Abstract

Introduction

The human gastrointestinal tract represents an ideal habitat for the growth and proliferation of diverse microbial communities. Complex, cross-kingdom interactions underlie the ecology of these communities, which involve prokaryotes such as bacteria and archaea, eukaryotic viruses and phages, and eukaryotes such as protozoa and fungi. Large-scale efforts have made inroads in the characterization of the bacterial communities in the microbiota. Despite the enormous impact of fungi on healthcare, the environment and vegetation (Fisher et al., 2018), understanding of the impact of the fungal communities -the mycobiota- in mammalian health and disease has lagged behind. Historically, immunity to fungi has been explored in the context of fungal infections, where specific causative fungal agents and associated pathologies are better understood, as reviewed in recent publications (Brown et al., 2012; Lionakis et al., 2017; Lionakis and Levitz, 2018; Salazar and Brown, 2018). However, recent evidence suggests that commensal fungi at barrier surfaces can influence host immunity during homeostasis, and furthermore, can impact the course and severity of several immune-mediated diseases of inflammatory origin (Iliev and Leonardi, 2017; Limon et al., 2017; Mukherjee et al., 2015; Richard and Sokol, 2019).

Systematic studies of the mycobiota is limited by the paucity of technologies and bioinformatics platforms for the characterization of host-associated fungal communities. While this is an ongoing issue, there has been significant recent progress towards the characterization of the “mycobiome” in different human cohorts (Auchtung et al., 2018; Bittinger et al., 2014; Chehoud et al., 2015; Findley et al., 2013; Fujimura et al., 2016; Hoarau et al., 2016; Kalan et al., 2016; Lewis et al., 2015; Mar et al., 2016; Nash et al., 2017; Oh et al., 2016; Sokol et al., 2017; Tipton et al., 2018; Zuo et al., 2018). These efforts have focused on certain barrier surface sites, such as the gut and the skin, and data on other mucosal surfaces is limited. Different research groups use distinct methodologies for fungal nucleic acids isolation, sequencing and analysis; this has posed challenges in terms of data comparison. Nevertheless, data have become available for the same type of diseases in cohorts collected around the world, allowing for the definition of the core mycobiomes at different human body sites and how they change during inflammation (Auchtung et al., 2018; Bittinger et al., 2014; Chehoud et al., 2015; Findley et al., 2013; Fujimura et al., 2016; Hoarau et al., 2016; Iliev and Leonardi, 2017; Kalan et al., 2016; Lewis et al., 2015; Limon et al., 2017; Mar et al., 2016; Nash et al., 2017; Oh et al., 2016; Sokol et al., 2017; Tipton et al., 2018; Zuo et al., 2018).

Despite being an ongoing effort, the characterization of mycobiota composition has advanced further than the much-needed mechanistic studies on diseases of inflammatory origin where the mycobiota emerges as a factor influencing immunity-driven phenotypes. In this review, we will discuss recent progress made in this direction, with emphasis on the importance of establishing causative links and defining the mechanisms underlying the impact of intestinal fungi in host immunity in health and disease.

Fungi in the context of the gut microbiome

Fungi, bacteria and viruses are present at human barrier surfaces and are all affected by pathophysiological conditions (Gilbert et al., 2018). Shotgun metagenomics suggests that fungi constitute 0.01–0.1% of the human gut microbiome (Huffnagle and Noverr, 2013; Nash et al., 2017; Qin et al., 2010). However, there are no accurate estimates of the amount of fungi in the gut, and differences in biomass, genomic size and annotated genomes complicate this endeavor (Richard and Sokol, 2019). The different members of the gut microbiota are integrated into a functional trans-kingdom community, with complex interactions that determine population abundance and metabolic function (Fan et al., 2015; Lamas et al., 2016; Maraki et al., 1999; Mason et al., 2012; Samonis et al., 1993; Sovran et al., 2018; Zuo et al., 2018). Bacteria and fungi are exceptional producers of primary and secondary metabolites with biological and anti- or pro-microbial activities (Rooks and Garrett, 2016). The presence of intestinal bacteria limits fungal colonization of the intestine (Maraki et al., 1999; Samonis et al., 1993). This phenomenon is well described in the murine gut where colonization resistance prevents persistent colonization with the opportunistic pathogen Candida albicans, which can be achieved only following treatment with antibacterial agents (Fan et al., 2015; Jiang et al., 2017; Kim et al., 2014; Leonardi et al., 2018; Noverr et al., 2004; Shao et al., 2019). Antibiotics alter the composition of the intestinal mycobiota in both humans and mice (Lewis et al., 2015; Sovran et al., 2018). Importantly, the effect that members of the mycobiota exert on the host varies depending on the targeted bacteria (Sovran et al., 2018), supporting the existence of strong functional connections between bacterial and fungal communities in the gut.

Bacterial metabolites activate G protein-coupled receptors (GPCRs) and inhibit histone deacetylases (HDACs) to modulate both host immunity as well as other members of the intestinal microbiota (Chang et al., 2014; Morita et al., 2019; Smith et al., 2013). Some of these metabolites inhibit fungal colonization and/or growth. In particular, lactic acid and butyrate partially inhibit the growth of Candida species (Bulgasem et al., 2016; Nguyen et al., 2011). The HIF-1α agonist L-mimosine, released by gut commensal bacteria, can stimulate host intestinal epithelial cells (IECs) to produce anti-microbial peptides that inhibit Candida spp. growth in the intestine (Fan et al., 2015). Tryptophan-metabolizing bacteria, including lactobacilli, produce metabolites that activate the transcription factor Aryl hydrocarbon receptor (AhR). AhR activation stimulates interleukin (IL) −22 release by group 3 innate lymphoid cells (ILC) and Th cells thus promoting resistance to Candida spp. colonization in the gut (Kiss et al., 2011; Lamas et al., 2016; Zelante et al., 2013).

