, CLRs recruit the intracellular Syk kinase that leads to the formation of a signaling complex consisting of the Caspase recruitment domain-containing protein 9 (Card9), B cell lymphoma 10 (Bcl10), and mucosa-associated lymphoid tissue lymphoma translocation protein 1 (Malt1) [
8]. Subsequent activation of Card9 results in the induction of the canonical transcriptional activator nuclear factor kappa beta (NF-kB) and activation of mitogen-activated protein kinases (MAPK). The result of this cascade is to stimulate the production of inflammatory mediators such as Il—6, Il—12, Csf2, Tnf, and Il—1β that are crucial for initiation of antifungal immunity.
Genetic deficiency in Card9 is considered an autosomal recessive primary immunodeficiency disorder (PID) which uniquely manifests as an extreme susceptibility to the development of severe fungal infection [
7,
9]. To date at least 24 missense and nonsense mutations have been identified in the promotor or protein-coding region of human Card9 [
10,
11]. Card9 deficiency was first reported in 2009 in a consanguineous family with severe infections caused by the normal human commensal organism
Candida sp. [
12]. Card9-deficient humans may develop persistent and recurrent mucosal
Candida sp. infections that are collectively termed chronic mucocutaneous candidiasis (CMC), and/or systemic infections that primarily manifest as fungal meningoencephalitis.
Candida infections of the kidney, liver, or spleen that are typically affected in patients with iatrogenic immunosuppression have not been reported in Card9-deficient patients [
13]. It is notable that Card9 deficiency is the only reported human genetic disorder where mucosal and systemic
Candida disease occur together in the absence of other non-fungal infections. In humans, mucosal candidiasis control requires IL-17 signaling via Th17 cells, gamma delta T-cells and innate lymphoid cell populations [
14]. Conversely, neutrophils are the most important effector cell in host defense against systemic
C. albicans infection and neutropenia is a well-established risk factor for this disease. Interestingly, human neutrophils require Card9 for killing unopsonized yeast cells via a reactive oxygen species-independent PI3Kγ-dependent pathway, while neutrophil-mediated killing of opsonized
C. albicans is mediated by FcRg and PKCd and is largely independent of Card9 [
15]. Thus, it appears that Card9 deficiency impairs both lymphoid and myeloid host defenses that mediate different mechanisms of protective immunity at mucosal and systemic sites, respectively [
9]. Card9 deficiency has also been associated with infection by a variety of other fungi including
Trichophyton sp.,
Phialophora verrucosa,
Exophiala sp.,
Aspergillus fumigatus,
Corynespora cassiicola, and others [
11].
Genetic deletion of Card9 in mice confers susceptibility to infection with several fungal pathogens [
7]. The role of Card9 has been most extensively studied using a
C. albicans infection model and has confirmed the crucial role that neutrophils play in host defense [
16].
In vivo, Card9 promotes neutrophil recruitment into fungal-infected organs via CXC chemokine production. For example, in the mouse brain where opsonization is naturally low, Card9 is required for the appropriate induction of CXC chemokines by resident macrophages (i.e., microglia) and glial cells, as well as recruited neutrophils [
16]. Elegant mechanistic studies have recently demonstrated that impaired control of
C. albicans infection in the mouse brain is due to defective neutrophil recruitment mediated by a lack of Card9-dependent Il—1b and Cxcl1 production by microglia following recognition of the fungus-secreted toxin Candidalysin [
16,
17].
