ScOpi1p is a well-characterized transcriptional repressor and master regulator of inositol and phospholipid biosynthetic genes in the baker’s yeast Saccharomyces cerevisiae. An ortholog has been shown to perform a similar function in the pathogenic fungus Candida glabrata, but with the distinction that CgOpi1p is essential for growth in this organism. However, in the more distantly related yeast Yarrowia lipolytica, the OPI1 homolog was not found to regulate inositol biosynthesis, but alkane oxidation. In Candida albicans, the most common cause of human candidiasis, its Opi1p homolog, CaOpi1p, has been shown to complement a S. cerevisiae opi1∆ mutant for inositol biosynthesis regulation when heterologously expressed, suggesting it might serve a similar role in this pathogen. This was tested in the pathogen directly in this report by disrupting the OPI1 homolog and examining its phenotypes. It was discovered that the OPI1 homolog does not regulate INO1 expression in C. albicans, but it does control SAP2 expression in response to bovine serum albumin containing media. Meanwhile, we found that CaOpi1 represses filamentous growth at lower temperatures (30°C) on agar, but not in liquid media. Although, the mutant does not affect virulence in a mouse model of systemic infection, it does affect virulence in a rat model of vaginitis. This may be because Opi1p regulates expression of the SAP2 protease, which is required for rat vaginal infections.
Citation: Chen Y-L, de Bernardis F, Yu S-J, Sandini S, Kauffman S, Tams RN, et al. (2015) Candida albicans OPI1 Regulates Filamentous Growth and Virulence in Vaginal Infections, but Not Inositol Biosynthesis. PLoS ONE 10(1): e0116974. https://doi.org/10.1371/journal.pone.0116974
Academic Editor: Julian R. Naglik, King’s College London Dental Institute, UNITED KINGDOM
Received: June 1, 2014; Accepted: December 17, 2014; Published: January 20, 2015
Copyright: © 2015 Chen et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was supported in part by grants NSC 102-2320-B-002-041-MY2 (YLC) and AHA 0765366B, NIH-1R03AI071863, NIH-1 R01AL105690 (TBR). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Candida albicans is a commensal organism that lives as a benign resident of the microflora of the human oral, gastrointestinal, and vaginal tracts as well as the skin. It can shift from a commensal to a pathogenic state in response to environmental stimuli that trigger developmental programs that induce the expression of virulence factors. Virulence factors exhibited by C. albicans include growth at 37°C, dimorphism, and production of secreted hydrolases such as proteases, lipases, and phospholipases [1, 2].
The pathways that regulate the transcription of secreted aspartyl protease (SAP) virulence factors in C. albicans are beginning to be understood, but much remains to be learned. SAPs are encoded by a family of 10 related genes (SAP1 to SAP10) [3, 4]. Unlike SAP1 to SAP8, which encode secreted proteases, SAP9 and SAP10 encode GPI-anchored proteases, located at the cell membrane or cell wall, and both are required for virulence . Among this family of genes, Sap2p is the most well-studied protease since it is the major secreted protease during in vitro growth conditions. SAP2 is expressed in in vitro conditions where bovine serum albumin is the main nitrogen source [3, 4], and its regulation in these conditions has been well characterized. SAP2 is under the control of the STP1 transcription factor and STP1’s upstream GATA transcription factors GLN3 and GAT1 [6, 7].
The importance of SAP2 in pathogenesis has been discussed by several groups. For instance, De Bernardis et al. demonstrated that SAP2 is a major virulence contributor in the rat vaginitis model [8, 9]. Schaller et al. showed that SAP2 is required to cause tissue damage in an in vitro model of vaginal candidiasis . In addition, Hube et al. demonstrated that SAP2 was required for virulence in a rodent model of systemic infection . In contrast, Naglik et al. and Lermann and Morschhäuser found that SAP2 was not required to invade and damage oral or vaginal reconstituted human epithelium [12, 13]. Meanwhile, the effect of the aspartic protease inhibitor pepstatin A on reducing tissue damage caused by C. albicans in the reconstituted human epithelium model remains elusive. Naglik et al. showed that pepstatin A can attenuate tissue damage, while Lermann and Morschhäuser demonstrated no effect, leaving the role for the Sap family in inducing epithelial damage controversial [12, 13]. Thus, there is contradictory evidence about the role of SAP2 amd other SAPs in pathogenesis.
S. cerevisiae OPI1 (ScOPI1) is a negative regulator of inositol biosynthesis, and acts to inhibit the transcription of ScINO1 along with other phospholipid biosynthetic genes in response to extracellular inositol levels [14–17]. ScINO1 encodes the inositol-3-phosphate synthase (ScIno1p) that catalyzes the conversion of glucose-6-phosphate to inositol-3-phosphate, which is then dephosphorylated by INM1 or INM2 to form inositol [17–19]. Inositol and cytidyldiphosphate-diacylglycerol (CDP-DAG) are precursors for the essential phospholipid phosphatidylinositol (PI). ScOpi1p acts as the master regulator of ScINO1 and other target genes by inhibiting the transcriptional activators ScIno2p and ScIno4p. The mechanism by which it regulates these genes in response to extracellular inositol has been well described [15–17, 20–22].
