The gut, the bad and the harmless: Candida albicans as a commensal and opportunistic pathogen in the intestine

Summary

Candida albicans is a regular member of the intestinal microbiota in the majority of the human population. This underscores C. albicans’ adaptation to life in the intestine without inducing competitive interactions with other microbes or immune responses detrimental to its survival. However, specific conditions such as a dysbalanced microbiome, a suppression of the immune system, and an impaired intestinal barrier can predispose for invasive, mostly nosocomial C. albicans infections. Colonization of the intestine and translocation through the intestinal barrier are fundamental aspects of the processes preceding life-threatening systemic candidiasis. Insights into C. albicans’ commensal lifestyle and translocation can thus help us to understand how patients develop candidiasis and may provide leads for therapeutic strategies aimed at preventing infection. In this review, we discuss the commensal lifestyle of C. albicans in the intestine, the role of morphology for commensalism, the influence of diet, and the interactions with bacteria of the microbiota.

Introduction

Invasive infections with Candida species are by far the most common cause of nosocomial fungal infections threatening patients of intensive care units (ICUs) and immunocompromised patients or patients with dysfunctional epithelial barriers [1,2]. Most human pathogenic fungi of the genus Candida are not ubiquitously present in the environment like, for example, Aspergillus, or Cryptococcus species, but associated with animals and humans. Systemic candidiasis of humans, therefore, originates from commensal body niches. In Westernized countries, most individuals carry Candida species as commensals in their intestinal microbiota [3,4] and patients show enrichment of Candida species in the majority of cases during prolonged stays in the ICU [5] or following transplantation [6]. Abdominal surgery or pathology to organs of the gastrointestinal (GI) tract is a risk factor for the development of intraabdominal candidiasis [7]. With regard to candidemia, or Candida bloodstream infections, Candida cells can be introduced directly into the bloodstream by contaminated central venous catheters. However, such infections are more commonly associated with C. parapsilosis, which, in contrast to other Candida species, colonizes the skin [8,9]. The evidence is overwhelming that C. albicans in the gastrointestinal (GI) tract is a major source of systemic candidiasis [6,8,10] and numerous experimental models have revealed that this fungus can enter the bloodstream by translocating through the intestinal barrier [11–13]. Therefore, it is essential to comprehend the pathogenesis of systemic candidiasis originating from the intestinal tract. Elucidating C. albicans’ role as a commensal, as well as a pathogen, can help us understand how this otherwise harmless commensal can undergo a shift towards an opportunistic pathogen. Further, it may aid to identify predisposing factors and to provide valuable insights for the development of strategies aimed at preventing candidiasis, an approach that will have a drastic impact on candidiasis-associated mortality [14].

C. albicans as a commensal of the gut

Even though a great deal is known about C. albicans pathogenicity [15], studies dealing with the commensal lifestyle of this fungus have only recently come into focus. [for further recent reviews see: 16,17–20].

As the GI of the majority of humans in Westernized societies is colonized by C. albicans [3,4] it can be concluded that this fungus has evolved to interact and compete with numerous bacterial cells and species that colonize this niche. However, as demonstrated in mouse models, the composition of the microbiota is critical for successful establishment, and some bacterial consortia can inhibit or prevent C. albicans colonization. In fact, C. albicans is not a normal colonizer of many mice strains and an intact murine bacterial microbiota can resist the ability of C. albicans to become established as a member of the community [21,22], whereas mice treated with antibiotics, in particular antibiotics targeting anaerobic bacteria [12,16,23], and germ-free mice [24] can be colonized by C. albicans yeast with ease. Colonization resistance can be accomplished by modifying the metabolite milieu in a way that diminishes fungal growth [25]. Additionally, commensal bacteria can stimulate the immune response, increasing the production of the antimicrobial peptide LL-37, which can contribute to colonization resistance [21]. On the other hand, colonization can be promoted by a specific diet (see below).

