Negative regulation of Candida glabrata Pdr1 by the deubiquitinase subunit Bre5 occurs in a ubiquitin independent manner

Abstract

The primary route for development of azole resistance in the fungal pathogen Candida glabrata is acquisition of a point mutation in the PDR1 gene. This locus encodes a transcription factor that upon mutation drives high level expression of a range of genes including the ATP-binding cassette transporter-encoding gene CDR1. Pdr1 activity is also elevated in cells that lack the mitochondrial genome (ρ^° cells), with associated high expression of CDR1 driving azole resistance. To gain insight into the mechanisms controlling activity of Pdr1, we expressed a tandem affinity purification (TAP)-tagged form of Pdr1 in both wild-type (ρ⁺ ) and ρ^° cells. Purified proteins were analyzed by multidimensional protein identification technology mass spectrometry identifying a protein called Bre5 as a factor that co-purified with TAP-Pdr1. In Saccharomyces cerevisiae, Bre5 is part of a deubiquitinase complex formed by association with the ubiquitin-specific protease Ubp3. Genetic analyses in C. glabrata revealed that loss of BRE5, but not UBP3, led to an increase in expression of PDR1 and CDR1 at the transcriptional level. These studies support the view that Bre5 acts as a negative regulator of Pdr1 transcriptional activity and behaves as a C. glabrata-specific modulator of azole resistance.

Figure 1.. Phenotypic analysis of BRE5 and UBP3 in C. glabrata and S. cerevisiae.

A. Isogenic C. glabrata (left side) or S. cerevisiae (right side) strains containing the indicated alleles of BRE5 and UBP3 were grown to mid-log phase and then 1000 cells/spot placed on rich medium containing gradients of fluconazole (FLC). The maximum concentration of FLC is noted at the top of each panel with increasing drug concentration indicated by the bar of increasing width. B. The same strains from above were streaked on rich medium (YPD) or this same medium containing 250 μg/ml brefeldin A (BFA). C. Complementation of drug resistant phenotype of a bre5Δ null. A low-copy-number plasmid (pBM16 (Zordan et al., 2013)) was used as an empty vector (EV) or a full-length BRE5-containing plasmid (pBM16-BRE5) and transformed into a bre5Δ strain. Transformants were grown to mid-log phase and spotted onto YPD medium containing 50 μg/ml nourseothricin (selectable marker in pBM16) and a gradient of fluconazole as indicated. Plates were incubated at 30° C and then photographed.

Figure 2.. Epistatic relationship of BRE5 with PDR1 and CDR1.

An isogenic set of wild-type or bre5Δ strains were further modified where noted by disruption of the PDR1 or CDR1 genes. Representative transformants were grown to mid-log phase and compared for fluconazole (FLC) resistance by gradient plates as described above. Note the large differences in the gradients used for testing cells that contain Pdr1 and Cdr1 versus those that lack either of these drug resistance loci. The bottom panel shows the relationship between Bre5, Pdr1 and Cdr1. The T-bar line indicates inhibition of Pdr1 by Bre5 while the arrow indicates the stimulation of CDR1 gene expression by Pdr1.

Figure 3.. Bre5 associates with Pdr1 in vivo.

Isogenic strains containing a wild-type (Bre5) or C-terminally Myc-tagged form of Bre5 (Bre5-Myc) were grown to mid-log phase and non-denaturing total protein extracts prepared. Aliquots of these total protein extracts were reserved and run as controls (Input) while the remainder was subjected to immunoprecipitation using anti-Pdr1 antiserum. An equal fraction of each immunoprecipitate was electrophoresed through SDS-PAGE in parallel, along with input controls, and the gels were blotted with either anti-Pdr1 antiserum (Left panel) or an anti-Myc antibody (Right panel). Prior to western blot analysis, the transferred proteins were visualized by staining the membranes with Ponceau S (Top panels). Molecular mass standards are indicated at the edge of both western blots.

Figure 4.. Loss of Bre5 stimulates Pdr1-regulated gene expression.