In contrast, other bacteria appear to promote the persistence of fungal species in the intestine. Escherichia coli and other Enterobacteriaceae can promote intestinal colonization by both Saccharomyces cerevisiae and C. albicans (Sovran et al., 2018). Similarly, Clostridium difficile infections (CDI) are accompanied by the expansion of C. albicans at the expense of other fungal species (Raponi et al., 2014; Zuo et al., 2018). Oral Candida administration in a Clostridium difficile mouse model worsened disease severity (Panpetch et al., 2019). As a dimorphic fungus, C. albicans can switch between yeast and hyphal form, depending on various environmental and nutritional factors. It is unclear which form predominates in the human gut although a recent study suggests that C. albicans is present in both hyphal and yeast forms in the murine gut (Witchley et al., 2019). In vitro, co-culture with C. albicans allows C. difficile to persist and proliferate under aerobic conditions (van Leeuwen et al., 2016). In turn, C. difficile produces p-cresol that reduces C. albicans virulence by inhibiting yeast-to-hypha transition and biofilm formation. Influencing the morphology of fungi might allow C. difficile to affect the interaction between Candida and other members of the microbiota. In vitro, C. albicans can form heterogeneous biofilms with various bacteria including E. coli (Bandara et al., 2009; El-Azizi et al., 2004) and hindering their formation might allow C. difficile to inhibit gut colonization by other bacteria. Consistently, C. difficile eradication following fecal transplant (FMT) is stymied by the presence of C. albicans in both the recipient and the donor feces (Zuo et al., 2018). Its ability to create optimal niches for the growth of particular bacterial species might also explain how gut colonization with C. albicans affects the recovery of the bacterial microbiota following antibiotic treatment (Mason et al., 2012; Panpetch et al., 2019). These findings suggest that members of the gut mycobiota can exert a form of colonization resistance preventing the colonization of the intestine by exogenous bacteria – and possibly other microbes. However, the vast majority of the mechanistic studies, however have been performed with a limited set of fungal strains. The influence that the mycobiome as a whole exerts on the host is still in its infancy, as is the question of how the collective microbiome interacts to affect pathophysiological conditions.

Cellular mechanisms of fungal sensing and innate recognition in the gut

The dense microbial colonization at barrier surfaces presents the host immune system with the challenge of preventing aberrant host immune responses towards these organisms and their metabolic products while ensuring a balanced microbial composition. Intestinal epithelial cells are the first crucial physical barrier interacting with commensals and are also an important producer of anti-microbial peptides and immune mediators that regulate immune homeostasis, microbiota composition and host defense (Allaire et al., 2019; Peterson and Artis, 2014). In recent years, many studies have paved the way to a better understanding of how the immune equilibrium between commensal fungi and the oral epithelium is maintained (Moyes et al., 2016; Naglik et al., 2017; Richardson et al., 2018; Verma et al., 2017). The dimorphic fungus C. albicans is the most common member of the oral and intestinal mycobiota and is associated with several human inflammatory conditions (Bacher et al., 2019; Leonardi et al., 2018). C. albicans yeasts interacts poorly with epithelial cells, while C. albicans hyphae invade epithelial cells and induce pro-inflammatory cytokines and host defense peptides (Moyes et al., 2016; Verma et al., 2017). Interestingly, transition to a hyphal reduces the colonization potential of C. albicans (Roman et al., 2018) possibly through the increased expression of hyphal-specific genes that promote anti-fungal immunity (Witchley et al., 2019).

How C. albicans and other fungi interact with IECs in the gut remains largely unknown. Candidalysin, a pore-forming peptide encoded exclusively by C. albicans hyphae-associated ECE1 gene, is a crucial virulence factor that directly induces oral and vaginal epithelium damage, as well as innate immune cells activation in the host (Moyes et al., 2016; Richardson et al., 2017). Upon epithelial cells damage, candidalysin induces the activation of MAPK/cFos/MPK1 pathway, which further lead to the production of pro-inflammatory cytokines, including IL-1α/β, IL-6, IL-8, G-CSF and GM-CSF in both oral and vaginal squamous epithelial cell lines in vitro (Moyes et al., 2016; Richardson et al., 2017). The gastrointestinal columnar epithelium is very different from both the oral and vaginal squamous epithelium. Although candidalysin was recently found to induce cell damage in human epithelial colorectal adenocarcinoma (Caco-2) cells in vitro (Allert et al., 2018), the potential effects of candidalysin in the gut warrants further investigation (Figure 1).

Open in a separate window

Figure 1:

Cellular mechanisms of innate and adaptive immune responses to gut commensal fungi.

Gut resident CX3CR1⁺ MNPs sense and phagocytose intestinal fungi, and respond by producing cytokines that support antifungal Th17 responses in a Syk dependent manner. This process further recruits neutrophils to the intestine that together with CX3CR1⁺ MNPs prevent the expansion of opportunistic fungi. Fungal-specific Treg cells against gut commensal fungi, such as C. albicans, have been detected in humans), but not in mice and might be involved in the maintenance immune homeostasis. Intestinal disease results in bacterial and fungal dysbiosis, with changes in the abundance of various fungal species. Intestinal fungi such as C. albicans expand and directly interact with intestinal epithelial cells (IECs). In vitro studies suggest that hyphal form of C. albicans can induce lECs damage via the toxin candidalysin. C. albicans can induce the expansion of fungal antigen-specific Th17 cells that can cross-react to other fungi such as A. fumigatus and are enriched in several inflammatory diseases targeting the gut and the lung. Increase of anti-fungal antibodies is associated with intestinal inflammation and the induction of these antibodies is partially dependent on CX3CR1⁺ MNPs. * Phenomenon observed in human, but not in mouse studies.

Intestinal mononuclear phagocytes (MNPs) play an important role in the regulation of mucosal immune homeostasis and host defense through sensing food and commensal antigens from the intestinal lumen (Hadis et al., 2011; Leonardi et al., 2018; Mazzini et al., 2014; Medina-Contreras et al., 2011; Niess et al., 2005). CX3CR1⁺ MNPs, derived from CCR2⁺Ly6C^hi blood monocytes cells (Bain et al., 2014), represent the majority of intestinal macrophages in the healthy intestine (Figure 1). CX3CR1⁺ MNPs ensure the establishment and maintenance of immune tolerance by expressing high levels of IL-10 and promoting the differentiation and generation of Tregs (Kim et al., 2018; Viladomiu et al., 2017). However, in response to specific intestinal bacteria, such as Salmonella typhimurium and segmented filamentous bacteria (SFB), CX3CR1⁺ MNPs activate Th17 adaptive immune responses (Kim et al., 2018; Panea et al., 2015). Besides their ability to recognize intestinal bacteria CX3CR1⁺ MNPs express anti-fungal C-type lectin receptors (CLRs), such as dectin-1, dectin-2, and Mincle, and can recognize and intake intestinal yeast and filamentous fungi. In the gut, CX3CR1⁺ MNPs induce fungal antigens specific Th17 responses to C. albicans via the activation of Syk signaling (Leonardi et al., 2018) (Figure 1). Other intestinal phagocytes, such as lysozyme-expressing dendritic cells (Lyso-DCs) and CD11b⁺ DCs (Figure 1), express Mincle that can sense mucosal bacteria commensals to induce Th17 differentiation in a Syk-dependent manner (Martinez-Lopez et al., 2019). Whether these phagocytes play a role in the regulation and priming of T cells responses to gut commensal fungi remains to be investigated. How intestinal phagocytes recognize and sense the commensal and pathogenic fungi is clinically relevant to the pathogenesis of several inflammatory diseases, a topic discussed later in this review.

