Fig 1.
AP-MS approaches identify novel interactors for C. albicans Hsp90.
AP-MS of affinity-tagged Hsp90E36A and nine Hsp90 co-chaperones identified physically interacting proteins. Cells were grown in YPD at 30°C, and statistically significant interactions were defined through SAINTexpress analysis compared with an unrelated tagged protein. Pink squares represent the 10 bait proteins. Purple squares represent prey proteins that were statistically significant interactors by SAINTexpress, and are grouped based on GO term annotation. Blue squares represent known stress granule or P-body proteins in S. cerevisiae. The width of the edges corresponds to peptide counts (peptide counts provided in S1 Table), and green edges represent physical interactions that have also been previously identified in S. cerevisiae, as annotated in BioGRID. AP-MS, affinity purification-mass spectrometry; GO, gene ontology; P-body, processing body; SAINTexpress, Significance Analysis of INTeractome; YPD, yeast extract peptone dextrose.
Fig 2.
Global proteomic changes upon perturbation of Hsp90.
(A) Global proteome analysis in C. albicans. Non-denaturing IEX fractionation and MS experiments were performed on lys2Δ/lys2Δ tetO-HSP90/hsp90Δ cells that were untreated, treated with DOX to repress HSP90 expression, or treated with 15 μM GdA to inhibit Hsp90 function. For DOX treatment, cells were treated with 0.05 μg/mL DOX overnight and then subcultured into medium with 5 μg/mL DOX for 4 hours before harvesting. Displayed is the co-fractionation (IEX-HPLC) profile of the C. albicans untreated extracts. Shading for the untreated condition indicates spectral counts recorded in LC-MS/MS, with intensity normalized on a per-gene basis. Clustering order maximizes the correlation of adjacent intensity profiles from the untreated sample. Raw data for this figure can be found in S2 Table. (B) Co-fractionation (IEX-HPLC) reveals a 20S proteasome cluster. This co-fractionation cluster is enriched for the 20S proteasome, with proteins in this category highlighted in orange. Shading for the untreated condition indicates spectral counts recorded in LC-MS/MS, with intensity normalized on a per-gene basis. For the treatment conditions, shading indicates per-gene scaled log-fold change compared with the untreated sample, with blue indicating a decrease and red indicating an increase in peptide abundance. Raw data for this figure can be found in S2 Table and S3 Table. (C) Proteasome protein levels increase upon Hsp90 perturbation. To repress Hsp90, cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 expression in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours. For pharmacological inhibition of Hsp90, wild-type cells were grown overnight in YPD and then subcultured into medium with 15 μM GdA or DMSO vehicle control for 4 hours. Proteins were extracted and immunoblotted with an antibody specific to a conserved peptide in the 20S proteasome proteins Pre1 (22.0 kDa), Pre2 (31.2 kDa), Pre3 (23.3 kDa), Pre5 (31.4 kDa), Pre6 (27.4 kDa), and Pre7 (27.4 kDa). Protein levels were normalized to the tubulin loading control and quantified compared with their respective vehicle controls. (D) Alteration in kinase levels upon Hsp90 depletion. Protein abundance for predicted kinase proteins that showed >1.5-fold log change in abundance across two or more fractions in cells with DOX-mediated transcriptional repression of HSP90 compared with untreated cells in the tetO-HSP90/hsp90Δ strain. Shading indicates per-gene scaled log-fold change compared with the untreated sample, with blue indicating a decrease and red indicating an increase. Raw data for this figure can be found in S3 Table. (E) Mck1 protein levels decrease upon transcriptional repression of HSP90. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction and western blotting. Protein levels were normalized to the tubulin loading control and quantified compared with the vehicle controls. DOX, doxycycline; GdA, geldanamycin; IEX-HPLC, ion exchange–high performance liquid chromatography; LC-MS/MS, liquid chromatography–tandem mass spectrometry; YPD, yeast extract peptone dextrose.
Fig 3.
The Hsp90 physical interaction network is modulated by antifungal drug stress.
