Efg1 and Cas5 Orchestrate Cell Wall Damage Response to Caspofungin in Candida albicans

Echinocandins are recommended as the first-line drugs for the treatment of systemic candidiasis. Cas5 is a key transcription factor involved in the response to cell wall damage induced by echinocandins.

ABSTRACT

Echinocandins are recommended as the first-line drugs for the treatment of systemic candidiasis. Cas5 is a key transcription factor involved in the response to cell wall damage induced by echinocandins. In this study, through a genetic screen, we identified a second transcription factor, Efg1, that is also crucial for proper transcriptional responses to echinocandins. Like CAS5, deletion of EFG1 confers hypersensitivity to caspofungin. Efg1 is required for the induction of CAS5 in response to caspofungin. However, ectopically expressed CAS5 cannot rescue the growth defect of efg1 mutant in caspofungin-containing medium. Deleting EFG1 in the cas5 mutant exacerbates the cell wall stress upon caspofungin addition and renders caspofungin-resistant Candida albicans responsive to treatment. Genome-wide transcription profiling of efg1/efg1 and cas5/cas5 using transcriptome sequencing (RNA-Seq) indicates that Efg1 and Cas5 coregulate caspofungin-responsive gene expression, but they also independently control induction of some genes. We further show that Efg1 interacts with Cas5 by yeast two-hybrid and in vivo immunoprecipitation in the presence or absence of caspofungin. Importantly, Efg1 and Cas5 bind to some caspofungin-responsive gene promoters to coordinately activate their expression. Thus, we demonstrate that Efg1, together with Cas5, controls the transcriptional response to cell wall stress induced by caspofungin.

INTRODUCTION

Candida albicans has emerged as one of the most prevalent opportunistic fungal pathogens in humans (1). As a part of the commensal microbiota, it colonizes multiple mucosal sites, including the gastrointestinal (GI) tract, and can cause superficial infections of the oropharynx, vagina, skin, and nails. In susceptible patients, C. albicans can enter the bloodstream and cause a frequently fatal disseminated infection that is characterized by the formation of microabscesses in most organs, with mortality rates approaching 40% despite treatment (2, 3).

The current first-line therapy for the treatment of systemic candidiasis is administration of the echinocandins, which target β-1,3-glucan synthase, encoded by FKS1 (4). Targeting Fks1 disrupts the synthesis of the major cell wall biopolymer β-1,3-glucan, resulting in loss of cell wall integrity and imparting a severe cell wall stress on the fungus (4). Stress signals at the cell surface are transmitted to Rho1 GTPase, which mobilizes a variety of effectors to induce a set of genes from the cell wall integrity signaling pathway (5). In C. albicans, the zinc finger transcription factor, Cas5, serves as a key transcriptional regulator of responses to cell wall stress (6). It is required for induction of ∼60% of caspofungin-responsive genes, and deletion of CAS5 confers hypersensitivity to caspofungin (7). Cas5 is also crucial for proper cell cycle dynamics in response to echinocandins (7). It is activated by the phosphatase Glc7 and can regulate the expression of target genes in concert with the transcriptional regulators Swi4 and Swi6 (7). Cas5 lacks an identifiable ortholog in humans but is required for drug resistance and virulence in C. albicans (8, 9), suggesting the possibility of Cas5 as a potential target for antifungal therapy. Although Cas5 has a profound impact on global transcriptional responses to cell wall stress, Cas5-independent caspofungin-responsive signaling pathways may also exist.

The transcription factor Efg1 is a member of the APSES family of basic helix-loop-helix transcriptional regulators that regulate cellular differentiation in ascomycetes. Members of this family share a conserved DNA binding domain. Efg1 is one of the best-characterized regulators of hyphal morphogenesis in C. albicans (10, 11). It is suggested to function downstream of the cAMP/protein kinase A (PKA) pathway to induce hyphal transcriptional program (12–14). Numerous in vitro and in vivo studies with efg1 mutants have demonstrated that Efg1 is important for C. albicans virulence and for the interactions of C. albicans with endothelial and epithelial cells (10, 15–18). In addition, Efg1 is required for chlamydospore formation (19), the white-cell-specific transcriptional profile (20, 21), and biofilm formation and regulation of cell wall proteins (22, 23). Efg1 has also been implicated in drug susceptibility, including to azoles and polyenes, in C. albicans (24, 25). Although Efg1 has been shown to control caspofungin-induced cell aggregation (26), the extent of its involvement in caspofungin susceptibility and induction of caspofungin-responsive genes is far less clear.

