ABSTRACT
We report here the identification and characterization of a previously uncharacterized, two-component response regulator gene (orf19.5843) from Candida albicans. Because of its apparent functions in stress adaptation, we have named this gene SRR1 (stress response regulator 1). Disruption of SRR1 causes defects in hyphal development, reduced resistance to stress, and severe virulence attenuation in the mouse model of disseminated candidiasis.
TEXT
Two-component signaling systems are widely used for signal transduction by bacteria, eukaryotic microorganisms, and plants. A two-component phosphorelay system in fungi includes a membrane-bound sensor histidine kinase (HK) protein, which is autophosphorylated on a histidine residue in the histidine kinase domain and then transfers the phosphoryl group to an internal receiver domain. The phosphoryl group is then shuttled through a histidine phosphotransfer (HPT) protein to a terminal response regulator (RR). When stress signals (oxidants, high salt, etc.) are detected by yeast cells, the RR protein is dephosphorylated and is able to activate the downstream mitogen-activated protein kinase (MAPK) pathway to adapt cells to stress (18, 19). The most extensively studied of the two-component MAPK systems in fungi is the hyperosmotic glycerol (HOG) MAPK pathway in Saccharomyces cerevisiae, (19, 25). In Candida albicans, Hog1 MAPK regulates glycerol accumulation and adaptation to high osmolarity, as well as oxidative stress, morphogenesis, and cell wall biosynthesis (2, 3, 38). Homologues of the Hog1 pathway, as well as other HKs and the response regulator protein (Ssk1p) of C. albicans, have been extensively studied (21, 27–30, 39, 41).
In C. albicans, there are three HK proteins, Chk1p, Sln1p, and Nik1/Cos1p (1, 12, 17, 32, 43). Sln1p is the homologue of S. cerevisiae Sln1p. While Chk1p is not found in S. cerevisiae, two homologues have been identified in Schizosaccharomyces pombe (9), but a functional pathway has not been assigned. Nik1/Cos1p is a homologue of the Nik1 protein of Neurospora crassa and is also not found in S. cerevisiae. The genes that encode RR proteins include SSK1 and SKN7. Ssk1p of S. cerevisiae activates the Hog1 MAP kinase pathway during osmotic stress, while Skn7p in S. cerevisiae acts as a transcription factor when phosphorylated. While homologues of these proteins are found in C. albicans, it is important to realize that C. albicans homologues have functions that are different from those of their counterparts in S. cerevisiae. We have shown previously that Ssk1p of C. albicans is required for adaptation to oxidative stress and regulates some aspects of cell wall biosynthesis (21), functions not reported for S. cerevisiae SSK1.
We report here that C. albicans possesses an additional response regulator gene, SRR1 (orf19.5843), besides the previously reported response regulators SSK1 and SKN7 (18). The newly identified response regulator SRR1 appears to be unique to the Candida clade (10). Homology searches did not reveal any orthologues of this protein in S. cerevisiae, Candida glabrata, or Candida krusei. However, orthologues of SRR1 are present in pathogenic fungi belonging to the CUG clade of Saccharomycotina (10). In C. albicans, previous studies by us and other groups have shown that both Ssk1 and Skn7 response regulators have important functions in morphogenesis and oxidative stress adaptation (14, 21, 40). In the present study, we demonstrate that a Δsrr1 mutant is defective in hyphal development and sensitive to both osmotic- and oxidative-stress growth conditions. Disruption of SRR1 causes severe virulence attenuation in the mouse model of disseminated candidiasis. These data imply that C. albicans utilizes SRR1 to regulate a number of critical processes to adapt to the host environment, including stress adaptation, hyphal development, and pathogenicity.
