Signaling through adenylyl cyclase is essential for hyphal growth and virulence in the pathogenic fungus Candida albicans

Abstract

The human fungal pathogen Candida albicans switches from a budding yeast form to a polarized hyphal form in response to various external signals. This morphogenetic switching has been implicated in the development of pathogenicity. We have cloned the CaCDC35 gene encoding C. albicans adenylyl cyclase by functional complementation of the conditional growth defect of Saccharomyces cerevisiae cells with mutations in Ras1p and Ras2p. It has previously been shown that these Ras homologues regulate adenylyl cyclase in yeast. The C. albicans adenylyl cyclase is highly homologous to other fungal adenylyl cyclases but has less sequence similarity with the mammalian enzymes. C. albicans cells deleted for both alleles of CaCDC35 had no detectable cAMP levels, suggesting that this gene encodes the only adenylyl cyclase in C. albicans. The homozygous mutant cells were viable but grew more slowly than wild-type cells and were unable to switch from the yeast to the hyphal form under all environmental conditions that we analyzed in vitro. Moreover, this morphogenetic switch was completely blocked in mutant cells undergoing phagocytosis by macrophages. However, morphogenetic switching was restored by exogenous cAMP. On the basis of epistasis experiments, we propose that CaCdc35p acts downstream of the Ras homologue CaRas1p. These epistasis experiments also suggest that the putative transcription factor Efg1p and components of the hyphal-inducing MAP kinase pathway depend on the function of CaCdc35p in their ability to induce morphogenetic switching. Homozygous cacdc35 Delta cells were unable to establish vaginal infection in a mucosal membrane mouse model and were avirulent in a mouse model for systemic infections. These findings suggest that fungal adenylyl cyclases and other regulators of the cAMP signaling pathway may be useful targets for antifungal drugs.

Figure 1

Deletion of CaCDC35 in C. albicans. (A) Restriction endonuclease map of CaCdc35p. The white rectangle indicates the coding region of the gene. The conserved central leucine rich repeat (LRR) motif, the protein phosphatase 2Cα (PP2Cα) domain, and the catalytic domain (CD) are indicated. The PCR with the divergent oligonucleotides OCR1 and OCR2 was used to delete the coding sequence of CaCDC35. A hisG-URA3-hisG cassette was then inserted, and a two-step procedure was used to delete both alleles of CaCDC35 by homologous recombination (see MATERIAL AND METHODS). (B) Southern blot analysis with the use of a 0.8-kb SphI-SpeI probe of the CaCDC35 coding region. The genomic DNA samples, digested with SphI and BspMII, were prepared from strains CAI4 (CaCDC35/CaCDC35; lane 1), CR20 (CaCDC35/cacdc35Δ::hisG-URA3-hisG, lane 2), CR20.1 (CaCDC35/ cacdc35Δ::hisG, lane 3), CR216 (cacdc35Δ::hisG-URA3-hisG/ cacdc35Δ:: hisG, lane 4), and CR276 (cacdc35Δ::hisG/cacdc35Δ::hisG, lane 5). The wild-type band was 7.25 kb, whereas the KO band with the URA3 blaster was 6.1 kb. After looping out the URA3 fragment, the band was reduced to 3.2 kb. (C) Northern blot analysis of poly(A)⁺ RNA isolated from strains SC5314 (CaCDC35/CaCDC35; lane 1), CR20 (CaCDC35/cacdc35Δ; lane2), CR216 (cacdc35Δ/cacdc35Δ; lane 3), and CR323 (cacdc35Δ/cacdc35Δ [CDC35]; lane 4). The blot was probed with fragments specific for CaCDC35 or the actin gene (CaACT1) and quantified by phosphorimaging. The ratios at the bottom of each lane represent the amount of the CaCDC35 transcript relative to the CaACT1 transcript in the same lane. The relative overexpression of the reintegrant construct may be due to the consequence of having only a limited region of the promoter on the plasmid as well as to the influence of the site of integration on general expression levels.