Adaptive immune responses to gut commensal fungi

Historically, one of the first readouts of the immune effect of intestinal fungi has been the induction of antibodies reactive to S. cerevisiae mannan (ASCA). Serum ASCA titers are increased in Crohn’s disease (CD) patients and in a variety of other immune mediated diseases (IMDs) (Muratori et al., 2003; Papp et al., 2010; Standaert-Vitse et al., 2006; Standaert-Vitse et al., 2009) (Table 1). To date, the physiological role of anti-fungal antibodies and the mechanism of their induction remain unknown. ASCA can be triggered by C. albicans mannosides. However, the levels of C. albicans in the feces correlate to ASCA titers in healthy subjects but not in their first-degree CD relatives (Schaffer et al., 2007; Standaert-Vitse et al., 2009). ASCA antibodies development does not appear to be generated as a result of a genetic defect but rather occur during the early stages of disease (Arbuckle et al., 2003; Israeli et al., 2005). This suggests that the alterations in the mucosal physiology occurring during CD and other IMD might alter the mechanisms of anti-fungal antibodies generation.

Analysis of ASCA-high fraction of CD serum unveiled their cross-reactivity with several fungal antigens including more abundant (C. albicans) and less abundant fungal genera (Aspergillus and Malassezia) (Leonardi et al., 2018; Limon et al., 2019; Standaert-Vitse et al., 2009). Antifungal IgG antibodies with specificity to C. albicans detectable in the serum of CD patients co-occurred with antibodies recognizing several mycobiota members belonging to the Saccharomycetaceae family including C. parapsilosis, S. cerevisiae and P. kudriavzevii (Leonardi et al., 2018). Yet, these antibodies do not co-occur with common airborne and skin fungi that have low presence in the gut, such as Aspergillus spp. and Malassezia spp. (Leonardi et al., 2018), suggesting gut priming by members of the gut mycobiota that have the ability to induce an unique immune response.

A recent study described a positive correlation between elevated uric acid (UA) levels and increased ASCA titers in healthy individuals (Chiaro et al., 2017). Follow up experiments in murine model of colitis suggested a role of S. cerevisiae in UA induction and aggravated intestinal inflammation (Chiaro et al., 2017). However, the decreased levels of S. cerevisiae in the feces and mucosa of IBD patients (Chehoud et al., 2015; Hoarau et al., 2016; Li et al., 2014; Liguori et al., 2016; Sokol et al., 2017) as well as the lack of correlation between UA and ASCA levels during human intestinal disease (Sendid et al., 2018) suggest that this phenomenon is limited to homeostatic conditions.

Type 17 immunity is the characteristic response to several gut “pathobionts” and commensals including C. albicans (Atarashi et al., 2011; Leonardi et al., 2018) (Figure 1). Multiple studies in mice and humans have demonstrated the importance of Th17 cells in multiple models of mucosal fungal infections: a topic extensively reviewed elsewhere (Conti and Gaffen, 2015; Iliev and Leonardi, 2017; McGeachy et al., 2019). In vivo, most experiments have focused on C. albicans and further research is needed to assess whether other members of the intestinal mycobiota can also induce Th17 responses at steady state. Importantly, Th17 cells that are induced in the gut and mesenteric lymph nodes upon intestinal Candida colonization aroused in the absence of apparent signs of gut fungal infection (Atarashi et al., 2015; Leonardi et al., 2018; Shao et al., 2019). This suggests that in addition to their role in antifungal immunity during mucosal fungal infections, fungi-induced Th17 cells might play additional roles in homeostasis and inflammation.

Indeed, recent studies have begun to uncover an increased level of plasticity and cross-reactivity of IL-17-producing CD4⁺ and CD8⁺ cell-mediated immunity to commensals. S. epidermidis or C. albicans induced type 17 program in skin-resident commensal-specific T cells. Deeper characterization of these commensal-specific T cells revealed the co-expression of both type 17 and type 2 programs and the production of IL-17, IL-22, IL-5 and IL-13 (Harrison et al., 2019). The induction of these classically antagonistic responses in the same T cells likely arises to ensure both type 17 immunity–mediated antimicrobial defense as well as type 2 immunity-related tissue repair upon tissue injury at a barrier surface(Harrison et al., 2019).

Interestingly, similar plasticity and functional heterogeneity of C. albicans specific memory T cells occur in healthy humans (Becattini et al., 2015) where C .albicans–specific Th17/ Th1 and Th17/Th2 share a high extent of clonotype (Becattini et al., 2015). Although the function of these cells is unknown, their presence in both humans and mice might suggest a possible function in tissue repair. Interestingly, C. albicans induces Th17 cells in the mouse gut upon intestinal colonization independently of the preexisting gut mycobiome composition (Atarashi et al., 2015; Doron et al., 2019; Leonardi et al., 2018), while Th2 cells appear in the gut and lung during filamentous fungal expansion followed by allergy and lung damage (Li et al., 2018). Cross-reactivity of fungal antigens is yet another mechanism by which fungal antigen-specific Th17 cell might expand. Circulating C. albicans-specific Th17 are found in the peripheral blood of healthy individuals (Bacher et al., 2013; Becattini et al., 2015) and Candida spp. are present in the gut of many healthy individuals (Iliev and Leonardi, 2017; Liguori et al., 2016). C. albicans is the major inducer of Th17 cells in humans among 30 members of the human mycobiome (Bacher et al., 2019). These C. albicans-specific Th17 cells are further cross-reactive to other fungal species, including the airborne fungus A. fumigatus (Bacher et al., 2019).