As C. albicans can colonize the majority of humans, colonization resistance is often considered only an aspect of adult murine models. It seems clear, however, that, in humans, the bacterial microbiome plays an essential role in controlling the level of C. albicans colonization since the use of broad-spectrum antibiotics is recognized as a major risk factor for candidemia [1,2]. Hematopoietic stem-cell transplant recipients treated with antibiotics targeting anaerobic bacteria show enrichment of Candida species in the intestinal fungal microbiome (the mycobiome) [6]. Similar enrichment of C. albicans (or C. glabrata) is observed in ICU patients treated with antibiotics leading to ultra-low-diversity microbiomes [5]. Overall these findings highlight a role for the microbiota in suppressing C. albicans overgrowth and pathogenicity in experimental models as well as patient cohorts. In contrast, C. albicans colonization itself can have an influence on the intestinal micro- and mycobiome [18] and can promote bacterial dysbiosis that promotes invasive infection [26]. Only recently has the role of the mycobiome come into focus [6,27–29], and it has been shown to play crucial roles in human diseases [30–32] (Reviewed in this issue by Hohl and colleagues).

Studies of direct interactions under laboratory conditions between bacterial species and C. albicans show that probiotic bacteria can antagonize the growth and pathogenicity of C. albicans. For example, a probiotic cocktail was able to reduce polymicrobial biofilm formation by C. albicans in combination with Escherichia coli and Serratia marcescens [33]. Bacterial metabolites such as fatty acids can inhibit C. albicans hyphal formation [34] and Lactobacillus rhamnosus reduces C. albicans overgrowth and prevents contact of the fungus with the intestinal barrier through shedding [35]. In models lacking bacteria, host epithelial cells provide some resistance to C. albicans [36]. However, C. albicans employs various strategies to successfully invade and translocate through the epithelial barrier [Reviewed in 37], unless it is genetically defective for these attributes [38]. The presence of only a single antagonistic bacterial species, e.g. L. rhamnosus, suppressed C. albicans to such an extent that reduced translocation through the epithelial barrier was observed in an in vitro intestine-on-chip model [39].

Influence of diet

The composition and the metabolism of our intestinal microbiota are highly influenced by diet. It is presumed that diet also influences C. albicans colonization of the gut. A Western diet, together with the use of antibiotics, has been posited as one of the reasons that the majority of the Western population is colonized by C. albicans [16]. Non-Western societies, for example, can exhibit much lower colonization rates as compared to many Western countries [18,40]. In murine models, a diet that does not support growth and effector functions of antagonistic bacteria like Lactobacillus species can be used to achieve colonization, which can lead to translocation under immunosuppressive treatment [11]. Of note, these diets consist of cornstarch, sucrose, and soybean oil, similar to a Western diet. Inclusion of fatty acid-rich coconut oil, however, reduces C. albicans colonization, even when other nutritional sources that support colonization are present [41]

Metabolic Adaptation to the GI tract

Niche-specific attributes, such as adaptation for effectively utilizing the nutrients that are present, make C. albicans highly efficient in colonizing the GI tract. Glucose levels in the GI tract, particularly the distal GI tract, are low and colonizers are thus obliged to utilize alternative carbon sources [42]. C. albicans has noted nutritional flexibility allowing it to adapt readily to the nutritional milieu that it encounters [43,44]. The organism is capable of simultaneously using different carbon sources and can continue to utilize lactate even when provided with glucose (in contrast to the baker’s yeast Saccharomyces cerevisiae). This nutritional flexibility is essential for colonization of the GI tract and a mutant unable to grow on less preferred carbon sources such as glycerol, citrate or lactate is out-competed by wild type cells in the GI tract [45].

As nutrient sources in the intestine are limited, tight regulation of nutrient acquisition pathways yields a fitness benefit by ensuring that genes are expressed only when absolutely necessary. N-acetyl glucosamine was postulated as a limiting nutrient that regulates nutrient sensing in the GI tract [46]. C. albicans undergoes “sugar-induced cell death” when grown with GlcNAc and no other nutrients but not when grown with glucose alone, consistent with the notion that GlcNAc is an important signal and a preferred carbon source [46].

Of note, nutrition sources affect stress resistance in C. albicans. A mutant lacking the transcription factor Rtg3 was found to exhibit hyper-susceptibility to cationic, oxidative and nitrosative stresses when grown in lactate but not glucose [42]; interestingly, Rtg3 is required for colonization of the murine GI tract [47] and plays a role in galactose [48] and sphingolipid metabolism [49].

The picture that emerges is that C. albicans is well adapted to growth without glucose. Remarkably, changing environmental conditions such as the availability of carbon sources (e.g. lactate instead of glucose) impacts fungal adaptation, which can influence colonization and pathogenicity [50]. For example, modification of cell wall architecture can either lead to immune evasion or hyper-activation of the host response [51] and altered cell wall morphology significantly modulates colonization in murine models [52].

