MIG1 Regulates Resistance of Candida albicans against the Fungistatic Effect of Weak Organic Acids

Abstract

Candida albicans is the leading cause of fungal infections; but it is also a member of the human microbiome, an ecosystem of thousands of microbial species potentially influencing the outcome of host-fungal interactions. Accordingly, antibacterial therapy raises the risk of candidiasis, yet the underlying mechanism is currently not fully understood. We hypothesize the existence of bacterial metabolites that normally control C. albicans growth and of fungal resistance mechanisms against these metabolites. Among the most abundant microbiota-derived metabolites found on human mucosal surfaces are weak organic acids (WOAs), such as acetic, propionic, butyric, and lactic acid. Here, we used quantitative growth assays to investigate the dose-dependent fungistatic properties of WOAs on C. albicans growth and found inhibition of growth to occur at physiologically relevant concentrations and pH values. This effect was conserved across distantly related fungal species both inside and outside the CTG clade. We next screened a library of transcription factor mutants and identified several genes required for the resistance of C. albicans to one or more WOAs. A single gene, MIG1, previously known for its role in glucose repression, conferred resistance against all four acids tested. Consistent with glucose being an upstream activator of Mig1p, the presence of this carbon source was required for WOA resistance in wild-type C. albicans. Conversely, a MIG1-complemented strain completely restored the glucose-dependent resistance against WOAs. We conclude that Mig1p plays a central role in orchestrating a transcriptional program to fight against the fungistatic effect of this class of highly abundant metabolites produced by the gastrointestinal tract microbiota.

INTRODUCTION

Over 60% of fungal infections are caused by Candida albicans, and these infections range from debilitating skin or mucosal infections to potentially lethal disseminated disease, most especially in immunocompromised individuals and in nosocomial settings (1, 2). At the same time, this species is also the fungus most frequently isolated from the oral cavity, the vaginal mucosa, and the gastrointestinal (GI) tract of most healthy individuals (2), indicating a commensal nature of the host-microbe interaction in unperturbed conditions. A switch from asymptomatic colonization to opportunistic infection is supported by evidence of strain relatedness between C. albicans isolates from blood cultures and from the GI tract of the same patients (3). Hence, controlling C. albicans growth in the GI tract might limit one of the primary sources of systemic candidiasis.

Antibiotic treatment, which is well known to profoundly modify the GI microbiome (4, 5), is a strong risk factor for both vulvovaginal and systemic candidiasis in humans (6, 7). Moreover, most mouse models of C. albicans GI colonization rely on oral antibiotic treatment (8–10). Other models rely on the use of germ-free mice (11), infant mice (12, 13), which harbor a significantly different GI microbiome than adults (14), or specific dietary modifications associated with altered GI microbiota composition (15). Overall, these observations suggest that the microbiota plays a primary role in limiting the colonization of C. albicans in the mammalian GI tract and indicate that dietary interventions could alter this relationship. The underlying mechanisms, however, are currently unclear.

Our working hypothesis is that C. albicans growth can be controlled by metabolites produced by GI microbiota. Weak organic acids (WOAs), primarily produced by anaerobic bacteria via fermentation of undigested complex carbohydrates, are among the most abundant metabolites found on mucosal surfaces and the lumen of the gut (16). Vaginal lactobacilli secrete large amounts of lactic acid (∼55 to 111 mM), concomitantly lowering the mucosal pH to ∼4.5 (17, 18). Short-chain fatty acids (SCFAs), such as acetic, propionic, and butyric acid, are produced by a large spectrum of GI bacteria and reach total concentrations of up to 140 mM (16, 19). However, with the exception of the stomach, the pH of the GI tract is generally higher than that of the vagina throughout most of its length (from pH 5.5 to 7 in the colon to pH 7 to 9 in the jejunum) (20). Consistent with our hypothesis, WOAs suppressed C. albicans growth and colony formation in vitro (15, 21, 22); however, only a few concentrations have been tested so far, and the mechanism of inhibition was not addressed. Moreover, a combination of a high lactic acid concentration and low pH is thought to be responsible for restricting the colonization of C. albicans in the vagina of healthy women (22, 23); whether WOAs might also limit C. albicans growth at the pH levels normally present in the GI tract has not been addressed.