Recent evidence suggests that, besides bacteria, gut commensal fungi might also play a role in the induction of immune tolerance (Figure 1). Regulatory T cells (Tregs) are implicated in the maintenance of mucosal immune homeostasis. In addition to Tregs developing in the thymus (nTregs), peripheral Treg cells (iTregs) differentiate in the mouse and human gut in response to oral antigens and commensal microbiota (Cebula et al., 2013; Coombes et al., 2007; Iliev et al., 2009; Josefowicz et al., 2012; Lathrop et al., 2011; Mucida et al., 2007; Sun et al., 2007). A mixture of Clostridia strains belonging to cluster IV, XlVa, and XVIII and derived from mice or healthy human donors is a potent inducer of Tregs in the murine colon (Atarashi et al., 2013; Atarashi et al., 2011).

Oral, vaginal and systemic infection with the C. albicans are associated with the induction of Treg cells and protection from disease. In these cases (De Luca et al., 2013; De Luca et al., 2007; Pandiyan et al., 2011), Foxp3⁺ Tregs exert their protective effect against C. albicans infection by enhancing antifungal T helper 17 (Th17) cell responses leading to increased fungal clearance (Pandiyan et al., 2011; Whibley et al., 2014). These findings suggest that the expansion of Treg cells induced by gut Candida might have a supportive role to the more dominant antifungal Th17 responses. In contrast to oral, vaginal and systemic infection, intestinal colonization with C. albicans do not have noticeable impact on the frequency of total Tregs in the murine intestine (Leonardi et al., 2018; Shao et al., 2019). Whether C. albicans colonization can affect the total number of fungal-specific Tregs is still unclear.

Interestingly, high numbers of CD4⁺CD25⁺CD127⁻Foxp3⁺ Tregs specific to both the gut commensal C. albicans and the airborne Aspergillus fumigatus circulate in the blood of healthy individuals (Bacher et al., 2016; Bacher et al., 2014). These cells present phenotypic, epigenetic and functional features of thymus-derived nTreg cells but are outnumbered by fungal-antigen specific Th2 cells in patients with Aspergillus-associated lung allergy. A. fumigatus-specific Tregs from adult peripheral blood have a memory phenotype and are undetectable in cord blood (Bacher et al., 2016), indicating their development later in life upon fungal antigen exposure. Interestingly, Tregs specific to several gut commensal bacteria are rare and variable among different donors (Bacher et al., 2016). Despite being focused on two fungal species, these studies reveal the presence of fungal-specific Tregs in the periphery with the lungs and the gut being possible routes for their induction (or of antigen exposure).

Gut mycobiota regulates immune homeostasis: mechanisms of interplay to consider

Evidence for the immunoprotective role of the gut mycobiota comes from recent studies showing that targeted perturbation of gut fungi by anti-fungal drugs has persistent effects on host immunity and health through mechanisms discussed later in this review (Li et al., 2018; Wheeler et al., 2016). Macrophages and monocytes possess memory properties to fungal cell wall constituents such as β-glucan and chitin, and can protect mice from a secondary challenge with fungi: a phenomenon referred to as “trained immunity” (Figure 2) (Cheng et al., 2014; Rizzetto et al., 2016; Saeed et al., 2014). β-glucan signaling through dectin-1 induce long non-coding RNAs (IncRNAs) that increase the expression of immune genes such as IL-6 and CSF1 (Fanucchi et al., 2019). Aerobic glycolysis is considered the metabolic basis for trained immunity and is induced through the dectin-1/AKT/mTOR/HIF-1α pathway where a new study showing NFAT dependency (Cheng et al., 2014). Trained immunity is suppressed in monocytes from dectin-1–deficient patients (Cheng et al., 2014). Interestingly, trained immunity can be induced in mice by intestinal colonization with a human blood isolate of C. albicans that has been adapted to the murine gut by sequential passages through the gastrointestinal tract (GI) of antibiotic-treated mice (Tso et al., 2018). In conditions promoting fungal intestinal domination triggered by antibiotics, multiple microscale genetic changes increase the in vivo fitness of C. albicans within the host, leading to the acquisition of commensal-like traits (Ene et al., 2018; Liang et al., 2019; Tso et al., 2018). This gut-adapted C. albicans strain confers enhanced protection against systemic challenge with several fungal and bacterial pathogens including C. albicans itself(Tso et al., 2018). The protection mechanism bares several features of trained immunity (Cheng et al., 2014; Saeed et al., 2014), as the effect is short lasting, IL-6 –dependent and can be recapitulated in T- and B-cell deficient-Rag1 knockout mice. However, in a similar model, a human blood C. albicans isolate also protects against a subsequent systemic C. albicans infection, largely due to the induction of systemic adaptive Th17 responses (Shao et al., 2019) (Figure 2). In addition to protection from systemic fungal infections, gut colonization with C. albicans also provides protection against systemic Staphylococcus aureus infection; whether this protection depends on trained immunity or adaptive Candida-specfic-Th17 cells, or a combination of both, remains unknown (Shao et al., 2019; Tso et al., 2018). These results show that both classical adaptive immune responses and innate immune memory can be primed by C. albicans in the gut and suggest that strain specific features might inform the host immune response. The relative importance of innate versus adaptive immunity in conferring this protective phenotype is still unclear.

Open in a separate window

Figure 2:

Gut mycobiota initiate trained immunity and adaptive immune responses to protect against systemic bacterial and fungal infections.

Members of the gut mycobiota, such as C. albicans and S. cerevisiae, are protective against systemic bacterial and fungal infections via trained immunity and/or activation of adaptive Th17 immunity. Fungal cell wall constituents such as b-glucan and chitin can enhance aerobic glycolysis in monocytes through the activation of dectin-1/ AKT/mTOR/HIF-1α pathway, leading to pro-inflammatory cytokines production, such as IL-6 and CSF1. In addition, S. cerevisiae derived mannans can initiate systemic protection against DSS induced intestinal inflammation in mice in the absence of commensal bacteria. Thus, mycobiota mediated trained immunity provides systemic immune-protection against systemic infection with bacterial and fungal pathogens. Gut commensal C. albicans also provides protection against systemic C. albicans, A. fumigatus and S. aureus by priming adaptive Th17 mediated-immune responses.