The structural gene encoding ScINO1 is conserved between S. cerevisiae and C. albicans, and shares similar function . C. albicans and S. cerevisiae INO1 homologs are similarly regulated in response to exogenously provided inositol . The ScOPI1 ortholog in C. albicans (OPI1) can complement Scopi1Δ for INO1 regulation in S. cerevisiae . We therefore hypothesized that OPI1 might function as an INO1 negative regulator in C. albicans as it does in Candida glabrata . However, a report regarding the ScIno2p and ScIno4p homologs suggested that the regulation of INO1 expression in S. cerevisiae and C. albicans might not be conserved . The C. albicans heterodimeric transcription factors INO2 and INO4 (related to ScINO2 and ScINO4 from S. cerevisiae) did not regulate INO1, but instead activated ribosomal protein genes such as RPL32. These results indicate that inositol regulation might be transcriptional rewired between these two related eukaryotic organisms. A previous report comparing S. cerevisiae and C. albicans Gal4p transcription factor homologs that control sugar metabolism suggests that these proteins have been rewired between these two organisms . In C. albicans the Gal4p homolog activates the gluconeogenesis gene LAT1 instead of galactose metabolism genes such as GAL10, and surprisingly GAL10 was activated by another transcription factor, CPH1. Therefore, we wished to investigate if C. albicans OPI1 has a similar role in inositol regulation to ScOPI1 and CgOPI1, or if it has possibly been transcriptional rewired.
In this communication, we report that C. albicans OPI1 does not regulate the inositol biosynthetic gene INO1, but affects the SAP2 expression and virulence of C. albicans in a rat vaginitis model. In addition, OPI1 affects morphogenesis at 30°C. These results illustrate that the regulation of inositol biosynthesis in C. albicans and S. cerevisiae is different. From now on, in this paper, all genes from C. albicans will be referred to by their simple names such as OPI1 or INO1, whereas genes from other organisms such as S. cerevisiae will be referred to as ScOPI1 or ScINO1.
Materials and Methods
Mouse model of systemic infection studies were conducted in the animal facility at University of Tennessee (UT) in good practice as defined by the United States Animal Welfare Act and in full compliance with the guidelines of the UT Institutional Animal Care and Use Committee (IACUC). The mouse experiments were reviewed and approved by the UT IACUC under protocol number L016. Procedures involving rats and their care were conducted in conformity with national and international laws and policies. The study has been approved by the Committees on the Ethics of Animal Experiments of the Istituto Superiore di Sanita’, Rome, Italy (Permit Number: DM 227/2009-B). All experimental procedures were carried out according to the ARRIVE (Animal Research: Reporting In Vivo Experiments; http://www.nc3rs.org.uk/page.asp?id=1357) and NIH (National Institutes of Health) guidelines for the ethical treatment of animals.
Strains and growth media
C. albicans strains used in this study are shown in Table 1. Media used in this study include YPD (yeast extract-peptone-dextrose: 1% yeast extract, 2% peptone, 2% glucose), defined medium 199 (M199, Invitrogen, pH7.0 adjusted by 150mM HEPES), Spider (1% nutrient broth, 1% mannitol, 0.2% dipotassium phosphate, 1.35% agar), YPD containing 10% fetal bovine serum, YCB-BSA (1.17% yeast carbon base-Difco, 0.2% bovine serum albumin-Sigma),YCB-BSA-YE (1.17% yeast carbon base, 0.2% bovine serum albumin, 0.1% yeast extract, pH 5.0) [28, 29]. Unless otherwise stated, agar plates were solidified with 2% agar (granulated, Fisher).
The C. albicans OPI1 gene (OPI1) was disrupted by using the CaNAT1-FLP cassette  (Table 2). For the OPI1 disruption construct, the 379 base pair (bp) 5’ non-coding region (NCR) of OPI1 was amplified with primers TRO522 and TRO526 (Table 3), and cloned as a KpnI-ApaI digested 228bp fragment into pJK863 5’ of the CaNAT1-FLP cassette (Fig. 1A). The 448 bp 3’ NCR of OPI1 was amplified with primers TRO524 and TRO525 which introduced SacII and SacI sites, and was cloned into pJK863 3’ of the CaNAT1-FLPcassette (Fig. 1A). This created the OPI1 knock out construct plasmid pYLC36 (Table 2, Fig. 1A), which was cut with KpnI and SacI to release the disruption construct (5’ NCR of OPI1-CaNAT1-FLP-3’ NCR of OPI1) which was transformed into the wild type SC5314 strain by electroporation . The disruption construct was used to sequentially disrupt both alleles of OPI1. The OPI1 reconstitution construct was made by amplifying a 1.7 Kb fragment containing the OPI1 ORF and 5’ NCR from SC5314 genomic DNA using primers (JCO12 and JCO14) that introduced KpnI and SalI sites. This fragment was ligated into the pRS316 vector along with another 1.7 Kb fragment containing the NAT1–3’NCR of OPI1 amplified from plasmid pYLC36 using primers JCO50 and TRO42 which introduced SalI and SacI sites. This resulted in the OPI1 reconstitution plasmid pYLC37 (Table 2, Fig. 1B). The 3.4 Kb KpnI-SacI fragment from pYLC37 was transformed into the opi1Δ/Δ mutant (YLC88) in order to create the reconstituted opi11Δ/Δ::OPI1 strain (YLC117). The SAP2 constitutive expression construct was made by cloning the dominant selectable marker NAT1 with primers JCO129 and JCO130 to NdeI-digested pAU34  and resulted in pYLC219. The SAP2 ORF was then cloned to XmaI-digested pYLC219 with primers JCO131 and JCO132, and resulted in pYLC221, which can constitutively express SAP2 under the control of the ACT1 promoter. To transform this constitutive construct into wild type and opi1Δ/Δ strains, a PpuMI-digested linear plasmid pYLC221 was integrated at the URA3 site of Candida genome, and resulted in OPI1/OPI1 URA3::PACT1-SAP2 (YLC223) and opi1Δ/Δ URA3::PACT1-SAP2 (YLC226).
