jueves, 17 de noviembre de 2016

The bacteriome–mycobiome interaction and antifungal host defense


Large communities of microorganisms, collectively termed the microbiome, inhabit our body surfaces. With the advent of next-generation sequencing, the diversity and abundance of these communities are being unravelled. Besides an imporant role in metabolic processes, the microbiome is essential for proper functioning of our immune system, including the defense against fungi. Despite the progress of the past years, studies aimed at characterizing our fungal colonizers (the mycobiome) are limited; nevertheless fungi are important players of the microbiome, either as a cofactor in disease or as potential pathogens. In this review, we describe the role of the bacterial microbiome in antifungal host defense. On the one hand, bacteria provide colonization resistance to fungi, inhibit Candida virulence by preventing yeast-hyphal transition and contribute to epithelial integrity, all factors are important for the pathogenesis of invasive fungal disease. On the other hand, several bacterial species modulate mucosal (antifungal) immune responses. Murine studies demonstrate important effects of the microbiome on the antifungal responses of T-helper 17 cells, regulatory T cells and innate lymphoid cells. Inferred from these studies, perturbation of the healthy microbiome should be avoided and microbiome manipulation and interventions based on bacteria-derived pathways involved in immunomodulation are attractive options for modulating antifungal host defense.


After a germ-free start at birth, acquisition of bacteria from the birth canal of the mother, and later from the environment, leads to the formation of complex communities of microorganisms on our body surfaces (oral cavity, throat, skin, the gastrointestinal, and distal part of the urogenital tract), collectively referred to as the microbiome [1]. The diversity and abundance of the human microbiome is strongly determined by the body habitat, with particularly dense populations of bacteria occurring in the lower digestive tract [1]. From a modestly diverse community with a low number of different species (low richness) and many shifts in composition during infancy, the gut microbiome evolves during the lifetime of an individual into a more stable organization throughout adulthood [2]. This relative temporal stability and the high interpersonal variability, most likely due to genetic and immune factors, lead to a specific individual “microbiome fingerprint” [3, 4], although external factors such as diet and antibiotic use have an important impact on the microbial communities [5, 6].
Millions of years of coevolution have led to a mutually beneficial relationship between the host and its colonizing microorganisms [7]. The human microbiome largely outnumbers our own body cells [8] and the functions encoded by their collective genome (metagenome) play an important role in host physiology. Short-chain fatty acids, bile acids, vitamins, lipids, and many other metabolites produced by gut bacteria are essential for many host metabolic pathways [7]. Given the critical function of microbiota-derived products in human metabolism, it is thus not surprising that dysbiosis has been linked to human diseases such as obesity or type 2 diabetes [9]. In line with this, transplantation studies in mice have showed a causal role for the microbiome in the pathogenesis of insulin resistance: transplantation of the gut microbiome from mice with the metabolic syndrome due to Toll-like receptor (TLR)5 deficiency to healthy, germ-free recipients, conferred the disease state [10]. In humans, Vrieze et al. [11] demonstrated an improved insulin sensitivity after transfer of the intestinal microbiome from lean, healthy volunteers to obese individuals with metabolic syndrome, characterized by an increased waist circumference and a decreased insulin sensitivity, which could be linked to a significant change in intestinal bacterial composition, particularly an increase in butyrate-producing species.
The cross-talk between the microbes and the host immune system, either directly or through metabolic intermediates, is also essential for the functioning of the immune system. In addition to the genetic make-up of the host, the colonizing microbiota has an important role in determining the host response to endogenous and exogenous infectious insults. Extensive developmental and functional defects of the intestinal immune system have been shown to arise in germ-free mice born and raised in a sterile environment [12]. Furthermore, the extraintestinal immune system is underdeveloped in germ-free animals, as evidenced by reduced numbers of CD4+ T cells and fewer and smaller germinal centers in the spleens of germ-free animals [13].
One important aspect that only very recently has started to gain interest and attention is that of the fungal communities inhabiting our skin and mucosa, the so-called ‘‘mycobiome,’’ and the modulation of antifungal immunity by the colonizing microorganisms. In this review we present a short overview of the present knowledge on the mycobiome, and we discuss the modulation of host defense by the microbiome, with a focus on antifungal immunity. We conclude with the implications of these complex interactions for human fungal infections and outline potential targets for control of such infections.

