A large-scale complex haploinsufficiency-based genetic interaction screen in Candida albicans: analysis of the RAM network during morphogenesis

Abstract

The morphogenetic transition between yeast and filamentous forms of the human fungal pathogen Candida albicans is regulated by a variety of signaling pathways. How these pathways interact to orchestrate morphogenesis, however, has not been as well characterized. To address this question and to identify genes that interact with the Regulation of Ace2 and Morphogenesis (RAM) pathway during filamentation, we report the first large-scale genetic interaction screen in C. albicans.Our strategy for this screen was based on the concept of complex haploinsufficiency (CHI). A heterozygous mutant of CBK1(cbk1Δ/CBK1), a key RAM pathway protein kinase, was subjected to transposon-mediated, insertional mutagenesis. The resulting double heterozygous mutants (6,528 independent strains) were screened for decreased filamentation on SpiderMedium (SM). From the 441 mutants showing altered filamentation, 139 transposon insertion sites were sequenced,yielding 41 unique CBK1-interacting genes. This gene set was enriched in transcriptional targets of Ace2 and, strikingly, the cAMP-dependent protein kinase A (PKA) pathway, suggesting an interaction between these two pathways. Further analysis indicates that the RAM and PKA pathways co-regulate a common set of genes during morphogenesis and that hyperactivation of the PKA pathway may compensate for loss of RAM pathway function. Our data also indicate that the PKA–regulated transcription factor Efg1 primarily localizes to yeast phase cells while the RAM–pathway regulated transcription factor Ace2 localizes to daughter nuclei of filamentous cells, suggesting that Efg1 and Ace2 regulate a common set of genes at separate stages of morphogenesis. Taken together, our observations indicate that CHI–based screening is a useful approach to genetic interaction analysis in C. albicans and support a model in which these two pathways regulate a common set of genes at different stages of filamentation.

Figure 1. Schematic of screening strategy.

In vitro mutagenesis of C. albicans genomic library WO-1 using a Tn7-based transposon containing the CaURA3-dpl200 auxotrophic marker yielded a library of plasmids from which genomic inserts were released by restriction endonuclease digestion and transformed into a cbk1Δ/CBK1 heterozygote strain. The resulting library was screened on SM for altered filamentation relative to the parental strain.

Figure 2. CHI–based screening identifies synthetic genetic interactions with CBK1 during morphogenesis.

(A) Examples of primary screening data for complex heterozygotes showing synthetic genetic interactions with CBK1; each strain was spotted on SM and incubated at 37°C for 3 days. Mutants with decreased peripheral invasion and decreased central wrinkling were selected. Representative positive scoring mutants from the primary screen are shown. An example of a strain complemented by re-integration of plasmid-borne CBK1 is shown. (B) Representative examples of independently constructed complex heterozygote strains showing complex haploinsufficient genetic interactions with cbk1Δ. (C) The ratio of pseudohyphal∶hyphal cells for the indicated strains was determined by light microscopy after 3 hours incubation in liquid SM at 37°C. The bars indicate mean values of two-three independent replicates of at least 100 cells. Error bars indicate standard deviation. Brackets indicate the results of Student's t test evaluation of differences between the indicated mutants; p<0.05 indicates a statistically significant difference. (D) Micrograph of filaments isolated from colonies of the parental cbk1Δ/CBK1 strain and the complex heterozygote cbk1Δ/CBK1 snz1Δ/SNZ1. Arrowheads indicate areas of hyphal-like morphology in the cbk1Δ/CBK1 mutant and pseudohyphae-like morphology in the cbk1Δ/CBK1 snz1Δ/SNZ1 double mutant.

Figure 3. Summary and bioinformatic analysis of screening data.

(A) Summary of screening results and list of CBK1-interacting genes. (B) List of CBK1-synthetic genetic interactions during morphogenesis grouped according to three most common GO terms. Colors indicate the number of times each insertion was isolated. (C) Venn diagram depicting the number of genes putatively co-regulated by the RAM and PKA pathways. (D) List of CBK1-interacting genes with Ace2 and both Ace2/Efg1 consensus binding sites within the region 1000 bp upstream of the start codon.