(A) Structure of the opi1::CaNAT1-FLP disruption construct. Approximately 500 base pairs of non-coding DNA flanking the 5’ and 3’ ends of the OPI1 gene (5’-and 3’-NCR, respectively) were cloned onto either flank of the CaNAT1-FLP cassette. The thick dark arrows represent the FRT sites of the FLP recombinase. The ball-and-stick symbol represents the ACT1 terminator (ACT1t), and the thinner bent arrow represents the SAP2 promoter (PSAP2). (B) The OPI1-NAT1 construct used to reintegrate OPI1 into the opi1Δ/Δ mutant. (C) Southern blotting was used to confirm the opi1Δ/Δ disruptions. The genomic DNA of the wild type and mutants was cut by KpnI and SphI restriction enzymes. A PCR product containing the ~500 bp 3’ NCR of OPI1 (amplified with primers TRO524 and TRO525) was used as a probe (red line) for Southern blot confirmation. Lanes: 1, wild type; 2, opi1Δ::NAT1-FLP/OPI1 strain; 3, opi1Δ/OPI1 strain; 4, opi1Δ/Δ::NAT1-FLP strain; 5, opi1Δ/Δ strain; and 6, opi1Δ/Δ::OPI1-NAT1 strain.
Northern blot analysis
Northern blotting for SAP2 and INO1 expression was performed as described [33, 34] with the following exceptions. Strains grown in YCB-BSA or YCB-BSA-YE medium at 37°C for 12 hrs (for SAP2) and in liquid medium 199 (pH 7.0) at 37°C for 2 hrs (for INO1) were collected for total RNA extraction by the hot phenol method. The PCR product containing bps 17–571 of the SAP2 ORF (primers JCO35 and JCO36) and bps 76–581 of the INO1 ORF (primers TRO562 and TRO563) were used as probes. Expression was normalized against C. albicans ACT1 gene expression probed on the same membrane. The ACT1 probe was generated with the primers JCO48 and JCO49.
RT real-time PCR
Strains were cultured overnight in YPD at 37°C, washed twice with dH2O, then diluted to 0.2 O.D600/ml and incubated in liquid YCB-BSA medium (1.17% yeast carbon base, 0.2% BSA) for 12 hrs at 37°C with shaking at 200 rpm. The 50 ml cultures were pelleted at 3000rpm at 4°C and immediately frozen with liquid nitrogen to stop cellular processes. Total RNAs were extracted with a RiboPure Yeast RNA Purification Kit (Ambion), treated with TURBO DNA-free Kit (Invitrogen), and 2 μg of DNA-free total RNAs was reverse transcribed to cDNA using High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). Real-time PCR reactions of 20 μl included 6 ng cDNA (in 6 μl), 10 μl of 2x qPCR master mix (Fast SYBR Green Master Mix; Applied Biosystems), 2 μl of 2.5 μM forward primer (JC798 for SAP2), 2 μl of 2.5 μM reverse primer (JC799 for SAP2). Primer design for detecting SAPs expression was based on previous publication by Naglik et al . Quantitative PCR conditions were shown below 95°C /10 min for denaturation; 95°C /3 sec, 60°C /30 sec (40 cycles); 95°C /15 sec, 60°C /60 sec, 95 °C /15 sec (melting curve). The StepOnePlus System and StepOne v2.2 (Applied Biosystems) were used to determine ∆∆Ct. The bar graphs of ACT1 normalized relative quantity compared with wild-type (SC5314) were created with Prism 5.03.
Southern blot analysis
Hybridization conditions for the Southern blot analysis were similar to those for Northern blot analysis, except that the Techne Hybrigene oven was set to 60°C for the incubation step, and 42°C and 60°C for washing steps. The cells were grown in liquid YPD at 30°C overnight. The genomic DNA was extracted using the Winston-Hoffman method  and 20μg of genomic DNA were subjected to Southern blotting. The genomic DNA of the wild type and opi1Δ/Δmutants was cut by KpnI and SphI restriction enzymes. PCR products containing the ~500bp 3’ NCR of OPI1(primers TRO524 and TRO525) were used as probes for Southern blot confirmation (Fig. 1C).
Mouse bloodstream infection studies
Five- to six-week-old male CD1 mice (18 to 20 g) from Charles River Laboratories were used in this study. Mice were housed at five per cage. For infection, colonies from each C. albicans strain were inoculated into 20 ml of YPD. Cultures were grown overnight at 30°C with shaking in YPD, washed twice with 25 ml of sterile water, counted by hemocytometer, and resuspended at 107 cells per ml in sterile water. Mice were injected via the tail vein with 0.1 ml of the cell suspension (106 cells), and the course of infection was monitored for up to 14 days. The survival of mice was monitored twice daily, and moribund mice (body weight reduced by 30%, unable to eat/drink, or severely hunched) were euthanized with CO2. Cells were also plated on YPD to determine the viability. At least two independent infections were performed for each strain. The statistical analysis was done using Prism 5.03 software (GraphPad Software). For the mouse model of systemic infection, Kaplan-Meier survival curves were compared for significance using the Mantel-Haenszel log rank test. Statistical significance was set at P< 0.05.