Recent studies suggest that innate and adaptive immune mechanisms likely collaborate to maintain homeostasis. Mono-colonization of germ-free mice with C. albicans or S. cerevisiae, two species recognized by CX3CR1⁺ MNPs in the gut (Leonardi et al., 2018), promote the establishment of intestinal homeostasis (Jiang et al., 2017). Both of these fungal species exert a protective effect against virus-induced lung inflammation in mice when commensal bacteria are depleted with a combination of broad-spectrum antibiotics (Jiang et al., 2017). However, in mice treated with only ampicillin, C. albicans did not confer additional protection (Shao et al., 2019). This suggests that some fungi might “substitute” the beneficial effect of commensal bacteria in at least some cases such as during profound perturbations of the bacterial microbiota. In such case, C. albicans protects mice in a TLR4 dependent manner from sepsis-like syndrome induced by DSS treatment of mice exposed to broad-spectrum antibiotics (Jiang et al., 2017). Similar protection is conferred to mice by eukaryotic viruses (Kernbauer et al., 2014), although different pathways appear to be involved. In additional studies, daily administration with C. albicans exacerbated the outcome of DSS-induced colitis in SPF mice but had no effect following antibiotic administration (Sovran et al., 2018). Similar discrepancies in protective vs detrimental effects have been reported upon colonization with probiotic S. cerevisiae that is protective during DSS-induced colitis following antibiotic treatment (Jiang et al., 2017) but detrimental in conventional SPF mice (Chiaro et al., 2017). Enterobacteria colonization might contribute to the protective effect of S. cerevisiae in DSS-induced colitis (Sovran et al., 2018), suggesting that the pre-existing enteric microbiota (fungi and bacteria) might further modulate the outcome of intestinal inflammation and protective immunity. How these findings apply to humans that are constantly exposed to fungal and bacterial antigens, remains unclear.

Altogether, these studies suggest that a balanced gut mycobiota contributes to the maintenance of host immune homeostasis that the protective effect of intestinal bacteria can be mimicked by fungal over-colonization during severe dysbiosis and that cross-kingdom interactions might play important role in this process.

Immunity to gut mycobiota and Inflammatory Bowel Disease

Immune-mediated diseases are diseases of complex etiology and presentation that affect 7-10% of the western-population with few effective therapies (El-Gabalawy et al., 2010). A growing number of studies have reported shifts in the composition of the gut fungal communities in patients affected by IMD (Sokol et al., 2017). Yet, these diseases are accompanied by profound physiological changes (such as inflammation, changes in gut permeability, bacterial and viral dysbiosis) that profoundly alter both the host immunity and the environmental niche where the fungi reside.

The healthy human gut is colonized by a diverse community of fungi, mostly belonging to the Ascomycota and Basidiomycota Phyla (Hallen-Adams and Suhr, 2017). Among the taxa consistently identified in several human cohorts, Candida and represent the most prevalent and abundant genera. Further genera include Pichia, Saccharomyces, Cladosporium, Malassezia, Aspergillus and Penicillium. Whereas the Candida spp, observed in the human gut seem to be exclusively found in the mammalian gut, the other species are commonly found in food, skin or in the environment and might therefore not be true colonizer (Hallen-Adams and Suhr, 2017). Both CD and UC patients are clearly affected by mycobiota dysbiosis (Table 1) (Hoarau et al., 2016; Sokol et al., 2017). IBD patients often experience fungal outgrowth (Sokol et al., 2017) and changes in gut mycobiota richness and diversity (Chehoud et al., 2015; Li et al., 2014). The characterization of the fungal communities has been performed predominantly in CD patients or in a mixed UC and CD patient population. The changes in gut fungal communities in UC patients have not been independently studied. Active inflammation seems to promote changes in the mycobiota composition (Liguori et al., 2016; Sokol et al., 2017), an observation replicated in mouse models (Lamas et al., 2016; Qiu et al., 2015). Importantly, these changes are also observed in the mucosa-associated mycobiota (Liguori et al., 2016). Mycobiota dysbiosis is observed also in treatment naive CD patients (Chehoud et al., 2015). In these newly diagnosed patients, CD associated taxa belonged to the Basidiomycota phylum and all depleted taxa belonged to the Ascomycota phylum. This shift is also found in the general IBD patient population (Sokol et al., 2017), suggesting an inflammation-driven perturbation of the core fungal community that is independent of the IBD therapeutic regimen. Malassezia sequences are increased in the intestinal washings from a cohort of CD patients with respect to healthy controls. M. restricta gastrointestinal delivery in mice aggravates the outcome of DSS-induced colitis in a CARD9-dependent manner (Limon et al., 2019). In the mouse skin, Malassezia colonization induces Th17 responses that are essential to prevent fungal outgrowth (Sparber et al., 2019). Consistent with their site of induction, Malassezia-responsive memory Th17 cells are also found in the circulation of atopic dermatitis patients (Sparber et al., 2019).

Among the gut fungi differentially affected by IBD, Candida spp. appear to be consistently increased in IBD and to be associated with active disease (Li et al., 2014; Sokol et al., 2017). In particular, C. albicans (Mar et al., 2016; Sokol et al., 2017) and C. parapsilosis appear consistently increased in various cohorts, including CD patients with different ethnicity (Chehoud et al., 2015; Li et al., 2014; Ott et al., 2008) (Table 1). In contrast to C. albicans, Saccharomycesspp. including S. cerevisiae is reduced in the feces of IBD patients, and its abundance is decreased in patients with active inflammation (Sokol et al., 2017). S. cerevisiae abundance is also reduced in the intestinal mucosa of CD patients (Chehoud et al., 2015; Hoarau et al., 2016; Li et al., 2014; Liguori et al., 2016). These findings suggest the fungal dysbiosis is associated with IBD and that Candida species are consistently associated with the inflamed gut.

Importantly, CD patients have higher frequencies of circulating C. albicans reactive T cells when compared to healthy subjects (Bacher et al., 2019) (Table 1). These cells appear to be cross-reactive to the airborne fungus A. fumigatus. These findings suggest that intestinal C. albicans can induce cross-reactive Th17 cells in patients with CD and might be associated with the progression and development of intestinal inflammation. In addition to the direct impact on gut innate immune pathways and on the induction of humoral immunity, the recognition of C. albicans by CX3CR1⁺ MNPs, also triggers antifungal Th17 responses (Figure 1)(Leonardi et al., 2018) and CX3CR1⁺ MNPs might be involved in the induction of these cross-reactive Th17 cells.