The SRR1 gene was identified by searching the Candida genome database (CGD) for proteins containing a receiver domain by performing BLASTP analysis. Because of its apparent functions in stress (oxidative and osmotic) adaptation, this response regulator gene was named SRR1 (stress response regulator 1). The SRR1 gene has an open reading frame of 849 bp, which encodes a 282-amino-acid (aa) protein with an estimated molecular mass of 32 kDa. Srr1p contains a receiver domain at the C terminus (154 to 270 aa) of this protein. The analysis of the promoter region of SRR1 by MatInspector (16) revealed that it contains a consensus sequence (underlined in 5′-TGCAACAGGGGGAGG-3′) of the fungal stress response elements (STRE); it remains to be seen if it is involved in the modulation of expression of SRR1. A putative peroxisomal targeting signal 1 (PTS1) sequence, NKI, is located at the C terminus of this protein. PTS1 motif prediction was performed by using the PTS1 predictor software (http://psort.hgc.jp/form2.html). A similar motif has been reported in Ssk1p of C. albicans (13), but its functional characterization has not yet been reported.
To determine the function(s) of SRR1 in C. albicans, we constructed a Δsrr1 mutant by following the method of Noble and Johnson (36) with the SN148 strain background (see Fig. S1 in the supplemental material). To generate strains with matched auxotrophic requirements, the URA3, HIS1, and LEU2 auxotrophic markers were restored by integration of an empty CIp10 (31) and a modified CIp20 (23) plasmid containing the LEU2 marker CIp20-Leu2 (see Tables S1 and S2 in the supplemental material) at the RPS10 locus. When these plasmids are integrated at the RPS10 locus, restoration of wild-type virulence and in vitro phenotypes to ura3 strains occurs (7). The C. albicans strains (see Table S3 in the supplemental material) generated in this way were auxotrophically (Ura+ His+ Leu+ Arg−) identical. The strains thus created were all Arg−. Arginine auxotrophy has been shown to be neutral for C. albicans virulence (36). All the experiments reported in the current study were performed by using strains NC1 (SRR1/SRR1), NC2 (srr1/srr1), and NC3 (srr1/srr1/SRR1).
To determine the effect of SRR1 disruption on growth rates, we compared the generation time of SRR1 deletion strain NC2 (srr1/srr1) with those of wild-type strain NC1 (SRR1/SRR1) and gene-reconstituted strain NC3 (srr1/srr1/SRR1). When incubated in yeast extract-peptone-dextrose (YPD) at 30°C, the NC2 strain grew significantly more slowly than the wild-type strain NC1, whereas the growth rate of the gene-reconstituted strain NC3 was similar to that of the wild-type strain. The calculated generation time for the null strain NC2 (2.13 h) was significantly higher than the 1.55-h and 1.59-h generation times of the wild type (NC1) and gene-reconstituted strain (NC3), respectively.
The two previously reported response regulators have functions related to the stress resistance, morphogenesis, and virulence of C. albicans (14, 21, 40). Thus, our initial experiments with SRR1 were focused to determine its role in stress response. The NC2 (srr1/srr1) mutant was sensitive to osmotic stress (1.5 M NaCl), a phenotype not observed with a C. albicans Δssk1 or Δskn7 mutant (21, 40), implying that Srr1p is critical to the response of cells to osmotic stress (Fig. 1). The NC2 (srr1/srr1) null mutant strain was also sensitive to hydrogen peroxide at a concentration of 8 mM (Fig. 1), suggesting that Srr1p is required for resistance to oxidative stress in vitro and, in this regard, is somewhat similar to other response regulators in C. albicans, Skn7p and Ssk1p (21, 40). We did not observe any difference between wild-type and null mutant strains in their sensitivities to menadione, potassium superoxide, sodium dodecyl sulfate (SDS), caffeine, Congo red, or calcofluor white (data not shown). Data from MIC assays verify (Table 1) that the null mutant strain NC2 (srr1/srr1) is more sensitive to oxidative and osmotic stress than the wild type (NC1) and reintegrant strain (NC3). In vitro killing assays also confirmed the hypersusceptibility of the NC2 mutant strain to H2O2 (data not shown). Thus, SRR1 is required for protection from oxidants, a result similar to that reported for the C. albicans Δssk1 null mutant (21). However, unlike the Δssk1 or Δskn7 response regulator mutant, the Δsrr1 mutant is sensitive to osmotic stress, suggesting that SRR1 is required for mediating resistance to osmotic stress.