Rat vaginitis studies
The protocol of estrogen-dependent rat vaginal infection model adapted from De Bernardis et al.  was used throughout this study. Briefly, oophorectomized female Wistar rats (80–100 g; Charles River, Calco, Italy) were injected subcutaneously with 0.5 mg of estradiol benzoate (Benzatrone; Samil, Rome). Six days after the first estradiol treatment, the animals were inoculated intravaginally with 107 yeast cells of each C. albicans strain in 0.1 mL. The inoculum was dispensed into the vaginal cavity through a syringe equipped with a multipurpose calibrated tip (Combitip; PBI, Milan, Italy). The yeast cells had been previously grown in YPD broth at 28°C on a gyratory shaker (200 rpm), harvested by centrifugation (1500 g), washed, counted in a hemocytometer, and suspended to the required number in saline solution. The results of two independent experiments are each represented separately. A third experiment involving all of the strains is not shown, but gave similar trends. In each experiment, each Candida strain was inoculated into 5 rats. Kinetics of C. albicans growth in, and clearance from, the vaginal cavity was measured by colony forming unit (CFU) enumeration after culturing 100 μl of vaginal samples, taken by washing the vaginal cavity by gentle aspiration of 100 μl of sterile saline solution, repeated four times, at 1:10 serial dilutions on Sabouraud agar containing chloramphenicol (50 μg/ml). CFUs were enumerated after incubation at 28°C for 48 h.
C. albicans OPI1 does not regulate INO1 expression
When heterologously expressed in an S. cerevisiae Scopi1∆ mutant, the C. albicans OPI1 gene has been demonstrated to repress expression of a reporter gene that contains the inositol/choline responsive element (ICRE) found in ScINO1 and other ScOpi1p-ScIno2p-ScIno4p target genes . This data suggested that C. albicans Opi1p may regulate the cognate C. albicans INO1 gene, as its homolog does in S. cerevisiae. In order to test this both copies of the C. albicans OPI1 gene were disrupted in C. albicans using the CaNAT1-FLP cassette  as described in Fig. 1.
The wild type and opi1Δ/Δ strains were then compared to see if the opi1∆/∆ mutant would fail to repress INO1, as expected, if it acts like the homologous S. cerevisiae mutant, Scopi1∆ . First, the strains were grown in Medium 199, pH 7.0, which contains low levels of inositol (~10 μM), which should result in high expression of INO1, and it was found that both upregulated INO1 to similar levels (Fig. 2). Then, they were grown in the same medium supplemented with 75 μM inositol, which should repress INO1 expression in wild-type, but not in the opi1∆/∆ mutant, if it cannot repress the gene. However, in both strains, INO1 was similarly repressed, suggesting that inositol biosynthesis is regulated by different transcription factors in C. albicans.
The opi1Δ/Δ mutant exhibits hyperfilamentous growth in filament-inducing media at 30°C
It has been shown that ScOPI1 is necessary to activate invasive growth and ScFLO11 expression in S. cerevisiae . It was therefore hypothesized that OPI1 would affect filamentous growth in C. albicans. Three filament-inducing media were used to test this hypothesis. In contrast to the situation with the Scopi1∆ mutant in S. cerevisiae, it was found that the opi1Δ/Δ mutant exhibited hyperfilamentous growth rather than hypofilamentous growth, but only at 30°C on solid filament-inducing agar plates (Fig. 3). This effect was not observed at 37°C on similar media. These phenotypes were also not seen in liquid forms of the same filament-inducing media at either 30°C or 37°C. In order to control for a possible effect from some other unlinked mutation, a copy of the C. albicans OPI1 gene was reintegrated into the opi1∆/∆ mutant (Fig. 1B), and it was found that the phenotype was restored when the wild-type copy of OPI1 was present (Fig. 3), indicating that the hyperfilamentous growth at 30°C is linked to the loss of OPI1 gene.
OPI1 does not affect virulence in a mouse model of systemic infection
The opi1∆/∆ mutant appears to affect the ability of the fungus to repress filamentation at lower temperatures. Some hyperfilamentous mutants such as nrg1∆/∆ and tup1∆/∆ have been found to be attenuated in virulence in mouse models of infection . Therefore, a mouse model of systemic infection was used to test the role of OPI1 in virulence. However, the OPI1 gene does not contribute to the virulence in this model since the opi1Δ/Δ mutant exhibits a similar phenotype to wild-type on the survival curves of mice (Fig. 4).
Each strain was used to infect mice by injecting 106 C. albicans yeast-form cells into the tail-vein of each mouse. The mice were then assessed over the course of 14 days. The number of mice used for a specific strain is indicated in parentheses. The data obtained here are from a single experiment.
OPI1 is involved in establishing infection in the rat vaginitis model
In addition to infections of the bloodstream, C. albicans can also cause infections of mucosal surfaces including the vaginal tract . A rat vaginitis model was used to determine if the opi1∆/∆ mutation would play a role in the establishment of infection in this host niche. It was demonstrated that OPI1 was involved in establishing rat vaginitis. In this model C. albicans cells are injected into the rat vaginal tract, and then over time the level of colonization is measured based on the recovery of colony counts. It was discovered that the opi1Δ/Δ mutant is quickly cleared by the host compared to the wild type (Fig. 5). The opi1∆/∆::OPI1 reintegrant strain and opi1∆/OPI1 heterozygous mutant had an intermediate phenotype between the opi1Δ/Δ mutant and wild type (Fig. 5).
For each C. albicans strain, 5 rats were inoculated on day 0 with 107 blastospores, and vaginal colony forming units (CFUs) at particular time points were counted by plating at the indicated time points. The error bars represented the standard errors of the mean in each group. The number of rats used for a specific strain is indicated in parentheses and the data obtained here are from a single experiment.
