Candida albicans the chameleon: Transitions and interactions between multiple phenotypic states confer phenotypic plasticity

Abstract

The ability of microbial cells to exist in multiple states is a ubiquitous property that promotes adaptation and survival. This phenomenon has been extensively studied in the opportunistic pathogen Candida albicans, which can transition between multiple phenotypic states in response to environmental signals. C. albicans normally exists as a commensal in the human body, but can also cause debilitating mucosal infections or life-threatening systemic infections. The ability to switch between cellular forms contributes to C. albicans’ capacity to infect different host niches, and strictly regulates the program of sexual mating. We review the unique properties associated with different phenotypic states, as well as how interactions between cells in different states can further augment microbial behavior.

The white-opaque switch governs multiple aspects of C. albicans biology

C. albicans exists as a harmless commensal in the gastrointestinal tract and is also commonly isolated from the skin and oral cavity of healthy individuals [1]. However, it can transition to an infectious colonizer of virtually any organ in the body, particularly in immunocompromised patients. C. albicans is currently the most common cause of life-threatening fungal infections in these individuals, with an estimated mortality rate of 35–67% [2,3].

C. albicans exhibits several forms of phenotypic switching in which cells undergo a reversible and heritable transition between alternative cell states. The best characterized of these epigenetic transitions is the white-opaque switch, named for the appearance of colonies on agar media. White cells are round and form brighter, dome-shaped colonies, whereas opaque cells are elliptical and form flatter, duller colonies [4]. Miller and Johnson demonstrated that the switch to the opaque state is necessary for efficient sexual mating [5**], stimulating interest into how phenotypic diversity impacts important physiological traits in C. albicans. Although refractory to mating, white cells were subsequently shown to respond to mating pheromones resulting in ‘sexual’ biofilms [6**], as further discussed below. White and opaque states exhibit many other distinct properties as outlined in Table 1. For example, white cells are predicted to have a fermentative metabolism, whereas opaque cells express genes associated with an oxidative metabolism [7]. Opaque cells are better colonizers of the skin, whereas white cells are more virulent in a bloodstream model of infection [8,9]. Furthermore, white cells are more susceptible to phagocytosis by host macrophages [10] and secrete a leukocyte chemoattractant [11], whereas opaque cells do not. These findings raise the possibility that the switch to the opaque state could promote immune evasion, although opaque cells are also more sensitive to reactive oxygen species, a common defense mechanism of host phagocytes [12].

Table 1. Properties of phenotypic states in C. albicans.

Characteristic properties of the heritable cell states in C. albicans. Cells in each state have unique phenotypic properties. MTL, mating-type locus. Predicted metabolic preferences based on gene expression studies.

State	MTL Configuration	Major Regulator	Cues supporting formation of each state	Metabolism	Preferred niche	Pheromone response
White	Heterozygous or homozygous	Efg1 [13,14,19]	Stable at 25°C or 37° C [35], aerobic conditions [37]	Fermentative [7]	Commensal, Bloodstream [8]	Biofilm formation [6]
Opaque	Usually homozygous [5,28] but can be heterozygous [30]	Wor1 [16–18]	25°C [35], CO2 [36], anaerobiosis [37], N-acetyl glucosamine [38]	Oxidative [7]	Skin [9]	Mating [5]
Gray	Heterozygous or homozygous [44]	N/A	Stable at 25°C or 37°C on rich media (YPD) [44]	Unique carbohydrate metabolism [44]	Skin [44]	Mating at low efficiency [44]
GUT	Heterozygous [45]	Requires Wor1 over-expression [45]	Passage through the mammalian gastrointestinal tract [45]	Adapted to the gastro-intestinal tract, high iron [45]	Commensal in gastro-intestinal tract [45]	No response [45]

Regulation of Phenotypic Switching

C. albicans strains are typically isolated in the white phenotypic state, in which the regulatory circuit is dependent on the transcription factor Efg1 [13–15]. Conversely, expression of the Wor1 transcription factor is necessary and sufficient for stable transitioning to the opaque state [16–18]. Wor1 acts in concert with at least five other transcription factors as part of a network of positive and negative feedback loops to regulate bi-stability in C. albicans [15,19]. Chromatin-level changes control the switching frequency, including a role for hyperacetylation of histone H3K56 in promoting opaque cell formation [20–23] and deposition of H2A.Z in promoting the white state [24]. Mediator complex also regulates white-opaque switching, as deletion of certain mediator components can destabilize the white or opaque states [25]. More recently, experiments demonstrated that the long 5′ untranslated region (5′ UTR) of WOR1 regulates switching by reducing translational efficiency on the WOR1 transcript [26]. These experiments reveal complex transcriptional, post-transcriptional and chromatin-mediated regulation of the switch, with parallels to other heritable circuits including embryonic stem cell differentiation [27].