The aim of this study was to evaluate the ability of WOAs normally produced by microbiota to limit the growth of C. albicans and to investigate their fungistatic effects under physiologically relevant concentrations and pH values. A systematic genetic screen uncovered MIG1 as a central regulator of WOA resistance in C. albicans. These results suggest that C. albicans colonization in the human GI tract might at least in part be regulated by microbiota-derived metabolites and point toward nutritional interventions as a potential strategy to lower the risk of fungal infections.

MATERIALS AND METHODS

Strains and media.

All strains used in this study are reported in Table S1 in the supplemental material. The C. albicans transcription factor (TF) deletion library was acquired from the Fungal Genetics Stock Center (http://www.fgsc.net/). All stock cultures were preserved in 35% glycerol and maintained at −80°C. Unless otherwise specified, cells were grown in yeast extract-peptone-dextrose (YPD) medium (1% [wt/vol] yeast extract, 2% [wt/vol] peptone, and 2% [wt/vol] d-glucose, supplemented with 1.5% [wt/vol] agar for solid medium only) or De Man, Rogosa, Sharpe (MRS) medium (Sigma) (24) at 37°C in a shaking incubator at 150 to 200 rpm. The composition of YPM medium was similar to that of YPD except that 2% (wt/vol) d-glucose was replaced by 2% (wt/vol) maltose.

Quantification of fecal WOAs.

Human stool samples were obtained with informed consent according to protocols approved by the National University of Singapore (NUS) institutional review board (IRB), filed under NUS-IRB reference code 12-208 (approval number NUS 1615). Feces (∼0.1 g) were homogenized in 1 ml of distilled water and centrifuged for 5 min at 13,000 rpm. Supernatants were purified first through a 40-μm filter and then through a HyperSep SLE column (Thermo Scientific) for removal of phospholipids. The filtered supernatants were then stored at −20°C until they were analyzed by high-performance liquid chromatography (HPLC), using a protocol similar to a previously published method (25). Lactic, acetic, propionic, and butyric acid were analyzed using an HP series 1100 HPLC system (Palo Alto, CA) equipped with a G1312A binary pump, G1315A diode array detector (DAD), G1329A autosampler, and ChemStation software. WOAs were separated on a PL Hi-Plex H column (8-μm particle size, 300 by 7.7 mm; Polymer Laboratories) run isocratically with 6 mM sulfuric acid in deionized water at a flow rate of 0.3 ml/min at 60°C. WOAs were detected at a wavelength of 210 nm and quantified using external standard curves from 0.5 to 100 mM for the respective authentic acids (Fluka). HP ChemStation software was employed to pilot the HPLC and determine the concentrations of each individual organic acid.

High-throughput growth assays.

Strains were first revitalized by streaking out glycerol stocks onto fresh YPD agar and were subcultured into YPD medium overnight at 37°C. Overnight cultures of C. albicans were then used to inoculate 96-well deep-well blocks at a starting optical density at 600 nm (OD₆₀₀) of 0.2 in a final volume of 1.7 ml fresh medium with the concentrations of WOAs indicated in the respective figure. The pH was adjusted by the use of either 5 M hydrochloric acid or 5 to 10 M sodium hydroxide solutions. Deep-well blocks were inoculated by a Freedom EVO 150 automated liquid-handling robot (Tecan) and incubated at 37°C under orbital shaking at 200 rpm. Growth curves were generated by regularly measuring the OD₆₀₀ of each culture with a SpectraMax Plus384 absorbance microplate reader (Molecular Devices). Background ODs (inoculum-free wells with medium only, cultured under the same conditions as inoculated wells) were subtracted at each time point from the acquired OD readings. The changes in OD over time were then fitted against an exponential growth curve and used to determine the growth rates.