A missense mutation of the CX3CR1 gene, termed T280M, impairs the fractalkine-mediated survival of circulating blood monocytes in humans (Collar et al., 2018). This mutation is associated with an increased susceptibility to systemic candidiasis in patients (Brand et al., 2006) highlighting the important protective role of monocytes/macrophages in the control of blood stream fungal infection. In mice, fractalkine binding to CX3CR1 promotes the survival of monocyte and macrophage in various tissues, including the colon (Landsman et al., 2009; Medina-Contreras et al., 2011; Peng et al., 2015). Furthermore, this mutation is associated with the development of intestinal stenosis and ileocolonic involvement in CD patients (Brand et al., 2006). Importantly, the T280M mutation is also associated with impaired anti-fungal humoral responses in CD patients (Leonardi et al., 2018). In particular, T280M patients have decreased titers of serum IgG against various gut mycobiota members including species such as C. albicans and C. parapsilosis (Leonardi et al., 2018) that are associated with intestinal inflammation (Sokol et al., 2017). These finding are corroborated by studies in mice where the depletion of CX3CR1⁺ MNPs causes a reduction in serum anti-fungal IgG despite a concomitant increase of the fungal burden (Figure 1) (Leonardi et al., 2018). Whether the ability to mount antibody responses against fungi plays any role in disease outcome or/and the control of mycobiota.

Thus far, mechanistic mouse studies have been performed with a limited number of model fungal strains largely belonging to C. albicans, S. cerevisiae and M. restricta species. As discussed above, experimental studies have reported contrasting outcomes (detrimental or protective) of fungal colonization on the development of intestinal inflammation in immunosufficient C57BL/6J mice (Chiaro et al., 2017; Jiang et al., 2017; Leonardi et al., 2018; Limon et al., 2019; Sovran et al., 2018). Several characteristics of model C. albicans strains have been shown to influence disease outcomes (Jawhara et al., 2012; Marakalala et al., 2013).

When bacterial communities are not disturbed by antibiotic treatment, defects in specific antifungal immunity pathways are the main driver of susceptibility to intestinal inflammation in several mouse models of fungal colonization and colitis (Iliev et al., 2012; Leonardi et al., 2018; Tang et al., 2015; Wang et al., 2016). Even in these settings, the disease phenotype can be influenced by mycobiota composition at different facilities such as the presence or absence of opportunistic Candida species (Iliev, 2015). However colonization with specific Candida strains can overcome such differences and provide consistent phenotypes in mice with deficiency in antifungal immunity (Iliev et al., 2012; Leonardi et al., 2018; Tang et al., 2015; Wang et al., 2016).

Gut mycobiota and inflammatory diseases targeting gut-distal sites: lung and liver disease

Increasing evidence suggests that a variety of lung disorders are strongly linked to gut mycobiota dysbiosis. In mice, intestinal fungal dysbiosis can be induced by disrupting bacterial or fungal communities with either antibiotics or antimycotics, respectively (Kim et al., 2014; Li et al., 2018; Noverr et al., 2004; Wheeler et al., 2016). Upon bacterial dysbiosis, antibiotic treatment leads to the overgrowth of Candida spp. and overproduction of intestinal Candida-derived prostaglandins that exacerbated allergic airway inflammation by promoting M2 macrophage polarization in the lungs of mice (Kim et al., 2014) (Figure 3A). Therefore, translocation of intestinal fungal-derived metabolites to the lung might be one of the mechanism involved in aggravation of allergic airway diseases (Figure 3A). In contrast, antimycotic treatment induces the expansion of filamentous fungi such as Wallemia mellicola, Aspergillus amstelodami, and Epicoccum nigrum that aggravates the outcomes of allergic airway disease by supporting a type 2 immune environment in the lung (Kim et al., 2014; Li et al., 2018; Noverr et al., 2004; Wheeler et al., 2016). W. mellicola alone can also aggravate allergic airway disease (Skalski et al., 2018). Although the precise gut-systemic immune mechanisms remain unknown, specific targeting of CX3CR1⁺ MNPs in the gut without depletion of these cells in the lung revealed a role for gut CX3CR1⁺ MNPs and Syk-dependent signaling in these cells in sensing fungal dysbiosis and mediating the increase in Th2 and eosinophils during lung allergic inflammation (Li et al., 2018) (Figure 3A).

Open in a separate window

Figure 3:

Mechanisms of gut mycobiota mediated gut-to-systemic immune crosstalk.

A. Gut fungal dysbiosis characterized by the overgrowth of filamentous fungi or Candida species contributes to lung allergy via a gut-lung axis. C-type lectin receptor-Syk-dependent recognition of gut filamentous fungi by intestinal CX3CR1⁺ MNPs leads to the activation and increase of pro-allergenic Th2 cells in the lung. Gut-derived and Candida species-derived PGE2 exacerbates allergic airway inflammation during intestinal fungal colonization by promoting M2 macrophage polarization in the lung. Intestinal C. albicans over-colonization can increase the levels of systemic Th17 cells that might exacerbate airways inflammation. B. Fungal cell wall components, including β-glucan, translocate from the intestinal lumen to the liver. Binding of β-glucan to Dectin-1 on Kupffer cells and possibly other bone marrow–derived cells induces the secretion of IL-1β and promotes alcoholic liver disease by leading to steatosis and cell death of hepatocytes. T cells primed by fungal antigens in the intestinal mucosa might migrate to the liver and contribute to liver disease.