Despite the availability of excellent mouse models that accurately recapitulate the pathogenesis of
C. neoformans pneumonia and meningitis, a clear understanding of the role that Card9 plays in host defense against this prevalent human pathogen is relatively limited. The first report to implicate Card9 in host susceptibility to
C. neoformans demonstrated a higher lung fungal burden at day 5 and day 14 post-infection with 10
6 colony-forming units of
C. neoformans 3501 in Card9-deficient mice [
18]. This observation was associated with reduced recruitment of interferon gamma-producing NK and memory phenotype T cells and reduced pulmonary expression of inducible nitric oxide synthase at day 3, but not at day 7, post-infection. Card9-deficient mice also had reduced expression of Il17a and the associated transcription factor RORγτ in lung homogenates at day 7 post-infection compared to wild-type mice; however, Il—17a-deficient mice actually had a significantly lower lung fungal burden at day 14 post-infection. The consequences of Card9 deficiency on
C. neoformans 3501 dissemination to other organs, including the brain, were not described in this report. A subsequent investigation using an avirulent
C. neoformans LW10 vaccine strain that overexpresses the zinc finger 2 transcription factor showed that Card9 was required for protection against progressive pulmonary infection and dissemination to the spleen but not the brain. Notably, no differences in lung leukocyte recruitment to infected organs were demonstrable; however, Card9-deficient mice had decreased M1 and increased M2 lung macrophage polarization following LW10 infection. Finally, following challenge with the highly virulent
C. neoformans H99 strain, wild-type C57BL/6 mice survived longer than Card9-deficient mice (median survival 21 days versus 26 days) but all mice succumbed by day 29 post-infection. No significant differences of fungal burden in the lungs, brain, or spleen were observed between wild-type and Card9-deficient mice in this model, nor were there any identifiable differences in lung leukocyte recruitment, anticryptococcal activity of macrophages or dendritic cells, or macrophage polarization [
19].
Taken together, these findings strongly suggest that Card9 regulates various aspects of host immunity against pulmonary cryptococcal infection such as lung macrophage polarization. Conversely, these results also raise several questions about the underlying mechanisms of disease, including an explanation for the lack of differential lung leukocyte recruitment, despite selectively higher chemokine expression (
Ccl2,
Ccl3,
Ccl11) and a higher LW10 fungal burden in Card9-deficient mice. In addition, the absence of differential fungal growth in the brain despite significant differences in survival of wild-type and Card9-deficient mice following LW10 or H99 infection is unexplained. Integration of these findings into a coherent model is limited by the use of diverse cryptococcal isolates, including a laboratory strain (B3501), an avirulent strain (LW10), or a highly virulent strain (H99). Finally, all of these studies were performed in the naturally susceptible C57BL/6 inbred mice that developed progressive
C. neoformans infection in the presence of a functional Card9 gene [
20,
21]. Thus, to gain further insight into the role of Card9 during the pathogenesis of cryptococcal pneumonia we challenged wild-type and Card9-mutant mice on the Balb/c background with a moderately virulent serotype D strain of
C. neoformans (52D) using well-established experimental conditions and analyzed host resistance and immune responses at serial time points up to day 28 post-infection. Our data demonstrate that Card9 is a crucial regulator of pulmonary fungal burden that is associated with differential expression of lung inflammatory mediators, immune cell recruitment, and T cell polarization, as well as fungal dissemination and growth in the brain, and survival.
2. Materials and Methods
Balb/c mice were purchased from Charles River Laboratories and subsequently bred and maintained in our specific-pathogen-free (SPF) facility. To generate mutant mice, a gRNA (TCTACTACCCTCAGTTATAC) designed to target exon 2 of the Card9 gene was purchased from Integrated DNA Technologies Inc. (IDT). The gRNA and Cas9 protein were complexed and microinjected together into the pronucleus of Balb/c embryos at the McGill Integrated Core for Animal Modeling (MICAM). These embryos were then transferred into pseudo-pregnant CD-1 females (Charles River Laboratories, Saint-Constant, QC, Canada) to generate potential F0 mice. Live pups were genotyped by PCR amplification of genomic DNA extracted from ear biopsies. Primers flanking exon 2 of the gene Card9 were used for the genotyping: Card9-fwd (GCAGGGCGCCTTATTCAATG) and Card9-rev (GGCTCCCCTTCTAGAGACCA). The PCR products were visualized on a 1% agarose gel and the products were sent for Sanger sequencing (Génome Québec, Montréal, QC, Canada). A founder male was found to have a 177-base pair (bp) deletion in one allele of the Card9 gene that includes part of intron 1–2 and part of exon 2. This mutation was germline transmissible and was named Card9em1Sq. The male founder was outcrossed to wild-type Balb/c females and heterozygous Card9em1Sq F1 mice were subsequently intercrossed to generate homozygous Card9em1Sq mice. All animals were maintained in compliance with the Canadian Council on Animal Care, and all experiments were approved by the McGill University animal care and use committee. All experimental groups described in this report include an equal number of male and female mice.