Through screening of 83 homozygous insertion mutants of putative transcription factors, Bruno et al. identified Cas5, which is responsible for the expression of many caspofungin-induced genes (6). In the present study, we screened 167 C. albicans deletion mutants of transcription factors (27) and found that in addition to the cas5 mutant, deletion of EFG1 confers hypersensitivity to caspofungin. Efg1 interacts with Cas5 to coordinately regulate gene expression upon caspofungin induction. Efg1 and Cas5 also control the expression of subsets of caspofungin-responsive genes independently of each other. We further show that deleting both EFG1 and CAS5 resensitizes echinocandin-resistant strains to caspofungin. Our work highlights the importance of transcription networks controlled by Efg1 and Cas5 in the response of C. albicans to caspofungin.

RESULTS

Deletion of EFG1 or CAS5 confers hypersensitivity to caspofungin in C. albicans.

In an effort to identify transcriptional activation programs that are important for caspofungin response, we sought to screen a transcription factor deletion library (27) for mutants that displayed increased caspofungin susceptibility, as measured on YPD (2% Bacto peptone, 1% yeast extract, 2% glucose) plates containing 500 ng/ml of caspofungin. Through screening of 167 deletion mutants of transcription factors, we found that deletion of Efg1 or Cas5 led to the most sensitivity to caspofungin (Fig. 1A). This sensitivity was not because of a growth defect of these mutants, as they grew as normally as the wild-type strain on medium without caspofungin (Fig. 1A). The observation that disruption of CAS5 increased susceptibility to caspofungin was not surprising, as this was reported previously (6, 7), validating the results of our screen. However, the function of Efg1 in transcriptional response to caspofungin is not well defined so far. Next, we compared the impact of Efg1 with that of Cas5 on caspofungin sensitivity. As shown in Fig. 1B, deletion of CAS5 resulted in an ∼16-fold decrease in MIC for caspofungin in YPD broth, whereas only an ∼8-fold decrease in MIC for caspofungin was displayed in the efg1 mutant compared to that in wild-type cells. In addition to Cas5, we identified another transcription factor, Efg1, that plays a critical role in caspofungin response in C. albicans.

FIG 1.

Efg1 is critical for caspofungin tolerance and is required for CAS5 induction in response to caspofungin. (A) Cells of the wild-type (WT; SN250) strain and the efg1 and cas5 mutant strains were serially diluted 10-fold and spotted onto YPD solid medium with or without 500 ng/ml of caspofungin (CSF). Photographs were taken after 2 days of growth at 30°C. (B) Caspofungin susceptibility assays were conducted in YPD medium. Growth was measured by absorbance at 600 nm after 48 h at 30°C. Optical densities were averaged from duplicate measurements and normalized relative to those of caspofungin-free controls. Data are quantitatively displayed in heat map format (see color bar). (C) Western blot analysis shows defective induction of Cas5 in the efg1 mutant in response to caspofungin. Cells were grown in YPD at 30°C with or without 250 ng/ml of caspofungin for 2 h. A protein extract of each sample was quantified and separated in an 8% SDS-PAGE gel. (D) Constitutively expressed CAS5 is not sufficient to rescue the growth defect of the efg1 mutant in caspofungin-containing medium. The assay was performed and analyzed as described for panel B.

Efg1 is required for the upregulation of CAS5 expression in response to caspofungin.

We next examined whether Efg1 and Cas5 protein levels changed in response to caspofungin. The protein level of Cas5-Myc under the control of its endogenous promoter was low in the absence of caspofungin, but increased upon caspofungin induction in an Efg1-dependent manner (Fig. 1C). However, Efg1 protein levels remained largely unchanged regardless of caspofungin or Cas5 (Fig. 1C). Since upregulation of CAS5 in response to caspofungin is dependent on Efg1, we wondered whether the major function of Efg1 in the regulation of caspofungin response is induction of CAS5 expression. Ectopically expressed Cas5-FLAG under the control of the TDH3 promoter was transformed into the efg1 mutant, as well as the cas5 mutant. The growth defect of the cas5 mutant in caspofungin was rescued by TDH3p-Cas5-FLAG (Fig. 1D), validating that Cas5-FLAG is functional. However, overexpressing CAS5 could not bypass the requirement of Efg1 in caspofungin tolerance, although it restored the growth defect of the efg1 mutant in caspofungin-containing medium to some extent (Fig. 1D). Our results indicate that Efg1 regulates caspofungin response at least partially independently of Cas5.

Deletion of both EFG1 and CAS5 exacerbates cell wall stress and leads to caspofungin susceptibility.