Growth of NC1 (SRR1/SRR1), NC2 (srr1/srr1), and gene-reconstituted strain NC3 (srr1/srr1/SRR1) at 30°C for 24 h on YPD agar (control plate) (A), YPD agar containing 8 mM H2O2 (B), and YPD agar containing 1.5 M NaCl (C). Five-microliter cell dilutions (5 × 105 to 5 × 101 cells) were spotted on the YPD agar compounds.
Susceptibilities of C. albicans strains to hydrogen peroxide and sodium chloride
Because the two previously reported response regulators Ssk1 and Skn7 have functions in hyphal formation, it was of interest to determine the effect of SRR1 disruption on morphogenesis by growing C. albicans strains on Spider, 10% serum, and RPMI (pH 7.5) growth media. The NC2 (srr1/srr1) null mutant strains displayed suppressed filament formation in hyphae, inducing growth comparable to that of the wild-type and reconstituted strains, indicating that SRR1 is critical for hyphal development in C. albicans under the growth conditions used in our experiments (Fig. 2).
Phenotypes of the NC1 (SRR1/SRR1), NC2 (srr1/srr1), and NC3 (srr1/srr1/SRR1) strains grown on solid media (Spider medium, RPMI medium, and 10% serum), which induce hyphal development. All the plates were incubated for 5 days at 30°C, and representative colonies were photographed.
To investigate the role of Srr1p in virulence, wild-type NC1, null mutant NC2, and gene-reconstituted strain NC3 were inoculated intravenously into immunocompetent BALB/c mice. The survival of mice over a 21-day period of time was determined for each strain (Fig. 3). We found that mice infected with the wild-type strain rapidly succumbed to the infection within 3 days, whereas mice infected with the null mutant strain NC2 (srr1/srr1) survived for at least 3 weeks (the duration of experiment). Mice infected with the gene-reconstituted strain NC3 (srr1/srr1/SRR1) survived longer than mice infected with the wild type. The colonization (CFU/g) of kidneys was determined for each strain at 24 to 72 h postinfection (Table 2), except for animals infected with the wild-type strain NC1, since all were moribund by 72 h. Fungal burdens, expressed as numbers of CFU/g of kidney (Table 2), were relatively constant for both the wild-type and gene-reconstituted strains over the course of 72 h. However, for the NC2 (srr1/srr1) null mutant strain, the highest number of CFU/g kidney was significantly lower (P < 0.0001; NC2 versus all strains) than for the wild-type strain, suggesting that the NC2 (srr1/srr1) null mutant strain was rapidly cleared from the kidneys. Histopathology of kidney sections supports this observation (Fig. 4). Taken together, the data clearly demonstrate that Srr1p is required for virulence in a murine model of systemic candidiasis.
Survival of mice infected with C. albicans strains in a murine model of hematogenously disseminated candidiasis. Mice were infected with 1 × 106 CFU of each strain, namely, NC1 (SRR1/SRR1), NC2 (srr1/srr1), and NC3 (srr1/srr1/SRR1). The percent survival of mice infected with each strain was determined 21 days postinfection.
Log10 CFU/g counts in mice infected with C. albicans strains
Periodic acid-Schiff (PAS)-stained kidney sections from mice infected with C. albicans strains NC1 (SRR1/SRR1), NC2 (srr1/srr1), and NC3 (srr1/srr1/SRR1). Arrows indicate the locations of yeast cells.
