UPC2A is required for high-level azole antifungal resistance in Candida glabrata

Abstract

Candida glabrata, the second most common cause of Candida infections, is associated with high rates of mortality and often exhibits resistance to the azole class of antifungal agents. Upc2 and Ecm22 in Saccharomyces cerevisiae and Upc2 in Candida albicans are the transcriptional regulators of ERG11, the gene encoding the target of azoles in the ergosterol biosynthesis pathway. Recently two homologs for these transcription factors, UPC2A and UPC2B, were identified in C. glabrata. One of these, UPC2A, was shown to influence azole susceptibility. We hypothesized that due to the global role for Upc2 in sterol biosynthesis in S. cerevisiae and C. albicans, disruption of UPC2A would enhance the activity of fluconazole in both azole-susceptible dose-dependent (SDD) and -resistant C. glabrata clinical isolates. To test this hypothesis, we constructed mutants with disruptions in UPC2A and UPC2B alone and in combination in a matched pair of clinical azole-SDD and -resistant isolates. Disruption of UPC2A in both the SDD and resistant isolates resulted in increased susceptibility to sterol biosynthesis inhibitors, including a reduction in fluconazole MIC and minimum fungicidal concentration, enhanced azole activity by time-kill analysis, a decrease in ergosterol content, and downregulation of baseline and inducible expression of several sterol biosynthesis genes. Our results indicate that Upc2A is a key regulator of ergosterol biosynthesis and is essential for resistance to sterol biosynthesis inhibitors in C. glabrata. Therefore, the UPC2A pathway may represent a potential cotherapeutic target for enhancing azole activity against this organism.

FIG 1

Representation of the ergosterol biosynthesis pathway in Candida glabrata. Genes (in italics) whose expression was monitored in this study are underlined. Sites of action for specific drug classes are indicated by dashed box. CoA, coenzyme A.

FIG 2

Enhanced activity of fluconazole in an SDD clinical isolate, SM1, with a disruption of UPC2A. (A) Minimum fungicidal concentrations were determined by spotting cultures on solid media after 48 h of growth in liquid RPMI medium. The MFC is the lowest concentration of fluconazole that inhibited growth completely when the strain was transferred to solid medium. (B) Broth microdilution assays were conducted according to CLSI guidelines. As a measure of the ability of the strains to regrow, the plates were allowed to incubate for 72 h prior to measurement of the optical density at 600 nm. (C) Serial dilutions of the indicated strains were incubated for 24 or 72 h on YPD solid media containing 0, 5, or 10 μg/ml of fluconazole. (D) Time-kill curves for UPC2A disruptants and revertants of a susceptible clinical isolate, SM1, treated with 10 μg/ml of fluconazole (open symbols) or a DMSO control (closed symbols). Cultures were grown in RPMI medium, with aliquots removed at the indicated time points and plated on potato dextrose agar for colony counts.

FIG 3

Enhanced activity of fluconazole in a resistant clinical isolate, SM3, with a disruption of UPC2A. (A) Minimum fungicidal concentrations were determined by spotting cultures on solid media after 48 h of growth in liquid RPMI medium. The MFC is the lowest concentration of fluconazole that inhibited growth completely when the strain was transferred to solid medium. (B) Broth microdilution assays conducted according to CLSI guidelines. As a measure of the ability of the strains to regrow, the plates were allowed to incubate for 72 h prior to measurement of the optical density at 600 nm. (C) Serial dilutions of the indicated strains were incubated for 24 or 72 h on YPD solid media containing 0, 5, or 10 μg/ml of fluconazole. (D) Time-kill curves for UPC2A disruptants and revertants of a resistant clinical isolate, SM3, treated with 10 μg/ml of fluconazole (open symbols) or a DMSO control (closed symbols). Cultures were grown in RPMI media, with aliquots removed at the indicated time points and plated on potato dextrose agar for colony counts.

FIG 4

Effect of UPC2A disruption on constitutive and fluconazole-induced gene expression. RNA was isolated from mid-log-phase cultures of parent clinical isolates, Δupc2A mutants, and revertants, and qRT-PCR analysis of relative gene expression was performed. Light bars represent untreated controls and dark bars represent treatment with 32 μg/ml of fluconazole for 3 h. (A) PDR1; (B) CDR1; (C) SNQ2; (D) PDH1. Data are shown as means ± SEs (n = 3).

FIG 5

Ergosterol quantification for UPC2A disruptants in an SDD clinical isolate, SM1 (A, C, and E), and matched resistant isolate SM3 (B, D, and F). Genotypes studied were as follows: wild-type UPC2A allele (solid circles), Δupc2a (open squares) (A and B), Δupc2b (open triangles) (C and D), and Δupc2a Δupc2b (open diamonds) (E and F). The results for the UPC2A disruptants shown in panels A, B, E, and F showed statistically significantly lower ergosterol content than was found in the wild type (P < 0.05). Knockout strains with the wild-type allele complemented back in showed no difference in ergosterol content compared to those of the wild-type isolates (data not shown). Strains were grown in RPMI medium for 16 h, followed by a heptane extraction and spectrophotometric scan between 240 and 300 nm (41). Each isolate was extracted and analyzed in three independent experiments.

FIG 6

Effect of UPC2A disruption on both baseline and sterol biosynthesis inhibitor-induced expression of ergosterol biosynthesis genes. RNA was isolated from mid-log-phase cultures in parent clinical isolates, Δupc2A mutants, and revertants. qRT-PCR analysis of relative gene expression was performed. (A) SM1, Δupc2A mutant, and revertant gene expression. Gene expression was normalized to that in the untreated SM1 sample for each ergosterol biosynthesis gene. (B) SM3, Δupc2A mutant, and revertant gene expression. Gene expression was normalized to that in the untreated SM3 sample for each ergosterol biosynthesis gene.