The transcription factor homolog CTF1 regulates {beta}-oxidation in Candida albicans

Abstract

Carbon starvation is one of the many stresses to which microbial pathogens are subjected while in the host. Pathways necessary for the utilization of alternative carbon sources, such as gluconeogenesis, the glyoxylate cycle, and beta-oxidation of fatty acids, have been shown to be required for full virulence in several systems, including the fungal pathogen Candida albicans. We have investigated the regulatory network governing alternative carbon metabolism in this organism through characterization of transcriptional regulators identified based on the model fungi, Saccharomyces cerevisiae and Aspergillus nidulans. C. albicans has homologs of the ScCAT8/AnFacB and ScADR1/AnAmdX transcription factors that regulate induction of genes encoding the proteins of gluconeogenesis, the glyoxylate cycle, and ethanol utilization. Surprisingly, C. albicans mutants lacking CAT8 or ADR1 have no apparent phenotypes and do not regulate genes for key enzymes of these pathways. Fatty acid degradation and peroxisomal biogenesis are controlled by nonhomologous regulators, OAF1/PIP2 in S. cerevisiae and FarA/FarB in A. nidulans; C. albicans is missing OAF1 and PIP2 and, instead, has a single homolog of the Far proteins, CTF1. We have shown that CTF1 is required for growth on lipids and for expression of genes necessary for beta-oxidation, such as FOX2. ctf1Delta/ctf1Delta (ctf1Delta/Delta) strains do not, however, show the pleiotropic phenotypes observed for fox2Delta/Delta mutants. The ctf1Delta/Delta mutant confers a mild attenuation in virulence, like the fox2Delta/Delta mutant. Thus, phenotypic and genotypic observations highlight important differences in the regulatory network for alternative carbon metabolism in C. albicans compared to the paradigms developed in other model fungi.

FIG. 1.

Two distinct families of transcription factors regulating fatty acid catabolism and peroxisome biogenesis in ascomycetes. Homologs were identified based on BLAST similarity to C. albicans CTF1 (A) or S. cerevisiae OAF1 (B). Only hits with a BLAST e value of <10 to 50 were included, except for C. albicans CTA4, which is the closest C. albicans homolog of ScOAF1. Proteins were aligned using ClustalW, and the phylogenetic tree was built using the neighbor-joining algorithm. A subset of homologs was included, but those chosen represent the species diversity of these family members. Boldface type indicates a homolog from a species pathogenic to humans. F. oxysporum, Fusarium oxysporum; Y. lipolytica, Yarrowia lipolytica; D. hansenii, Debaryomyces hansenii; L. elongisporus, Lodderomyces elongisporus; C. guilliermondii, Candida guilliermondii; C. immitis, Coccidioides immitis; A. niger, Aspergillus niger; C. glabrata, Candida glabrata; V. polyspora, Vanderwaltozyma polyspora; A. gossypii, Ashbya gossypii; K. lactis, Kluyveromyces lactis.

FIG. 2.

ctf1Δ/Δ and adr1Δ/Δ mutant strains have growth defects on specific carbon sources. Fivefold serial dilutions of the indicated strains were plated onto solid YNB media with various carbon sources and incubated at 30°C. Carbon sources were present at 2%. (A) Growth on glucose (2 days), ethanol, glycerol, or acetate (4 days). (B) Growth on glucose (2 days), oleate, linolenic acid, or olive oil (6 days). The strains used were the wild-type (SC5314), icl1Δ/Δ (MRC10), cat8Δ/Δ (MRC79), cat8Δ/Δ + CAT8 (MRC77), adr1Δ/Δ (MRC24), adr1Δ/Δ + ADR1 (MRC89), ctf1Δ/Δ (MRC41), ctf1Δ/Δ + CTF1 (MRC49), and fox2Δ/Δ (MRC6) strains. The ADR1- and CAT8-complemented strains were omitted from panel B for space reasons, but behaved identically to the parent mutants and the wild type (data not shown).

FIG. 3.

(A) ctf1Δ/Δ strains cannot metabolize Tween 20. Fivefold serial dilutions of the indicated strains were plated onto solid YNB media with the indicated carbon source at 2% and incubated at 30°C for 2 days (glucose) or 6 days (others). Strains used were as indicated in the legend for Fig. 2. (B) Chemical structure of the Tween compounds. (C) Predominant fatty acid side chains of the different Tween compounds.

FIG. 4.

CTF1 is induced by fatty acids and regulates the β-oxidation gene FOX2. (A) Northern analysis of CTF1 indicates that it is present only in cells using oleate as the carbon source. Cells were grown for 1 hour in YNB media with glucose (G), potassium acetate (A), or oleate (O) as the carbon source (see Materials and Methods). Ethidium bromide staining of the rRNA (lower panels) was used as a loading control. (B) FOX2 is not induced in ctf1Δ/Δ mutant strains. The strains were grown as described for panel A. (C) ADR1 and CAT8 have no apparent role in carbon source-dependent gene expression. Cells of the indicated genotypes were grown as described for panel A, and Northern blots were probed for FOX2, ICL1, or FBP1. Blots were stripped and reprobed for the 18S rRNA as a loading control. Strains used were as described in the legend for Fig. 2.

FIG. 5.

CTF1 regulates several β-oxidation and peroxisome biogenesis genes. Cells of the indicated genotype were grown in the presence of glucose (G), acetate (A), or oleate (O) for 1 hour. Northern blots were probed for the gene indicated on the right. Blots were stripped and reprobed for the 18S rRNA as a loading control.

FIG. 6.

ctf1Δ/Δ mutant strains localize peroxisomally targeted reporter proteins to peroxisome-like structures. Fusions were constructed in which the five C-terminal codons from C. albicans ICL1 or FOX2 or S. cerevisiae MLS1 (ScMLS1) were appended to the carboxy terminus of GFP under the control of the ACT1 promoter in pACT1-GFP, a CIp10-based vector, and integrated at the RPS10 locus in the wild-type (CAI4-F2), ctf1Δ/Δ (MRC40), fox2Δ/Δ (Can128), or pex5Δ/Δ (MRC113) strain. Cells were grown into logarithmic phase in YNB-glucose medium, collected by centrifugation, washed with water, and resuspended in YNB medium with 2% glucose or 2% oleate for 1 hour before examination by fluorescence microscopy.

FIG. 7.

Loss of CTF1 causes a mild attenuation of virulence. Outbred ICR mice were injected with 10⁶ cells of the wild type (SC5314), the ctf1Δ/Δ mutant (MRC79), or the ctf1Δ/Δ + CTF1 complemented (MRC77) strain via the tail vein. Animals were monitored for signs of infection (see Materials and Methods). Data were analyzed using the log-rank test (Prism5; GraphPad Software). The comparison of the wild-type and ctf1Δ/Δ mutant survival curves showed the difference to be statistically significant, with a P value of 0.013. The difference between the ctf1Δ/Δ mutant and complemented strain was not significant (P = 0.14).

FIG. 8.

A simplified comparison of the regulation of gene expression by carbon source in three model fungi. Transcriptional regulators from S. cerevisiae (green boxes), C. albicans (red ovals), and A. nidulans (blue diamonds) are shown with their effects on the expression of either glyoxylate cycle genes (left), such as ICL1/AcuD, or fatty acid catabolic genes (right), such as FOX2 or AcuJ (encoding a carnitine acetyltransferase). PIP2 (an OAF1 partner) and FarB (a FarA partner) were omitted for clarity. See text for discussion.