Transcription factor screen.

The C. albicans library of 165 TF knockout strains was generated and described previously (26) and contains two independently derived knockouts for each TF, designated set X and set Y. Our genetic screen focused on set X of the deletion library and assayed each deletion strain for sensitivity or resistance to WOAs based on growth rates in liquid cultures. The library was first revitalized by spotting out the glycerol stocks (which were provided in a 96-well format) onto fresh YPD agar plates with the help of the liquid-handling robot before subculturing the strains into 96-well deep-well blocks containing YPD medium overnight at 37°C. Growth inhibition studies were then performed in YPD medium at pH 5.5 using the same protocol described above. The concentrations of each WOA and of NaCl chosen for use in this screen were the corresponding 50% inhibitory concentrations (IC₅₀s) for the wild-type (WT) control strain provided in the library which we previously determined in YPD medium at pH 5.5 (data not shown). Assays were repeated with at least four independent biological replicates. Growth rates were obtained by fitting a series of OD readings against an exponential curve over a 6- to 8-h time course, omitting time points corresponding to the lag phase or the stationary phase of growth.

RESULTS

Weak organic acids are fungistatic under physiological conditions.

To test under which circumstances weak organic acids (WOAs) directly affect C. albicans growth, we set up high-throughput assays based on liquid-handling robotics to quantitatively assess fungal growth rates in a wide range of combinations of WOA concentrations and pH values and in different growth media (YPD and MRS). A clear dose-response curve was obtained for each WOA tested, in spite of the fact that the pH of the medium was adjusted to the same value (Fig. 1A; see also Fig. S1A and B in the supplemental material). Nonlinear regression of the dose-response curves found significantly different values for the IC₅₀ (concentration required to inhibit growth rate by 50%) of each acid tested. In particular, butyric acid had the most potent activity against C. albicans growth, followed by propionic, acetic, and lactic acid (Fig. 1B; see also Fig. S1B). With the exception of that of lactic acid, the IC₅₀s were in the same order of magnitude of previously reported (19) or herein tested (Fig. 1C) physiological concentrations in human stool, which contained high levels of acetic acid (30 to 100 μmol/g), lower levels of propionic and butyric acid (5 to 20 μmol/g), and often, undetectable amounts of lactic acid (Fig. 1C). Moreover, WOAs were consistently more potent than HCl at pH 4, 5.5, and 6.5 (Fig. 1B; see also Fig. S1B), thus covering a range of acidity levels found throughout the human GI tract (20, 27). Overall, these data demonstrate that WOAs possess a specific growth inhibition activity that goes beyond their pH-modulating property, consistent with our previous results on the effects of WOAs on the C. albicans transcriptome (28). Importantly, the inhibition of fungal growth by WOAs was not restricted to C. albicans but was conserved across other yeast species, including the other CTG clade member, Candida parapsilosis, and the non-CTG clade members Candida glabrata and Saccharomyces cerevisiae (Fig. 1D). The range of IC₅₀s was comparable across fungi, with butyric acid scoring as the most potent of the four WOAs tested against all four species tested (Fig. 1D), suggesting that it might be active even at the relatively smaller concentrations of this particular WOA normally found in the human gut (Fig. 1C) (19). The same was true for propionic acid and, to a lesser extent, acetic acid, which scored as the second- and the third-most potent WOAs, respectively, for all four species tested (Fig. 1D). Interestingly, C. parapsilosis, a yeast species less frequently associated with humans than C. albicans (29), appeared to be the most sensitive of the four species tested to acetic, propionic, and butyric acid.

FIG 1.

