Regulation of the CgPdr1 transcription factor from the pathogen Candida glabrata

Abstract

Candida glabrata is an opportunistic human pathogen that is increasingly associated with candidemia, owing in part to the intrinsic and acquired high tolerance the organism exhibits for the important clinical antifungal drug fluconazole. This elevated fluconazole resistance often develops through gain-of-function mutations in the zinc cluster-containing transcriptional regulator C. glabrata Pdr1 (CgPdr1). CgPdr1 induces the expression of an ATP-binding cassette (ABC) transporter-encoding gene, CgCDR1. Saccharomyces cerevisiae has two CgPdr1 homologues called ScPdr1 and ScPdr3. These factors control the expression of an ABC transporter-encoding gene called ScPDR5, which encodes a homologue of CgCDR1. Loss of the mitochondrial genome (ρ(0) cell) or overexpression of the mitochondrial enzyme ScPsd1 induces ScPDR5 expression in a strictly ScPdr3-dependent fashion. ScPdr3 requires the presence of a transcriptional Mediator subunit called Gal11 (Med15) to fully induce ScPDR5 transcription in response to ρ(0) signaling. ScPdr1 does not respond to either ρ(0) signals or ScPsd1 overproduction. In this study, we employed transcriptional fusions between CgPdr1 target promoters, like CgCDR1, to demonstrate that CgPdr1 stimulates gene expression via binding to elements called pleiotropic drug response elements (PDREs). Deletion mapping and electrophoretic mobility shift assays demonstrated that a single PDRE in the CgCDR1 promoter was capable of supporting ρ(0)-induced gene expression. Removal of one of the two ScGal11 homologues from C. glabrata caused a major defect in drug-induced expression of CgCDR1 but had a quantitatively minor effect on ρ(0)-stimulated transcription. These data demonstrate that CgPdr1 appears to combine features of ScPdr1 and ScPdr3 to produce a transcription factor with chimeric regulatory properties.

Fig. 1.

Phenotypic and reporter gene analysis of C. glabrata multidrug resistance genes. (A) Isogenic ρ⁺ and ρ⁰ C. glabrata strains lacking the indicated genes were grown to mid-log phase and then placed on rich medium containing the indicated concentrations of fluconazole. The plates were incubated at 37°C and then photographed. WT, wild type. (B) The CgCDR1- and CgCDR2-RLUC fusion plasmids were integrated into isogenic ρ⁺ and ρ⁰ C. glabrata strains containing or lacking the CgPDR1 gene. Appropriate transformants were grown to mid-log phase and then assayed for luciferase activity using the coelenterazine substrate. The ρ⁰ transformants were plotted on a separate scale due to the high-level expression of the reporter genes. The error bars indicate standard deviations.

Fig. 2.

Mapping the CgCDR1 promoter. (A) Diagram of the upstream −1,500 bp of the CgCDR1 promoter. The vertical solid lines indicate the predicted locations of PDREs. The two thick horizontal solid lines indicate the DNA contained in the two CgCDR1 promoter deletion constructs analyzed here. The −407 fragment contains the site 5 PDRE, while the −364 fragment does not. The thin line labeled with an asterisk indicates the bounds of the PCR fragment used in the EMSA described for panel B. (B) CgCDR1-RLUC fusion genes containing the indicated amounts of 5′ noncoding DNA were integrated into isogenic ρ⁺ and ρ⁰ C. glabrata strains. Transformants were assayed for luciferase activity as described for panel A. The error bars indicate standard deviations. (C) Two versions of the EMSA probe described above were prepared by PCR. They were either the wild-type (wt) form or a mutant (Mut) variant in which several base pairs in the putative PDRE were mutagenized. Both fragments were radiolabeled with ³²P and used as a probe in an EMSA reaction. The purified recombinant CgPdr1 DBD fragment was either omitted (−) or included (+) in the binding reaction mixture. These reaction mixtures were then electrophoresed through a nondenaturing gel. The gel was dried and exposed to X-ray film.

