Systematic analysis of the Candida albicans kinome reveals environmentally contingent protein kinase-mediated regulation of filamentation and biofilm formation in vitro and in vivo

Abstract

Protein kinases are critical regulatory proteins in both prokaryotes and eukaryotes. Accordingly, protein kinases represent a common drug target for a wide range of human diseases. Therefore, understanding protein kinase function in human pathogens such as the fungus Candida albicans is likely to extend our knowledge of its pathobiology and identify new potential therapies. To facilitate the study of C. albicans protein kinases, we constructed a library of 99 non-essential protein kinase homozygous deletion mutants marked with barcodes in the widely used SN genetic background. Here, we describe the construction of this library and the characterization of the competitive fitness of the protein kinase mutants under 11 different growth and stress conditions. We also screened the library for protein kinase mutants with altered filamentation and biofilm formation, two critical virulence traits of C. albicans. An extensive network of protein kinases governs these virulence traits in a manner highly dependent on the specific environmental conditions. Studies on specific protein kinases revealed that (i) the cell wall integrity MAPK pathway plays a condition-dependent role in filament initiation and elongation; (ii) the hyper-osmolar glycerol MAPK pathway is required for both filamentation and biofilm formation, particularly in the setting of in vivo catheter infection; and (iii) Sok1 is dispensable for filamentation in hypoxic environments at the basal level of a biofilm but is required for filamentation in normoxia. In addition to providing a new genetic resource for the community, these observations emphasize the environmentally contingent function of C. albicans protein kinases.IMPORTANCECandida albicans is one of the most common causes of fungal disease in humans for which new therapies are needed. Protein kinases are key regulatory proteins and are increasingly targeted by drugs for the treatment of a wide range of diseases. Understanding protein kinase function in C. albicans pathogenesis may facilitate the development of new antifungal drugs. Here, we describe a new library of 99 protein kinase deletion mutants to facilitate the study of protein kinases. Furthermore, we show that the function of protein kinases in two virulence-related processes, filamentation and biofilm formation, is dependent on the specific environmental conditions.

Keywords: Candida albicans; biofilm; filamentation; protein kinase.

Fig 1

Construction of protein kinase homozygous deletion collection in C. albicans. (A) Genotypes of parental and homozygous protein kinase deletion strain. (B) Venn diagram showing protein kinases targeted (yellow) and homozygous protein kinase mutants constructed (blue). (C) The 20 protein kinases for which only heterozygous deletion mutants were obtained. Red font indicates a protein kinase with previously reported to be potentially essential. Black font indicates a protein kinase for which a homozygous deletion mutant has been reported in the literature.

Fig 2

Competitive fitness assay of protein kinase deletion mutants. (A) Schematic of flow cytometry-based competitive fitness assay. (B) Representative competitive fitness data for ckb1∆∆ compared to SN95. (C) Spot dilution assay showing phenotype of ckb1∆∆ mutant relative to ckb2∆∆ and SN95. (D) Representative spot dilution growth phenotypes for strains without competitive fitness defects by flow cytometry (black font) and those with reduced fitness (red font).

Fig 3

Competitive fitness profile of protein kinase deletion mutants under 10 stress conditions. (A) Hierarchical clustering of the protein kinase deletion mutants with the components of specific pathways highlighted. (B) Confirmation of Sak1-Snf1 phenotypes on spot dilution assays.

Fig 4

Summary of filamentation phenotypes of protein kinase deletion mutants and representative phenotypes. (A) Table of protein kinase mutants with reduced filamentation on all filament-inducing conditions tested. Black font indicates protein kinases previously implicated in filamentation and blue font indicates protein kinases not previously reported to play a role in filamentation. (B) Representative phenotypes of the reference strain SN95 and protein kinase deletion mutants on RPMI medium incubated at 37°C.

Fig 5

Genetic interactions of protein kinase mutants with nrg1∆∆ mutant reveal role for cell wall integrity pathway during hyphal extension. (A) Strategy to assess the timing of protein kinase function during filamentation using genetic interactions with nrg1∆∆ mutant. (B and C) The indicated strains were incubated in YPD at 30°C (B) or RMPI + 1% bovine calf serum (C) and the distribution of yeast, pseudohyphae, and hyphae as determined by microscopy on fixed samples. The bars indicate the means of two independent experiments with greater than 100 cells counted per experiment. Chi-squared tests were used to evaluate the statistical significance of comparisons (P < 0.05): * indicates comparison to protein kinase single deletion mutant; # indicates comparison to nrg1∆∆ mutant; ∆ indicates comparison to SN95. (D) The length of hyphae formed after 4-hour induction in RPMI + 1% bovine calf serum was determined as described in Materials and Methods. Statistical significance was determined by Mann-Whitney; * indicates P < 0.05. (E) Western blot analysis of Mkc1 phosphorylation in the indicated strains using anti-p42/44 antibody in the indicated strains after incubation in YPD at 30°C or RMPI + 10% BCS at 37°C for 4 hours. Ponceau staining was used to assess loading. ** indicates a non-specific band observed in serum-induced samples. The blots are representative of three independent experiments. (F) The percentage of filamentous cells in WT or cbk1∆∆ mutant strains 24 hours after inoculation of mouse ears.