A. Total RNA was isogenic wild-type and bre5Δ cells grown to mid-log phase. Levels of mRNA for genes of interest was determined using gene-specific primers and reverse transcription-quantitative PCR (RT-qPCR) analyses. All mRNA levels were normalized to expression of TEF1 mRNA which served as an unaffected control. Values listed inside each bar are the averages of at least 3 independent determinations of mRNA and represent a ratio of transcription in bre5Δ cells/transcription in the wild-type strain. B. Isogenic strains corresponding to the wild-type, bre5Δ and ubp3Δ were grown to early log phase and then either treated with fluconazole (+FLC) or allowed to continue to grow for 90 minutes with no treatment (-FLC). Total RNA was prepared after this time and analyzed for expression of either PDR1 or CDR1 by RT-qPCR. Levels of these transcripts were normalized to TEF1 as above. Values presented are a ratio of expression in presence of fluconazole relative to expression in the absence of drug.

Figure 5.. Pdr1 polypeptide responds to Bre5 and drug induction.

A. Isogenic PDR1 and pdr1Δ cells containing wild-type or null versions of BRE5 and UBP3 as indicated were grown to mid-log phase in rich medium. Total protein extracts were prepared and equal amounts of protein separated on SDS-PAGE. Levels of proteins were analyzed by western blot using either α-Pdr1 antiserum or α-tubulin (α-Tub1) antibody. The band labeled NS is a non-specific protein that cross reacts with the α-Pdr1 antiserum. The position of Pdr1 is indicated by the arrow at the right-hand side of the figure. Molecular mass standards are labeled on the right-hand side. Quantitation of the western blot is shown on the right-hand side. B. Isogenic wild-type and bre5Δ cells were grown to early log phase and then induced with the addition of fluconazole (FLC) as described above. After 90 minutes of growth in the presence (+) or absence (−) of fluconazole, total protein extracts were prepared and analyzed for levels of Pdr1 protein by western blot as above.

Figure 6.. Synthesis of Pdr1 increases in the absence of Bre5.

A. The indicated strains were grown to early log phase and then labeled with the addition of ³⁵S-methionine for 12 minutes. After this labeling period, a large excess of unlabeled methionine was added and time points removed immediately (0 time) or at 35 and 70 minutes. All time points were processed simultaneously to prepared a cell-free protein extract. Anti-Pdr1 immunoprecipitations were then carried out on each time point.

Figure 7.. Mitochondrial status regulates Pdr1:Bre5 association.

Isogenic wild-type (BRE5) or a derivative carrying the BRE5-13X Myc fusion gene (Bre5-Myc) were generated that contained (ρ+) or lacked (ρ0) the mitochondrial genome. A ρ⁺ pdr1Δ strain (pdr1Δ) was also processed as a control for the location of the immunoprecipitated Pdr1. All strains were grown to mid-log phase in rich medium and whole cell protein extracts prepared. Aliquots (5%) of the whole cell extract were retained as input controls (Input for BRE5 and BRE5-Myc strains, In for the pdr1Δ strain). The remaining whole cell protein extracts were subjected to immunoprecipitation using the anti-Pdr1 antiserum. After washing, immunoprecipitates were electrophoresed along with the input controls through 8% SDS-PAGE and analyzed by western blotting using anti-Pdr1 (α-Pdr1) and anti-Myc (α-Myc) antisera. Locations of the immunoprecipitated proteins are indicated by the arrows at the left.

Figure 8.. Promoter association of Bre5 responds to mitochondrial genome status.

The strains indicated at the right all contain an integrated copy of the BRE5-13X Myc allele. These strains were grown to mid-log phase and then processed for chromatin immunoprecipitation using an anti-Myc antibody. Immunoprecipitated promoter fragments were analyzed by quantitative PCR using primer sets specific for the promoters of the genes listed at the bottom. Levels of immunoprecipitated DNA are plotted on the ordinate. Neither ACT1 or ERG11 (encodes the lanosterol α−14 demethylase target of azole drugs) is responsive to Pdr1 under these conditions.