Genetic Analysis of NDT80 Family Transcription Factors in Candida albicans Using New CRISPR-Cas9 Approaches

Abstract

Ndt80 family transcription factors are highly conserved in fungi, where they regulate diverse processes. The human fungal pathogen Candida albicans contains three genes (NDT80, REP1, and RON1) that encode proteins with similarity to Saccharomyces cerevisiae Ndt80, although the homology is restricted to the DNA binding domain. To better understand their role in virulence functions, we used clustered regularly interspaced short palindromic repeat/CRISPR-associated gene 9 (CRISPR/Cas9) to delete the three NDT80-family genes. An ndt80Δ mutant showed strong defects in forming hyphae in response to serum or N-acetylglucosamine (GlcNAc), which was linked to the ability of Ndt80 to regulate the expression of RAS1, an upstream regulator of hyphal signaling. Conversely, the ndt80Δ mutant formed hyphal cells on glycerol medium, indicating that Ndt80 is not required for hyphal growth under all conditions. In contrast to our previously published data, a ron1Δ single mutant could grow and form hyphae in response to GlcNAc. However, deleting RON1 partially restored the ability of an ndt80Δ mutant to form hyphae in response to GlcNAc, indicating a link to GlcNAc signaling. REP1 was required for growth on GlcNAc, as expected, but not for GlcNAc or serum to induce hyphae. The ndt80Δ mutant was defective in growing under stressful conditions, such as elevated temperature, but not the ron1Δ mutant or rep1Δ mutant. Quantitative assays did not reveal any significant differences in the fluconazole susceptibility of the NDT80-family mutants. Interestingly, double and triple mutant analysis did not identify significant genetic interactions for these NDT80 family genes, indicating that they mainly function independently, in spite of their conserved DNA binding domain.IMPORTANCE Transcription factors play key roles in regulating virulence of the human fungal pathogen C. albicans In addition to regulating the expression of virulence factors, they also control the ability of C. albicans to switch to filamentous hyphal growth, which facilitates biofilm formation on medical devices and invasion into tissues. We therefore used new CRISPR/Cas9 methods to examine the effects of deleting three C. albicans genes (NDT80, REP1, and RON1) that encode transcription factors with similar DNA binding domains. Interestingly, double and triple mutant strains mostly showed the combined properties of the single mutants; there was only very limited evidence of synergistic interactions in regulating morphogenesis, stress resistance, and ability to metabolize different sugars. These results demonstrate that NDT80, REP1, and RON1 have distinct functions in regulating C. albicans virulence functions.

Keywords: Candida albicans; NDT80; REP1; RON1; hyphae; morphogenesis.

FIG 1

Strategy for multiple deletion of NDT80-family genes using transient CRISPR/Cas9 and the SAT1-FLP system. (A) In Candida albicans (Ca), there are three genes that encode proteins with similarity to S. cerevisiae (Sc) Ndt80. The amino acid sequence similarity is restricted to the DNA binding domain. The phylogenetic analysis of putative Ndt80 family proteins in Ascomycota indicates that Ndt80-like proteins can be assigned to two groups. One superbranch contains direct orthologs of S. cerevisiae Ndt80, such as Ca Ndt80 and Ca Ron1. Ca Rep1 clusters in the other superbranch, which features orthologs from Pezizomycotina. Colored stars indicate S. cerevisiae Ndt80 and C. albicans Ndt80 family proteins as follows. Color code: Candida CTG clade, blue; Saccharomycetaceae, green; Pezizomycotina, red. (B) The CAS9 gene and the sgRNA were expressed transiently after transformation and were not integrated into the genome. The sgRNA targets Cas9 protein to produce a double-strand break (DSB) at a defined target sequence. The double-strand breaks can be repaired by homology-directed recombination with the SAT1-FLP cassette DNA fragment, which has homology on the ends to the target gene, to create a homozygous deletion of the gene of interest (GOI). The SAT1-FLP cassette confers nourseothricin (NAT) selection and marker recycling. Marker excision of the SAT1 gene is mediated by the maltose-inducible FLP recombinase, leaving a single FLP recombinase target (FRT) site in place of the each GOI.

FIG 2

The REP1 gene is needed for growth on galactose, GlcNAc, and glucosamine. Dilutions of cells were spotted onto minimal medium plates containing the indicated sugar. The genotype of the strain in each row is indicated on the left. The sugars were present at 50 mM, except for the plates containing glycerol, which was present at a higher concentration (300 mM) to promote better growth of the strains. The plates were incubated at 30°C for (A) 2 days or (B) 6 days and then photographed. The deletion mutants of REP1 (strains rep1Δ, rep1Δ ndt80Δ, ron1Δ rep1Δ, and ron1Δ rep1Δ ndt80Δ) were specifically defective in growth on galactose, GlcNAc, and glucosamine. Note that the deletion mutants of NDT80 (strains ndt80Δ, rep1Δ ndt80Δ, ron1Δ ndt80Δ, and ron1Δ rep1Δ ndt80Δ) grew slightly better on nonfermentative carbon sources (glycerol, lactate, and acetate). (C) The ndt80Δ cells grown on glycerol medium were distinct in that there were filamentous outgrowths of cells from the edges of the colonies. Scale bars indicate 1 mm. The strains used are listed in Table 1.