There are functional differences between the three response regulator proteins of C. albicans, Srr1p, Ssk1p, and Skn7p, that may have an impact on the contribution of each protein to disease. The characterization of the NC2 (srr1/srr1) mutant strain clearly demonstrates a role for SRR1 in resistance to osmotic and oxidative stress. The avirulence of the Δsrr1 null mutant is similar to that of the Δssk1 mutant, while the virulence of the Δskn7 mutant was only slightly attenuated (40).
Previous reports have suggested that C. albicans mutants with growth defects tend to have reduced virulence in a mouse model of systemic candidiasis (37). Because the NC2 (srr1/srr1) null mutant has a longer generation time than the wild type, there is a concern that its avirulence might be attributed to its longer generation time. In this context, it has been reported that mutants with long generation times still cause disease and death in mice at slightly higher inoculum densities than those used here (20), indicating that the generation time is less critical than the inoculum density to the development of candidiasis. A recent study has also revealed that growth rate is not essential for C. albicans pathogenicity, as previously thought (35). In that study, it was reported that mutants with low growth rates showed normal infectivity in a mouse model of disseminated candidiasis (35). There are other examples in the literature about C. albicans mutants exhibiting normal virulence despite an in vitro growth defect phenotype. For example, the C. albicans protein kinase cka2 null mutant strain has a growth defect but still exhibits normal virulence, like the wild-type strain, in the mouse model of hematogenously disseminated candidiasis (22). Taken together, data from these studies suggest that C. albicans virulence is a complex interplay of multiple (host as well as pathogen) factors. In vitro assays cannot replicate complex in vivo systems, and growth rate assays performed in vitro cannot be the exclusive determinants of C. albicans virulence.
The virulence of C. albicans is multifactorial, including the requirement for adhesins, yeast- to hyphal-phase transition (morphogenesis), and expression of extracellular invasive enzymes, such as secreted aspartyl proteases and phospholipases (11). The ability to overcome stress has also been reported to be important for C. albicans virulence. Several studies have also shown that stress adaptation is essential for C. albicans virulence (8). C. albicans strains with mutations in genes that encode the transcription factor Cap1p (5), superoxide dismutase (26), catalase (33, 42), the stress-activated MAPK Hog1p (3), the upstream two-component histidine kinase Chk1p (15), and the response regulator Ssk1p (14) are reported to be avirulent in a murine model of disseminated candidiasis. Our findings, with SRR1, also reveal a positive relationship between stress adaptation and virulence. Data presented in this paper strongly support a role for SRR1 in morphogenesis, stress adaptation, and virulence of C. albicans. Two-component proteins have also been reported to be essential for stress adaptation and virulence in other pathogenic fungi, including Aspergillus fumigatus (24), Cryptococcus neoformans (4, 6), and Blastomyces dermatitidis (34), suggesting that their exploitation could lead to the development of drugs that may have broad spectra.
ACKNOWLEDGMENTS
This work is supported in part by an American Heart Association National Scientist Development Grant (0635108N) to N.C.
We thank Suzanne Noble, UCSF, for the generous gift of pSN40 and pSN52 plasmid vectors and the wild-type C. albicans SN148 strain for this work. We also thank Al Brown, University of Aberdeen, and Michael Lorenz, University of Texas Health Science Center at Houston, for plasmid CIp20. We gratefully acknowledge Guillaume Delmas, PHRI Center at UMDNJ, for performing animal experiments.
FOOTNOTES
- Received 28 July 2011.
- Accepted 8 August 2011.
- Accepted manuscript posted online 12 August 2011.
- ↵*Corresponding author. Mailing address: Public Health Research Institute and Department of Microbiology and Molecular Genetics, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, 225 Warren Street, Newark, NJ 07078. Phone: (973) 854-3470. Fax: (973) 854-3101. E-mail: chauhan1{at}umdnj.edu.
† Supplemental material for this article may be found at http://dx.doi.org/10.1128/EC.05188-11.
REFERENCES
- Copyright © 2011, American Society for Microbiology. All Rights Reserved.