To understand the genetic relationship between Efg1 and Cas5 in the regulation of caspofungin response, EFG1 was deleted in a cas5 mutant. We first compared the cell sizes of the double mutant and each single mutant with that of wild-type cells. The efg1 mutant cells are, on average, significantly smaller in volume, whereas the cas5 mutant cells are larger (Fig. 2A). The effect of EFG1 or CAS5 deletion on cell morphology has already been described (6, 10). The cell size of the efg1 cas5 double mutant was intermediate (Fig. 2A). We then determined the sensitivity of these strains to caspofungin using the spot test on YPD plates containing caspofungin with different concentrations ranging from 0 to 80 ng/ml. As shown in Fig. 2B, the efg1 cas5 double mutant was more susceptible to caspofungin than one or the other single mutant. This phenotype was further confirmed by susceptibility testing for caspofungin in YPD liquid medium, in which the viability of double mutant was at least 5-fold and 16-fold lower than those of the cas5 and efg1 single mutants, respectively, when known amounts of cells were plated on YPD plates after caspofungin treatment for 48 h (Fig. 2C). Our data demonstrate that deletion of EFG1 and CAS5 has an additive effect on the deficiency in cell wall stress responses induced by caspofungin.

FIG 2.

Deleting EFG1 in the cas5 mutant enhances susceptibility to caspofungin. (A) Cell morphologies of the efg1 mutant, cas5 mutant, and cas5 efg1 double mutant compared to the wild type (SN250). Cells were grown in YPD at 30°C. (B) Cells of the wild type (SN250), efg1 mutant, cas5 mutant, and cas5 efg1 double mutant were serially diluted 10-fold and spotted onto YPD solid plates containing caspofungin (0, 20, 40, and 80 ng/ml). Photos were taken after 2 days of growth at 30°C. (C) Quantification of caspofungin susceptibility. A total of ∼10³ cells of the indicated strains was incubated in YPD liquid medium with or without 20 ng/ml of caspofungin. Aliquots were then plated onto YPD plates. The viability was calculated as a ratio of colonies formed in plates of caspofungin-treated cells versus untreated cells. Data represent means and SEM from three independent experiments. Significance was measured with unpaired t test in GraphPad Prism. ****, P < 0.0001. (D) An assay for susceptibility of the indicated strains to caspofungin was performed and analyzed as described for Fig. 1B.

The most common mechanism of resistance to echinocandins is amino acid substitutions that occur in hot spot regions in the drug target β-1,3-glucan synthase Fks1 gene (28). Cas5 was reported to have an impact on echinocandin resistance (7). Given that the efg1 cas5 double mutant exhibited greater susceptibility to caspofungin than that of cas5 single mutant, we tested whether deleting EFG1 in the cas5 mutant could further reduce echinocandin resistance. A strain carrying the fks1^S645F mutant constructed by us previously, which exhibited hyperresistance to caspofungin (29), was used for susceptibility testing. As shown in Fig. 2D, deleting either EFG1 or CAS5 in the fks1^S645F mutant resulted in increased susceptibility, as the MIC for caspofungin decreased ∼2-fold or ∼4-fold, respectively. However, the MIC for caspofungin of these strains was still much higher than that of wild-type cells (Fig. 2D). Importantly, we found that fks1^S645F mutation in the efg1cas5 double mutant exhibited ∼2-fold decreased MIC values for caspofungin compared to that of wild-type cells (Fig. 2D). Thus, deleting both EFG1 and CAS5 results in cells losing fitness and renders the fks1^S645F resistant strain responsive to caspofungin treatment.

The overlapped and divergent roles of Efg1 and Cas5 in the regulation of expression of caspofungin-responsive genes.

Given that the efg1 cas5 double mutant displayed a more severe deficiency in caspofungin tolerance than that of one or the other single mutant, one would predict a difference between Efg1 and Cas5 in the regulation of caspofungin-responsive genes. To test this hypothesis, we performed three independent transcriptome sequencing (RNA-Seq) analyses to profile each mutant and reference strain SN250, both treated with caspofungin for 2 h. Using a 1.5-fold cutoff, there are 1,687 genes differentially expressed in wild-type cells upon caspofungin induction (see Table S1 in the supplemental material). The heat map shown in Fig. 3A displays 424 caspofungin-responsive genes whose expression was regulated by Cas5, 278 genes regulated by Efg1, and 137 genes coregulated by Efg1 and Cas5. To identify the physiological roles of Cas5 and Efg1 on response to caspofungin, we subjected these gene sets to pathway analysis using Gene Ontology (GO). As shown in Fig. 3B, the caspofungin-responsive genes specific to the regulators had different physiological functions. The sets that had changed expression upon deletion of CAS5 were significantly enriched in genes for nutrient metabolism and biosynthesis, whereas genes whose expression was regulated specifically to Efg1 functioned in diverse processes, including chromatin organization, DNA conformation change, and iron importation. Among those caspofungin-responsive genes that were regulated by Efg1 or Cas5, only 24% with differences in expression were common to both regulators, most of which were involved in cell wall and external encapsulating structure organization (Fig. 3B). Our RNA-Seq data suggest that Efg1 and Cas5 function differentially in the regulation of transcriptional response to caspofungin.

FIG 3.

