Identification of the Putative Protein Phosphatase Gene PTC1 as a Virulence-Related Gene Using a Silkworm Model of Candida albicans Infection

Strains and growth conditions.The C. albicans strains used or developed during cloning and disruption of the C. albicans PTC1 gene are listed in Table 1. Disruption of 20 other putative phosphatase genes of C. albicans (strains not listed) was attempted using the methods described below for C. albicans PTC1. Escherichia coli XL-1 Blue and cloning vector pUC19 were used for DNA manipulation. SD-URA medium (synthetic dextrose medium without uracil) containing 0.67% yeast nitrogen base (YNB) without amino acids (Difco, Detroit, MI), 2% dextrose, and 0.072% CSM-URA (complete synthetic mixture without uracil) (QBiogene, Irvine, CA) was used for C. albicans transformation. For examination of C. albicans morphology, yeast or hyphal cells were produced by manipulation of the medium composition and temperature as follows. Yeast cells were produced in liquid medium by growth in 1% yeast extract-2% peptone-2% dextrose (YPD, pH 5.6; QBiogene) or SD-AU (0.67% YNB without amino acids-2% dextrose-0.072% CSM-ARG-URA [complete synthetic mixture without arginine and uracil] [QBiogene]) at 30°C with shaking. Yeast growth was assessed by measuring the optical densities at 600 nm (OD₆₀₀) of cultures. Hyphal cells were produced in liquid medium from a yeast cell inoculum by growth in YPD (pH 7.2) plus 10% filtered, heat-inactivated bovine serum (GIBCO Invitrogen [Carlsbad, CA] cell culture) at 37°C with shaking. Hyphal growth was assessed by measuring the length of hyphae for 100 cells in three separate experiments. Filamentous growth on solid medium was obtained by inoculating agar plates (2%) containing 10% filtered, heat-inactivated bovine serum and incubating at 37°C for 7 days. Microscopic observation was performed by using a conventional fluorescence microscope (IX81; Olympus, Tokyo, Japan) equipped with a DP70 digital camera (Olympus).

Disruption of PTC1 in C. albicans and plasmid construction for the generation of revertant strains expressing either PTC1 allele.The primers used for PTC1 are listed in Table S1 in the supplemental material. The primers used for the attempted disruption of 25 other putative phosphatase genes are listed in Table S2 in the supplemental material. The nucleotide sequences of cloned fragments were confirmed using an Applied Biosystems (Foster City, CA) genetic analyzer, model 3130. C. albicans was transformed as described by Umeyama et al. (53). C. albicans PTC1 was disrupted in the ura⁻arg⁻ strain TUA4 by using a strategy described previously (21). Briefly, two auxotrophic markers, URA3 and ARG4, were used to disrupt both alleles of the PTC1 gene, producing the null mutant strain PTC103 (see Fig. S1 in the supplemental material). Primers ΔcheckPTC1-3′ and Δcheck-5′ were designed for PCRs to confirm that the expected constructs were obtained (see Fig. S1 in the supplemental material). At least three independently constructed null mutants were produced for each gene, and the consistency of phenotypes (growth rate, colony morphology and yeast-to-hypha transition properties) was confirmed before one of each construct was selected for further analyses. To produce revertant C. albicans strains containing either allele (A or B) of PTC1, plasmids p3HA-PTC1A and p3HA-PTC1B were constructed as follows. DNA fragments containing either allele A or allele B of PTC1 were PCR amplified from TUA4 genomic DNA using two primers, PTC1-N and PTC1-C. Following digestion with BamHI and SphI, the fragments were cloned into the BamHI and SphI sites of plasmid p3HA-ACT1 (53) to generate p3HA-PTC1A or p3HA-PTC1B. Then PTC1 alleles derived either from StuI-digested p3HA-PTC1A or from StuI-digested p3HA-PTC1B were reintegrated at the high-expression RP10 locus (40) of the null mutant PTC103, generating strain PTC1A or PTC1B. The StuI-digested empty vector p3HA-ACT1 was integrated into strain PTC103 as a null mutant control strain (PTC1C). The reason for choosing the integrative plasmid p3HA-ACT1 containing the ACT1 promoter, rather than a homologous promoter, was that it provided consistent expression of each of the alleles introduced, allowing direct comparison of the phenotypes of the recombinant strains.