The fungal microbiome (mycobiome) in humans

Whereas culture-dependent techniques were only able to detect a minority of the microbial species of the microbiome, high-throughput sequencing techniques using the 16S ribosomal subunit gene as a target have greater sensitivity and studies using such techniques have provided comprehensive information about the composition and function of the healthy human (bacterial) microbiome [1]. Bacteria nearly completely constitute the microbial load of our feces. However, the less abundant components of the microbiome (termed the ‘‘rare biosphere’’), including fungi, also have significant impact on human health, either as a reservoir for potential pathogens (such as Candida species) or as a cofactor in diseases such as inflammatory disorders [14, 15]. Thus far, compared to the bacteriome, the mycobiome and its interaction with the commensal bacteria and host have received less attention. However, analogous to the bacteriome, culture-independent techniques offer far greater sensitivity in detecting these rarer species and enable the unraveling of the fungal commensal communities. We discuss the mycobiome in the oral cavity, intestines, female urogenital tract, the skin, and the lower airways (summarized in Table 1). The composition of the mycobiome differs between the different body surfaces, and therefore may play different roles in disease, either causally or as a marker of underlying disease.
Table 1. Dominant fungal species at different body sites
Oral cavity Candida, Cladosporium, Amobasidium, species from the family of Saccharomycetales (other than Candida), Aspergillus, Fusarium, Cryptococcus
Intestinal tract Saccharomyces, Candida, Cladosporium
Vagina Candida
SkinArms, body, faceMalassezia
 ScalpAcremonium, Dydimella
 Foot (nail, toe web, plantar heel)High fungal diversity
Lower airways Nearly absent. Possibly Aspergillus
Compared to a culture-based approach, which showed that 25 to 60% of the individuals tested were orally colonized by fungi (almost exclusively with Candida species) with an age-dependent increase in colonization rate [16], a different study using PCR analysis of oral rinse samples from 20 healthy volunteers aged 21–60 years with different ethnic backgrounds, revealed fungi to be present in all individuals [17]. One hundred and one species were detected with a diverse individual distribution: Candida species were most prevalent (isolated in 75% of the study participants), followed by Cladosporium (65%), Aurobasidium (50%), species belonging to the family of Saccharomycetales, other than Candida (50%), Aspergillus (35%), Fusarium (30%), and Cryptococcus (20%). A study among 12 healthy volunteers showed similar results, with Pichiaspecies being a prominent representative of the family of Saccharomycetales [18]. The differences in mycobiome composition between the various niches in the oral cavity, such as the gingiva, dental plaques, palate, tongue, and throat, remain to be determined, however. In the elderly, microbiomes with a high oral Candida load (relative to the bacterial burden) have been shown to be less diverse and to have a predominance of streptococci [19].