Fig. 3.

Evidence for positive autoregulation of the CgPDR1 gene. (A) The CgPDR1-RLUC fusion gene was integrated into ρ⁺ and ρ⁰ strains containing or lacking the wild-type CgPDR1 gene. Transformants were assayed for luciferase activity as described in the legend to Fig. 2. The error bars indicate standard deviations. (B) A fragment containing the two putative PDREs present in the CgPDR1 promoter was end labeled with ³²P. This fragment was then incubated with the indicated volumes (in microliters) of bacterially expressed CgPdr1 DNA binding domain. After binding, the protein-DNA complexes were treated with DNase I as described previously (17). The digestion products were separated on a urea-polyacrylamide gel and detected by exposure to X-ray film. The radiolabeled fragment was untreated or digested separately with RsaI and BspMI restriction enzymes and electrophoresed in parallel to provide the size markers indicated on the left side of the autoradiogram. A graphic representation of the two protected regions is shown on the right, with the approximate position of each PDRE indicated.

Fig. 4.

Roles of CgGal11 homologues in control of CgCDR1 in ρ⁺ and ρ⁰ cells. (A) Disruption alleles of CgGAL11A or CgGAL11B were transformed into isogenic ρ⁺ and ρ⁰ cells. Transformants were tested for the ability to grow in the presence of fluconazole as described previously. (B) Plasmids containing 13× Myc-tagged versions of either CgGAL11A (lanes A) or CgGAL11B (lanes B) were integrated in the wild-type versions of both genes into ρ⁺ and ρ⁰ C. glabrata strains. Correct transformants were grown to mid-log phase, and protein extracts were prepared and analyzed by Western blotting using antibodies directed against the Myc epitope. Protein extracts from the untagged strains (−) were analyzed as a negative control for the presence of the Myc tag. (C) The strains in panels A and B were transformed with the CgCDR1-RLUC plasmid. The transformants were assayed for luciferase activity as before.

Fig. 5.

Transcription profile of C. glabrata multidrug resistance gene transcripts. Isogenic ρ⁺ and ρ⁰ strains containing the indicated alleles of the CgGAL11 genes were grown to mid-log phase, and total RNA was isolated from each. Each RNA sample was assayed by quantitative reverse transcription-PCR analysis for the steady-state level of each indicated multidrug resistance gene. Each sample was assayed for CgACT1 and normalized to the level seen in ρ⁺ wild-type cells.

Fig. 6.

Fluconazole induction of CgCDR1 and CgPDR1 expression requires CgGal11A. The RLUC fusion gene at the top of each panel was integrated into strains containing the indicated alleles of CgGAL11. Transformants were grown to early log phase, challenged with various concentrations of fluconazole in the medium, and then assayed for luciferase activity. The error bars indicate standard deviations.

Fig. 7.

Overexpression of CgPsd1 induces fluconazole resistance in C. glabrata. (A) Plasmids containing either the S. cerevisiae PSD1 promoter or the C. glabrata PGK1 promoter driving expression of a CgPsd1-HA fusion protein were integrated into the wild-type CgPSD1 gene or the CgPGK1 promoter, respectively. This single crossover led to production of a C. glabrata strain producing HA-tagged CgPsd1 under the control of the wild-type CgPSD1 promoter (CgPsd1-HA) or the CgPGK1 promoter (CgPGK1-Psd1-HA). Appropriate transformants were grown to mid-log phase and then analyzed by Western blotting using anti-HA antibodies. Molecular mass standards (kDa) are indicated on the left. P denotes the presence of the unprocessed precursor form of CgPsd1-HA, while M indicates the position of the processed C-terminal CgPsd1-HA fragment. (B) The plasmids described in the legend to panel A were integrated into the strains indicated. Desired transformants were tested for their relative growth on rich medium lacking (YPD) or containing (FLC) fluconazole.