Virulence studies.Animal experiments were performed by following the provisions of the Principles of Morality for animal experiments of the National Institute of Infectious Diseases, Japan. For the mouse model of disseminated C. albicans infection, groups of six CD-1 (ICR) mice were inoculated intravenously with 1 × 10⁶ CFU of the indicated C. albicans strain (53). Silkworm infection experiments were performed as described previously (20) with slight modifications. Briefly, larvae after the fourth molt, purchased from Ehime Sansyu (Ehime, Japan), were fed with Silkmate (Nihon Nosan Kogyo Corporation, Yokohama, Japan) at 27°C until they developed to fifth-instar larvae. On the second day of the fifth-instar stage, silkworms were anesthetized by contact with ice-cold water prior to inoculation. Late-log-phase C. albicans yeast cells grown in YPD were washed and resuspended in phosphate-buffered saline. Portions (0.05 ml) of the cell suspensions containing 1 × 10⁶ CFU were injected into the hemolymph through the dorsal surface (abdominal segment) of 10 silkworms for each C. albicans strain by using a 30-gauge needle. Silkworms were not fed after the injections, and mortality was checked at 24-h intervals. Silkworms were kept at 27°C throughout the experiments. The experiments were repeated twice more (total, 30 silkworms). To determine the morphology of C. albicans cells of either wild-type, null mutant, or revertant strains during the progress of silkworm infections, transverse sections of infected silkworms (12 h postinoculation) were examined by microscopy. Silkworms (n = 5) were fixed in 10% phosphate-buffered formaldehyde before embedding, transverse sectioning, and staining with periodic acid-Schiff stain.

RNA isolation and real-time PCR.The cells were collected by centrifugation, washed twice with ice-cold water, frozen with liquid nitrogen, and stored at −80°C until use. Total RNA was extracted from the cells resuspended in TES buffer (10 mM Tris-HCl [pH 7.5], 10 mM EDTA, 0.5% sodium dodecyl sulfate) with acid phenol (Sigma-Aldrich, St. Louis, MO) at 65°C for 45 min. The aqueous-phase solution was purified with acid phenol and then chloroform and was precipitated with ethanol. cDNA was synthesized with SuperScript III reverse transcriptase in a SuperScript III Platinum two-step qRT-PCR kit with Sybr green (Invitrogen, Carlsbad, CA). mRNA expression levels were examined by quantitative real-time PCR (QRT-PCR) using ABI Prism 7000 (Applied Biosystems) and Sybr Premix ExTaq (Takara, Otsu, Japan) and were standardized with ACT1 mRNA. All primers used for QRT-PCR are listed in Table S1 in the supplemental material.

Measurement of secreted hydrolytic enzyme activities of C. albicans. C. albicans secreted hydrolytic enzymes were induced in modified medium as previously described by Bramono et al. (7) and Tsuboi et al. (52). Proteinases were induced on a 2% agar plate containing 1.2% yeast carbon base (YCB; Difco) and 0.5% bovine serum albumin at 30°C for 3 days. Lipases were induced on a 2% agar plate containing 0.67% YNB without amino acids (Difco) and 2.5% Tween 80 at 30°C for 3 days. Hydrolytic enzyme activities were measured as the ratio of the diameter of the colony to the total diameter of the colony plus the zone of precipitation according to the method of Price et al. (45).

Susceptibilities of C. albicans PTC1 mutants to antifungals and chemical and environmental stressors.The antifungals used in this study were terbinafine (Wako Pure Chemical Industries, Osaka, Japan), micafungin (Astellas Pharma Inc., Osaka, Japan), fluconazole (Pfizer), and amphotericin B (Sigma-Aldrich). Chemical stressors included incubation of cultures with the indicated concentrations of NaCl, H₂O₂, and Congo red. Environmental stressors included growth at a high temperature (42°C) and in a medium with a low or a high pH (pH 2.5 or 9.0, respectively). The effects of antifungal drugs or stressors on the yeast cell growth of each C. albicans strain were determined using an agar plate dilution assay (43) and a broth microdilution assay adapted from the standard CLSI (formally NCCLS) document M27-A2 (42), in which the medium was inoculated with C. albicans yeast cells from a 16-h culture (YPD; pH 5.6; 30°C). For the agar plate dilution assay, the inoculum cells were washed and resuspended in sterile distilled water to an OD₆₀₀ of 0.1 before the suspension was serially diluted 10 times. Five microliters of cell suspensions (containing 10⁴, 10³, 10², and 10 CFU, respectively) was spotted onto YPD agar plates containing antifungals or stressors at the indicated concentrations or under the indicated conditions. Cell growth was monitored after incubation at 30°C for 48 h or at the indicated temperature for the indicated period. For the broth microdilution assay, 4 × 10³ washed yeast cells were resuspended in YPD (100 μl) and inoculated into YPD (100 μl) containing an antifungal drug or chemical stressor at the concentrations indicated in flat-bottom microtiter plates. Plates were incubated at 30°C for 48 h, and the OD₅₉₅ of the cells was measured using a Beckman Coulter (Fullerton, CA) DTX880 multimode detector.