The white-opaque switch is also controlled by transcription factors encoded at the MTL (mating-type locus). Switching from white to opaque typically requires that diploid strains are homozygous at this locus (i.e., either “MTLa/a” or “MTLα/α”) [5] [28], as expression of Wor1 is repressed by a complex of a1 and α2 transcription factors that is present in MTLa/α cells [29]. However, recent work has shown that certain a/α isolates can also switch to the opaque state under specific environmental conditions [30]. A similar white-opaque switch has been discovered in the related species Candida tropicalis [31**,32] and Candida dubliniensis [33]. In the case of C. tropicalis, white-opaque switching appears to be independent of MTL regulation [32,34]. The fact that the switch exists exclusively in three commensal fungi that co-evolved with the mammalian host suggests that this transition provides an advantage for survival in vivo.

Multiple environmental factors influence C. albicans switching between white and opaque states. Conditions that affect switching in vitro include temperature [35], carbon dioxide [36], anaerobiosis [37], N-acetylglucosamine [38] and genotoxic and oxidative stress [39]. In vivo, opaque-to-white switching occurs en masse during systemic infection [40], whereas increased switching in the opposite direction, from white to opaque, was observed in one strain during gastrointestinal colonization [41]. The sensitivity of the switch to a diverse array of environmental factors is reflective of multiple pathways impinging on white-opaque signaling. Some of these have been defined and include the pheromone MAPK cascade [42**], the cAMP/PKA pathway [38], and the Hog1 stress-activated protein kinase pathway [43].

Recently, C. albicans was shown to stably exist in two additional phenotypic states, termed the “gray” and “GUT” states, that are related to white and opaque forms (Figure 1). Gray cells displayed a distinct metabolic profile from white and opaque cells, and were more successful than either of these states in an ex vivo mouse tongue infection model [44**]. Cells in the GUT state were observed following gastrointestinal colonization using a strain that overexpressed the WOR1 gene [45*]. GUT cells were hypercompetitive when retested in this niche, indicating a potential role for this phenotypic state in promoting commensalism [45]. The metabolic reprogramming of cells can therefore generate specialized forms that enable colonization and infection of different host niches.

Figure 1. Candida albicans cells can exist in four related heritable states.

The white state is positively regulated by the transcription factor Efg1, while the opaque state requires expression of the transcription factor Wor1. Other transcription factors that co-regulate the white-opaque switch with Wor1 and Efg1 are not shown. Opaque cells often display characteristic ‘pimples’ on the cell surface when analyzed by scanning electron microscopy. Deletion of both of Wor1 and Efg1 can promote formation of the grey state. White cells overexpressing the Wor1 transcription factor can be induced to adopt the GUT (gastrointestinally induced transition) phenotypic state after passage through the murine gastrointestinal tract. GUT cells were characterized in a/α cell types and appear similar in shape to opaque cells, but unlike opaque cells they show increased fitness in colonizing the gastrointestinal tract.

C. albicans mating necessitates switching to the opaque state

The most striking difference between white and opaque cells is their sexual fecundity. Opaque cells are the mating competent form of C. albicans, mating a million times more efficiently than white cells [5]. Conventional heterothallic mating occurs between diploid a and α opaque cells and is driven by the secretion and response to sex-specific pheromones (Figure 2A). A conserved MAPK cascade transduces the pheromone signal resulting in activation of the transcription factor Cph1 (ortholog of Saccharomyces cerevisiae Ste12). This triggers polarized growth, cell fusion and karyogamy, resulting in the formation of tetraploid mating products [46–49]. Meiosis has not been observed in C. albicans and instead tetraploid cells return to the diploid state by a series of reductional mitotic divisions [50–52]. It was revealed that C. albicans diploid cells can reduce their ploidy to that of true haploids by a similar parasexual process, further extending the range of ploidy states that can be adopted by this species (also see Gerstein and Berman review, this issue). Haploids are homozygous at the MTL locus and can readily switch to the opaque state and mate to reform diploids, or can undergo spontaneous diploidization [53]. The transient formation of haploid a and α cells could potentially support mating events even in individuals that appear to be carrying a single a/α diploid strain.