(A) AP-MS was performed on HSP90E36A-GFP/HSP90 cells grown at 30°C in the presence or absence of 8 μg/mL fluconazole (Flu) or 100 nM caspofungin (Casp). Statistically significant Hsp90 interaction partners are shown as a dot plot in which node color represents the absolute number of spectral counts, node size represents relative abundance between samples, and node edges represent the SAINTexpress BFDR rate at which a given prey protein was observed. Raw data for this figure can be found in S4 Table. (B) Bait versus bait comparison shows the highest correlation is between interactors identified in the YPD and fluconazole conditions. Color scale represents the degree of correlation. AP-MS, affinity purification-mass spectrometry; AvgSpec, Average Spectral Count; BFDR, Bayesian false discovery rate; GFP, green fluorescent protein; SAINTexpress, Significance Analysis of INTeractome.
Fig 4.
Pbs2 is a client of Hsp90 in C. albicans.
(A) Pbs2 and Hsp90 physically interact. Immunoprecipitation of Pbs2-HA with anti-HA agarose co-purified Hsp90, while Hsp90 did not co-purify with anti-HA agarose in control cells lacking tagged Pbs2. There was no difference in Hsp90 levels between input samples. (B) Pbs2-HA is destabilized upon transcriptional repression of HSP90. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction and western blotting. Protein levels were normalized to the tubulin loading control and quantified compared with the no DOX control. (C) Pbs2-HA protein levels decrease upon fluconazole but not caspofungin stress. Cells were incubated with 8 μg/mL fluconazole (Fluc) or 100 nM caspofungin (Caspo) before protein extraction and western blotting. Protein levels were normalized to the tubulin loading control and quantified compared with the no drug control. (D) PBS2 transcript levels are not altered by transcriptional repression of HSP90 or treatment with antifungal drugs. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before RNA extraction and qRT-PCR. Cells were incubated with 8 μg/mL fluconazole or 100 nM caspofungin before RNA extraction and qRT-PCR. PBS2 transcript levels were normalized to PMA1 and TEF1. Significance was determined by one-way ANOVA. **** indicates P-value < 0.001. Raw data for this figure can be found in S1 Data. Caspo, caspofungin; DOX, doxycycline; Fluc, fluconazole; HA, hemagglutinin; IP, immunoprecipitation; qRT-PCR, quantitative reverse transcription PCR; YPD, yeast extract peptone dextrose.
Fig 5.
Hsp90 represses the sumoylation of Rvb1.
(A) Chaperone proteins vary in their interaction with Hsp90 upon drug treatment. AP-MS was performed on HSP90E36A-GFP/HSP90 cells grown at 30°C in the presence or absence of 8 μg/mL fluconazole (Flu) or 100 nM caspofungin (Casp). Hsp90 interaction partners are shown as a dot plot in which the node color represents the absolute spectral count, node size represents relative abundance between the conditions, and node edge color represents the SAINTexpress BFDR rate. Raw data for this figure can be found in S4 Table. (B) Rvb1 migrates as two bands upon fluconazole treatment. Cells were incubated with 8 μg/mL fluconazole (Flu) or 100 nM caspofungin (Casp) before protein extraction and western blotting. Rvb2 migration is unaffected by antifungal drug treatment. Molecular weight markers are shown (kDa). (C) Rvb1 migrates as two bands upon Hsp90 depletion. Wild-type and tetO-HSP90/hsp90Δ cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction and western blotting. Wild-type cells were also incubated with 8 μg/mL fluconazole (Flu). Molecular weight markers are shown (kDa). (D) The change in Rvb1 migration depends on the Smt3 SUMO protein. Cells were incubated with or without 8 μg/mL fluconazole (Flu) before protein extraction and western blotting. Molecular weight markers are shown (kDa). (E) Rvb1 is sumoylated. Smt3-specific antibodies immunoprecipitate the high-molecular weight band of Rvb1. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction. Samples were immunoprecipitated with an anti-Smt3 antibody and probed with an anti-Rvb1 antibody. The anti-Rvb1 antibody only identified a band upon Hsp90 depletion. Molecular weight markers are shown (kDa). (F) Ulp1-GFP is destabilized upon transcriptional repression of HSP90. Wild-type and tetO-HSP90/hsp90Δ cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction and western blotting. Protein levels were normalized to the tubulin loading control and quantified compared with the no DOX control. (G) Migration of Rvb1 and Rvb2 upon fluconazole treatment in S. cerevisiae. Cells were incubated with 0, 4, or 8 μg/mL fluconazole before protein extraction and western blotting with anti-Rvb1 or anti-Rvb2 antibodies. AP-MS, affinity purification-mass spectrometry; BFDR, Bayesian false discovery rate; Casp, caspofungin; DOX, doxycycline; Flu, fluconazole; GFP, green fluorescent protein; SAINTexpress, Significance Analysis of INTeractome; SUMO, small ubiquitin-like modifier.
