Systematic Complex Haploinsufficiency-Based Genetic Analysis of Candida albicans Transcription Factors: Tools and Applications to Virulence-Associated Phenotypes

Abstract

Genetic interaction analysis is a powerful approach to the study of complex biological processes that are dependent on multiple genes. Because of the largely diploid nature of the human fungal pathogen Candida albicans, genetic interaction analysis has been limited to a small number of large-scale screens and a handful for gene-by-gene studies. Complex haploinsufficiency, which occurs when a strain containing two heterozygous mutations at distinct loci shows a phenotype that is distinct from either of the corresponding single heterozygous mutants, is an expedient approach to genetic interactions analysis in diploid organisms. Here, we describe the construction of a barcoded-library of 133 heterozygous TF deletion mutants and deletion cassettes for designed to facilitate complex haploinsufficiency-based genetic interaction studies of the TF networks in C. albicans We have characterized the phenotypes of these heterozygous mutants under a broad range of in vitro conditions using both agar-plate and pooled signature tag-based assays. Consistent with previous studies, haploinsufficiency is relative uncommon. In contrast, a set of 12 TFs enriched in mutants with a role in adhesion were found to have altered competitive fitness at early time points in a murine model of disseminated candidiasis. Finally, we characterized the genetic interactions of a set of biofilm related TFs in the first two steps of biofilm formation, adherence and filamentation of adherent cells. The genetic interaction networks at each stage of biofilm formation are significantly different indicating that the network is not static but dynamic.

Keywords: Candida albicans; biofilm formation; complex haploinsufficiency; disseminated candidiasis; haploinsufficiency; hyphal morphogenesis.

Figure 1

Schematic for construction of heterozygous transcription factor deletion library and LEU2-marked deletion cassettes.

Figure 2

Representative haploinsufficent and haploproficient strains. Ten-fold dilution series of the indicated strains were spotted on agar plates and incubated at either 37°C (A) or 30°C for 1-3 days and photographed. Two independent isolates of each heterozygous strain were plated as noted by -1 and -2. A. The indicated strains were plated on YPD (yeast peptone dextrose) and SM (Spider medium). B. The strains were plated on YPD and YPD+15 mM caffeine. C. The indicated strains were plated on YPD+1%DMSO and YPD+1%DMSO+fluconazole.

Figure 3

TF heterozygous strains have altered competitive fitness at early stages of dissemination in a murine model of candidiasis. A. CDR1 mice were infected with pools of orthogonally barcoded heterozygous deletion mutants and harvested at 24 or 48 hr post infection (N = 5 for each time point). The competitive fitness was calculated as described in Materials and Methods. Competitive fitness data from representative pool of 48 heterozygous strains is shown. Mutants with statistically significant (P < 0.05, Student’s t-test with Benjamani-Hochberg correction for multiple comparisions) decreased (red points) or increased (green points) competitive fitness relative to the median pool R/I. B and C. Competitive fitness of the indicated strain relative to a wild type reference strain. Bars indicate means of 3-5 independent mice with error bars indicate standard error of the means.

Figure 4

Haploinsufficiency and genetic interactions of biofilm TFs during the initial adherence stage of biofilm formation. A. The indicated strains were incubated in microtiter plates in YETS medium at 37°C for 90 min before non-adherent cells were removed by careful washing with PBS. The bars indicate the mean OD₆₀₀ for each strain (triplicate) with error bars indicating standard deviation. Strains with a statistically significant (Student’s t-test P < 0.05) change from the reference strain are indicated by red bars. B. The expected adherence fitness (AF = Mutant OD₆₀₀/Reference OD₆₀₀) for a given double mutant (AF_mutA X AF_mutB) is plotted on x-axis for each biofilm TF double heterozgote and the observed AF for the double mutant is plotted on the y-axis. Mutants that fall within the zone defined by the diagonal (ε = 0 ± estimated error) have no genetic interaction while those above the diagonal (green) display alleviating interactions while those below (red) show aggravating interactions. C. Network of biofilm TF interactions during the adhesion step. Green edges indicate alleviating interactions; red edges indicate aggravating interactions. D. Aggravating interaction network centered on Tec1. E. Alleviating interaction network.

Figure 5

Overexpression of TEC1 increases adhesion of WT and suppresses adhesion defects of other biofilm TF deletion mutants. The mean adhesion of the indicated strains with and without an allele of TEC1 under the control of the strong, constitutive TDH1 promoter is shown by the bars with error bars indicating the standard deviation of duplicates or triplicates.

Figure 6

The genetic interactions of biofilm TFs during filamentation of adherent cells are distinct from adhesion and mature biofilm formation. A. Representative micrographs used to measure the filament length of the indicated strains. The lengths were determined as described in the Materials and Methods with the mean and standard deviation for the indicated strains derived from the measurement of at least one hunderd cells. B. The normalized filament length (mutant length/WT length) expected for a given double mutant was calculated as the product of the normalized mutant lengths of the two single mutants and is plotted on the y-axis. The observed filament length for a given mutant is plotted on the x-axis. Mutants that fall within the zone defined by the diagonal (ε = 0 ± estimated error) have no genetic interaction while those above the diagonal (green) display alleviating interactions while those below (red) show aggravating interactions. C. The genetic interaction network of adherent cell filamentation centered on BCR1. Green indicates alleviating interactions.

Figure 7

The genetic interaction network of biofilm TFs during planktonic filament formation is distinct from adherent cell filamentation. A. Plot of the distribution of filament lengths for WT and rob1Δ tec1Δ derived ImageStream-based analysis along with mean length (µm) ± SD for both strains. B. The normalized filament length (mutant length/WT length) expected for a given double mutant was calculated as the product of the normalized mutant lengths of the two single mutants and is plotted on the y-axis. The observed filament length for a given mutant is plotted on the x-axis. Mutants that fall within the zone defined by the diagonal (ε = 0 ± estimated error) have no genetic interaction while those above the diagonal (green) display alleviating interactions while those below (red) show aggravating interactions. C. Genetic interaction network for alleviating interactions centered on BRG1. D. Genetic interaction network for aggravating interactions centered on ROB1.

Figure 8

The genetic interaction network of biofilm TFs affecting the proportions of yeast and filamentous cells under planktonic conditions is similar to that for length under planktonic conditions. A. ImageStream gating diagrams showing WT and rob1Δ tec1Δ yeast and filamentous cells. B. The expected normalized proportion of filamentous cells (mutant/WT) for the double mutants was calculated as the product of the normalized filament proportion of the two single mutants and is plotted on the y-axis. The observed normalized filament proportion for a given mutant is plotted on the x-axis. Mutants that fall within the zone defined by the diagonal (ε = 0 ± estimated error) have no genetic interaction while those above the diagonal (green) display alleviating interactions while those below (red) show aggravating interactions. C. The genetic interaction network for alleviating interactions.

Figure 9

The genetic interactions of the biofilm TFs are dynamic and are temporally reorganized during biofilm formation. The genetic interaction networks for BCR1 and TEC1 are shown for adhesion, adherent cell filamentation, and mature biofilm stages of biofilm formation. The interaction networks for the mature biofilm stage are derived from Glazier et al. 2017.