C. neoformans 52D (ATCC 24067) was grown and maintained on Sabouraud dextrose agar (SDA) (BD, Becton Dickinson and Company, Mississauga, ON, Canada). To prepare an infectious dose, a single colony was suspended in Sabouraud dextrose broth (BD) and grown to early stationary phase (48 h) at room temperature on a rotator. The stationary phase culture was then washed with sterile phosphate-buffered saline (PBS) and diluted to 2 × 105 CFU per ml in sterile PBS. The fungal concentration of the experimental dose was confirmed by plating a dilution of the inoculum on SDA and counting the CFU after 72 h of incubation at room temperature.
For intratracheal administration of C. neoformans, 8-week-old mice were anesthetized with 128 mg/kg of ketamine (Vetoquinol) and 6.4 mg/kg of xylazine (Elanco, Mississauga, ON, Canada) subcutaneously. Briefly, a 30½ needle mounted on a 1 mL syringe containing 2 × 105 CFU C. neoformans/mL was inserted into the trachea under direct vision and 50 mL of inoculum followed by 50 mL of air was dispensed into the lungs. The needle was removed immediately after injection and the incision was closed using coated vicryl 5–0 resorbable suture (Ethicon, Markham, ON, Canada). The mice were monitored daily following surgery.
After mice were euthanized with CO2, their infected lungs and brain were excised and placed in sterile, ice-cold PBS. Tissues were then homogenized using a glass tube and pestle attached to a mechanical tissue homogenizer (Glas-Col, Terre Haute, IN, USA), and plated at various dilutions on Sabouraud dextrose agar. Plates were incubated at room temperature for 72 h, and CFU were counted.
Following euthanasia, lungs were perfused with ice-cold PBS via the right ventricle of the heart. Using 10% neutral buffered formalin acetate (Sigma, Oakville, ON, Canada), the lungs were inflated to a pressure of 25 cm H2O and fixed overnight. Subsequently lungs were embedded in paraffin, sectioned at 5 μm, and stained with hematoxylin eosin (H&E), periodic acid-Schiff (PAS), or mucicarmine reagents at the Histology Facility of the Goodman Cancer Research Centre (McGill University, Montreal, QC, Canada). Representative photographs of lung sections were acquired with a ZEISS Axio Imager M2 microscope and ZEISS Zen 3.3 (blue edition) software (Zeiss, Dorval, QC, Canada).
For analysis of total lung cytokine and chemokine production, mice were euthanized and lungs flushed with 10 mL of ice-cold PBS. Whole lungs were homogenized in 2 mL PBS with Halt protease and phosphatase inhibitor cocktail (Thermo Scientific, Mississauga, ON, Canada) using a sterilized glass tube and pestle attached to a mechanical tissue homogenizer (Glas-Col, Terre Haute, IN, USA) and spun at 12,000 rpm for 20 min. Supernatants were collected, and aliquots were stored at −80 °C for further analysis. Total protein concentration of each sample was measured using Pierce BCA Protein Assay kit (Thermo Scientific). The whole-lung level of IFN-γ (DY485) was analyzed using a DuoSet enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN, USA). Levels of Kc/Cxcl1, Lif, Lix/Cxcl5, Mip-2/Cxcl2, Mip-1α/Ccl3, Eotaxin/Ccl11, G-Csf, Mcp-1/Ccl2, Vegf-a, Il—4, Il—5, Il—6, Il—1β, Il—10, Il—13, Il—17A and Tnf-α were quantified by multiplex ELISA using the MILLIPLEX Mouse Cytokine/Chemokine Magnetic Bead Panel (MCYTOMAG-70K, MilliporeSigma, Oakville, ON, Canada) according to the manufacturer’s protocol. Multiplex ELISA data were acquired with Luminex MAGPIX instrument and xPONENT platform, and analysis was performed on MILLIPLEX Analyst software (version 5).