It has been shown that deletion of the C. albicans SAP2 proteasegene  causes a similar clearance to the opi1∆/∆ mutant, and a sap2∆/∆ mutant (SAP2MS4A)  was included in this experiment as a control. Our results confirmed that an independently constructed sap2∆/∆ mutant (gift from Joachim Morschhäuser), behaved like a previously constructed sap2∆/∆ mutant, and exhibits reduced colonization in the rat vaginal tract (Fig. 5), suggesting the importance of SAP2 in the rat vaginitis model. Our results also indicate that OPI1 plays a critical role in establishing infection in the rat vaginal tract (Fig. 5).
OPI1 affects rat vaginal establishment through regulating SAP2
The similarity of the phenotypes of the opi1∆/∆ mutant with the sap2∆/∆ mutant suggested that OPI1 might act through SAP2. In wild-type cells, SAP2 is upregulated in bovine serum albumin (BSA) media. We performed reverse transcriptase (RT) real-time PCR to detect if OPI1 controls SAP2 expression in YCB-BSA medium. The opi1∆/∆ mutant showed 5.5 fold reduced SAP2 expression compared to wild type (Fig. 6), indicating that OPI1 controls SAP2 expression. The opi1∆/∆::OPI1 reintegrant strain can restore the SAP2 expression and actually shows ~ 3 fold higher expression of SAP2 than the wild type.
RT real-time PCR was used to assess the SAP2 expression levels in the wild-type (SC5314), opi1Δ/Δ, opi1Δ/Δ::OPI1 and sap2 Δ/Δ mutants. Strains were cultured overnight in YPD at 37°C, washed twice with dH2O. Then strains were diluted to 0.2 O.D600/ml and incubated in liquid YCB-BSA medium (1.17% yeast carbon base, 0.2% BSA) for 12 hrs at 37°C. The error bars represented the standard errors of the mean. The data obtained here are from a representative single experiment with technical triplicates. P value was determined by t tests and < 0.05 was considered statistically significance.
In order to test if OPI1 affects colonization of the rat vaginal tract through SAP2, an epistasis experiment was performed in which the SAP2 gene was overexpressed in the opi1∆/∆ mutant via the ACT1 promoter (PACT1-SAP2). This overexpression was confirmed by Northern blotting (S1 Fig.). If opi1∆/∆ blocked rat colonization by compromising SAP2 expression, then overexpression of SAP2 from an independent promoter should suppress the phenotype. In contrast to the opi1∆/∆ mutant, the opi1Δ/ΔURA3::PACT1-SAP2 mutant was suppressed for its defect in rat vaginal infection, and behaved similarly to the wild type (Fig. 7). This implicates the OPI1 gene as a regulator of SAP2 in the vaginal tract of the rat.
For each strain 5 rats were inoculated on day 0 with 107 blastospores, and vaginal CFUs were counted by plating at the indicated time points. The error bars represented the standard errors of the mean in each group. The number of rats used for a specific strain is indicated in parentheses and the data obtained here are from a single experiment.
Our results show that the OPI1 gene of C. albicans, unlike its homologs in S. cerevisiae and C. glabrata [22, 25], does not affect INO1 expression, but does repress filamentous growth at low temperature (Fig. 3) and regulates virulence in the rat vaginitis model (Figs. 5 and 7). The latter phenotype appears to be mediated by changes in SAP2 expression. The opi1Δ/Δ mutant exhibits reduced SAP2 expression compared with the wild type in liquid YCB-BSA medium (Fig. 6), and in vivo SAP2 overexpression can restore the opi1Δ/Δ mutant’s vaginal colonization defect, when under the control of the constitutive ACT1 promoter. Epistasis experiments are inherently challenging to interpret, as overexpression of a target gene such as SAP2 could lead to enhanced colonization by a mechanism that bypasses the actual defect caused by the opi1∆/∆ mutation. Based on our data, this possibility cannot be completely ruled out. It is also possible that OPI1 controls colonization in the rat vaginal tract by regulating one of the other SAPs (e.g. SAP1, SAP3–10). However, as the differential expression of nine other SAPs in the opi1∆/∆ mutant compared to the wild type using the same condition (i.e. YCB-BSA liquid medium and RT real time PCR) was not detected, we do not know if these others are affected in vivo, and this remains to be examined (unpublished data). Meanwhile, further studies will be needed to test if SAP4, SAP5 and SAP6 genes are regulated by OPI1 under hypha-inducing conditions since SAP4–6 are hypha-specific genes [38, 39].
In S. cerevisiae, ScOpi1p is the master regulator of ScINO1 and other phospholipid genes [15–17, 20]. ScOpi1p controls expression in response to cellular inositol levels by binding to Ino2p in the Ino2p-Ino4p heterodimer and repressing its activation of ScINO1, among other targets. When inositol is plentiful, PI is efficiently synthesized from CDP-DAG and inositol by the ScPis1p enzyme [40, 41]. In this circumstance, the endoplasmic reticulum (ER) localized pool of phosphatidic acid (PA), which is the precursor for CDP-DAG, is consumed, and ScOpi1p is translocated to the nucleus. There, it binds ScIno2p and represses ScINO1 with help from the global repressor Sin3p via a direct interaction involving the N-terminal Sin3p binding domain of ScOpi1p [16, 24]. When inositol is not plentiful in the environment, cellular stores drop and PI synthesis slows causing a build-up of precursors including PA. ScOpi1p binds to PA in the ER via its basic domain, and the ER membrane protein Scs2p via its FFAT domain [20, 42]. This sequesters ScOpi1p to the ER, and then Ino2p-Ino4p activate transcription of ScINO1 so inositol can be synthesized for PI production.
