Discovery of a "white-gray-opaque" tristable phenotypic switching system in candida albicans: roles of non-genetic diversity in host adaptation

Abstract

Non-genetic phenotypic variations play a critical role in the adaption to environmental changes in microbial organisms. Candida albicans, a major human fungal pathogen, can switch between several morphological phenotypes. This ability is critical for its commensal lifestyle and for its ability to cause infections. Here, we report the discovery of a novel morphological form in C. albicans, referred to as the "gray" phenotype, which forms a tristable phenotypic switching system with the previously reported white and opaque phenotypes. White, gray, and opaque cell types differ in a number of aspects including cellular and colony appearances, mating competency, secreted aspartyl proteinase (Sap) activities, and virulence. Of the three cell types, gray cells exhibit the highest Sap activity and the highest ability to cause cutaneous infections. The three phenotypes form a tristable phenotypic switching system, which is independent of the regulation of the mating type locus (MTL). Gray cells mate over 1,000 times more efficiently than do white cells, but less efficiently than do opaque cells. We further demonstrate that the master regulator of white-opaque switching, Wor1, is essential for opaque cell formation, but is not required for white-gray transitions. The Efg1 regulator is required for maintenance of the white phenotype, but is not required for gray-opaque transitions. Interestingly, the wor1/wor1 efg1/efg1 double mutant is locked in the gray phenotype, suggesting that Wor1 and Efg1 could function coordinately and play a central role in the regulation of gray cell formation. Global transcriptional analysis indicates that white, gray, and opaque cells exhibit distinct gene expression profiles, which partly explain their differences in causing infections, adaptation ability to diverse host niches, metabolic profiles, and stress responses. Therefore, the white-gray-opaque tristable phenotypic switching system in C. albicans may play a significant role in a wide range of biological aspects in this common commensal and pathogenic fungus.

Figure 1. Three distinct phenotypes (white, gray, and opaque) of C. albicans on YPD medium.

G, gray; G-Sec, gray sector; O, opaque; Op-Sec, opaque sector; W, white. The strain (BJ1097) was used. (A) Morphologies of white, gray, and opaque colonies on YPD agar without phloxine B. (B) Morphologies of white and sectored colonies on YPD agar containing phloxine B. A white colony initially grown on Lee's medium was replated onto YPD agar. The colonies were imaged after 5 days of growth at 25°C. The dye phloxine B stained opaque sectors dark pink and gray sectors light pink. (C) Colony and cellular morphologies of the three phenotypes of C. albicans on YPD agar with phloxine B. Colonies were grown at 25°C for 5 days. Scale bar, 10 µm.

Figure 2. Switching frequencies of the white-gray-opaque tristable switching system in C. albicans.

Gr, gray; Op, opaque; Wh, white. Colonies (strain BJ1097) were grown under the conditions indicated in the figure. Colonies were counted for switching frequency calculations. (A) Switching frequencies in air at 25°C for 5 days. (B) Switching frequencies in air at 37°C for 4 days. (C) Switching frequencies in 5% CO₂ at 25°C for 5 days.

Figure 3. Differential gene expression profiles in white, gray, and opaque cells.

Differentially expressed genes were defined as ones with greater than or equal to 4-fold difference of relative expression levels revealed by RNA-Seq analysis between two different cell types. The numbers of overlapping genes less than 4-fold difference of relative expression levels are also show. (A) Distinct gene expression patterns of white-gray, white-opaque, and gray-opaque cell types. The numbers indicate differentially expressed genes between the two cell types compared. (B) Distinct gene expression patterns of white, gray, and opaque cell types. The numbers indicate white (69)-, gray (45)-, or opaque (129)-enriched genes. (C) Verification of cell type-enriched genes by quantitative real-time PCR assays. WH11, white-enriched; OP4 and WOR1, opaque-enriched; HSP31, LIP9, HMS1, CAS2, and PGA26, gray-enriched.

Figure 4. Differential Sap activities in white, gray and opaque cells.