FIG 3

Deletion of NDT80 caused defects in cell separation and hyphal growth. The strains indicated at the top were grown in the medium indicated on the left, and then cell morphology was assessed microscopically. Interestingly, the mutant strains lacking NDT80 (strains ndt80Δ, rep1Δ ndt80Δ, ron1Δ ndt80Δ, and ron1Δ rep1Δ ndt80Δ) displayed a cell separation defect in minimal media containing 50 mM glucose, whereas the other mutants were similar to the wild-type control in that respect. Cells were also grown in liquid medium containing 15% serum or 50 mM GlcNAc to induce hyphal growth. The wild-type control and the ron1Δ mutant showed the formation of filamentous hyphal cells. In contrast, the ndt80 mutant strains were defective, as they formed swollen, slightly elongated cells in the presence of the serum or GlcNAc. Since the rep1Δ mutant could not grow on GlcNAc, 5 mM glucose was added to the GlcNAc medium to support growth of the rep1Δ strains. The rep1Δ and ron1Δ rep1Δ mutants both grew well under these conditions and were induced in vitro to form hyphae by GlcNAc. Cells were incubated at 37°C for 4 h and then photographed. Scale bars indicate 20 µm.

FIG 4

NDT80 is needed for invasive hyphal growth into agar. The cells indicated at the top were spotted onto the agar plates listed on the left. Serum was present in the medium at 15% (vol/vol), and GlcNAc was present at 2.5 mM. The plates were incubated at 37°C and then photographed after 5 days to record the extent of invasive growth emanating from the edges of the colonies. The deletion mutants lacking NDT80 (mutants ndt80Δ, rep1Δ ndt80Δ, ron1Δ ndt80Δ, and ron1Δ rep1Δ ndt80Δ) were strongly defective in invasive hyphal growth into agar media containing serum, Spider, or GlcNAc. Note that the ron1Δ ndt80Δ double mutant showed an improved ability to form hyphae on GlcNAc (arrow). The deletion mutants of REP1 (mutants rep1Δ, rep1Δ ndt80Δ, ron1Δ rep1Δ, and ron1Δ rep1Δ ndt80Δ) grew poorly and were not detectably induced on GlcNAc. Scale bars indicate 1 mm.

FIG 5

Sensitivity of mutants to antifungal drugs. The sensitivity to amphotericin B and fluconazole was determined by a disk diffusion assay in which cells were spread onto the surface of an RPMI 1640 medium plate and then filter discs containing 25 µg of the drugs were placed on the surface of the plate. The ron1Δ rep1Δ ndt80Δ triple mutant was slightly more sensitive to amphotericin B (AMB) than the wild-type SC5314 strain but was not more sensitive to fluconazole (FLC). None of the mutants showed significant differences in the zone of 50% growth inhibition (RAD₅₀) in fluconazole compared to the wild-type strain. However, the deletion mutants lacking NDT80 (mutants ndt80Δ, rep1Δ ndt80Δ, ron1Δ ndt80Δ, and ron1Δ rep1Δ ndt80Δ) did not show the trailing growth around the fluconazole disks that was seen for the other strains. The plates were incubated at 30°C for 48 h and then photographed. Image analysis software was used to measure the zone of growth inhibition to determine the average radius that corresponded to a 50% growth reduction (RAD₅₀). Double asterisks show statistically significant differences (P < 0.01) from the wild-type strain based on a Student's t test.

FIG 6

Sensitivity to cell wall stress and heat stress. The absence of NDT80 resulted in increased sensitivity to Congo red and inhibition of growth at an elevated temperature (44°C). The deletion mutants of NDT80 were also weakly sensitive to SDS. Dilutions of cells were spotted onto YPD plates containing the indicated chemicals. Congo red was present at 140 µM, and SDS was present at 0.06%. The plates were incubated at 30°C and then photographed after 3 days. To examine heat sensitivity, YPD cultures were incubated at 44°C and then photographed after 4 days.

FIG 7

Ndt80 regulates expression of RAS1. Deletion of NDT80 resulted in significantly reduced RAS1 expression compared to wild-type strain SC5314. Complementation of an ndt80Δ mutant with NDT80 restored RAS1 expression to nearly wild-type levels. Cells were cultured under hypha-inducing conditions (YNBNP medium, pH 7.0, 37°C) for 1 h prior to RNA extraction and cDNA synthesis for qRT-PCR analysis. Values are shown as ratios of RAS1 transcription to ACT1 transcription. Asterisks show statistically significant differences (*, P < 0.05; **, P < 0.01) from the wild-type strain based on one-way analysis of variance (ANOVA) with multiple comparisons.