WOA exposure reduces yeast growth at physiologically relevant concentrations and pH values. Quantitative growth assays of C. albicans strain SC5314 (A to C) and other yeast species (D) under a range of different concentrations of lactic, acetic, propionic, or butyric acid. Experiments were carried out using a high-throughput liquid-handling robot, and growth rates determined via a turbidimetric assay. (A) Dose-response curve in YPD adjusted to pH 5.5 by the addition of various amounts of NaOH. Growth rates are expressed as doublings per hour. Data are represented as average results and standard errors of the means (SEM) from 4 biological replicates. (B) IC₅₀ of each individual acid as obtained from the data in panel A at pH 4, 5.5, and 6.5 by nonlinear regression. Error bars represent the SEM. (C) HPLC measurements of concentrations of lactic, acetic, propionic, and butyric acid in fecal samples from 24 healthy human volunteers. (D) Fitted IC₅₀s of each individual acid for the four indicated yeast species grown in YPD at pH 5.5. Data are represented as the average results and SEM from ≥4 biological replicates.

Significant reductions in optical density were observed at both 6 and 24 h after WOA treatment initiation, indicating that these compounds have a long-lasting and not just an acute effect on C. albicans growth (28). Moreover, little or no significant decrease in cell viability was observed with any of the acids tested, indicating a fungistatic rather than a fungicidal effect (Fig. 2A; see also Fig. S2A in the supplemental material). Consistent with the notion that acidic conditions favor the yeast form of C. albicans (30) and that some WOAs might actually inhibit hyphal morphogenesis (31), no induction of germination was observed under any of the conditions tested (Fig. 2B; see also Fig. S2B). Hence, the observed changes in optical density cannot be attributed to a switch to hyphal growth. In conclusion, WOAs exhibit long-lasting fungistatic properties toward C. albicans at physiologically relevant concentrations and pH values. Furthermore, each WOA tested had a distinct potency, with lactic acid being the weakest and the SCFAs propionic and butyric acid being the most potent inhibitors of yeast growth.

FIG 2.

WOAs do not affect the viability or morphology of C. albicans. (A) Passive propidium iodide incorporation was used to measure the percentage of viable cells during exposure to the IC₅₀ of the indicated WOA in comparison to the percentage in control YPD culture at 6 h and 24 h of incubation at pH 5.5. Data are represented as the average results and SEM from 3 biological replicates. (B) Cell morphologies in control YPD cultures and after 6 h of exposure to the IC₅₀ of each indicated WOA at pH 5.5. Magnification, ×400. *, P < 0.05; **, P < 0.01.

MIG1 is required for WOA resistance in C. albicans.

C. albicans senses and adapts to its environment via numerous pathways (32). To understand the genetic requirements of this fungus for its response to WOAs, we took advantage of a previously published library of C. albicans mutant strains, each of which had had 1 of 165 transcription factors (TFs) deleted (26). The library was subjected to robotically assisted high-throughput growth assays, with the medium adjusted to pH 5.5 and each compound used at its respective IC₅₀ (Fig. 3). To increase specificity in identifying TFs involved in the WOA response, as opposed to a more general response to pH or to fungistatic stress, we performed control experiments in which the pH of the medium was adjusted with HCl alone or the IC₅₀ of sodium chloride (NaCl) was used, respectively. Consistent with prior reports (33, 34), we were unable to calculate a growth rate in the control medium for five strains (with DPB4, HFL1, RBF1, orf19.6888, or NRG1 deleted) and had to exclude them from further analysis. An additional strain (tup1Δ) showed strong growth impairment in YPD (below 50% of the median growth rate for all of the strains in the library), probably due to its exclusively filamentous growth (35), and was thus not considered further. We next calculated the ratios of the growth rates in the presence versus the absence of treatment for both the WT and all mutant strains to measure the intrinsic sensitivity of each strain to each compound. Finally, we calculated a ratio of growth ratios (quotient of growth ratio of a tested strain over the growth ratio of the control strain) for each mutant, to detect strains with altered sensitivity to a given compound in comparison to that of the WT (see Data Set S1 in the supplemental material). Ratios of growth ratios of >1.25 or <0.75 and associated with a one-sample t test P value of <0.05 were considered significant (Fig. 3A to F). By this analysis, 22 strains showed significantly altered levels of sensitivity to one or more WOAs, while no hits were recovered from the NaCl or the pH control experiments (Fig. 3G). Of these 22 TF mutants, 13 showed an altered sensitivity to acetic acid, 10 to propionic acid, 6 to butyric acid, and 7 to lactic acid, but only one strain, in which the MIG1 gene (orf19.4318) was deleted, displayed higher sensitivity to all four WOAs (Fig. 3G).