Figure 2. Interactions between white and opaque cells in a population.

A. Conventional heterothallic mating occurs between diploid opaque C. albicans cells of opposite mating types. Pheromone signaling stimulates mating projection formation and cell fusion to generate a/α tetraploid products.

B. Same-sex opaque cells can undergo mating in the presence of an opposite-sex white cell (top) or opaque cell (bottom). Pheromone signaling between these cell types leads to amplification of the pheromone response and conjugation between cells of the same sex.

C. Pheromone signaling from opaque cells can stimulate responding white a or α cells to adhere to a surface and undergo formation of a ‘sexual biofilm’ that includes filamentous cells and secretion of an extracellular matrix. Within this immobilized structure, pheromone gradients between opaque cells are stabilized thereby leading to more efficient chemotropism and conjugation.

The population structure of C. albicans is predominantly clonal [54,55], although there is evidence of a mixed evolutionary history consistent with a sexually or parasexually reproducing species [56,57]. In addition to heterothallic mating, experiments have shown that C. albicans is capable of unisexual, homothallic mating [58*]. In particular, C. albicans opaque a cells lacking the α pheromone-degrading protease Bar1 experience autocrine pheromone signaling, resulting in same-sex a-a mating. Same-sex mating was also detected in ménage-à-trois matings, where conjugation between cells of the same sex was induced by pheromone secreted by cells of the opposite sex (see Figure 2B) [58]. Subsequent experiments established that activation of the pheromone response was sufficient to cause opaque cells to undergo same-sex conjugation [59].

Despite being unable mate, C. albicans white cells can respond to pheromones produced by opposite-sex opaque cells, upregulating production of their own pheromones in response. Synergistic signaling between the two cell types can enhance mating between rare opaque cells due to the amplification of pheromone levels [60**]. White cells can thereby facilitate the mating of opaque cells, despite being essentially sterile themselves (Figure 2B). Interestingly, the role of such “helper” white cells promotes both opposite-sex and same-sex mating between minority opaque cells [60]. Thus, cooperation between white and opaque cells can support mating even in populations where only one cell type is in the “mating-competent” opaque state.

Pheromone signaling drives biofilm formation

Biofilms are communities of cells that adhere to inert or biological surfaces. They are often associated with an extracellular matrix and communal cells exhibit distinct phenotypic properties from free-floating planktonic cells [61]. C. albicans causes a variety of device-associated infections that involve the formation of biofilms on catheters, heart valves, pacemakers, dentures, and prosthetic joints [62]. Biofilm formation is initiated when a basal layer of unicellular yeast adheres to the substrate, upon which a layer of pseudohyphal and hyphal cells grows during maturation [63]. These microbial communities are often drug resistant due to the occlusion of antifungal drugs, the upregulation of drug efflux pumps, and the existence of persister cells within the biofilm [64,65]. Conventional studies of biofilm formation in C. albicans have utilized MTLa/α cell types; these make up the majority of clinical isolates whereas less than 10% of strains are naturally homozygous for MTLa or MTLα [56,66].

More recently, it has been demonstrated that pheromone signaling in C. albicans can promote the formation of ‘sexual’ biofilms. These were first observed in white a and α cells responding to pheromones secreted by opaque cells of the opposite sex [6]. Responding white cells were induced to undergo substrate adhesion, filamentation, and development of a complex biofilm structure. Subsequently, it was shown that white cells also respond to pheromones secreted by opaque cells of the same sex, using a self-activation mechanism similar to that that drives same-sex mating [67]. Sexual biofilms contain a basal layer of unicellular yeast cells upon which a layer of filamentous cells is formed, but appear to lack the drug resistance and small-molecule impermeability characteristic of ‘conventional’ biofilms generated by MTLa/α cells [68]. Interestingly, the permeability of sexual biofilms supports the build-up and maintenance of pheromone gradients within the biofilm. Thus, pheromones secreted by minority opaque cells form stable gradients within this structure, facilitating chemotropism between opaque a and α partners [6]. Presumably as a result of this mechanism, mating between minority opaque cells is 5- to 100-fold higher when these cells are seeded in a sexual biofilm compared to a conventional biofilm [69*]. These findings suggest that C. albicans has evolved two types of biofilms that are influenced by their MTL configuration; conventional a/α biofilms that are ‘pathogenic’ based on their general impermeability and resistance to antifungals, and sexual biofilms that may provide an appropriate setting for mating to occur in vivo [68,70,71]. It is also hypothesized that the white-opaque switch evolved to generate sexual biofilms that facilitate mating of Candida strains in the host [71].