An insight into the fungal microbiome of the intestinal tract comes from the recent study from Hoffmann et al. [20] Fungi were present in all stool samples, with the two phyla Ascomycota and Basidiomycota having reciprocal patterns of abundance. At genus level, Saccharomyces(89%), Candida (57%), and Cladosporium (42%) were most prevalent, and the fungal burden, but not the types of fungi, correlated positively with the Prevotella/Bacteroides ratio. Interestingly, recent consumption of carbohydrates, as assessed by three 24 h diet recalls within 1 week before stool collection, was associated with high Candida abundance. Like the bacteriome, the mycobiome also is perturbed in disease. Crohn's disease is associated with an increased diversity and relative abundance of Candida species, Aspergillus clavatus, and Cryptococcus neoformans and with a decreased presence of Saccharomyces cerevisiae in the fecal mycobiome [21].
A large variety of fungal species reside in the healthy vagina of Caucasian women of reproductive age, with 70% of the samples harboring Candida species, 82% of which are represented by Candida albicans [22]. However, assessing true fungal diversity was limited by possible airborne contamination, during either the collection or the sequence procedure (presence of Cladosporium species), and the significant proportion of unspecified fungi identified through sequencing. Besides these limitations, one should realize that cross-sectional studies ignore the shifts in microbiome composition over time, which has been shown to be an important factor for the healthy vaginal bacterial microbiome [23].
Differences in the microenvironments at different body sites strongly influence the microbiome of the skin [1]. Malassezia species, including as-yet-unclassified phylotypes, constitute most of the fungal microbiota of the normal facial skin and at the forearm (with relative species stability over time and host specificity) [24, 25]. Non-Malassezia and filamentous fungi form the remaining 20% of the fungal colonizers of the face. On the scalp, Malassezia is present, but with low relative abundance, and Acremonium species dominate the mycobiome at this location [26]. These data are supported by a recent comprehensive assessment of the skin mycobiome at 11 core-body and arm sites, showing the large presence of the genus Malassezia, with only 11 species-level classifications revealing differences between sites in fungal composition. A different picture was apparent at three foot sites — plantar heel, toenail and toe web — which displayed a high diversity of fungi, with physiologic characteristics and topography of the skin most likely shaping the composition of these fungal communities [27]. The skin mycobiome is affected by skin diseases: dandruff-afflicted scalps harbor more fungi of the Basidiomycota phylum (mainly Filobasidium floriforme, but also Malassezia species), and less of the Ascomycota (Acremonium and Didymella species). The mycobiome in atopic dermatitis is characterized by a greater diversity in non-Malassezia fungi, contained Candidaand Cryptococcus species and is affected by disease severity [25].
While previously deemed sterile, based on culture, DNA-based techniques have recently demonstrated the presence of bacterial communities in the lung [28, 29]. Very little is known about the existence and composition of the lung mycobiome. Culture of exhaled breath condensate and bronchial brushing in healthy volunteers with a mean age of 65 years revealed no fungi, as opposed to Aspergillus nigerA. ochraceus, and Penicillium species found in lung cancer patients [28]. The authors of this study also found that sequencing of bronchoscopic alveolar lavage samples from healthy volunteers detected scant fungal amplification. In another study, a low abundance of Aspergillus was detected in bronchoscopic alveolar lavage samples, however the presence of predominantly environmental fungi in these samples and in simultaneously obtained oropharyngeal wash samples, suggests carry-over from the upper respiratory tract [29].