Other key nutrients also affect the ability of C. albicans to colonize. For example, although the bioavailability of iron in the GI tract is higher than in tissue, Mamouei et al. showed that a mutant defective in iron permeases (ftr1Δ ftr2 Δ double null mutant) was outcompeted by wild type cells in a competitive GI colonization study [53] indicating that iron uptake is important in the gut. Further, C. albicans cells modify the GI tract metabolite milieu, for example, by producing the prostaglandin PGE2 from host-derived arachidonic acid. PGE2 production may change the environment of the GI tract, providing conditions that are more conducive for C. albicans colonization [54].

Further metabolites such as bile acids also have a significant impact on C. albicans’ behavior and, therefore, can influence colonization and/or infection. Secondary bile-acids such as lithocholic acid and deoxycholic acid have direct antifungal activity by limiting growth, filamentation, and adhesion to host cells [55], aspects that may contribute to the commensal lifestyle of C. albicans. However, antibiotic treatment prevents the conversion of primary into secondary bile acids by commensal bacteria. The consequent accumulation of primary bile acids such as taurocholic acid can increase C. albicans outgrowth [56], and stimulate filamentation [56,57].

In general, adaptation to gut metabolites and other specific conditions in the gut, such as hypoxia and diverse pH values, requires sophisticated regulatory networks, some of which may also contribute to pathogenicity [47,50,58].

Phenotypic adaptation to the GI tract

Invasive candidiasis is often associated with filamentous growth, which facilitates the invasion of host tissues [59].

Despite being essential for invasive disease, evidence is mounting that hyphae and expression of hyphal-associated proteins are detrimental for the commensal lifestyle of C. albicans [60–63]. Even though it was revealed that hyphal morphology is abundant during colonization of the intestine and is not associated solely with invasive disease, knockout of the hyphal-specific transcription factor Ume6 or the hyphal-associated factors Sap6 or Hyr1 is beneficial for colonisation [61]. In competition colonization experiments, strains adapted to the GI tract through long-term colonization easily outcompeted strains that were expanded in vitro or short-term colonizing strains [64], illustrating the flexibility to adapt to a commensal lifestyle. In fact, long-term experimental colonisation of the gut poses a selective pressure towards spontaneously mutated strains incapable of forming hyphae [60].

Since hyphae and their associated proteins seem detrimental to commensalism, it seems conceivable that the intestinal environment during homeostasis suppress filamentation. In addition to previously discussed bile acids that suppress filamentation, the hypoxic environment of the intestine has been postulated to play a role in supporting the restriction of hyphal morphogenesis detrimental to commensalism through modulation of Efg1 [65]. Furthermore, the expression of Efg1, which is a crucial determinant for commensalism, relies on the immune status of the host [62].

Finally, although hyphal formation does not seem to support GI colonization, the ability to form hypha is conserved among the vast majority of clinical isolates suggesting an evolutionary pressure to maintain this key attribute of C. albicans [60]. In this context, it should be noted that knockout of hyphal associated genes, which increases colonization fitness (above), results in a loss of filamentation in vitro, but not in vivo in the intestine [61].

C. albicans also forms a specialized cell type termed “GUT” cells during colonization [66]. GUT cells can be mimicked genetically and, during GI colonization, GUT-mimetic cells exhibit a transient decrease in competitive fitness versus WT followed by a transient enhancement [67]. GUT-mimetic cells thus have a competitive advantage once a proinflammatory host immune response against C. albicans has been initiated (>1 week of colonization). GUT-mimetic cells exhibit several differences from WT cells, including increased ability to adhere to the GI mucosa ex vivo, increased susceptibility to bile acids and altered respiratory metabolism. These differences may contribute to their altered fitness in the GI tract.