The transcriptional regulation of both conventional and sexual biofilms has been examined in C. albicans. In conventional a/α biofilms, a network of six major transcription factors regulates expression of ~2,200 genes [72*]. A comparison between the transcriptional control of sexual biofilms and conventional biofilms revealed that four of the six transcriptional regulators involved in conventional biofilms are also required for sexual biofilms, indicating partial overlap of the two regulatory circuits [73*]. Upstream signaling in conventional biofilms involves the Ras1/cAMP pathway, which activates the master transcriptional regulator Efg1 [68,74]. However, there have been conflicting reports concerning how the pheromone signal is transduced via the MAPK pathway to activate sexual biofilm formation. Experiments in our lab and others have shown that pheromone signaling in white cells requires the Cph1 transcription factor [42,60,73], and that this factor is essential for the transcriptional response and adhesive properties of pheromone-treated white cells [73]. In contrast, a separate set of studies indicated that pheromone signaling acts through the Tec1 transcription factor, and that Cph1 is not required for sexual biofilm formation [75]. Differences in experimental conditions may explain some of these differences [71], although all groups agree that Tec1 makes a significant contribution to both conventional and sexual biofilm formation [42,73,75,76], which reflects, at least in part, a key role for Tec1 in promoting hyphal growth [77,78]. Overall, it is clear that questions remain concerning the precise signaling mechanisms contributing to the establishment of sexual biofilms, as well as the role of these biofilms during infection of the host.

Recent studies in C. tropicalis have revealed striking differences with C. albicans with respect to the properties of white and opaque cells. For example, while switching between states is influenced by multiple environmental conditions in C. albicans, only N-acetylglucosamine has been shown to also affect the transition in C. tropicalis [31,32]. Furthermore, while pheromone signaling induces the formation of sexual biofilms in C. tropicalis as in C. albicans, it is responding opaque cells, not white cells, that contribute to biofilm development in the former [79*]. Thus, opaque cells are both the mating competent form and the biofilm forming cell type in C. tropicalis. These observations raise intriguing questions as to the evolution of the white-opaque switch in Candida species. At face value, they suggest that the switch did not simply evolve so that biofilms generated by white cells could facilitate mating between opaque cells, as white cells do not form sexual biofilms in C. tropicalis. It is possible that this mechanism represents a further refinement of the mating system in C. albicans that occurred after it diverged from C. tropicalis. There is also an incomplete understanding of the in vivo niches supporting the growth of white and opaque cells in the mammalian host. Given recent advances in our understanding of the spatial and temporal heterogeneity of the human mycobiome (see review by Xu and Dongari-Bagtzoglou, this issue), it is likely that we will soon learn more about the niches favored by white and opaque forms in the natural host. Distinguishing cells in different phenotypic states is likely be driven by emerging techniques such as RNA FISH (fluorescence in situ hybridization) to quantify the in vivo expression of state-specific gene transcripts.

Conclusions

C. albicans generates diversity by undergoing heritable and reversible transitions between multiple alternative states. The creation of phenotypic diversity is an established mechanism to ensure the stability and health of microbial populations, providing an ‘insurance policy’ against changing environmental conditions [80,81]. Morphological plasticity is a common attribute of both pathogenic and non-pathogenic species. In the model yeast S. cerevisiae, different phenotypic states can be generated by changes in chromosomal copy number or by prions [82,83], while phenotypic variation has been observed in multiple human fungal pathogens [84](see also review by Wang and Lin, this issue). In the case of C. albicans, not only do different states offer potential fitness advantages, but also interactions between cells in different states can provide additional benefits, such as increasing mating efficiency or induction of sexual biofilm formation. The result of this synergy is that the ‘whole is greater than the sum of its parts’ when it comes to evaluating phenotypic diversity. Further investigation of different phenotypes are certain to lead to greater understanding of C. albicans biology, as well as providing new insights into cell specialization in microbial species.

Highlights.

• Phenotypic switching between states occurs in the fungal pathogen C. albicans.

• Switching provides an adaptive strategy for different host environments.

• Interactions between cells in different states can promote unique responses.

• These include facilitating mating and formation of sexual biofilms.