Interaction between microbiome and the immune system

The bacteriome-mycobiome interaction

Candida is a member of the healthy microbiome, but under certain circumstances this commensal state can change into pathogenicity and give rise to either life-threatening systemic infections or to less dangerous, yet still morbidity-inducing mucosal infections. Both host and microbial factors influence the derailment of fungal colonization, leading to invasion and culminating in a fungal infection (Fig. 1). There are several steps in the pathogenesis of candidiasis: (i) colonization of the mucosal surface, (ii) tissue invasion and mucosal infection and, in the case of systemic disease, (iii) dissemination into the bloodstream (usually from the gastrointestinal tract). Important modulating factors influencing the stages of the infection include decreased epithelial integrity, deficiencies in host immune defenses, and the virulence factors of Candida. However, the interplay between fungus and the host is not a simple two-way interaction and an additional important player intervenes: the bacteria of the human microbiome also influence fungal infections, as outlined below (Fig. 1).
Figure 1.
The protective role of commensal bacteria in the pathogenesis of candidiasis. Following colonization and adhesion to the mucosal surface, Candida hyphae can penetrate the epithelial layer after integrity loss and cause mucocutaneous infection (top right) or disseminate in the case of impaired phagocyte function. Commensal bacteria interfere with colonization through competition for surface and nutrients and the secretion of inhibitory substances (top). By inducing low pH and the secretion of SCFAs and quorum sensing molecules, commensal bacteria also inhibit hyphenation (top). Bacteria also strengthen epithelial barrier function through increasing mucus production and integrity by IL-22 and epithelium-derived LL-37 (top right). Abbreviations: IL, interleukin, M, macrophage; N, neutrophil, QSM, quorum sensing molecules, SCFA, short-chain fatty acids.
Competitive exclusion from mucosal niches by residing bacteria (colonizing resistance) is an important protective mechanism against potential invading microorganisms, among which are fungal pathogens. As a healthy microbiome inhibits the overwhelming growth of fungi, this is of particular importance for host defense against Candida, as the invasion is directly dependent on the fungal load colonizing the mucosa [30]. Several mechanisms are responsible for these effects including competition for surfaces and substrates, as well as the secretion of bacteria-derived substances that inhibit fungal adhesion and growth [30]. Such a toxic effect has been demonstrated in vitro in the case of H2O2 and bacteriocin-like compounds released by lactobacilli obtained from healthy premenopausal women [31]. On the other hand, synergistic interkingdom relations also exist, as exemplified by the intermicrobial binding (coaggregation) between C. albicans and streptococci, which is considered crucial for the colonization by C. albicans of the oral cavity [32].
The ability to form hyphae is one of the main virulence factors of Candida and is required for invasiveness [33]. Pseudomonas aeruginosa and Enterococcus faecalis have been shown to inhibit hyphal morphogenesis of C. albicans and thereby promote fungal commensalism [34]. Short chain fatty acids (SCFAs) are by-products of anaerobic fermentation of bacteria and SCFAs produced by lactobacilli, important inhabitants of the intestinal tract and the vagina, have also been shown to inhibit the transition from budding to hyphal growth [35]. Furthermore, lactobacilli contribute to low vaginal pH, which counteracts the transition from the yeast form to the filamentous form, and in this way potentially prevent Candida-related disease [36].
Epithelial cells form a natural barrier against invasion, and disruption of the (intestinal) epithelial lining has been shown to be a major risk factor for invasive candidiasis [37]. SCFAs have also shown to be involved in preserving this protective barrier by inducing production of cathelicidin LL-37 by intestinal epithelial cells, a well-known antimicrobial peptide with an important function in maintaining epithelial integrity [38, 39]. Additional evidence for bacteria-induced epithelial resistance stems from the observation that IL-22-mediated intestinal protection and integrity is increased after interaction of lamina propria dendritic cells with bacterial flagellin [40]. Furthermore, in mice the intestinal bacteria influence the mucus layer, which strengthens the mechanical epithelial barrier [41].
One important aspect in the transition of a fungus from commensalism to a pathogenic state is the status of the immune system, or more specifically the presence of immune deficits. The recruitment of neutrophils and production of defensins induced by Th17 cell-derived IL-17 and IL-22 has been shown to be essential for controlling growth and invasion of Candida at the level of the skin and mucosa in humans [42], while phagocytes such as neutrophils and macrophages have been shown to prevent fungal dissemination in human cancer patients [43].
The individual contributions of the immune system (immune deficiency) on the one hand and the protective role of bacteria (colonization resistance and maintaining epithelial barrier function) on the other hand, are difficult to unravel. Most likely a contribution of both factors is responsible for the effect on Candida colonization or invasion: for example, in a mouse model of gastrointestinal colonization, neutropenia or mucosal damage alone are insufficient to cause fungal pathology; both factors are required to produce fungemia [44]. One may speculate that the role of an immune deficit in fungal pathology is larger in the case of a healthy (and thus protective) bacteriome and vice versa. It is also important to observe that immune deficiency itself also affects colonization loads, as illustrated by increased oral colonization by fungi in CD4+ T-cell lymphocytopenia during HIV infection [45] and in hematopoietic stem cell transplant recipients with a loss-of-function dectin-1 polymorphism [46].
A similar three-way interaction between immune deficiency, colonization, and disease is present in the lungs. The presence of low numbers of commensal bacteria at this location offers little resistance against fungal colonization and antifungal immunity mainly relies on the immune system itself (i.e. mucus, ciliary transport, defensins, alveolar macrophages). It should be noted, however, that pulmonary microbiome research, let alone the mycobiome and the implication for fungal infections, is still in its infancy and future research will have to elucidate the importance of each pathway involved.