To perform flow cytometry, the lungs and trachea were excised using sterile technique after flushing the blood with 10 mL of sterile PBS. Lungs were transferred to a sterile Petri dish and the lobes were injected with 5 mL of enzyme cocktail composed of 2 mg/mL Collagenase D (Roche, Laval, QC, Canada) and 80 U/mL DNase I (Roche, Laval, QC, Canada) in HBSS (Wisent, Saint-Jean-Baptiste, QC, Canada) through the trachea. Lung tissue (without the trachea) and any extra enzyme cocktail was transferred to a gentleMACs C tube (Miltenyi, Gaithersburg, MD, USA) and kept on ice until processing. Subsequently, a single cell suspension was obtained using the gentleMACS Octo Dissociator with heaters (Miltenyi, Gaithersburg, MD, USA) and its program 37C_m_LDK_1. Following dissociation, single cell suspensions were filtered through a 100-micrometre and 40-micrometre cell strainers (BD). Red blood cells were removed using 1× RBC lysis buffer (BioLegend, San Diego, CA, USA) before the cells were counted. Lung cells (3 × 106 cells) were stained with fixable viability dye eFluor780 (eBioscience [eBio], Mississauga, ON, Canada) at the concentration of 1:10,000 for 30 min (4 °C). The cells were then washed with PBS supplemented with 0.5% BSA (Wisent, Saint-Jean-Baptiste, QC, Canada) and incubated with anti-CD16/32 (93, eBio) at a concentration of 1:100 in PBS/0.5% BSA at 4 °C for 10 min. Single-cell suspensions were subsequently stained at 4 °C for 30 min with either an adaptive or innate immune system panel composed of fluorescence-conjugated anti-mouse monoclonal antibodies purchased from eBio, BD, and BioLegend (BL). Adaptive immune system panel: CD45-BV506 (30-F11; eBio), CD3e-PE-Cy7 (17A2; eBio), CD4-V421 (GK1.5; BL), CD8-PerCP-eF710 (53—6.7; eBio). Innate immune system panel: CD45-PE-Cy7 (30-F11; BL), CD11b-BV785 (M1/70; BL), CD11c-APC (N418; eBio), Ly-6G-BV421 (1A8; BL), Ly-6C-BV711(HK1.4; BL), F4/80-BUV395 (T45—2342; BD), SiglecF-PerCP-eF710 (1RNM44N; eBio), and NK1.1 (PK136; BL). All cells were subsequently washed with PBS/0.5% BSA and resuspended in 1% paraformaldehyde. Cells were acquired on BD flow cytometers (Canto II or Fortessa X-20) with FACSDiva Software (version 8.0.3). Analyses were performed using FlowJo software v.10.1 (TreeStar, Ashland, OR, USA). Gating was conducted using Fluorescence Minus One (FMO) controls.
For intracellular cytokine staining of T cells, lungs were processed as described above. Cells (5 × 106 cells) were plated and stimulated for 3 h with 1× of Cell Activation Cocktail (with Brefeldin A) (BL) that contains PMA (phorbol 12-myristate-13-acetate), ionomycin, and Brefeldin A. A control stimulation was also carried out without the presence of PMA/ionomycin in PBS but in the presence of brefeldin A (GolgiPlug; BD) for 3 h. Cells were then washed, blocked with anti-CD16/32 antibodies (93; eBio), and stained with a surface antibody cocktail consisting of the fluorescence-conjugated anti-mouse monoclonal antibodies CD3-BUV737 (17A2; eBio), CD4-FITC (RM4—5; BL), CD8-PerCP-eF710 (H35—17.2; eBio), and CD45-BUV395 (30-F11; BD). The cells were then fixed, permeabilized, and stained with IL-13-PE-Cyanine7 (eBio13A, eBio), IFN-g-APC (XMG1.2; BL), and IL-17A-PE (eBio1787; eBio). Cells were acquired on BD flow cytometer Fortessa X-20 with FACSDiva Software. Analyses were performed using FlowJo software v.10.1 (TreeStar). Gating was performed using Fluorescence Minus One (FMO) controls.
For statistical analysis of all experiments, the mean and standard error of the mean (SEM) are shown unless otherwise stated. To test the significance of single comparisons, an unpaired Mann–Whitney test was applied with a threshold p value of 0.05. A Log-rank (Mantel-Cox) test was performed to analyze the survival curves. The significance of the difference in brain dissemination rates was determined using a Fisher’s exact test. All statistical analyses were performed with GraphPad Prism software version 10.1 (GraphPad Software Inc., Toronto, ON, Canada).
4. Discussion
Infection by
Cryptococcus neoformans begins with the inhalation of poorly encapsulated infectious propagules that are detected by cells of the innate immune system [
19,
29]. Lung resident alveolar macrophages and dendritic cells mediate this function through expression of various pattern recognition receptors including members of the C-type lectin receptor (CLR) family [
30]. CLRs are a heterogeneous superfamily of soluble and transmembrane proteins with a characteristic C-type lectin domain that recognizes carbohydrate structures of the fungal cell walls such as glucans, mannans, and chitin.