The regulatory network of Efg1 and Cas5 on global transcriptional profile in response to caspofungin treatment. (A) Heat map showing the genes that exhibited a 1.5-fold expression difference cutoff (P < 0.05) in each comparison group. “wt-/wt+,” ‘‘efg1+/wt+,’’ and ‘‘cas5+/wt+ represent differentially expressed genes of the untreated wild type, treated efg1 mutant, treated cas5 mutant relative to the treated wild type, respectively. Blue and red indicate higher and lower expression levels in each comparison group. The degree of color saturation represents the magnitude of the expression ratio. (B) Bar graph showing the relative percentage of caspofungin-responsive genes whose expression is dependent on Cas5 (blue bars), dependent on Efg1 (red bars), or dependent on both (white bars) in the indicated GO term processes. The total number of genes belonging to each GO term process is indicated. (C) Gene expression Venn diagram. Gene sets controlled by Cas5 and Efg1 based on RNA-Seq analysis of the respective knockout mutants are shown. Activation was defined by a minimum 1.5-fold decrease of gene expression in the deletion mutant relative to wild type and repression by a minimum 1.5-fold increase. The efg1 mutant and cas5 mutant were assessed in medium with caspofungin. (D) Impacts of Efg1 and Cas5 on the expression of caspofungin-responsive genes. Transcription levels were monitored by qRT-PCR and normalized to CDC28. Error bars represent standard deviations from the means of triplicate samples. Significance was measured with unpaired t test in GraphPad Prism. ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *P < 0.05.

Venn diagrams in Fig. 3C show that deletion of CAS5 resulted in downregulation of 319 genes and upregulation of 105 genes relative to the wild type upon caspofungin induction, whereas the numbers with regard to the efg1 mutant are 173 and 105, respectively, indicating that the cas5 mutant had a more severe gene expression defect than that of the efg1 mutant. This is consistent with the result that the cas5 mutant is more sensitive to caspofungin than the efg1 mutant (Fig. 1B). Notably, Efg1 and Cas5 regulate a common set of genes associated with cell wall organization, suggesting that Efg1 is also involved in the response to cell wall damage induced by echinocandins. Although 100 upregulated genes and 37 downregulated genes are coregulated by Efg1 and Cas5, each transcription factor also governed induction of some caspofungin-responsive genes independently of the other (Fig. 3C). The genes activated only by Cas5 upon caspofungin addition were enriched for GO terms associated with cell wall organization, consistent with the major role of Cas5 in cell wall response. Efg1 but not Cas5 regulated members of the SAP gene family and iron ion transport, which are previously demonstrated roles in nutrient acquisition and virulence (30–32).

Based on our transcription profiling data, we picked out some representative genes from each group whose expression in response to caspofungin was dependent on Cas5, Efg1, or both, including ECM331, ALS1, PGA13, and ORF19.7504. Among these, ECM331, PGA13, and ALS1 encode cell wall-associated proteins (33–35), whereas ORF19.7504 encodes a component of the protein phosphatase type 2A complex (36). The induction of these transcripts in caspofungin was confirmed by quantitative reverse transcription-PCR (qRT-PCR). Consistent with a previous report (6), the induction of the ECM331 gene in response to caspofungin was dependent on Cas5 but less dependent on Efg1 (Fig. 3D). In contrast, Efg1 was required for the increased expression of ALS1 upon caspofungin exposure, while Cas5 played minor role in it (Fig. 3D). Both Efg1 and Cas5 were critical for the induction of PGA13 and ORF19.7504. The induction of all of these genes except for ECM331 was abolished in the efg1 cas5 double mutant (Fig. 3D). Therefore, Efg1 and Cas5 regulate expression of caspofungin-responsive genes in a coordinated and complementary manner.

Efg1 interacts with Cas5 in vivo.

Given that Efg1 and Cas5 coregulated expression of several caspofungin-responsive genes, we wanted to determine whether Efg1 interacts with Cas5 in vivo. We first determined the interaction between Efg1 and Cas5 using a yeast two-hybrid assay. We cloned EFG1 into pGBKT7 to generate an EFG1-BD fusion and cloned CAS5 into pGADT7 to generate a CAS5-AD fusion. The interaction between BD-Efg1 and AD-Cas5 fusions was detected in the yeast two-hybrid assay (Fig. 4A). The BD-Efg1 and AD-Cas5 fusions did not interact with the control (Fig. 4A). Therefore, Efg1 could interact with Cas5 by two-hybrid assay. An immunoprecipitation experiment with C. albicans was performed to determine whether Efg1 interacts with Cas5 in vivo. Cas5 was tagged at its C terminus with Myc, and Cas5-Myc was expressed in a strain containing Efg1-FLAG. Immunoprecipitation of Cas5-Myc was able to pull down FLAG-tagged Efg1 in the presence or absence of caspofungin (Fig. 4B). The interaction was specific because the interaction was not detectable in the control strain that carried only Efg1-FLAG (Fig. 4B). Therefore, Efg1 can interact with Cas5 in vivo, and the interaction is not regulated by caspofungin.