FIG 3.

Requirement of MIG1 for WOA resistance in C. albicans. (A to E) Volcano plots of the results of the quantitative high-throughput growth assays of 165 TF mutant strains in YPD adjusted to pH 5.5 and supplemented with one of four WOAs or NaCl as indicated, with each used at its respective IC₅₀ (as determined using the WT control). Ratios of growth ratios and P values were obtained by comparing growth data from the conditions described above with growth data obtained in parallel in YPD adjusted to pH 5.5. (F) Same as experiment whose results are shown in panels A to E but comparing growth data from YPD adjusted to pH 5.5 against growth data from unadjusted YPD. In all panels, dashed red lines represent the P value cutoff of 0.05; dashed gray lines are plotted at ratio of growth ratios values of 0.75 and 1.25; red circles represent mutant strains passing both significance thresholds; filled gray symbols represent WT control strains; and mig1Δ strain data points are labeled “MIG1.” (G) Four-way Venn diagram of the significant hits obtained under each WOA treatment. n = 4.

To validate the role of MIG1 in sensitivity to WOAs, we inactivated both MIG1 alleles in a strain SC5314 background by using the SAT1-flipping strategy (36). After confirming the inactivation of the MIG1 gene and the apparent ploidy of the resulting mig1Δ mutant (see Fig. S3 in the supplemental material), we confirmed the role of MIG1 in the resistance to WOAs by using more traditional liquid cultures grown in flasks and measured with a standard cuvette spectrophotometer. Consistent with the results of the large-scale screen, mig1Δ cells displayed higher levels of sensitivity to lactic, acetic, propionic, and butyric acid, with growth reductions of 24, 79, 19, and 29%, respectively, compared to the growth of strain SC5314 (P < 0.05), whereas no significant difference was observed for growth with NaCl (Fig. 4A). Overall, these results point toward Mig1p as a central transcriptional regulator of an intrinsic resistance mechanism of C. albicans to WOAs.

FIG 4.

Effect of glucose on WOA-mediated fungistatic mechanism. (A) Quantitative growth assays of control strain (SC5314) and mig1Δ homozygous mutant performed in YPD adjusted to pH 5.5 and supplemented with one of four WOAs or NaCl, each used at its respective IC₅₀ as determined with the control strain in the corresponding medium. Results are expressed as the ratio of the growth rate of the strain grown in the presence of acid or NaCl over the growth rate of the strain grown in the respective control medium. (B) Quantitative growth assays of SC5314 grown in YPD or YPM medium supplemented with one of four WOAs or NaCl. (C) Quantitative growth assays of SC5314 (WT), mig1Δ heterozygous mutant (mig1 hetero.), mig1Δ homozygous mutant (mig1 homo.), and MIG1-complemented strain (mig1 compl.) grown in YPD or YPM medium supplemented with acetic acid at the IC₅₀ calculated for the WT in the respective medium. If not indicated, differences between samples are not significant. Data are represented as the average results and SEM from ≥3 biological replicates. **, P < 0.01; *, P < 0.05.

Glucose antagonizes the WOA-mediated fungistatic mechanism.