A previous report demonstrated that OPI1 from C. albicans could complement an Scopi1∆ mutant in S. cerevisiae, and it could repress the ICRE promoter element found on ScINO1 and other phospholipid biosynthetic genes when expressed heterologously in S. cerevisiae . However, we found that in C. albicans, OPI1 does not regulate INO1 expression. This overlap in function of CaOPI1 when expressed heterologously in S. cerevisiae, but not endogenously in C. albicans, may be due to the conservation of some key domains required for ScOpi1p function, but not the conservation of other domains (S2 Fig.). In particular, the C. albicans Opi1p has very little conservation with ScOpi1p in the large N-terminal ScSin3p binding domain . However, CaOpi1p does have some conserved sequences with the C-terminal ScIno2p interaction domain of ScOpi1p, including two out of three residues (ScOpi1p aas 358–360) that were shown to be crucial for ScIno2p-ScOpi1p interactions in S. cerevisiae . In contrast, CaOpi1p shares very few residues in common with ScOpi1p in the PA-binding basic domain, and no residues of the FFAT domain that binds to ScScs2p [20, 42]. It does, however, carry a leucine zipper motif with some isoleucine substitutions that has been shown to be crucial for ScIno2p-ScOpi1p interactions . Thus, this conservation of some domains, but not others may help explain why CaOpi1p can complement a Scopi1∆ mutant for ScINO1 repression , but not act the same within C. albicans itself. Further support for our findings comes from the observation that CaINO2 and CaINO4 do not appear to regulate CaINO1 either, but may actually regulate ribosomal genes . This is in marked contrast to CgOPI1 from C. glabrata, which does regulate CgINO1 with help from CgINO2 and CgINO4 . Consistently, CgOpi1p has close conservation of all of the important regulatory domains of ScOpi1p (S2 Fig.). Interestingly, one other ScOpi1p homolog has been characterized, and this is Yas3p from Yarrowia lipolytica. Yas3p also does not have a number of domains conserved with ScOpi1p, and like CaOpi1p does not regulate YlINO1, but does, along with Ino2p and Ino4p homologs Yas1p and Yas2p, respectively, regulate hexane metabolism genes . The C. albicans regulators of CaINO1 are currently unknown, and this will be interesting to elucidate, as expression of CaINO1 is regulated by extracellular inositol levels, but not by CaOpi1p and apparently not by CaIno2p or CaIno4p either.
Finally, the role Opi1p in repressing filamentation at 30°C on solid media remains elusive (Fig. 3). The opi1Δ/Δ mutant exhibits hyperfilamentous growth in filament-inducing agar plates including medium 199, spider, and 10% serum at 30°C, but not 37°C. These results indicate that OPI1 might be a low temperature repressor of filamentous growth. It has been demonstrated that C. albicans CPP1, a tyrosine phosphatase, is required tor repress the yeast to hyphal transition at 23°C in contact with solid surfaces [44, 45]. The cpp1Δ/Δ mutant exhibited hyperfilamentous growth on spider and a wide variety of rich and defined solid media including Lee’s medium, YPD, YPM, and 10% serum at 23°C, but not at 37°C. The germ tube formation defect of the cpp1Δ/Δ mutant was not observed at liquid culture at 37°C, an effect similar to opi1Δ/Δ mutant. In contrast to opi1Δ/Δ, the cpp1Δ/Δ mutant exhibited reduced virulence in mouse systemic infection and mouse mastitis models [44–46]. The relationship between Opi1 and Cpp1 is unknown and needs further studies in C. albicans. Taken together, our data suggest that, when compared to its homolog in S. cerevisiae, C. albicans has a transcriptionally rewired regulator, OPI1, which does not regulate INO1 expression but affects morphogenesis, SAP2 expression and virulence in a rat vaginitis model. It also makes it clear that identification of ScOpi1p homologs in other fungi does not clearly implicate them for roles in regulating inositol biosynthesis in these microbes. Rather, the Opi1p family members, which are conserved in a wide variety of fungi appear to have a diversity of functions.
S1 Fig. SAP2 is overexpressed in the PACT1-SAP2 construct.
Expression was tested by Northern blotting in YPD media, and it was confirmed that the PACT1-SAP2 construct overexpressed SAP2.
S2 Fig. Multiple sequence alignment comparing Opi1p homologues of S. cerevisiae (S.C.), C. glabrata (C.G.), Y. lipolytica (Y.L.), and C. albicans (C.A.).
This alignment was performed using Clustal W. 184.108.40.206. Asterisk represents conservation among all four species. Various domains represented by different colors, and boxes highlight particularly conserved regions between species. Blue: Opi1-Sin3 interaction domain. Gold: phosphatidic acid (PA)-binding domain. Red: Leucine zipper. Green: FFAT (2 phenylalanines and an acid tract). Purple: Polyglutamine tract. Orange with black boxes: Ino2p activator interaction domain.
We are grateful to Julia Koëhler, Alexander Johnson, and Joachim Morschhäuser for plasmids and strains. We also thank Antonio Cassone and Jeffrey M. Becker for their advice, comments, and support in this project.
Conceived and designed the experiments: YLC TBR. Performed the experiments: YLC FB SJY SS SK RT EB. Analyzed the data: YLC FB SJY. Wrote the paper: YLC FB SJY TBR.
- 1. Heitman J, Filler SG, Edwards JE, Mitchell AP (2006) Molecular principles of fungal pathogenesis. Washington DC: ASM Press.
- 2. Calderone R, Clancy C (2012) Candida and Candidiasis. Washington DC: ASM Press.