Influence of the microbiome on human antifungal host defense

A second important aspect of the relationship between the microbiome, host defense, and fungal infection is represented by the modulation of host antifungal immune responses by bacterial commensals (Fig. 2). A study by Villena et al. [47] provides evidence for the causative relationship between the administration of probiotic microorganisms and the immune response in C. albicans infection: in a mouse model of disseminated candidiasis, supplementation of the diet with Lactobacillus casei to previously malnourished mice prevented mortality, while in the control group of mice replenished with a balanced conventional diet without supplemental Lactobacilli, mortality was 57% [47]. The resistance to the fungal challenge was explained by increased phagocytic and fungicidal activity of peritoneal macrophages, higher numbers and function of neutrophils, and the re-establishment of a normal and well-balanced systemic cytokine response [47].
Figure 2.
Schematic representation of major interactions between antifungal host defense and the microbiome. Besides antagonistic and synergistic relations between bacteria and fungi, several bacterial species have been identified that stimulate protection through mainly soluble factors, such as SCFAs, PSA, IAld and SAA. SCFAs and PSA induce Treg cells, which have protective effects against Candida in the early phase of an infection and maintain homeostasis in later phases. SCFAs also augment macrophage function. IAld stimulates IL-22 production and epithelial-mediated protection, whereas SAA stimulates Th17-cell differentiation. On the other hand, fungus-derived PGE2 inhibits TNF-α production and inhibits the induction of IL-10. The interactions between macrophages, dendritic cells and innate lymphoid cells and lymphocytes are not shown. Abbreviations: D, dendritic cell; IAld, indole-3-aldehyde; IL, interleukin; ILC, innate lymphoid cell; M, macrophage; N, neutrophil, Th, T helper cell; Treg, regulatory T cell; TLR, Toll-like receptor; SAA, serum amyloid A; SCFA, short-chain fatty acids; SFB, segmented filamentous bacteria; PGE2, prostaglandin E2; PSA, polysaccharide A. Arrows visualize stimulatory effects and capped lines inhibitory effects.
A recent study illustrated the complex and multidirectional interactions between host genetics, the bacteriome, and the antifungal host defense in human individuals [48]. The skin microbiome was sampled in patients with chronic mucocutaneous candidiasis (CMC) and hyper IgE syndrome (HIES), immunodeficiency syndromes characterized by defective T-helper (Th)17 immune responses and leading to an increased susceptibility to skin and mucosal infections with fungal pathogens and Staphylococcus aureus. This study showed that the skin microbiome of CMC and HIES patients contains reduced numbers of the normal skin microbiota (such as Corynebacteria) and more Gram-negative bacteria (Acinetobacter), which in peripheral blood mononuclear cell stimulation experiments secondarily suppressed the expected cytokine response to C. albicans and S. aureus, the two major pathogens in these patients [48]. The importance of this study is supported by a second study among immunocompromised individuals, predominantly HIES, which showed that these patients had increased ecological permissiveness of the skin, with a reduction in the relative abundance of the commonly found skin yeast Malassezia and increased representation of opportunistic pathogens such as Candida and Aspergillus [49]. These findings imply that the skin serves as a potential niche for recurrent fungal infections in highly susceptible patients, such as those with primary immunodeficiencies.
Several commensal bacterial species, such as LactobacillusClostridium species, segmented filamentous bacteria (SFB), and Bacteroides fragilis [50-53], have been described to influence the immune system, and therefore are also likely to play a role in modulating antifungal immune responses. Most of the knowledge regarding this topic has been obtained from studies involving the gut microbiome, but a similar mechanism may relate to other sites such as the skin, a location where T-cell function has been shown, in mice, to be dependent on local skin commensals [54]. It should be noted, however, that these studies were mostly performed in mice [50-53], and that the bacterial species involved have not yet been shown to modulate antifungal host defense or been linked to fungal susceptibility in humans. Nevertheless, they show how bacterial species modulate the immune system, including the pathways involved in the defense against fungi [50-53].
As has been reviewed elsewhere [55], a coordinated and well-balanced interplay between proinflammatory cells, which include Th1 cells, Th17 cells, innate lymphoid cells (ILCs), and anti-inflammatory regulatory T (Treg) cells is essential not only for fungal clearance, but also for ensuring a homeostatic environment [55]. Most studies have focused on IL-22, IL-17 (produced by Th17 cells and ILCs) and Treg cells, which we discuss hereafter, followed by a discussion of the (limited) studies available about IgA and antifungal immunity.