Mig1p has been extensively studied in the budding yeast S. cerevisiae within the context of glucose repression. It is posttranslationally activated in the presence of glucose and represses the expression of genes required for the utilization of alternative carbon sources (37). To test whether the presence of glucose was required to protect C. albicans from the fungistatic effect of WOAs, we replaced the 2% (wt/vol) glucose in YPD medium with 2% (wt/vol) maltose, a disaccharide known not to trigger glucose repression, yielding YPM medium. In accordance with our hypothesis, the fungistatic effects of acetic, propionic, and butyric acid were significantly increased in YPM medium compared to their effects in YPD medium (Fig. 4B), indicating that the presence of glucose is required for intrinsic resistance of C. albicans to SCFAs. The absence of glucose, however, did not significantly alter the sensitivity of cells to lactic acid, indicating the potential existence of a glucose-independent function of MIG1 in conferring resistance to this particular WOA on C. albicans.

To more directly test whether Mig1p was involved in glucose-dependent modulation of WOA resistance, the carbon source experiment described above was repeated with the previously mentioned mig1Δ mutant, as well as with a MIG1-complemented strain, obtained by reintroducing a wild-type MIG1 allele into the mig1Δ strain. Consistent with glucose acting through MIG1, no significant difference in acetic acid-mediated growth inhibition was observed when the mig1Δ homozygous mutant was incubated in YPD or YPM medium. Wild-type, mig1Δ heterozygous, and MIG1-complemented cells instead were all hypersensitive to acetic acid when grown in YPM medium, to levels that phenocopied the mig1Δ homozygous mutant. In the presence of glucose, however, the MIG1-complemented strain displayed resistance to acetic acid that was restored to levels indistinguishable from those of wild-type or mig1Δ heterozygous cells (Fig. 4C), thereby excluding the possibility that the previously observed hypersensitivity phenotype of the mig1Δ mutant was due to bystander mutations.

DISCUSSION

Besides immunodeficiency and immune suppression, the use of antibiotics represents an important risk factor for human candidiasis (38). Since antibiotics are well known to alter the GI microbiome (4), they are also expected to alter the production of microbiota-derived metabolites, such as WOAs. Could these metabolites normally restrict C. albicans growth? With the exception of a few studies (21, 22, 28, 31, 39), little was known about the effects of WOAs on C. albicans. Here, we extended these previous studies by systematically and quantitatively analyzing the effects of lactic, acetic, propionic, and butyric acid on fungal growth via robotically assisted high-throughput assays. By this means, we found each of these four molecules to negatively affect several yeast species via a fungistatic rather than a fungicidal mechanism. While we noticed significant differences in the potency of each molecule, all except lactic acid had specific inhibitory effects at physiologically relevant concentrations and pH values, indicating that WOAs have the potential to inhibit C. albicans growth in various body sites. Furthermore, we recently identified a pH-independent core transcriptional response to all four acids tested (28), suggesting the existence of a common transcriptional regulatory circuit responding to WOAs in C. albicans. According to our genetic screen, only a single TF mutant strain (hms1Δ) displayed a higher resistance to one of the WOAs (i.e., propionic acid), while 21 had significantly increased levels of sensitivity to one or more WOAs (Fig. 3). The transcriptional regulatory circuit underlying the response of C. albicans to WOAs might therefore be primarily dedicated to mediating resistance against these abundant microbiota metabolites. This observation is consistent with the long history of association with the human body of this fungus, which hence might have evolved mechanisms to sense and respond to various challenges issued by GI microbiota (32, 40).