FIG 4.

Efg1 interacts with Cas5 in vitro and in vivo. (A) Yeast two-hybrid assays. AH109 was transformed with plasmids as indicated. (B) Coimmunoprecipitation of Efg1-3FLAG with Cas5-13Myc in the presence or absence of caspofungin. Strains with indicated Efg1-3FLAG and Cas5-13Myc were grown in YPD at 30°C treated with or without 125 ng/ml of caspofungin for 30 min. Protein lysates were subjected to immunoprecipitation with anti-Myc antibody and protein A-Sepharose, and the precipitated proteins were separated by 8% SDS-PAGE and probed with anti-FLAG antibody. As input control, cell lysates were analyzed by Western blotting with anti-FLAG antibody.

Efg1 and Cas5 bind to the promoters of caspofungin-responsive genes to induce their expression.

Efg1 interacts with Cas5 in vivo and both regulators are critical for the induction of caspofungin-responsive genes. To determine whether they are directly involved in the induction of caspofungin-responsive genes, chromatin immunoprecipitation (ChIP) was performed to determine whether a Myc-tagged Efg1 or a Myc-tagged Cas5 is present at the promoters of ECM331 and ALS1. These two genes were chosen because they represented genes whose expression was mainly controlled by Cas5 and Efg1, respectively. Ectopically expressed Cas5-Myc under the control of the ACT1 promoter was used in ChIP experiments to circumvent the effects of Efg1 on CAS5 expression, and Efg1-Myc was under the control of its endogenous promoter. Cross-linked chromatin of either Efg1-Myc, Cas5-Myc, or no tag (as a control) in the presence or absence of caspofungin was precipitated with anti-Myc and protein A beads. The precipitated DNA was analyzed by quantitative PCR at ≈250-bp intervals across the whole promoters of ECM331 and ALS1. CDC28 coding sequence served as a negative control. Among them, a specific region from each of the promoters, from −322 to −138 for ALS1 and −1172 to −979 for ECM331 showed higher ChIP signals for both Efg1-Myc and Cas5-Myc in the presence of caspofungin than the other regions. These regions were therefore used for subsequent ChIP experiments. As shown in Fig. 5A, the ChIP signal of Efg1-Myc increased by ∼1.5-fold upon caspofungin induction in wild-type cells, while an ∼4-fold increase in bound Cas5-Myc was detected at ECM331 promoter in response to caspofungin. Deletion of CAS5 diminished the increase of Efg1 binding in caspofungin (Fig. 5A). We also performed a reciprocal ChIP of Cas5-myc in an efg1 mutant, and detected a dramatic increase of Cas5 binding on ECM331 promoter in caspofungin-containing medium, compared to that in the medium without caspofungin (Fig. 5A), indicating that Efg1 is not necessary for the binding of Cas5 on ECM331 promoter upon caspofungin induction. These data are consistent with the result that Cas5, but not Efg1, played a major role in the induction of ECM331 by caspofungin.

FIG 5.

ChIP of Efg1 and Cas5 at ECM331 (A) and ALS1 (B) promoters in wild type (SN250) and indicated mutant cells, respectively. Cells of the wild type and indicated mutant containing EFG1p-Efg1-Myc or ACT1p-Cas5-Myc were grown in YPD medium at 30°C in the presence or absence of 50 ng/ml of caspofungin for 5 min. Input and ChIP products were quantified by using quantitative PCR with promoter-specific primers. Samples were normalized to a region of CDC28. Normalized values were used to calculate ChIP/input. The enrichment over that for untagged controls is shown. Samples were run in triplicate, and the error is indicated. Significance was measured with unpaired t test in GraphPad Prism. ***, P < 0.001; **, P < 0.01; *P < 0.05. ns, no significance.

Unlike for ECM331, the induction of ALS1 in response to caspofungin is dependent on both Efg1 and Cas5, and Efg1 plays a major role. Through ChIP experiments, we found that both Efg1-Myc and Cas5-Myc in the wild-type strain showed a higher enrichment at the ALS1 promoter in the presence of caspofungin than that in media without caspofungin (Fig. 5B). Deleting CAS5 did not completely block Efg1 binding to the ALS1 promoter, as Efg1-Myc was still detected at ALS1 promoter upon caspofungin induction in cas5 mutant cells, although the ChIP signal was not as high as that in wild-type cells (Fig. 5B). However, the binding of Cas5 to the ALS1 promoter was dependent on Efg1 (Fig. 5B). Our results indicated that Efg1 and Cas5 coordinately bind to the promoter of both caspofungin-responsive genes ECM331 and ALS1 yet play divergent roles in the regulation of their expression.