C. neoformans is detected by the mannose receptor (CD206) [
31], Dectin-2 [
32], and DC-SIGN [
33] but is not recognized by Dectin-1, Dectin-3, and Mincle [
34,
35,
36].
Many PRRs use signaling and/or adaptor proteins that are shared with other members of the same family in order to activate inflammatory/innate immune responses [
37]. Detection of fungal infection by several host CLRs including Dectin-1, -2, -3 and Mincle triggers the recruitment and activation of spleen tyrosine kinase (Syk) that leads to the formation of a caspase recruitment domain-containing protein 9 (Card9)–B-cell lymphoma 10 (Bcl10)–mucosa-associated lymphoid tissue lymphoma-translocation gene 1 (Malt1) scaffold complex [
38]. Formation of the Card9–Bcl10–Malt1 complex can activate NFkB and MAP kinases which result in phagocytosis, DC maturation, and induction of proinflammatory cytokines. Human
CARD9 mutations are particularly associated with susceptibility to
Candida albicans infection of the central nervous system, al-though infection with other uncommon fungal strains has also been reported [
11]. Similarly, Card9
−/− mice are more susceptible to
C. albicans infection of the brain owing to defective neutrophil recruitment that is mediated via Card9-dependent production of IL-1β and CXCL1 by microglia [
17].
Several groups have reported that Card9-deficient mice are susceptible to respiratory infection with various cryptococcal strains. The first report to implicate Card9 in host resistance following high dose infection with
C. neoformans B3501 identified early and transient reductions in the recruitment of interferon gamma-producing NK and memory phenotype T cells to the lung; however, the origin, specificity, and function of the latter cell subset was not defined and extrapulmonary fungal dissemination or survival differences were not reported [
18]. A subsequent study found that
C. neoformans LW10, an attenuated and normally avirulent vaccine strain, causes a progressive and disseminated infection in Card9-deficient mice [
39]. Nevertheless, an unexplained lack of differences in leukocyte recruitment at infected sites sharply contrasts with observations showing a profound Card9-dependent defect in neutrophil recruitment following
C. albicans infection [
16,
17]. Finally, the shorter survival time of Card9-deficient mice following
C. neoformans H99 infection compared to controls was not attributable to fungal burden differences in the lungs, brain, or spleen, nor was it associated with differences in lung leukocyte recruitment, macrophage or dendritic cell anticryptococcal activity, or macrophage polarization [
39]. The demonstration of comparable cryptococcal burden in the brains of Card9
−/− and wild-type mice contrasts with the significantly increased
C. albicans load in the brains of Card9
−/− mice [
17]. Taken together, these observations point to a non-redundant role for Card9 in anti-cryptococcal host resistance, yet they also raise several important questions about how Card9-dependent immune mechanisms contribute to the pathogenesis of cryptococcal disease.
To further characterize the consequences of Card9 deficiency on anti-cryptococcal immune responses, we used a well-established model of intratracheal infection with 10
4 CFU of
C. neoformans 52D, a moderately virulent serotype D clinical isolate. We hypothesized that the effect of Card9 deficiency might be confounded by the natural susceptibility of C57BL/6 inbred mice to progressive infection with
C. neoformans 52D [
23]. Therefore, we used CRISPR-Cas9 technology to generate mice lacking exon 2 of the Card9 gene on the naturally resistant Balb/c genetic background (Card9
em1Sq) and characterized fungal growth, survival, inflammatory mediator production, and lung leukocyte recruitment at serial time points after infection.
The first notable observation of this study was the highly susceptible phenotype of Card9
em1Sq mice compared to the inbred Balb/c strain. In response to infection with 10
4 CFU of
C. neoformans 52D, Card
9em1Sq mice began to die at 20 DPI, and all had succumbed by 36 DPI; conversely, none of the Balb/c mice died during the same observation period. According to the damage response framework, the mechanism of host death could be attributable to uncontrolled fungal replication caused by a lack of host immunity, or could be due an exuberant or dysregulated host inflammatory/immune response [
40]. To evaluate these possibilities, fungal burden was quantified in the lungs and brain as a marker of local and disseminated infection, respectively. The pulmonary fungal burden was comparable between Card9
em1Sq and Balb/c mice up to 14 DPI, suggesting that the Card9 mutation does not significantly alter the innate immune response to
C. neoformans 52D. Relative to 14 DPI, the fungal burden of Balb/c mice was significantly reduced at 21 DPI and was 100-fold lower at 28 DPI, while no reduction was observed in Card9
em1Sq mice, suggesting that Card9 activates adaptive immunity to effectively control lung fungal growth.