Fig 6.
Hsp90 is required for the stability of stress granule and P-body proteins.
(A) Stress granule and P-body proteins vary in their interaction with Hsp90 upon drug treatment. AP-MS was performed on HSP90E36A-GFP/HSP90 cells grown at 30°C in the presence or absence of 8 μg/mL fluconazole (Flu) or 100 nM caspofungin (Casp). Statistically significant Hsp90 interaction partners are shown as a dot plot in which the node color represents the absolute spectral count, node size represents relative abundance between the conditions, and node edge color represents the SAINTexpress BFDR rate. Raw data for this figure can be found in S4 Table. (B) Stress granule protein abundance upon Hsp90 depletion. Protein abundance across 120 IEX fractions in untreated cells or cells treated with DOX to transcriptionally repress HSP90 in the tetO-HSP90/hsp90Δ strain. Raw data for this figure can be found in S3 Table. (C) Stress granule proteins are required for caspofungin tolerance. Mutants of proteins involved in stress granules (Pab1, Kre30, and Sup35) are hypersensitive to caspofungin. MIC assays were performed in YPD medium at 30°C for 24 hours, and optical densities at 600 nm were averaged for two biological replicates performed in technical duplicate. Growth was normalized to the no drug well. To repress target gene expression, the strains were incubated in the indicated concentrations of DOX. Raw data for this figure can be found in S1 Data. (D) Stress granule and P-body protein levels are decreased upon transcriptional repression of HSP90. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before protein extraction and western blotting. Protein levels were normalized to the tubulin loading control and quantified compared with the no DOX control. (E) Stress granule protein transcripts are not dependent on Hsp90. Cells were grown overnight in ±0.5 μg/mL DOX to repress HSP90 in the tetO-HSP90/hsp90Δ strain and then subcultured into medium ± 5 μg/mL DOX for 4 hours before RNA extraction and qRT-PCR. Transcript levels were normalized to PMA1 and TEF1. Significance was determined by one-way ANOVA. ** indicates P value <0.01, * indicates P value <0.05. Raw data for this figure can be found in S1 Data. AP-MS, affinity purification-mass spectrometry; BFDR, Bayesian false discovery rate; DOX, doxycycline; GFP, green fluorescent protein; IEX, ion exchange; MIC, minimum inhibitory concentration; P-body, processing body; qRT-PCR, quantitative reverse transcription PCR; SAINTexpress, Significance Analysis of INTeractome; YPD, yeast extract peptone dextrose.
Fig 7.
Stress-dependent protein aggregation correlates with dependence on Hsp90 for stability.
(A) Identification of stress granules and P-bodies in C. albicans. Cells were grown overnight in YPD, diluted to an OD600 of 0.1, and subcultured for 4 hours before a 10-minute incubation at 46°C. Scale bar is 10 μm. (B) Protein aggregation depends on temperature. Cells were grown overnight in YPD, diluted to an OD600 of 0.1, and subcultured for 4 hours at 30°C before a 10-minute incubation at the indicated temperatures. Scale bar is 10 μm. Arrows indicate aggregates. GFP, green fluorescent protein; OD600, optical density at wavelength 600 nm; P-body, processing body; RFP, red fluorescent protein; YPD, yeast extract peptone dextrose.