DISCUSSION

A major determinant of caspofungin sensitivity in fungi is the cell wall integrity pathway. Cas5-mediated transcriptional activation program has been shown to play a key role in C. albicans cell wall damage response induced by caspofungin (6, 7). Following Cas5, we identified a second transcription factor, Efg1, that is critical for C. albicans caspofungin tolerance. Efg1 cooperates with Cas5 to activate the expression of caspofungin-responsive genes. However, these two regulators also independently regulate some genes induction in response to caspofungin. Deletion of EFG1 in the cas5 mutant results in increased sensitivity to caspofungin and enhances susceptibility of the echinocandin-resistant strain to caspofungin treatment. Our study provides insights into the transcription network controlled by Efg1 and Cas5 that is central to caspofungin response in C. albicans.

Among the mutants we have screened, efg1 and cas5 mutants exhibited the most sensitivity to caspofungin (Fig. 1A). Although Efg1 and Cas5 coregulated expression of a number of caspofungin-responsive genes, we suggest that the mechanisms of regulation by these two regulators of caspofungin response are distinct in at least some aspects. First, transcription profiling showed that Efg1 and Cas5 control induction of some caspofungin-responsive genes independently of each other (Fig. 3C). Second, Efg1 is required for the increase of Cas5 expression upon caspofungin induction (Fig. 1C), but overexpression of CAS5 could not rescue the growth defect of efg1 mutant in caspofungin-containing medium (Fig. 1D), suggesting that Efg1 regulates caspofungin response at least partially independent of Cas5. Third, in addition to regulating cell wall protein expression, Efg1 and Cas5 play different roles on cell wall integrity. For example, Efg1 was shown to have impact on the structural components of the cell wall (22), while Cas5 is required for appropriate cell cycle dynamics in response to cell wall stress (7). Except for caspofungin, both efg1 and cas5 mutant showed enhanced susceptibility to azole drugs (9, 24). It would be interesting to determine whether these two regulators have an additive effect in the regulation of azole response.

An appropriate transcription program in response to cell wall stress is critical to enable C. albicans proliferation in the presence of echinocandins. Cas5 has been shown to be a key transcriptional regulator in this process (6, 7). In this study, we found that Efg1 and Cas5 coordinately activate expression of caspofungin-responsive genes. Unlike Cas5, whose major function is in the response to cell wall stress, Efg1 also regulates other cellular processes and development, such as hyphal development (10, 11), white-opaque switch (20), and chlamydospore formation (19). A previous study reported that Efg1 interacted with Flo8 to regulate the hyphal transcription program (37). Like that of Cas5, the function of Flo8 is relative specific. It regulates a hypha-specific set of the Efg1-regulated genes. Therefore, we suggest that the multiple functions of Efg1 are achieved by interacting with different coregulatory proteins that modulate its functional specificities. Efg1 has been shown to recruit NuA4 histone acetyltransferase to hypha-specific promoters, which is required for subsequent binding of chromatin remodeling complex Swi/Snf and transcriptional activation (38). Efg1 may also recruit NuA4 to promoters of caspofungin-responsive genes. It is well known that chromatin modifiers play critical roles in the regulation of gene expression (39), which may explain why Efg1 is required for the induction of some Cas5-regulated caspofungin-responsive genes. Because the cas5 mutant has a stronger phenotype regarding caspofungin sensitivity than that of the efg1 mutant and Cas5 regulates expression of some caspofungin-responsive genes independently of Efg1, Cas5 may interact with additional regulators to regulate the caspofungin response transcription program.

Targeting transcription factors that are crucial for cellular stress responses may provide a powerful strategy for antifungal drug development (40). Both Efg1 and Cas5 are critical in the pathogenicity of C. albicans (8, 10, 41). These two transcription factors can be exploited as selective drug targets because they are evolutionarily divergent between fungi and humans. Another major benefit of targeting Efg1 and Cas5 is that their inhibition strongly enhances the susceptibility to both azole and echinocandin drugs, suggesting that they could be exploited as both single and combination therapeutic methods. We showed that deleting both EFG1 and CAS5 in an echinocandin-resistant strain resensitizes it to caspofungin treatment (Fig. 2D), providing the possibility of identification of natural or synthetic chemicals or peptidomimetics that inhibit the activity of Efg1 and Cas5 to reverse echinocandin resistance. Unlike targeting an enzyme which could simply bind to its active pocket, targeting these proteins would have to specifically disrupt protein-nucleic acid or protein-protein interactions. Recently, the small molecule iKIX1 was found to abrogate azole resistance in Candida glabrata in vivo by disrupting the interaction between the mediator complex and the Pdr5 efflux pump transcriptional activator Pdr1, whose constitutive activation confers azole resistance (42). Given that Efg1 interacts with Cas5 to regulate expression of caspofungin-responsive genes, it is feasible to identify small molecules that disrupt the interaction between Efg1 and Cas5 as a novel therapeutic strategy in fungal disease.