Statistics.The log rank test was used to analyze survival rates for silkworm and mouse model infections, and the t test was used to analyze fungal burdens in the kidneys of infected mice as well as proteinase activities among wild-type, revertant, and null mutant strains.

Identification of 28 putative protein phosphatase genes in C. albicans and disruption of 21 of these genes.The presence of at least 28 putative protein phosphatases in C. albicans (Table 2) was determined using sequence-searching tools and analysis of the C. albicans genome sequence (1) by homology to annotated S. cerevisiae protein phosphatases (http://www.yeastgenome.org/ ), as well as domain searches (consensus patterns) of protein phosphatases using PROSITE (http://au.expasy.org/prosite ).

Gene disruption using the split-Ura-blaster technique (17), also used previously for disruption of the C. albicans YVH1 gene (21), was attempted for 25 of the 28 putative protein phosphatase genes identified. The CYR1/CDC35, CDC14, and CPP1 genes were not included in the study, because they have been previously annotated in C. albicans (13, 14, 24, 36) and did not have reported effects on virulence. For four genes (orthologues of S. cerevisiae genes GLC7, PPG1, PPH21, and TEP1), one allele was deleted but homozygous null mutants could not be obtained. However, 21 genes were successfully disrupted in both alleles (Table 2). The collection of 21 disruptant strains included 4 (the yvh1Δ, cmp1Δ, sit4Δ, and ptc7Δ strains) derived from genes already annotated in the C. albicans genome database; 3 of these (the yvh1Δ, cmp1Δ, and sit4Δ disruptants) have been reported as showing attenuated virulence in a mouse model of systemic infection (2, 3, 21, 34).

Virulence of null mutants of putative protein phosphatase genes in the silkworm model of C. albicans infection.Twenty-one protein phosphatase disruptants were evaluated for their abilities to kill silkworms in comparison with the wild-type parental C. albicans strain TUA6. The survival of silkworms was evaluated at day 1 following inoculation into the hemolymph. The parental wild-type strain TUA6 killed all 30 silkworms within 24 h, a result consistent with the findings of a previous report of silkworm infection with C. albicans (20). Control silkworms inoculated with the diluent only (phosphate-buffered saline) showed 100% survival. Decreasing the inoculum of the wild-type strain 10-fold (1 × 10⁵ cells) increased the survival rate of the silkworms (data not shown). Six mutant strains (orthologues of S. cerevisiae genes PPZ1, PTC7, MIH1, LTP1, PTP1, and PPS1 [7033, one of the two orthologues]) possessed wild-type virulence properties (all silkworms died within 24 h), and 11 mutant strains (orthologues of S. cerevisiae genes SAL6, PPT1, PTC2/PTC3, PTC6, PTC5, PTC4, OCA1, SIW14, OCA6, PTP3, and PPS1 [4405, the other orthologue]) showed slight to moderate (up to 50% survival) reductions in virulence properties. However, these findings were preliminary, because we have not demonstrated that full virulence of the mutants is restored by complementation. Table 3 shows virulence data for four strains (sit4, yvh1, cmp1, and ptc1 null mutants) for which virulence in the silkworm model was considerably attenuated (>63% survival). These results revealed that the C. albicans PTC1 gene is involved in virulence for the silkworm model, as well as confirming the virulence-related nature of three genes (YVH1, CMP1, and SIT4) previously identified using a mouse model of systemic infection (2, 3, 21, 34).

Comparison of the virulence of the parental strain TUA6 with that of the ptc1Δ null mutant and heterozygous revertant strains in silkworm and murine infection models.To substantiate the observation that the C. albicans PTC1 gene is involved in virulence, the heterozygous revertant strains PTC1A and PTC1B were constructed. Either one of the alleles of PTC1 was reintroduced, under the control of the ACT1 promoter, into the RP10 locus of the C. albicans ptc1Δ null mutant as described in Materials and Methods. All strains were URA⁺ and ARG⁺. The loss and gain of PTC1 mRNA in the null mutant and revertant strains, respectively, were confirmed by QRT-PCR (Fig. 1A). The two revertant strains, the parent strain TUA6, and the null mutant strain PTC1C were compared in further infection experiments. As shown in Fig. 1B, all the silkworms in groups infected with either PTC1A, PTC1B, or the parental strain TUA6 were killed within 30 h of inoculation. In contrast, it took more than 70 h to kill all the silkworms in the group infected with the null mutant strain. Silkworms infected with either TUA6, PTC1A, PTC1B, or PTC1C were examined by microscopy of transverse sections at 12 h postinoculation as described in Materials and Methods. Hyphal growth morphology predominated in the silkworms inoculated with either the wild-type or the revertant strains, whereas hyphal growth morphology was significantly reduced in silkworms infected with strain PTC1C (Fig. 2).