Th17 cells are implicated in the pathogenesis of asthma, chronic obstructive pulmonary disease (COPD), or cystic fibrosis (CF) (Bacher et al., 2019; Iwanaga and Kolls, 2019; McAleer et al., 2016). Patients with lung immune disorder are frequently sensitized to environmental filamentous fungi, such as A. fumigatus, Alternaria and Cladosporium species (Knutsen et al., 2012; Sudfeld et al., 2010). Further, reports also revealed that Candida spp. are persistent colonizers of the respiratory tract of cystic fibrosis (CF) patients (Muthig et al., 2010). A recent characterization of the lung fungal mycobiome revealed that both Candida and Aspergillus are the most abundant species in the respiratory tract of patients with CF (Kim et al., 2015). Among different environmental and human commensal fungi, recent study identified C. albicans as the main inducer of Th17 cells in peripheral blood from over 100 healthy donors (Bacher et al., 2019). Interestingly, memory T cells with specificity to C. albicans showed high level of cross-reactivity with the inhaled environmental fungus A. fumigatus and vice versa. A. fumigatus was not the only fungus cross-reactive with these cells. Further analysis revealed that Th17 cell with specificity to several fungal species were highly cross-reactive with C. albicans, suggesting a broad modulation of human Th17 responses by this fungal commensal (Bacher et al., 2019). The study further observed fungal antigens cross-reactive Th17 cells in multiple lung inflammatory disorders including chronic pulmonary inflammation with asthma, chronic obstructive pulmonary disease (COPD), or cystic fibrosis (CF) (Bacher et al., 2019) (Table 1). Patients with lung immune disorders are frequently sensitized to environmental filamentous fungi, such as A. fumigatus, Alternaria and Cladosporium species (Kim et al., 2015; Knutsen et al., 2012; Muthig et al., 2010; Sudfeld et al., 2010). Indeed, the presence of C. albicans-induced Th17 cells in both lung (asthma, COPD and cystic fibrosis) and intestinal disease (such as CD) suggests a potential intestinal priming of these cells (Bacher et al., 2019; Shao et al., 2019) (Figure 3A). Additionally, A. fumigatus-reactive T cells in patients with acute allergic bronchopulmonary aspergillosis (ABPA) do not co-express IL-4 and IL-17A, indicating acquisition of IL-17-producing capacity through cross-reactivity with C. albicans antigens, but not through conversion of already existent lung Th2 cells (Bacher et al., 2019). Indeed, intestinal colonization with C. albicans induces a potent Th17 response (Atarashi et al., 2015; Leonardi et al., 2018). This increase in Th17 levels upon intestinal colonization is also found in the lung and is associated with an increased susceptibility to allergic airway inflammation (Shao et al., 2019). Altogether these recent data suggest that priming of fungal cross-reactive T cells by gut commensal fungi and selective recruitment to these cells to the lung might be an important and underappreciated factor in the pathogenesis of several human inflammatory airway diseases related to gut fungal dysbiosis.

Human studies have further shown a strong correlation between bacterial dysbiosis and the increased risk of allergic airway diseases (Arrieta et al., 2015; Fujimura et al., 2016). Both bacterial composition and bacteria-derived metabolites are associated with asthma in human cohorts (Arrieta et al., 2015; Fujimura et al., 2016). Reduction of bacterial genera, such as Lachnospira, Veillonella, Faecalibacterium, and Rothia during the first 100 days of life is associated to an increased risk of asthma (Arrieta et al., 2015). Furthermore, neonatal gut microbiome alterations are strongly associated with childhood asthma (Fujimura et al., 2016). In addition, fungal dysbiosis with an increased abundance of Candida spp occur in the high-risk asthma group (Figure 3A) (Fujimura et al., 2016).

The liver is constantly exposed to microbial-derived products and a disruption in the tolerogenic mechanisms that counteract this continuous exposure has the potential to trigger potent autoimmune responses. Autoimmune liver diseases (AILD) are chronic immune-mediated inflammation whose worldwide prevalence ranges from 1 to 50 per 100,000 individuals. The three major forms of AILD are primary biliary cirrhosis (PBC), primary sclerosing cholangitis (PSC) and autoimmune hepatitis (AIH). Although the triggers of AILDS remain unknown, novel studies in human and mice models suggest that an altered microbiota could be involved in the disease pathogenesis (Cai et al., 2017).

Recent evidence suggests that intestinal fungi could play a role in this detrimental gut-liver axis. Serum antibodies against ASCA are increased in auto-immune liver disease including AILD, primary biliary cirrhosis, and primary sclerosing cholangitis (PSC) when compared to healthy (Muratori et al., 2003; Papp et al., 2010) and to individuals with chronic viral hepatitis (Figure 3B) (Papp et al., 2010). Furthermore, the fungal burden in the feces, in particular, Candida spp., is increased in several forms of liver disease including alcoholic cirrhosis, NASH, viral hepatitis and cryptogenic cirrhosis (Figure 3B) (Krohn et al., 2018). PSC co-occurs with IBD in the majority of the patients. A recent study reported that, in addition to bacterial dysbiosis, mycobiota diversity is increased in PSC and PSC-IBD patients when compared to IBD patients or healthy controls (Lemoinne et al., 2019). Further correlations between the bacterial and fungal microbiota suggest that the inter-kingdom network might be altered in these patients (Lemoinne et al., 2019).

In humans, alcohol abuse is associated with Candida overgrowth and fungal dysbiosis independently of the stage of liver disease. ASCA levels in patients with alcohol abuse related cirrhosis are higher than in healthy controls and patients with viral cirrhosis and strongly correlated with mortality (Yang et al., 2017). In mice, chronic ethanol feeding promotes the expansion of intestinal fungi and translocation of fungal cell products (in particular, β-glucan) to the systemic circulation. In this model, mice lacking Clec7a in bone marrow-derived cells are protected from ethanol-induced liver disease (Yang et al., 2017).

These results suggest that sensing of intestinal fungi might modulate the extent of liver damage. Evidence suggests that direct sensing of β-glucan by phagocytic Kupffer cells in the liver triggers the production of IL-1β thus promoting liver inflammation in this model and similar mechanisms might be involved in other forms of liver pathologies (Figure 3B).

While gut mycobiota “dysbiosis” occurs in several inflammatory diseases as reviewed above, data from additional cohorts of patients with these diseases is needed to distinguish consistent changes from cohort-specific “noise” and to determine the key fungal players associated with the specific diseases and conditions. The data collected thus far suggest that intestinal fungi, and particularly Candida species, can induce innate and adaptive immunity to aggravate the outcome of several immune-mediated diseases. The human gut represents and unique niche for fungal adaptation, however very few studies have utilized human gut fungal isolates to determine the mechanisms of disease. Using such gut fungal isolates might help determine universal versus species/strain-specific immune mechanisms of anti-fungal immunity involved in the maintenance of immune homeostasis or the development of disease phenotypes.

Therapeutic approaches targeting fungal communities.

The mounting evidence of gut mycobiota involvement in human health and disease has increased the interest in the potential of mycobiota targeting that might represent a plausible therapeutic approach for several IMD. Direct targeting of fungi with antifungal drugs is the avenue most pursued thus far. Antifungal drugs present effective, at times live-saving treatments for susceptible and immunocompromised individuals (Brown et al., 2012; Eyre et al., 2018). Nevertheless, the same approaches and drugs might be only partially effective when attempting to target specific members of the gut mycobiota. Further, some antifungal drugs do not reach the lower Gl tractor are rapidly metabolized (Li et al., 2018). Although prolonged delivery of antifungal drugs can partially overcome this barrier, we have recently shown that this approach might have undesired outcomes on both commensal microbiota and inflammatory disease outcome (Li et al., 2018; Wheeler et al., 2016), suggesting that long-term antifungal treatments should be considered with caution, especially for patients with inflammatory diseases.