Figure 4. The set of CBK1-interacting genes includes transcriptional targets of Ace2.

The binding of Ace2-TAP to the promoter regions of 5 CBK1-interacting genes was assessed by ChIP in yeast and hyphae-phase cells (SM, 3 h, 37°C) containing a TAP-tag fused to the C-terminus of Ace2. Bars indicate the ratio of promoter DNA (determined by PCR) in tagged extracts relative to un-tagged extracts (error bars indicate SD of three replicates). Grey bars show promoters with increased abundance in tagged extracts, suggesting they are bound by Ace2. P _cht3, a known target of Ace2 , , and primers to a coding sequence (ORF cds) serve as positive and negative controls. A persistent contaminating band prevented accurate assessment of ACT1 in hyphae.

Figure 5. The PKA pathway compensates for decreased RAM pathway activity during morphogenesis.

(A) Transcript levels of ENO1 and PGK1 in each mutant were compared to wild type by qRT-PCR using the 2^−ΔΔCt method and are graphed as Log₂ change over wild type (three independent experiments performed in triplicate). Bars indicate mean value and error bars indicate standard deviation. The observed elevation in levels of each transcript in the indicated mutants relative to wild type were statistically significant by Student's t test (p<0.05). (B) Phosphorylation of fluorescent PKA substrate (PepTag, Promega) by cell extracts (10 µg protein) derived from wild type (WT) and ace2Δ/2Δ cells harvested after incubation in SM for 3 h at 37°C. The indicated time points represent PKA reaction time. (C) The ratio of pseudohyphal∶hyphal cells for the indicated strains was determined by light microscopy after 3 h incubation in liquid SM at 37°C. The bars indicate mean values of two-three independent replicates of at least 100 cells. Error bars indicate standard deviation. Brackets indicate the results of Student's t test evaluation of differences between the indicated mutants; p<0.05 indicates a statistically significant difference. (D) Hybrid pseudohyphae/hyphae cells of cbk1Δ/CBK1 tpk1Δ/1Δ following staining with Calcofluor white. Arrows indicate budneck localized septa (pseudohyphae-like) and block arrows indicate distal septa (hyphae-like).

Figure 6. Elevated PKA activity accounts for increase pseudohyphae in RAM mutants.

(A) ace2Δ/Δ cells were incubated in YPD at 30°C for 3 h −/+ PKA inhibitor MyrPKI (10 µM) and examined by light microscopy. (B) EFG1 expression was determined by semi-quantitative RT-PCR for each strain at the indicated time after transfer to SM at 37°C. ACT1 levels were used as loading control. The graph indicates the fold change in EFG1 levels for the mutant strains relative to wild type at the 180 min time point. The bars indicate the mean fold change in EFG1 relative to wild type for three independent replicates and the error bars indicate standard deviation. The brackets indicate that the difference between EFG1 transcript levels was statistically significant for each mutant relative to wild type (Student's t test, p<0.02). (C) The expression of ENO1 and PGK1 were examined in the indicated strains as described in Figure 5A. The brackets indicate that the difference between PGK1 transcript levels was statistically significant for the two mutants (Student's t test, p<0.02).

Figure 7. Efg1 and Ace2 are present in the nuclei at different time points during morphogenesis.

(A) The localization of Ace2-GFP was determined in stationary phase cells prior to initiation of hyphal induction (yeast form) and after 3 hr hyphal induction in SM. DAPI staining was used to identify the nuclei. (B) The localization of Efg1-Myc was determined by indirect immunofluorescence under conditions identical to those described for Ace2-GFP. (C) The binding of Efg1-Myc to promoter regions of ENO1 and PGK1 was examined by ChIP for cells corresponding to the time points examined in A and B. (D) ACE2 expression in wild type (WT) cells compared to ACT1 by RT-PCR at the indicated times after hyphal induction in SM at 37°C. (E) ACE2 expression was determined in wild type cells −/+ PKA inhibitor (MyrPKI, 10 µM) after 3 h induction in SM at 37°C.