MATERIALS AND METHODS

Media and growth conditions.

C. albicans and Saccharomyces cerevisiae strains were routinely grown at 30°C in either YPD (2% Bacto peptone, 1% yeast extract, 2% glucose) or YPDA (0.2% adenine-supplemented YPD). C. albicans transformants were selected on synthetic medium (2% glucose, 0.17% yeast nitrogen base without ammonium sulfate, 0.5% ammonium sulfate, and auxotrophic supplements) or YPD plates supplemented with 200 μg/ml of nourseothricin.

Strain construction.

The C. albicans strains used in this study are listed in Table S2. Primer sequences are listed in Table S3. EFG1 and CAS5 were deleted based on a SAT1-flipping strategy (43). The CAS5 coding sequence was amplified using primers 1 and 3. The resulting PCR product was digested with BamHI and MluI and inserted into the BamHI-MluI site of pPR673 (38) to create pACT1-CAS5-13MYC. The resulting plasmid was digested with StuI to express Cas5-13MYC under the control of the ACT1 promoter. A 1.2-kb PCR product (primers 2 and 3) containing the C-terminal CAS5 coding region and a 0.8-kb PCR product (primers 4 and 5) containing the C-terminal EFG1 coding region were inserted into the BamHI-MluI sites of pPR673 (38). These two plasmids were digested with ApaI and KpnI, respectively, to target integration into their own loci to express Cas5-13Myc and Efg1-13Myc. The pTDH3-CAS5-3FLAG and pTDH3-EFG1-3FLAG plasmids were constructed by amplifying CAS5 (primers 6 and 7) and EFG1 (primers 8 and 9) to insert into the BglII-MluI site of the pBES116-TDH3p-3FLAG plasmid (44). The resulting plasmids were digested with AscI to target integration into ADE2 locus to express Efg1-3FLAG and Cas5-3FLAG. C. albicans fks1^S645F mutant strains were constructed as described previously (29).

Caspofungin susceptibility tests.

For solid media, cell cultures were serially diluted 10-fold, spotted onto YPD medium containing different concentrations of caspofungin (0 to 80 ng/ml), and incubated for 2 days at 30°C. To quantify caspofungin susceptibility, ∼10³ cells were incubated in YPD liquid medium with or without 20 ng/ml of caspofungin at 30°C for 2 days. Cells were then plated on YPD plates at 30°C for 72 h. The viability was calculated as a ratio of colonies formed in plates of caspofungin-treated cells versus untreated cells. For liquid media, caspofungin susceptibility was measured in 96-well microtiter plates (Thermo) as described by Sun et al. (29), and assays were conducted in a total volume of 0.1 ml of YPD medium/well containing different concentrations of caspofungin. Briefly, caspofungin was serially diluted 2-fold from 4,000 to 7.8 ng/ml and 600 to 4.6 ng/ml. Cell cultures were prepared with ∼10³ cells in each well according to their optical densities and incubated at 30°C under dark condition for 2 days, and values for optical density at 600 nm (OD₆₀₀) were measured with a spectrophotometer (BioTek Instruments).

Western blotting.

Overnight cultures were 1:50 diluted into YPD medium. After incubation at 30°C for 4 h, cells were treated with 250 ng/ml of caspofungin or equal double-distilled water (ddH₂O) for 2 h, harvested, washed with precooling in 10 ml of phosphate-buffered saline (PBS), and resuspended in 350 μl of lysis buffer (50 mM Tris-HCl [pH 7.5],100 mM NaCl, and 0.1% NP-40), supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF; Biosharp) and one protease inhibitor cocktail tablet per 25 ml (Roche). Cell cultures were lysed by beads beating three times for 30 s with 5 min on ice during breaks; finally, about 400 μl supernatant was obtained and quantitative proteins were separated in an 8% SDS-PAGE gel, transferred to a polyvinylidene difluoride (PVDF) membrane (Bio-Rad), and blocked with 3% skim milk (Becton, Dickinson and Company) resolved in 10 ml of 0.05% Tween 20-supplemented phosphate-buffered saline (PBST) for 1 h at room temperature. Blots were hybridized with the following primary antibodies at room temperature for 1 h: anti-Myc (1:10,000; Sigma), anti-FLAG (1:3,000; Sigma), and anti-tubulin (1:3,000; Sigma). Anti-mouse antibody (1:5,000; Sigma) was used as a secondary antibody. Blots were washed three times with 10 ml of PBST and once with 10 ml of PBS; signals were detected by using Clarity Western ECL substrate (Bio-Rad) following the protocol.

RNA-Seq analysis.