• Open in new tab
• Download powerpoint

FIG. 1.

Expression levels of PTC1 mRNA in C. albicans wild-type and mutant strains and the virulence of these strains in silkworm and mouse infection models. (A) Relative expression levels of the PTC1 transcript were determined by QRT-PCR and standardized with ACT1 mRNA. mRNAs were extracted from exponentially growing TUA6 (wild-type), PTC1A and PTC1B (revertant), and PTC1C (null mutant) cells grown in YPD (pH 5.6) at 30°C. (B) C. albicans cells (1 × 10⁶ CFU), either TUA6 (○), PTC1A (□), PTC1B (▪), or PTC1C (▵), were inoculated into silkworms (n = 10), and the survival of the silkworms was monitored. Significant differences between the survival rates with TUA6 versus PTC1C (P < 0.001) were observed. (C) C. albicans strains (1 × 10⁶ CFU) (symbols as explained for panel B) were each injected into mice (n = 6) via the tail vein, and survival was monitored. Significant differences between the survival rates with TUA6 versus PTC1C (P < 0.005) and with TUA6 versus PTC1B (P < 0.01) were observed. (D) Fungal burdens in the kidneys of infected mice (n = 3) were determined at day 5 after inoculation. Asterisks indicate statistically significant differences between TUA6 and PTC1B (*, P < 0.01) and between TUA6 and PTC1C (**, P < 0.005).

• Open in new tab
• Download powerpoint

FIG. 2.

Morphology of C. albicans cells in the infected silkworm. C. albicans strains (1 × 10⁶ CFU) were each inoculated into silkworms (n = 5). Silkworms infected with each strain were obtained 12 h after inoculation, stained with periodic acid-Schiff stain, cross-sectioned, and examined by microscope. Representative silkworm cross-sections were photographed with an object lens of ×40. Bar, 10 μm.

The mutant strains were also tested in a mouse model of systemic infection. Mice that were injected with the revertant strains or wild-type strain were all killed within 15 days postinfection, whereas 60% of mice infected with the null mutant survived more than 30 days (Fig. 1C). The revertant strains showed slightly increased virulence compared to that of the parental strain. The reduced virulence of the null mutant compared to that of the parental or revertant strains in the mouse model of infection was also reflected in reduced fungal burdens in the kidneys of infected mice (Fig. 1D). Recovery of C. albicans from the kidneys of groups of infected mice was determined 2 days after inoculation. Significant differences in the tissue burden among the four C. albicans strains were observed (Fig. 1D); the number of C. albicans cells recovered from the kidneys of mice infected with the null mutant was an order of magnitude lower than that from those infected with the wild-type or revertant strains. However, kidneys were not examined by microscopy to determine whether the CFU recovery could have been affected by different levels of hyphal formation in kidneys infected with different strains. As noted below, the mutant strain showed a reduced ability to form hyphae, and this may have affected the CFU yield from mice infected with this strain.

Effects of PTC1 disruption on yeast and hyphal growth.The effect of PTC1 gene disruption on the growth of C. albicans in the yeast morphology was determined by measuring the mass doubling times of the wild-type, PTC1C disruptant, and PTC1A and PTC1B revertant strains grown in YPD (pH 5.6) at 30°C. Mean mass doubling times were 1.65, 1.51, 1.40, and 1.41 h, respectively. There were no significant statistical differences among these strains.

In order to investigate the effect of PTC1 disruption on the yeast-to-hypha transition in C. albicans, cells were incubated in either liquid or agar-solidified YPD medium supplemented with 10% serum (pH 7.2) at 37°C. In liquid medium, there was a difference in hyphal formation in liquid serum-YPD between the wild-type and null mutant strains (Fig. 3A). The lengths of hyphae of the wild-type and ptc1Δ strains in liquid YPD containing 10% serum after 3 h at 37°C were 49.0 ± 7.2 μm and 33.3 ± 12.1 μm, respectively. For the revertant strains, the mean hyphal length was 55.8 ± 9.4 μm. The ptc1Δ disruptant strain (PTC1C) was more clearly defective in hyphal formation on serum agar (Fig. 3B). The revertant strains PTC1A and PTC1B rescued the defect of hyphal formation of the null mutant on serum agar, although hyphae were longer than those of the wild-type strain on serum agar (Fig. 3B). These in vitro results of cell morphology were consistent with the observed difference in hyphal growth in vivo (Fig. 2). The expression of the PTC1 transcript was assessed by QRT-PCR of mRNA extracted from wild-type TUA6 cells grown under conditions inducing yeast or hyphal growth in liquid media. The expression level of PTC1 mRNA in cells growing as budding yeasts was higher than that in cells displaying hyphal morphology (Fig. 3C).