Fig 6

An extensive network of protein kinases governs biofilm formation. (A) Representative protein kinase mutants with severe or moderate phenotypes based on the imaging assay used to screen the deletion collection. Biofilms were formed in RPMI + 10% bovine calf serum at 37°C for 24 hours in microtiter plates, stained with Calcofluor white, and imaged as described in Materials and Methods. The cross-section images of the biofilm are shown. (B) Protein kinases whose deletion mutants have reduced biofilm formation in both Spider medium and RPMI + 10% bovine calf serum at 37°C. Red font indicates severe phenotype. (C) Venn diagram showing the overlap between protein kinase mutants with reduced filamentation and biofilm phenotypes in RPMI + 10% bovine calf serum at 37°C.

Fig 7

Confirmation of biofilm phenotypes in optical density assay. (A–C) The indicated strains were incubated in microtiter plates in RPMI medium at 37°C. For adhesion, the plates were washed with PBS at 90 minutes and the OD₆₀₀ determined. Mature biofilms were measured at 24 hours and 48 hours. Error bars indicate the mean of 4–5 replicates with error bars standard deviation. **** indicates statistical significance (P < 0.05) by two-way ANOVA with correction for multiple comparisons. (D) Cross-sectional image of the bub1∆∆ and SN95 biofilms indicating that the bub1∆∆ mutant forms hyphae.

Fig 8

Sok1 is dispensable for filamentation at the basal biofilm layer and in hypoxia. (A) Cross-section and apical views of the biofilms formed by SN95 and the sok1∆∆ mutant in RPMI + 10% BCS at 37°C after 24 hours. (B) The indicated strains were spotted on RPMI agar plates supplemented with 10%BCS and incubated for 4 days in ambient air or a GasPak hypoxic environment. (C) Time course of biofilm formation with SN95 and sok1∆∆ strains. (D) Quantification of biofilm thickness over time course. Bars are means of three independent experiments with error bars indicating standard deviation. * indicates statistical significant (P < 0.05) difference between SN95 and sok1∆∆ mutant by two-tailed Student’s t-test. (E) Apical views of SN95 and sok1∆∆ mutant biofilms at 2 hours indicate filament formation in both samples.

Fig 9

The role of Cbk1 in biofilm formation is independent of Ace2 and partially suppressed by overexpression of ALS3 and HGC1. (A) Apical and cross-section images of 24-hour biofilms formed by the indicated strains in RPMI + 10% BCS at 37°C. Please note that the images of the SN95 control strain are duplicated to allow ease of comparison to both the cbk1∆∆ mutants (top panel) and kic1∆∆ mutants (bottom panel). (B and C) show a comparison of the biofilm volume for cbk1∆∆ (B) and kic1∆∆ (C) mutants and corresponding P_RBT5-ALS3/HGC1 overexpression strains. **** indicates a statistically significant difference between the indicated groups by two-way ANOVA with multiple comparisons test. (D) Fluorescence microscopy of Calcofluor-stained cells from biofilms for the indicated strains. (E and F) show the fungal burden of jugular venous catheters placed in rats that were infected with the indicated strains. The catheters were removed at 24 hours; the fungal cells were processed and plated as described in Materials and Methods. Bars indicate the mean of three replicates with standard deviation noted by error bars. **** indicates a statistically significant difference (P < 0.05) between groups by two-way ANOVA with multiple comparisons test. The fungal burden for cbk1∆∆ and kic1∆∆ mutants and corresponding P_RBT5-ALS3/HGC1 overexpression strains are shown in E and F, respectively.

Fig 10

The HOG pathway plays a condition-dependent role in biofilm formation. The biofilm density for the hog1∆∆ (A) and pbs2∆∆ (B) mutants was determined by optical density in RPMI medium at 37°C at the indicated time points. Error bars indicate the mean of 4–5 replicates with error bars standard deviation. **** indicates statistical significance (P < 0.05) by two-way ANOVA with correction for multiple comparisons. (C) The fungal burden of jugular venous catheters placed in rats that were infected with either SN95 or the pbs2∆∆ mutant. The catheters were removed at 24 hours; the fungal cells were processed and plated as described in Materials and Methods. Bars indicate the mean of three replicates with standard deviation noted by error bars. **** indicates a statistically significant difference (P < 0.05) between groups by Student’s t-test. (D) Electron micrographs of biofilms formed in vascular catheters by SN95 and the pbs2∆∆ mutant.