Both IL-22 and IL-17 produced by Th17 cells and group 3 ILCs residing at barrier surfaces constitute a crucial component of the mucosal immune response against Candida [55]. The role of IL-17 is complex and has been associated with both fungal protection and increased immunopathology and with defective pathogen clearance [56]. In addition to the tryptophan catabolites induced by indoleamine 2,3-dioxygenase 1 (IDO1), which are potent anti-inflammatory metabolites and ensure that the mucosal is healthy and able to resist fungal colonization, the tryptophan derivative indole-3-aldehyde (IAld) also protects against fungal infection. Lactobacilli produce IAld and, by ligation of the aryl hydrocarbon receptor on ILCs by IAld, it induces IL-22 production, driving epithelial protection both in the intestine and in the vagina of mice [50]. Microbiome-driven IL-22 production has also been described as a response to epithelial penetration, simulated by systemic flagellin administration to mice. In response to this simulated breach, intestinal CD103+CD11b+ dendritic cells produce IL-23, which stimulates IL-22-mediated protection of the intestine [40]. On the other hand, studies in mice, either germ-free or with antibiotic-induced depletion of the intestinal microbiome, show that gut bacteria could decrease IL-22 production by group 3 ILCs via the epithelial expression of IL-25 [57]. One aspect, however, that is most often not considered in studies employing antibiotic-induced depletion of bacteria is the very well-known clinical effect of fungal overgrowth, showing the challenges inherent with studies aimed at assessing the complex interplay of the microbiome (bacteriome-mycobiome-virome) and host defense.
SFB are spore-forming, Clostridium-related, Gram-positive bacteria, which adhere tightly to the epithelial lining of the murine ileum and are capable of inducing the generation of Th17 cells [51], as well as other Th cells [58]. SFB-induced serum amyloid A has been shown to induce IL-6 and IL-23 from gut dendritic cells in mice, leading to the differentiation of Th17 cells and production of IL-17 through the intermediary action of gut dendritic cells [51]. Bacteria-derived adenosine 5′-triphosphate might be one mechanism by which bacteria induce this Th17 differentiation from naive CD4+ T cells [59]. Furthermore, IL-1β, but not IL-6, produced by lamina propria macrophages under the influence of commensal bacteria, acts directly on T cells and drives their differentiation into steady-state intestinal Th17 cells [60]. The implications of the immunomodulatory effects of SFB for humans has, however, to be confirmed since 16S rRNA sequencing has demonstrated SFB colonization only in young humans [61].
Bacteroides fragilis produces the molecule polysaccharide A (PSA), which induces Treg cells through TLR2 stimulation without the requirement of dendritic cells. This restrains the Th17-cell response and promotes Bacteroides colonization in colonic crypts [53]. Atarashi et al. [52] showed that Clostridium strains derived from human feces, and which lack prominent toxins and virulence factors, induce colonic Treg cells through the production of SCFAs (acetate, propionate, butyrate, isobutyrate) and by providing unspecified bacterial antigens (Treg cells specific to the strains of Clostridia) [52]. These findings may have the following implications for fungal infections. First, humans have an expanded Treg-cell population specific for mucosal fungi (A. fumigatus and C. albicans), which represses proinflammatory immune responses and maintains intestinal homeostasis (and thus commensalism) [62]. On the other hand, Treg cells also offer beneficial protective functions, while still exerting suppressive properties at later times, as demonstrated by the promotion of IL-17-dependent clearance of fungi during acute oral candidiasis in mice [63].
While most of the research to date has focused on cellular mechanisms of antifungal immunity as discussed above, much less is known concerning the humoral components of host defense, such as mucosal antibodies. Relatively little is known about the role of IgA, the representative of the humoral immune system at the mucosal level, in antifungal immunity. Human (plasma) IgA has been shown to bind yeast polysaccharides, pointing toward a potential role in fungus-driven immune responses [64]. In one study, after one month of consumption of the probiotics L. casei and Bifidobacterium breve, the salivary anti-Candida IgA levels in elderly humans were shown to be increased, and a small but significant reduction in fungal load was also demonstrated [65]. However, only two-thirds of the 62% of participants with an increase in IgA concentrations had a reduction in the amount of Candida in the oral cavity, so the evidence is not conclusive [65]. Furthermore, another study showed the opposite correlation between fungal colonization rate, mean CFU/mL counts and anti-Candida IgA levels [66]. Further work into the role of the humoral immune response to control fungal infections is required.