Dietary interventions are showing efficacy among several groups of patients (Albenberg and Wu, 2014; Lewis and Abreu, 2017). Diet modulates mycobiota composition: Carbohydrate-rich diets support Candida colonization in the gut (Vargas et al., 1993) and dietary short-chain fatty acids negatively correlate with Aspergillus spp load.(Hoffmann et al., 2013). Coconut oil-rich diet reduces C. albicans gastrointestinal colonization by decreasing intestinal availability of long-chain fatty acids and altering the fungal metabolic program (Gunsalus et al., 2016). A high-fat diet can change the fungal and bacterial communities in murine models, increasing the abundance of C. albicans and decreasing the abundance of S. cerevisiae. Diet can also be a source of fungi associated with vegetables, fruits and dairy products in humans (David et al., 2014; Suhr et al., 2016), and can be a factor influencing fungal community structure across different ethnicities (Mar et al., 2016).

Given the impact of diet on fungal communities, the notion of mycobiota modulation with probiotics and prebiotics is appealing but currently underexplored. In mice, Bacteroidetes thetaiotamicron and Blautia producta promote resistance to C. albicans gut colonization via mechanisms involving HIF-1α and the antimicrobial peptide CRAMP (Fan et al., 2015). Several probiotic lactobacilli species possess antifungal properties in vitro and compete with Candida at the vaginal mucosa (Boris and Barbes, 2000). Studies in humans are lacking. Recent studies, however, highlight the poorly characterized consequences of probiotics administration in individuals undergoing antibiotic treatments. Antibiotic perturbation modestly enhances probiotic colonization in the healthy human mucosa, delays the post-antibiotic recovery to the original microbiota composition and is associated with a different transcriptomal profile in host cells as compared to controls (Suez et al., 2018). A delay in the recovery of the initial microbiota status is not per se a drawback, especially in patients affected by dysbiosis. Yet, probiotics are constantly prescribed to patients with intestinal disorders, highlighting the urgency in better understanding whether and how probiotic affect microbiota recovery in these patients. Lactobacillus-secreted soluble factors, that presumably affect fungi (Zelante et al., 2013), further contribute to delayed microbiome recovery in vitro (Suez et al., 2018). These factors likely affect how probiotics affect fungal colonization and immune responses and might contribute to individual specific responses (Zmora et al., 2018).

Fecal microbiome transplantation (FMT) can provide sustained benefits for patients with recurrent Clostridium difficile infections (CDI) (Eiseman et al., 1958; van Nood et al., 2013), and its use has been successfully expanded to IBD (Paramsothy et al., 2017) and immune checkpoint inhibitor-associated colitis (Wang et al., 2018). In healthy volunteers, autologous FMT induces a near-complete post-antibiotic recovery of the microbiota (Suez et al., 2018). FMT might thus represent a better approach than the use of probiotics to achieve mucosal protection following antibiotic treatment in healthy individuals. However, autologous FMT is not a likely option in patients with immune-related pathologies were the intestinal microbiota is often affected by dysbiosis. In such cases, heterologous FMT might represent a valuable alternative. Because of the massive effect of FMT on intestinal ecology, it is reasonable to speculate that FMT would also affect, and be affected by, the mycobiota. Indeed, fungal colonization in recipients is associated with FMT response in CDI (Zuo et al., 2018). FMT responders display increased fungal diversity and high relative abundance of Saccharomyces, Aspergillus, and Penicillium, while non-responders show an increased abundance of C. albicans. Interestingly, high abundance of C. albicans in donor feces also correlates with reduced response to FMT, suggesting that presence of Candida in either donors or recipients might be associated with poor outcome of FMT in CDI (Zuo et al., 2018). Whether Candida domination is a consequence of reduced bacterial diversity in non-responders or a factor negatively affecting the response to FMT in CDI affected individuals requires further investigation. Intestinal fungi may play a role in the response to FMT in other inflammatory settings, such as UC. FMT has shown great promise in UC patients, improving several parameters including clinical remission and disease-related microbiota dysbiosis (Paramsothy et al., 2017; Paramsothy et al., 2018). Future studies in the context of UC may provide insight into the mechanisms whereby mycobiota impact recolonization after FMT.

Concluding Remarks

In this review, we outlined the recent advances on the mechanisms involved in gut mycobiota interaction with host mucosal immunity and on its consequences on inflammation-mediated diseases. We have also outlined challenges hindering progress in this area. The composition of intestinal mycobiota remains poorly characterized in some immune-mediated disorders while more datasets are becoming available for others. Altogether, discrepant results have been reported in different human cohorts. Therefore, a collaborative, standardized and comprehensive mapping of the gut mycobiome in human diseases is critically important to fill this gap and allow for a clear definition of the key fungal players in specific IMDs.

The current mechanistic studies are limited by the use of few model fungal strains of non-intestinal origin. Since strain-specific features can dramatically influence disease outcomes (Marakalala et al., 2013) important insights can be gained by using human fungal strains isolated from specific disease cohorts. Such studies should help determine universal versus strain-specific immune mechanisms involved in the maintenance of immune homeostasis or the development of specific inflammatory diseases.

Besides the gap in knowledge on the human mycobiota composition and lack of information on key fungal players involved in diseases with intestinal component, our understanding of the mechanisms by which specific gut fungi and inter-kingdom interactions influence host immunity lags further behind. Radically new approaches need to be developed to assess inter-kingdom aspects of microbiota interactions among different members of the intestinal community and the human host. Such approaches would pave the way to study how fungal metabolic activities affect host immunity and homeostasis.

A greater understanding of the immune mechanisms involved in the interaction with the mycobiota during inflammatory diseases is needed to direct the implementation of therapeutic fungal targeting outside of the classical infectious disease context. Among the possible approaches, antifungal drugs appear to be suitable to effectively target intestinal fungal communities. However, long-term treatment with antifungal drugs can dramatically alter the host mycobiota and might prove detrimental to the outcome of IMD. Other approaches, such as dietary interventions, probiotics, FMT and antifungal metabolites might provide suitable alternatives to restore a dysbiotic mycobiota. These approaches are still in their infancy and further studies are needed to assess their safety and efficacy.

Acknowledgments

Footnotes

References