The role of different yeast morphotypes in the human host is further reviewed in [59]

The benefit of Candida albicans as a commensal for human health

From an evolutionary perspective, it is tempting to speculate that C. albicans may have some beneficial effects for human health since carriage of a potentially dangerous commensal might otherwise be subjected to negative natural selection. It is generally believed that C. albicans colonization may be crucial to educate our immune system. C. albicans colonization of the GI tract of mice, specifically with strains evolved to colonize the intestine, offer immunological protection against systemic infection [60]. As even Rag1^−/− mice, lacking functional T and B cells could mount protective immunity [60], the intestinal colonization is believed to induce an innate immune memory called trained immunity. Yet intestinal C. albicans colonization is also instrumental in shaping the T-cell responses (Reviewed in this issue by Bacher and colleagues). Specifically, Th17 responses, which are essential in driving mucosal immunity, are polarized by C. albicans colonization and offer protection against systemic candidiasis [68]. Potentially, the evolutionary benefit of harboring C. albicans as a commensal lies not only in its ability to train the immune system against systemic candidiasis, but also to infections with other pathogens such as Staphylococcus aureus, [60,68], Aspergillus fumigatus, or Pseudomonas aeruginosa [60]. However, despite being able to protect against systemic infection, the Th17 polarization induced by intestinal C. albicans commensalism can aggravate airway inflammatory diseases [68,69].

Apart from educating our immune system, another evolutionary advantage of C. albicans colonization may be competition with other pathogens or an influence on the balance of the intestinal microbiota. In fact, targeting intestinal Candida spp. with antifungal drugs can result in a mycobiota dysbiosis that can drive the development of allergic airway disease [32] via activation of gut resident CX3CR1 mononuclear phagocytes [70]. The effect of C. albicans on Clostridium difficile infection (CDI) is an interesting example of an interaction of C. albicans with another opportunistic pathogen. C. albicans can support the survival of C. difficile under aerobic conditions, while C. difficile inhibits C. albicans filamentous growth [71], indicating a direct interaction between these two opportunistic pathogens. In murine models of CDI, administration of C. albicans immediately before challenge with C. difficile resulted in increased disease [72]. In contrast, administering C. albicans to mice and allowing the establishment of colonization and a return to homeostasis in the GI tract before C. difficile challenge, resulted in protection against lethal CDI [73]. These findings may reflect the fact that C. albicans colonization has anti-inflammatory effects under some conditions, but when inflammation is pre-existing, the introduction of C. albicans enhances inflammation [74]. The importance of the timing of colonization with C. albicans versus with C. difficile may contribute to differences observed in human studies; only some studies observed low levels of C. albicans colonization in CDI patients (reviewed in [75]). Another report has shown that C. albicans can reduce Pseudomonas aeruginosa pathogenicity [76].

Although C. albicans can overgrow in the intestine during antibiotic treatment as a result of the absence of antagonizing bacteria, C. albicans can actually promote increased bacterial diversity during recovery from antibiotic treatment [77]. C. albicans colonization following antibiotics supports recovery of Bacteroidetes, and promotes Enterococcus faecalis, but reduces lactobacilli in the long-term. Similarly, a high intestinal C. albicans burden before fecal microbiota transplantation (FMT) is also associated with increased bacterial diversity following FMT in ulcerative colitis patients [78]. Yet, C. albicans high levels of colonization post transplantation highlight negative impact outcome of FMT [79][78].

Overall this illustrates that C. albicans as a commensal is engaged in an intensive fungal-host-microbiota cross-talk that shapes immune responses and microbiota composition. Whether this cross-talk contributes to health or disease often is specific to conditions of each individual host.

Breaching the boundary: invasion of the epithelial barrier leading to systemic disease

As discussed above, a disturbed microbiota, a suppressed immune system, and/or a defective epithelial barrier can all promote C. albicans translocation through the epithelia barrier into the bloodstream. Uncovering the fungal and host aspects which influence epithelial barrier invasion and translocation may help to answer the question of why C. albicans originating from the gut can cause systemic disease under predisposing factors.

Dissection of C. albicans translocation revealed filamentous growth, damage capacity, and decreasing barrier integrity as crucial as translocation occurs via a transcellular pathway, associated with necrotic cell death [38]. Consequently, mutants incapable of adhesion, filamentation, or inflicting damage are severely impaired in translocation [38]. Of note, attenuated growth without impaired pathogenicity mechanisms does not necessarily reduce translocation. Further, filamentation per se is not sufficient for full translocation potential.