Implications for human fungal infections

As outlined above, the composition of the microbiome is of vital value to the defense of the host against infections. This becomes particularly apparent when the normal microbiome is disrupted, as occurs during treatment with antibiotics. A 2 week treatment with beta-lactam antibiotics has been shown to induce large fluctuations of both bacterial diversity and, importantly, the function of the intestinal bacteriome [67]. A 5 day course of ciprofloxacin in healthy adults had similar dramatic effects on microbiome composition [5]. Earlier studies from before the DNA-based microbiome analysis already showed the implications of these perturbations of the bacteriome for the mycobiome and susceptibility to fungal disease. A 3 day course of antibiotic administration, followed by an oral C. albicans challenge, showed that those mice that received antibiotics affecting anaerobic flora had more intestinal adherence of C. albicans [68]. A prospective study involving adult cancer patients treated with antibiotics for infectious complications also showed antibiotics with high anaerobic activity (ticarcillin-clavulanic acid) or high intestinal concentrations (ceftriaxon) to be responsible for increased colonization rate and density of Candida species [69]. As described above, a decreased colonization resistance by bacteria favors Candida colonization and increases the risk of tissue invasion (Fig. 1). This is illustrated by the fact that antibiotic treatment is an established risk factor for both mucocutaneous and disseminated candidiasis [30, 70].
Interestingly, murine gastrointestinal colonization models have revealed a role for C. albicansin the reassembly of the antibiotic-perturbed bacterial microbiome. The presence of Candida during re-colonization antagonizes Lactobacillus regrowth and promotes Enterococcus faecalispopulations in both the murine cecum and stomach [71, 72]. In the cecum, C. albicansincreases species diversity and Bacteriodetes population recovery [72].
Anticancer treatment triggers fungal infections in humans. For candidiasis, several factors involved in the pathogenesis, as discussed above, are often present: Candida colonization, mucosal barrier injury (mucositis), and immunodeficiency. As already discussed, these factors are influenced by the host microbiome, which in turn is perturbed by the use of prophylactic and/or therapeutic antibiotics. However, the role of chemotherapy as modulator of the microbiome has only recently started to be addressed. One study among eight patients showed the fecal flora to contain a reduced number of species and to have a decreased diversity, with a shift from firmicutes to bacteriodetes 7 days after receiving conditioning chemotherapy for bone marrow transplantation in the absence of antimicrobial drugs and nutritional support [73]. A second study involved nine pediatric patients with acute myeloid leukemia, each undergoing four cycles of chemotherapy, and found a reduced number of anaerobic bacteria compared with that of healthy volunteers, and a decrease in bacterial diversity and low similarity in species composition during treatment [74]. Although these studies do not directly assess the consequences for the mycobiome and fungal disease, this may be inferred from the decreased abundance of anti-inflammatory bacterial species and the increase in the firmicutes-to-bacteroidetes ratio after chemotherapy that, besides an impaired colonization resistance, may potentiate mucositis [75].