Overnight cultures of wild-type (SN250), efg1 mutant, and cas5 mutant strains were 1:50 diluted into YPD medium. After incubation at 30°C for 4 h, cells were treated with 125 ng/ml of caspofungin or equal ddH₂O for 2 h, harvested, and stored at −80°C. A sequencing library was constructed by using a NEBNext Ultra RNA library prep kit for Illumina (New England BioLabs [NEB], USA). The RNA-Seq library was assessed by the Agilent Bioanalyzer 2100 system and quantified by qRT-PCR before sequencing on the Illumina NovaSeq platform. Clean reads were selected from raw reads by removing reads containing adapter, ploy-N, and low quality. Q20, Q30, and GC content of clean data were calculated. Clean reads were mapped to a C. albicans reference genome (SC5314_A21), and mapped reads were counted and normalized to the number of fragments per kilobase of transcript per million mapped reads (FPKM) according to the gene length and total mapped reads. Differentially expressed genes were defined by an adjusted P value (P_adj) of <0.05 found by DESeq2 between any two conditions analyzed in this study. Only genes exhibiting a 1.5-fold difference (P_adj < 0.05) in response to caspofungin and relieved 1.5-fold difference (P_adj < 0.05) due to deletion of Efg1 or Cas5 in the wild-type strain were used for downstream analysis. GO term process analysis was performed on the Candida Genome Database (http://www.candidagenome.org/cgi-bin/GO/goTermFinder).

Quantitative RT-PCR.

Cell cultures were grown and treated as described above for RNA-Seq analysis. Total RNA was isolated by using the RNAprep pure tissue kit (TIANGEN), cDNA was synthesized in Maxima H Minus cDNA synthesis master mix with double-strand-specific DNase (dsDNase) (Thermo Scientific), and qRT-PCR was performed using the iQ SYBR green supermix (Bio-Rad) in 96-well plates. Primers 18 and 19 were used to amplify PGA13, primers 20 and 21 were used to amplify ALS1, primers 22 and 23 were used to amplify ORF19.7504, primers 24 and 25 were used to amplify ECM331, and primers 26 and 27 were used to amplify CDC28. Data were normalized to CDC28 and plotted using GraphPad Prism.

Yeast two-hybrid assays.

The full-length coding sequence of C. albicans Efg1 was cloned into the NdeI-BamHI site of pGBKT7 vector as a bait plasmid, and the full-length coding sequence of C. albicans Cas5 was cloned into the NdeI-BamHI site of pGADT7 vector as a prey plasmid; these plasmids were cotransformed into S. cerevisiae (AH109). Transformants were then streaked onto synthetic double-dropout medium (SD Trp⁻ Leu⁻) and SD quadruple-dropout medium (SD His⁻ Trp⁻ Leu⁻ Ade⁻) and incubated at 30°C for 2 days. Trp1 and Leu2 were the yeast selective markers on plasmid pGBKT7 and pGADT7, respectively, to confirm the successful transformation of pGBKT7 and pGADT7 plasmids. HIS3 and ADE2 were reporter genes activated by the interaction between pGBKT7 and pGADT7. pGBKT7-53 and pGADT7-T interaction served as a positive control, pGBKT7-Lam and pGADT7-T interaction served as a negative control, and pGBKT7-Lam and pGADT7-Cas5 and pGBKT7-Efg1 and pGADT7-T were used to confirm the specific interaction between Efg1 and Cas5.

Coimmunoprecipitation.

Protein extraction was performed as described previously (44), except that cells were treated with 125 ng/ml of caspofungin for 30 min, ∼400 μl of protein extract was incubated overnight at 4°C with 2 μg of anti-Myc antibody (Sigma). Protein A-Sepharose (GE) was washed three times with lysis buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 0.1% NP-40) and centrifuged for 30 s at 1,000 rpm. An ∼40-μl suspension of protein A-Sepharose was subjected to precipitation of the immunocomplex. After incubation at 4°C for 2 to 3 h, beads were washed three times with lysis buffer and proteins were eluted from the beads by boiling for 10 min; eluted proteins were then subjected to Western blotting.

Chromatin immunoprecipitations.

ChIP experimental procedures were performed as previously described, with modifications (45). Overnight cultures of C. albicans were diluted 1:75 into YPD medium and incubated at 30°C for 4 h; they were then either left untreated or treated with 50 ng/ml of caspofungin for 5 min. ChIP signals were quantified by qPCR; untagged strains were used as the control for cells expressing Efg1-MYC and Cas5-MYC. Efg1p-Efg1-MYC and ACT1p-Cas5-MYC enrichment is presented as a ratio of ALS1 and ECM331 promoter IP (bound/input) versus control locus IP (bound/input).

Data availability.

RNA-Seq raw data were uploaded to the NCBI SRA database under the accession number PRJNA662508.