Meningitis caused by dissemination of pulmonary C. neoformans infection to the central nervous system causes severe morbidity and mortality. To evaluate the contribution of Card9-dependent mechanisms to containment of pulmonary cryptococcal infection the incidence of dissemination and fungal burden in the central nervous system were determined following intratracheal C. neoformans infection. A comparable rate of dissemination and fungal burden were observed between Balb/c and Card9em1Sq mice at 7 and 14 DPI; however, at 21 and 28 DPI, Card9em1Sq animals had a significantly higher incidence of central nervous system dissemination as well as a markedly higher brain fungal burden. Thus, a lack of Card9-dependent mechanisms results in greater dissemination and replication of C. neoformans 52D in the brain of Balb/c mice. Previous reports have shown that human and mouse Card9 deficiency is associated with spontaneous and progressive cerebral nervous system candidiasis. Elegant mechanistic studies in mice demonstrated that Il-1b and Cxcl1 production by microglia in a Card9-dependent manner is required for recruitment of neutrophils that control cerebral C. albicans infection. Although Card9 deficiency has not been associated with human cryptococcal meningitis, the experimental findings in the current report suggest the role of Card9 in antifungal host defense of the brain may not be limited to Candida sp. Accordingly, it will be of interest to investigate the precise immune defects that are regulated by Card9 in the brain in response to disseminated cryptococcal infection and to compare these to experimental cerebral candidiasis.
Numerous studies have demonstrated that a Th1 or Th2 pattern of pulmonary inflammation is associated with resistance or susceptibility to progressive
C. neoformans infection (reviewed in [
41]). Specifically, classical (M1) macrophage polarization and interferon-gamma (Th1) production by CD4
+ T cells are two principal effector mechanisms for control and elimination of pulmonary cryptococcal infection while alternative (M2) macrophage polarization and interleukin-4 and/or interleukin-13 (Th2) production by CD4
+ T cells are permissive [
42]. Previous studies have shown that STAT1 activation within macrophages is required for M1 polarization and anti-
C. neoformans activity via the production of nitric oxide (NO) [
42,
43]. To determine whether progressive cryptococcal infection in Card9-deficient mice was associated with a Th2 pattern of host response, lung histology, inflammatory mediator production, and cellular recruitment to the lung were performed. The presence of numerous heavily encapsulated
C. neoformans in association with clear evidence of airway goblet cell metaplasia and mucus production in lung sections of Card9
em1Sq but not Balb/c mice is consistent with a Th2 response. The lungs of Card9
em1Sq mice also had significantly higher levels of Th2 cytokines and significantly lower levels of Th1 and Th17 cytokines compared to Balb/c mice. These observations suggest that Card9 may regulate M1 macrophage polarization in response to pulmonary cryptococcal infection through STAT1 activation. Characterization of STAT1 phosphorylation, M1- and M2-associated gene expression, as well as NO production, in macrophages from WT and Card9
em1Sq mice that have been infected with
C. neoformans could confirm this possibility. Th17-associated cytokines, including Il17a, are predominantly produced by neutrophils during pulmonary infection and contribute to disease resolution, but are not required for cryptococcal eradication [
44]. Neutrophil depletion prior to intranasal infection with
C. neoformans H99g (engineered to produce Ifn-γ) did not alter lung fungal burden but increased the level of Il17a in association with an increased frequency of Il17a
+ γδ T cells [
45]. Notably, Card9
em1Sq mouse lungs had significantly lower levels of all measured inflammatory mediators and chemokines, as well as a non-statistically significant increase in
Ccl2, indicating that Card9-dependent pathways regulate the overall intensity of the inflammatory and chemotactic response following pulmonary cryptococcal infection.