Similar to some intestinal bacteria, distinct intestinal epithelial cell types are important for C. albicans translocation. For example, intestinal epithelial models containing M-cells revealed that C. albicans preferentially interacts with and invades the intestinal barrier via M-cells [80]. As M-cells are part of the Peyers Patches, translocation via M-cells may also facilitate immune recognition. Other intestinal epithelial characteristics aid in resisting translocation. Mucins suppress C. albicans virulence [81]. Consequently, integration of mucus-producing goblet cells in intestinal epithelial infection models suppresses translocation [35]. During infection, intestinal epithelial cells respond by inducing NFκB signaling. Once activated, the NFκB signaling pathway induces resistance against C. albicans-induced cell damage, and its neutralization renders epithelial cells more susceptible to damage [36].

In the intestine, resident immune cells provide another layer of defense against C. albicans invasion and translocation. CX3CR1 positive phagocytes, for example, are highly efficient at recognizing fungi and initiating antifungal host defense [82].

Additional mechanisms that allow the intestinal epithelium to resist C. albicans invasion are reviewed in [17,37]. Importantly it remains to be elucidated how C. albicans initiates translocation. For example, does the fungus invade as a single hypha, or does it first form micro-colonies on the epithelium and then invade from the colony? Near-physiological modeling of C. albicans translocation using organ-on-chip technology [39] may be able to provide insights here.

Conclusion

The ability of C. albicans to colonize the human gut as a harmless commensal is also a vital aspect of its role as an opportunistic pathogen. Apart from other Candida species few pathogens exhibit similar strategies of causing infections in predisposed patients. A pressing question is whether or not in a healthy host C. albicans always tries to invade, but the absence of predisposing conditions does not permit invasion. Considering the currently available literature on C. albicans commensalism and pathogenicity summarized in this review (Fig. 1 and 2), several possibilities that drive the commensal-to-pathogenic shift can be considered. I) A commensal-to-pathogen shift is actively suppressed by the antagonistic bacterial microbiota and their metabolites, the intestinal barrier, and the host immune system. This model explains why C. albicans in simple model systems lacking these suppressing factors behaves like a vicious pathogen and invades and damages host cells. II) The combination of predisposing factors alters the chemical, physiological and nutritional intestinal environment to stimulate the commensal-to-pathogen shift, driving the fungus to engage the host actively. Deprivation of specific nutrients during dysbiosis could also be considered as a driving force of pathogenicity. III) The fungus actively supports a commensal growth mode and suppresses hyphal formation and thus invasion because it is detrimental in the gut as it causes inflammation leading to clearance. Studies of gene expression in C. albicans cells colonizing hosts of differing immune status support this possibility. Yet, hyphal formation must be beneficial at some stages of a commensal lifestyle or for survival in general. Possibly, “near-pathogenic” behavior is essential for commensalism and is tipped to “full pathogenicity” when predisposition occurs.

Fig. 1:

The key mechanisms preventing a commensal-to-pathogen shift of C. albicans.

Fig. 2:

Mechanisms which contribute to a commensal versus pathogenic life style of C. albicans. A healthy diet feeds a healthy microbiome, which antagonizes colonization and hyphal formation of C. albicans via modulation of chemical, nutritional and physiological conditions. However, C. albicans cells also seem to self-regulate a commensal growth (via the “GUT” phenotype or transcriptional down-regulation of hyphal growth) and some gut niches may be colonized by hypha, which potentially show a near-pathogenic behaviour. Mucus production inhibits hyphal formation and keeps the fungus in distance from epithelial cells, which show resistance to occasionally invading hypha, supported by a healthy immune response.

In contrast, treatment with antibiotics and a Western diet can cause a dysbiosis of the microbiome and fungal overgrowth. This normally does not lead to fungal pathogenicity in the gut and the microbiome will return to a balanced state, for example after antibiotic therapy is terminated. However, overgrowth may favour the formation of invasive hypha and further predisposing conditions may lead to translocation through the epithelial barrier. This often includes a lack of epithelial fitness and proliferation as a result of cytostatic therapy or physical damage of the epithelial barrier. Further, a compromised status of the innate immune system can cause a failure of the final line of resistance against invasion into the bloodstream and dissemination to vital organs.

The complexity and multifactorial aspect of interactions between the fungus, host, and microbiota makes it difficult (at this stage of our knowledge) to comprehend the commensal-to-pathogen shift. Nevertheless, significant progress in understanding the two different lifestyles of C. albicans, commensalism and pathogenicity, has been made by a combination of experimental and clinical research and observations, and improved methodology to study microbial, metabolic, and immune interactions. Hopefully, this increased understanding can foster the development of strategies aimed at preventing candidiasis from an intestinal origin.