(A) Sap activities in white, gray, and opaque cells cultured on solid YCB-BSA medium. 5×10⁶ cells of each cell type in 5 µl ddH₂O were spotted onto YCB-BSA medium plates and grown at 25°C for six days. The white precipitation zones (halos) around the cell spots indicate Sap-mediated BSA hydrolysis. Scale bar, 1 cm. (B) Sap activities in white, gray, and opaque cells cultured in liquid media. Cells were grown in liquid Lee's glucose or YCB-BSA medium. Quantitative activity assays are described in the Materials and Methods section. (C and D) Expression of GFP in the reporter strains of SAP1p-GFP (C) and SAP2p-GFP (D). Cells were grown on Lee's glucose and YCB-BSA plates for four days at 25°C in air. Scale bar, 10 µm. (E) Relative expression levels of SAP1 and SAP2 in white, gray, and opaque cells. Cells were grown in liquid Lee's glucose and YCB-BSA media for 24 hours at 25°C in air.

Figure 5. Virulence of white, gray, and opaque cells in systemic and cutaneous infections.

(A and B) Survival curves for white, gray, and opaque cells of strain BJ1097. 3.75×10⁶ (A) or 1×10⁶ (B) cells of each cell type were injected into each mouse via tail vein. For each cell type, ten mice were used for infection. Survival rates with significant difference (p<0.05, Student's t-test, two tails): wh>op, wh>gr, and op>gr in (A); wh>op and wh>gr in (B). (C) White, gray, and opaque cells differ in fungal burdens in different organs in a mouse systemic infection model. 5×10⁵ cells of each type (white, gray, or opaque) of BJ1097 were mixed with 5×10⁵ cells of SC5314N (Nou^R, locked in white phase) in 250 µl PBS and then injected into a mouse via tail vein. Mice were killed at 24 hours after injection. Different organs were used for fungal burden assays. Eight mice were used for each cell type. Competitive index = the ratio of CFU number of BJ1097 to CFU number of SC5314N in each organ. Each cycle represents a value of competitive index in a mouse (e.g., a white cycle represents the value of the ratio of BJ1097/SC5314N). Bar, median value. Competitive indexes with significant difference (p<0.01, Student's t-test, two tails): In the liver, op>wh and op>gr; in the kidney, wh>op and wh>gr; in the spleen, op>wh and gr>wh; in the lung, op>wh; and in the brain, op>wh and op>gr. Wh, white cells, gr, gray cells, and op, opaque cells. (D) Ex vivo tongue infection model. Tongues were excised from humanely killed mice and one tongue was added to each well of a 24-well polystyrene plate containing 1×10⁷ cells of white, gray, or opaque cells in 1 ml PBS. After 24 hours of incubation at 37°C, cells in the liquid and on the tongue (after homogenization) were plated onto YPD agar for CFU assays. The total cell number of each well (including cells in the liquid and attached to the tongue) is shown. “0 h” indicates initial inoculated cell number (1×10⁷) in each well. *p<0.05; **p<0.01 (Student's t-test, two tails). The experiment was repeated three times. For each time, three tongues were used for each cell type. The result of a representative experiment is shown.

Figure 6. Roles of Wor1 and Efg1 in the regulation of white-gray-opaque transitions in C. albicans.

NA, not available. The complete switching frequency data under different conditions are shown in Figure S10. (A) White-gray switching frequencies (%) on YPD medium. The wor1/wor1 mutant cannot switch to the opaque phenotype under all conditions tested. (B) Gray-opaque switching frequencies (%) on YPD medium. The efg1/efg1 mutant cannot switch to the white phenotype under all conditions tested. (C) Switching frequencies (%) on YPD medium. The wor1/wor1 efg1/efg1 double mutant is locked in the gray phase under all conditions tested. (D) Switching frequencies (%) on YPD medium of the wild type control (adapted from Figure 2A). (E) Regulatory model of the white-gray-opaque tristable phenotypic switching system.

Figure 7. White-gray-opaque transitions are independent of the MTL.

(A) Cellular morphologies of white, gray and opaque cells of BJ1097a (MTL a/Δ) and BJ1097α (MTLΔ/α). Scale bar, 10 µm. (B) White, gray, and opaque cells differ in mating efficiencies. Experimental strains with MTLΔ/α URA3 ⁺ sat1⁻ (or Clon−) genotype: the wor1/wor1, efg1/efg1, wor1/wor1 efg1/efg1 double mutants, and the wild type BJ1097α. Tester strain WTa (MTL a/a, ura3 ⁻ Clon+). Mating efficiency = average ± standard deviation (SD). The fold changes of mating efficiencies to that of WT white cells are also shown. The mean values were used for calculations.