- 3. Naglik JR, Challacombe SJ, Hube B (2003) Candida albicans secreted aspartyl proteinases in virulence and pathogenesis. Microbiol Mol Biol Rev 67: 400–428, table of contents. . pmid:12966142
- 4. De Bernardis F, Sullivan PA, Cassone A (2001) Aspartyl proteinases of Candida albicans and their role in pathogenicity. Med Mycol 39: 303–313. pmid:11556759
- 5. Albrecht A, Felk A, Pichova I, Naglik JR, Schaller M, et al. (2006) Glycosylphosphatidylinositol-anchored proteases of Candida albicans target proteins necessary for both cellular processes and host-pathogen interactions. J Biol Chem 281: 688–694. pmid:16269404
- 6. Dabas N, Morschhauser J (2008) A transcription factor regulatory cascade controls secreted aspartic protease expression in Candida albicans. Mol Microbiol 69: 586–602. pmid:18547391
- 7. Martinez P, Ljungdahl PO (2005) Divergence of Stp1 and Stp2 transcription factors in Candida albicans places virulence factors required for proper nutrient acquisition under amino acid control. Mol Cell Biol 25: 9435–9446. pmid:16227594
- 8. De Bernardis F, Arancia S, Morelli L, Hube B, Sanglard D, et al. (1999) Evidence that members of the secretory aspartyl proteinase gene family, in particular SAP2, are virulence factors for Candida vaginitis. J Infect Dis 179: 201–208. pmid:9841840
- 9. De Bernardis F, Cassone A, Sturtevant J, Calderone R (1995) Expression of Candida albicans SAP1 and SAP2 in experimental vaginitis. Infect Immun 63: 1887–1892. pmid:7729898
- 10. Schaller M, Bein M, Korting HC, Baur S, Hamm G, et al. (2003) The secreted aspartyl proteinases Sap1 and Sap2 cause tissue damage in an in vitro model of vaginal candidiasis based on reconstituted human vaginal epithelium. Infect Immun 71: 3227–3234. pmid:12761103
- 11. Hube B, Sanglard D, Odds FC, Hess D, Monod M, et al. (1997) Disruption of each of the secreted aspartyl proteinase genes SAP1, SAP2, and SAP3 of Candida albicans attenuates virulence. Infect Immun 65: 3529–3538. pmid:9284116
- 12. Lermann U, Morschhauser J (2008) Secreted aspartic proteases are not required for invasion of reconstituted human epithelia by Candida albicans. Microbiology 154: 3281–3295. pmid:18957582
- 13. Naglik JR, Moyes D, Makwana J, Kanzaria P, Tsichlaki E, et al. (2008) Quantitative expression of the Candida albicans secreted aspartyl proteinase gene family in human oral and vaginal candidiasis. Microbiology 154: 3266–3280. pmid:18957581
- 14. White MJ, Hirsch JP, Henry SA (1991) The OPI1 gene of Saccharomyces cerevisiae, a negative regulator of phospholipid biosynthesis, encodes a protein containing polyglutamine tracts and a leucine zipper. J Biol Chem 266: 863–872. pmid:1985968
- 15. Wagner C, Blank M, Strohmann B, Schuller HJ (1999) Overproduction of the Opi1 repressor inhibits transcriptional activation of structural genes required for phospholipid biosynthesis in the yeast Saccharomyces cerevisiae. Yeast 15: 843–854. pmid:10407264
- 16. Wagner C, Dietz M, Wittmann J, Albrecht A, Schuller HJ (2001) The negative regulator Opi1 of phospholipid biosynthesis in yeast contacts the pleiotropic repressor Sin3 and the transcriptional activator Ino2. Mol Microbiol 41: 155–166. pmid:11454208
- 17. Henry SA, Kohlwein SD, Carman GM (2012) Metabolism and regulation of glycerolipids in the yeast Saccharomyces cerevisiae. Genetics 190: 317–349. pmid:22345606
- 18. Donahue TF, Henry SA (1981) myo-Inositol-1-phosphate synthase. Characteristics of the enzyme and identification of its structural gene in yeast. J Biol Chem 256: 7077–7085. pmid:7016881
- 19. Murray M, Greenberg ML (2000) Expression of yeast INM1 encoding inositol monophosphatase is regulated by inositol, carbon source and growth stage and is decreased by lithium and valproate. Mol Microbiol 36: 651–661. pmid:10844654
- 20. Loewen CJ, Gaspar ML, Jesch SA, Delon C, Ktistakis NT, et al. (2004) Phospholipid metabolism regulated by a transcription factor sensing phosphatidic acid. Science 304: 1644–1647. pmid:15192221
- 21. Klig LS, Hoshizaki DK, Henry SA (1988) Isolation of the yeast INO4 gene, a positive regulator of phospholipid biosynthesis. Curr Genet 13: 7–14. pmid:2834106
- 22. Klig LS, Homann MJ, Carman GM, Henry SA (1985) Coordinate regulation of phospholipid biosynthesis in Saccharomyces cerevisiae: pleiotropically constitutive opi1 mutant. J Bacteriol 162: 1135–1141. pmid:3888957
- 23. Chen YL, Kauffman S, Reynolds TB (2008) Candida albicans uses multiple mechanisms to acquire the essential metabolite inositol during infection. Infect Immun 76: 2793–2801. pmid:18268031
- 24. Heyken WT, Wagner C, Wittmann J, Albrecht A, Schuller HJ (2003) Negative regulation of phospholipid biosynthesis in Saccharomyces cerevisiae by a Candida albicans orthologue of OPI1. Yeast 20: 1177–1188. pmid:14587102