Defective cellular immunity is a major risk factor for progressive and/or disseminated cryptococcal infection and is most commonly observed with HIV infection and other immunosuppressive conditions that impair the number and/or function of CD4+ T lymphocytes. To determine whether Card9 deficiency alters the pattern of lung immune cell recruitment, flow cytometry was performed at serial time points after C. neoformans infection. Compared to the Balb/c strain, Card9em1Sq mice recruited fewer CD4+ T lymphocytes to the lung from 14–28 DPI. Phenotypic characterization by intracellular cytokine staining showed a higher and lower frequency of Il-13+ CD4+ and Ifnγ+ CD4+ cells in Card9em1Sq compared to Balb/c mice, respectively.
Myeloid-derived cells including monocytes and neutrophils are rapidly recruited to the lung following intratracheal cryptococcal infection [
45,
46]. Most experimental studies have shown that circulating monocytes are important to cryptococcal clearance and differentiate into monocyte-derived macrophages and dendritic cells that are capable of phagocytosing and destroying
C. neoformans [
47]. Conversely, failure to recruit these cells leads to Th2-type responses with increased lung fungal burden [
46]. Compared to monocytes, macrophages, and dendritic cells, the role of neutrophils is less well understood [
48]. In vitro, human and mouse neutrophils are highly effective at killing
C. neoformans through oxidative and non-oxidative mechanisms [
49,
50]. In mice, neutrophils have also been shown to capture and remove cryptococci from the brain vasculature [
51]; however, antibody-based depletion studies suggest that neutrophils may have a detrimental effect on the control of cryptococcal infection [
52]. Card9 is mainly expressed by cells of the myeloid lineage; accordingly, recruitment of these cells was analyzed at serial time points after cryptococcal infection. Card9
em1Sq mice recruited significantly fewer monocytes, dendritic cells, and neutrophils to the lung following
C. neoformans infection compared to Balb/c mice, confirming that this adaptor regulates myeloid cell responses. Notably, Card9
−/− mice also have markedly impaired neutrophil recruitment to the brain following
C. albicans infection [
16], suggesting that this signaling pathway is activated by both pathogens and mediates myeloid cell recruitment to a broad range of fungi. Finally, a trend towards greater lung eosinophil recruitment was observed in Card9
em1Sq mice and is consistent with the observed cytokine production pattern and Th2 cell polarization.
The current study builds on previous reports that have implicated Card9 in host resistance to C. neoformans infection. The choice of Balb/c mice was based on their naturally occurring relative resistance to C. neoformans and underlines the importance of host genetic background when analyzing immune responses to infection. Through a comprehensive analysis of Balb/c mice that have wild-type or mutant Card9 we have elucidated broad effects of this adaptor molecule on the innate and adaptive immune response to C. neoformans 52D, a moderately virulent clinical isolate that has been widely used to study pulmonary disease pathogenesis. Our data indicate that fungal recognition and signaling through Card9 activate innate and adaptive immunity in the lungs that controls dissemination and replication in the brain. Some strengths of our report include the use of CRISPR-Cas9 gene editing technology to precisely introduce a defined genetic modification in Card9 without affecting adjacent sequences, and the use of a well-established intratracheal infection model to comprehensively characterize lung inflammatory and immune responses. The present data also extend the findings from previous reports that either did not analyze disseminated disease, or did not identify differences in fungal burden, leukocyte recruitment to affected sites, or mortality. Our data suggest that Card9 is required for the control of fungal burden in the lungs as well as dissemination and replication in the brain through various mechanisms that include the expression of cytokines and chemokines, as well as the recruitment of myeloid and lymphoid cell subsets. The consequence of these broad Card9-dependent defects is universal mortality from an otherwise non-lethal infection model.
Our study has several limitations that should be pointed out. First, we used a single inbred genetic background and one well-characterized C. neoformans isolate for our studies; accordingly, these results may not be generalizable to other inbred mice or cryptococcal strains. It would be informative to analyze the role of Card9 on host resistance to a highly virulent strain such as C. neoformans H99 or C. gattii VGII on the Balb/c or other genetic backgrounds. Second, while a marked increase in brain fungal burden was identified in Card9em1Sq mice, we did not characterize the immune response in the central nervous system and cannot propose potential mechanisms of susceptibility as we have managed for infection of the lungs. Third, we did not determine which CLRs activate Card9 in response to intratracheal infection with C. neoformans 52D.