• Open in new tab
• Download powerpoint

FIG. 3.

Cell and colony morphology of C. albicans strains TUA6 (wild type), PTC1A (revertant), and PTC1C (null mutant). (A) (Left) Yeast growth in fresh YPD (pH 5.6) with shaking at 30°C for 3 h. Bar, 10 μm. (Right) Hyphal growth in fresh YPD (pH 7.2) containing 10% serum with shaking at 37°C for 3 h. Bar, 20 μm. (B) Cells (1 × 10⁴ CFU) were spotted onto the surface of 10% serum agar and incubated at 37°C for 7 days. (C) Relative expression levels of the PTC1 transcript were determined by QRT-PCR and standardized against ACT1 mRNA. mRNAs were extracted from wild-type cells grown under different morphological growth conditions as described for panel A.

The PTC1 gene is required for proteolytic activity.Hydrolytic enzymes such as secreted aspartic proteinases and phospholipases are well-known virulence factors in C. albicans. We determined the secreted proteolytic, lypolytic, and hemolytic activities of the ptc1Δ strain relative to those of the parental and revertant strains by using agar plate assays. No significant differences were observed between the strains in the assay of phospholipase or hemolysin activity (data not shown). However, the null mutant had decreased proteolytic activity compared to that of the wild-type parental strain or the revertant strains PTC1A and PTC1B (P < 0.005 for TUA6 versus PTC1C) (Fig. 4).

• Open in new tab
• Download powerpoint

FIG. 4.

Proteinase activities of C. albicans strains TUA6 (wild type), PTC1A and PTC1B (revertants), and PTC1C (null mutant). Proteinase activities were measured by spotting cells (1 × 10⁴ CFU) onto a 1.2% yeast carbon base containing 0.5% bovine serum albumin at 30°C for 3 days. Relative activities were estimated as described in Materials and Methods. Data are means ± standard errors for three independent experiments. The asterisk indicates a statistically significant difference (P < 0.005) between TUA6 and PTC1C.

Susceptibilities to antifungals and stressors.The S. cerevisiae PTC1 gene has several specific cellular functions, including a regulatory role in the osmotically activated HOG pathway (19, 47-49, 54), and deletion of PTC1 in S. cerevisiae resulted in cells with pleiotropic stress phenotypes (19, 37, 48). Hence, we examined the susceptibilities of the C. albicans ptc1Δ mutant in comparison with those of the parent and revertant strains to a wide range of stress conditions by using plate dilution assays (Fig. 5A). Stressors such as high temperatures, high pH, high salt concentrations, and oxidative stress, to all of which the S. cerevisiae ptc1Δ strain was highly sensitive (19), did not affect the C. albicans ptc1Δ strain to any greater extent than the wild-type or revertant strains. However, the ptc1Δ strain showed greater resistance than the parental or revertant strains to the β-1,3-glucan-binding dye Congo red and to the squalene epoxidase inhibitor terbinafine. In contrast, the ptc1Δ strain was more susceptible to the β-1,3-glucan synthase inhibitor micafungin than the wild type. The differential responses of the parental strain TUA6, the revertants, and the null mutant to terbinafine and micafungin were confirmed by microdilution assays (Fig. 5B).

• Open in new tab
• Download powerpoint

FIG. 5.

Susceptibilities of C. albicans strains TUA6 (wild type), PTC1A and PTC1B (revertants), and PTC1C (null mutant) to antifungals and stressors. (A) Agar plate spotting assay. Tenfold serial dilutions of exponentially growing cells (5 μl) were spotted onto YPD agar plates containing antifungals or stressors at the indicated concentrations or under the indicated conditions. Cell growth was monitored after incubation at 30°C or the indicated temperature for 48 h. Representative results of an experiment repeated twice are shown. (B) Microdilution susceptibility assay. Susceptibilities to micafungin and terbinafine were determined for C. albicans TUA6 (○), PTC1A (□), PTC1B (▪), and PTC1C (▵) cells grown in YPD containing the indicated concentrations of antifungals at 30°C for 48 h.