Figure 2

Intracellular cAMP concentrations in strains SC5314 (wild-type) and three isolated colonies A, B, and C of CR216 (cacdc35Δ::hisG-URA3-hisG/cacdc35Δ). After addition of 100 mM glucose to the medium, cells were harvested at the indicated time points, and cAMP levels were determined as described in MATERIAL AND METHODS. The data represent mean values of two independent measurements.

Figure 3

Defects in hyphal formation caused by deletion of both CaCDC35 alleles. (A) The CaCDC35 wild-type strain SC5314 (WT) and strain CR216 deleted for both alleles of CaCDC35 (cacdc35Δ/cacdc35Δ) were grown for 2 or 6 h at 37°C in liquid Lee's medium with or without (+/−) 10% fetal calf serum (FCS) and with or without (+/−) 10 mM dibutyryl-cAMP. Photomicrographs were taken by Nomarski optics at 1000× magnification. Scale bar, 10 μm. (B) The CaCDC35 wild-type strain SC5314 (WT), the heterozygous strain CR20 (CaCDC35/cacdc35Δ), and the homozygous strain CR216 (cacdc35Δ/cacdc35Δ) were grown for 5 d at 37°C on either solid agar medium containing 10% fetal calf serum (FCS) or on solid Lee's medium (Lee et al., 1975). The same defects in hyphal growth were observed on solid low ammonia dextrose nitrogen starvation medium (SLAD; Gimeno et al., 1992) and Spider medium (Liu et al., 1994; our unpublished results). Scale bar, 2 mm. Photomicrographs were taken with the use of phase contrast at 20× magnification.

Figure 4

Epistasis analysis. (A) C. albicans strain CR216 deleted for both alleles of CaCDC35 was transformed with the empty control plasmid pVEC and plasmids pVEC-CaCDC35, pYPB1-ADHpL-CaRAS1, pYPBL-ADHpT-HST7, pYPBl-ADHpL-CPH1, and pRC2312–EFG1 carrying the indicated C. albicans genes. Transformants were grown for 6 h at 37°C in selective medium containing 10% fetal calf serum (FCS). To induce expression of EFG1, 2% casamino acids were added to the medium. (B) Strains CAI4 (WT) and CR216 deleted for both alleles of CaCDC35 were transformed with pYPBl-ADHpL-CaRAS1^G13V carrying the hyperactive mutant version of CaRAS1 and grown for 16 h in selective medium at 30°C (two top panels). The overnight cultures were transferred to fresh -URA medium containing 10% FCS and grown for 6 h at 37°C (two bottom panels). Photomicrographs were taken by Nomarski optics at a 1000× magnification (scale bar, 10 μm) and are representative of many cells.

Figure 5

Phagocytosis of C. albicans cells by macrophages. Macrophages incubated with wild-type C. albicans cells (SC5314) or C. albicans cells deleted for both alleles of CaCDC35 (CR216) for the indicated time periods. Micrographs were taken at a magnification of 1000× by phase contrast (left panels) or fluorescence after selective immunostaining of C. albicans cells with C. albicans–specific antibodies (right panels). The black arrows point to C. albicans cells phagocytosed by the macrophage cells. Scale bar, 10 μm.

Figure 6

Vaginal infection. Mice were intravaginally inoculated with 5 × 10⁴ cells of strain CR340 (cacdc35Δ/cacdc35Δ [pVEC-CaCDC35]; dark bars) and 4 × 10⁵ cells of strain CR323 (cacdc35Δ/cacdc35Δ [pVEC]; open bars). C. albicans cells were then recovered from the vaginal canal at the indicated time points and colony forming units (CFU) were determined. The data represent log mean values ± SD of 10 independent experiments.

Figure 7

Survival curves of mice (10 for each group) intravenously infected with 5 × 10⁵ cells of C. albicans strains CR340 (cacdc35Δ/cacdc35Δ [pVEC-CaCDC35]; light open circle) and 4 × 10⁶ cells of strain CR323 (cacdc35Δ/cacdc35Δ [pVEC]; bold open circle).