An important implication for human health is represented by the possibility to make the step from the association between a perturbed microbiome and disease susceptibility, to the manipulation of the microbiome as a therapeutic tool. Pioneering studies have shown that duodenal infusion of feces from healthy donors into patients with recurrent Clostridium difficileinfection restored microbiome diversity in the patients and cured 15 of the 16 included patients [76]. In fungal infections, manipulation of the microbiome is currently limited to the supplementation of beneficial (‘‘probiotic’’) bacterial strains. Lactobacilli application has been suggested to be successful in recurrent vulvovaginal candidiasis (RVVC) [77]. In a small, noncontrolled intervention study, women with RVVC with active infection applied one vaginal tablet composed L. fermentum LF10 and L. acidophilus AL02 every day for a week, followed by one tablet every 3 nights for the next 3 weeks, and thereafter one tablet every 7 days for the last month. Vaginal infection resolved in 87% and mild recurrences occurred in 12% of the patients [77]. These results are promising, however, randomized controlled trials, also including antifungal therapy, and with long-term follow-up, are needed to establish the place of lactobacilli supplementation in RVVC prevention and treatment. On the other hand, it has recently been suggested that a vaginal bacterial community with low abundance of lactobacilli is present in a proportion of asymptomatic women and that lactobacilli are abundantly present during most cases of vulvovaginal candidiasis [23, 78]. These data emphasize the difficulty of understanding the interaction between the host and the bacteriome that predisposes to infection [23, 78]. Besides the manipulation of the microbiome, the metabolic pathways involved in microbiome–host interactions offer another possibility to modulate the immune system. Although this strategy has not yet been applied in the prevention or treatment of human fungal infections, oral administration of PSA in mice protects against experimental colitis [79]. Similarly, interesting recent data suggest a potential role in the treatment of fungal infection for proteins secreted by the fungus Pichia, commensally present in the oral cavity. This protein inhibits Candida growth and inhibits germination, adhesion, and biofilm formation in vitro through nutrient limitation and shows a high cure rate in a mouse model of oral candidiasis [18].


High-throughput sequencing techniques have enabled unraveling of the composition of the microbiome of the human host and has shed light on the complexity and dynamics of host–fungal interactions. Besides the well-known cell–cell interface contacts between pattern-recognition receptors and pathogen-associated molecular patterns, bacterial commensals are important players in this host–fungal relationship. They shape our immune system and modulate specific components of the immune system involved in antifungal host defense. Further knowledge of the bacterial species involved and the metabolic and molecular mechanisms engaged provides relevant information for future therapeutic strategies. Development of pathogen-specific antibiotic therapies that would avoid the perturbation of beneficial bacterial–fungal interactions (such as those induced by broad-spectrum antibiotics) is essential for the prevention of clinical problems secondary to dysbiosis. Besides this, manipulation of the microbiome, either by microbiome ‘‘transplantation,’’ supplementation of beneficial strains or selective eradication of deleterious bacterial strains, or metabolic interventions derived from the pathways essential in bacteria–immune system interactions, can be envisaged as useful future adjuvant approaches to the conventional antifungal armamentarium.

No hay comentarios:

Publicar un comentario