MIG1 was uncovered by our screen as the only TF required for resistance to all four WOAs tested, and we thus propose this to be a central transcriptional regulator of a novel resistance mechanism of C. albicans to WOAs. Among the TF genes found to be required for resistance to butyric acid, WAR1 was previously described to promote resistance to the WOA sorbate in S. cerevisiae (41). MNL1, on the other hand, was previously described to regulate a stress response during exposure to acetic and butyric acid in C. albicans (39) but was not uncovered by our screen. This difference could be due to the fact that MNL1 was identified in minimal medium at pH 3, while we used rich medium at a pH above the pK_a of the molecules (pH 5.5). Interestingly, both MNL1 and MIG1 have known associations with a third TF, NRG1, which was described not only to antagonize MNL1 (39) but also to share with MIG1 the regulation of some target genes (42). However, due to its filamentation phenotype, we were unable to measure the growth rate of the nrg1Δ mutant strain in our liquid cultures, which explains why it was absent from our list of candidate TFs. Further experiments would be necessary to test a possible involvement of MNL1 and NRG1 in the MIG1-dependent mechanism of resistance to WOAs.

While the role of MIG1 has been thoroughly described within the context of glucose repression in both S. cerevisiae and C. albicans (37, 43, 44), its implication in WOA resistance is novel. Further investigation would be required to identify the MIG1-regulated pathways involved in C. albicans resistance to WOAs. In this regard, it has to be noted that in spite of MIG1 being principally a transcriptional repressor, it might also behave as a transcriptional activator (45). Therefore, MIG1 might theoretically mediate WOA resistance either by repressing the expression of hypothetical WOA importers or by activating the expression of an exporter, such as the S. cerevisiae Pdr12p (46). However, no clear ortholog of PDR12 exists in the C. albicans genome. Other possible resistance mechanisms include alteration of the plasma membrane lipid composition (47) or other as-yet-undocumented mechanisms.

Our observation that, in the absence of glucose, cells displayed significantly increased levels of sensitivity to acetic, propionic, and butyric acid shows that the availability of specific carbon sources can have important consequences on how cells cope with environmental insults. A similar phenomenon was previously reported for both C. albicans and S. cerevisiae under osmotic stress, whereby resistance of these fungi to high concentrations of NaCl was dependent on the presence of glucose in the medium (48). Here, we also observed a trend toward increased sensitivity to high salt in the absence of glucose (Fig. 4B), but our results show that MIG1 is unlikely to play a role in resistance to osmotic stress (Fig. 4A). The modulation by glucose of a MIG1-mediated resistance mechanism thus appears to be specific to WOA stress.

In any case, we report here that conditions that inhibit C. albicans growth in vitro are similar to those in host environments found in vivo. Thus, strategies aimed at controlling fungal growth under the physiological conditions of the GI tract might prove successful in regulating overall C. albicans colonization levels. In conclusion, our results predict that an antagonistic interplay between WOAs (which are highly abundant microbiota-derived metabolites) and glucose (a very common nutrient) might regulate C. albicans growth in the mammalian GI tract. Deeper understanding of the mode of action of these natural products in C. albicans growth inhibition should lead to innovative strategies for the control of this opportunistic pathogen. The use of probiotic bacterial strains designed to alter the composition of WOAs in specific host niches (e.g., the vagina and GI tract) represents a valuable avenue to explore, especially now that C. albicans strains with acquired resistance to several available antifungal drugs are being reported more frequently (49). Such strategies are already being applied by treating vulvovaginal candidiasis with probiotic lactobacilli (50), as well as for curing disease caused by other pathogens, such as Clostridium difficile, where fecal transplants have been proven to be highly effective (51). Alternatively or in conjunction with probiotic/prebiotic treatments, rationally designed dietary interventions, such as reduction of starches and simple sugars, might lead to similar results. As previously demonstrated in mice, a rationally designed diet aimed at reducing the level of lactobacilli and the ensuing production of lactic acid was sufficient to allow low levels of C. albicans GI colonization in mice untreated with antibiotics (15). While the effectiveness and safety of these potential therapies still need to be fully ascertained, further understanding of the mechanisms of action of bacterial metabolites against fungal pathogens is of primary importance to evaluate potential risks and to develop novel treatment opportunities.