- 25. Bethea EK, Carver BJ, Montedonico AE, Reynolds TB (2010) The inositol regulon controls viability in Candida glabrata. Microbiology 156: 452–462. pmid:19875437
- 26. Hoppen J, Dietz M, Warsow G, Rohde R, Schuller HJ (2007) Ribosomal protein genes in the yeast Candida albicans may be activated by a heterodimeric transcription factor related to Ino2 and Ino4 from S. cerevisiae. Mol Genet Genomics 278: 317–330. pmid:17588177
- 27. Martchenko M, Levitin A, Whiteway M (2007) Transcriptional activation domains of the Candida albicans Gcn4p and Gal4p homologs. Eukaryot Cell 6: 291–301. pmid:17158732
- 28. Chen YL, Montedonico AE, Kauffman S, Dunlap JR, Menn FM, et al. (2010) Phosphatidylserine synthase and phosphatidylserine decarboxylase are essential for cell wall integrity and virulence in Candida albicans. Mol Microbiol 75: 1112–1132. pmid:20132453
- 29. Styles C (2002) How to set up a yeast laboratory. Methods Enzymol 350: 42–71. pmid:12073328
- 30. Shen J, Guo W, Kohler JR (2005) CaNAT1, a heterologous dominant selectable marker for transformation of Candida albicans and other pathogenic Candida species. Infect Immun 73: 1239–1242. pmid:15664973
- 31. De Backer MD, Maes D, Vandoninck S, Logghe M, Contreras R, et al. (1999) Transformation of Candida albicans by electroporation. Yeast 15: 1609–1618. pmid:10572258
- 32. Uhl MA, Johnson AD (2001) Development of Streptococcus thermophilus lacZ as a reporter gene for Candida albicans. Microbiology 147: 1189–1195. pmid:11320122
- 33. Kohrer K, Domdey H (1991) Preparation of high molecular weight RNA. Methods Enzymol 194: 398–405. pmid:1706459
- 34. Reynolds TB (2006) The Opi1p transcription factor affects expression of FLO11, mat formation, and invasive growth in Saccharomyces cerevisiae. Eukaryot Cell 5: 1266–1275. pmid:16896211
- 35. Hoffman CS, Winston F (1987) A ten-minute DNA preparation from yeast efficiently releases autonomous plasmids for transformation of Escherichia coli. Gene 57: 267–272. pmid:3319781
- 36. Murad AM, Leng P, Straffon M, Wishart J, Macaskill S, et al. (2001) NRG1 represses yeast-hypha morphogenesis and hypha-specific gene expression in Candida albicans. EMBO J 20: 4742–4752. pmid:11532938
- 37. Staib P, Lermann U, Blass-Warmuth J, Degel B, Wurzner R, et al. (2008) Tetracycline-inducible expression of individual secreted aspartic proteases in Candida albicans allows isoenzyme-specific inhibitor screening. Antimicrob Agents Chemother 52: 146–156. pmid:17954688
- 38. Chen YC, Wu CC, Chung WL, Lee FJ (2002) Differential secretion of Sap4–6 proteins in Candida albicans during hyphae formation. Microbiology 148: 3743–3754. pmid:12427964
- 39. Felk A, Kretschmar M, Albrecht A, Schaller M, Beinhauer S, et al. (2002) Candida albicans hyphal formation and the expression of the Efg1-regulated proteinases Sap4 to Sap6 are required for the invasion of parenchymal organs. Infect Immun 70: 3689–3700. pmid:12065511
- 40. Chen M, Hancock LC, Lopes JM (2007) Transcriptional regulation of yeast phospholipid biosynthetic genes. Biochim Biophys Acta 1771: 310–321. pmid:16854618
- 41. Fischl AS, Carman GM (1983) Phosphatidylinositol biosynthesis in Saccharomyces cerevisiae: purification and properties of microsome-associated phosphatidylinositol synthase. J Bacteriol 154: 304–311. pmid:6300035
- 42. Loewen CJ, Roy A, Levine TP (2003) A conserved ER targeting motif in three families of lipid binding proteins and in Opi1p binds VAP. EMBO J 22: 2025–2035. pmid:12727870
- 43. Hirakawa K, Kobayashi S, Inoue T, Endoh-Yamagami S, Fukuda R, et al. (2009) Yas3p, an Opi1 family transcription factor, regulates cytochrome P450 expression in response to n-alkanes in Yarrowia lipolytica. J Biol Chem 284: 7126–7137. pmid:19131334
- 44. Schroppel K, Sprosser K, Whiteway M, Thomas DY, Rollinghoff M, et al. (2000) Repression of hyphal proteinase expression by the mitogen-activated protein (MAP) kinase phosphatase Cpp1p of Candida albicans is independent of the MAP kinase Cek1p. Infect Immun 68: 7159–7161. pmid:11083847
- 45. Csank C, Makris C, Meloche S, Schroppel K, Rollinghoff M, et al. (1997) Derepressed hyphal growth and reduced virulence in a VH1 family-related protein phosphatase mutant of the human pathogen Candida albicans. Mol Biol Cell 8: 2539–2551. pmid:9398674
- 46. Guhad FA, Csank C, Jensen HE, Thomas DY, Whiteway M, et al. (1998) Reduced pathogenicity of a Candida albicans MAP kinase phosphatase (CPP1) mutant in the murine mastitis model. APMIS 106: 1049–1055. pmid:9890266
- 47. Shepherd MG (1985) Pathogenicity of morphological and auxotrophic mutants of Candida albicans in experimental infections. Infect Immun 50: 541–544. pmid:3902649