The calcium transporter Pmc1 provides Ca2+ tolerance and influences the progression of murine cryptococcal infection

Authors


Abstract

The Ca2+-calcineurin signaling pathway in the human fungal pathogen Cryptococcus neoformans is essential for adaptation to the host environment during infection. Calcium transporters regulate cytosolic calcium concentrations, providing Ca2+ loading into storage organelles. The three calcium transporters that have been characterized in C. neoformans, Cch1, Eca1 and Vcx1, are required for fungal virulence, supporting a role for calcium-mediated signaling in cryptococcal pathogenesis. In the present study, we report the functional characterization of the putative vacuolar calcium ATPase Pmc1 in C. neoformans. Our results demonstrate that Pmc1 provides tolerance to high Ca2+ concentrations. The double knockout of C. neoformans PMC1 and VCX1 genes impaired the intracellular calcium transport, resulting in a significant increase in cytosolic calcium levels. Furthermore, Pmc1 was essential for both the progression of pulmonary infection and brain colonization in mice, emphasizing the crucial role of calcium signaling and transport for cryptococcal pathogenesis.

Abbreviations
CFU

colony forming unit

ER

endoplasmic reticulum

Fura-2 AM

Fura-2 acetoxymethylester

GXM

glucuronoxylomannan

WT

wild type

YPD

yeast extract–peptone–dextrose medium

Introduction

In eukaryotic cells, ionic calcium (Ca2+) is an intracellular regulator of a wide variety of signaling pathways. One of the key activities of intracellular Ca2+ involves the Ca2+ binding protein calmodulin, which activates the phosphatase calcineurin in response to increasing cytosolic calcium levels [1, 2]. The Ca2+ levels are coordinated by transporters located in the plasma, vacuolar or endoplasmic reticulum (ER) membranes. Activated calcineurin mediates nuclear translocation of specific transcription factors, which trigger the expression of calcineurin-responsive genes [3, 4]. These gene products allow the cell to cope with stress and maintain Ca2+ homeostasis. Ca2+ related pathways have been associated with crucial pathogenic processes in different fungal pathogens, including Cryptococcus neoformans, Cryptococcus gattii, Candida albicans and Aspergillus fumigatus [5-8].

The calcium-calcineurin signaling pathway in the human pathogenic fungus C. neoformans is essential for adaptation to the host milieu and establishment of infection [9]. Calcineurin is required for fundamental biological events of C. neoformans, such as mating, morphogenesis, growth at 37 °C and virulence [7, 10-13]. Moreover, the C. neoformans transcription factor Crz1 is calcineurin activated and regulates cell wall integrity [14]. C. neoformans calmodulin is critical for cell viability and for fungal response to high temperatures [15]. Other important components of the calcium-calcineurin signaling network in C. neoformans are the Ca2+ transporters Cch1, Eca1 and Vcx1. Cch1 mediates Ca2+ entry in fungal cells and is required for the uptake of this ion in low Ca2+ environments [16]. Eca1, a sarcoplasmic/endoplasmic reticulum Ca2+-ATPase, participates in stress tolerance [17]. Vcx1, a vacuolar Ca2+-exchanger, regulates Ca2+ tolerance [18]. Importantly, Cch1, Eca1 and Vcx1 are involved in C. neoformans virulence [16-18], demonstrating the importance of Ca2+-mediated signaling in this pathogen.

Ca2+ transporters mediate the transfer of free ions from the cytosol to storage organelles, preventing toxicity. The vacuole is the major Ca2+ storage organelle in many eukaryotes, including fungi. In this organelle, Ca2+-ATPases and Ca2+exchangers facilitate the accumulation of Ca2+ (reviewed in [19]). In Saccharomyces cerevisiae, the Ca2+-ATPase Pmc1 regulates cytosolic calcium levels, providing Ca2+ tolerance and vacuolar Ca2+ loading [19, 20]. We previously demonstrated that the PMC1 ortholog is upregulated in the C. neoformans vcx1 knockout strain, possibly because Pmc1 also transports Ca2+ into vacuoles, generating functional redundancy [18]. Here, we report the functional characterization of the C. neoformans PMC1 gene. For this purpose, a pmc1 null mutant, a double pmc1 vcx1 mutant and relevant complemented strains were constructed. The PMC1 knockout resulted in hypersensitivity to high Ca2+ concentrations, a phenotype that was even more pronounced in the pmc1 vcx1 double knockout strain. Furthermore, disruption of PMC1 influenced the relative intracellular Ca2+ concentration. Notably, lack of Pmc1 interfered with C. neoformans virulence in mice.

Results

In silico characterization of the vacuolar Ca2+-ATPase PMC1 ortholog in C. neoformans

The PMC1 gene was identified in the C. neoformans H99 genomic database at the Broad Institute (accession number CNAG_01232.2) based on its similarity to the vacuolar Ca2+-ATPase from S. cerevisiae. The C. neoformans PMC1 coding region is 4655 bp long, contains eight introns, and encodes a putative 1414-amino-acid protein. blast search using the Conserved Domain Database at NCBI revealed the presence of the following conserved domains in the C. neoformans Pmc1 ortholog: a cation ATPase C domain (PFAM00689), an E1–E2 ATPase domain (PFAM00122), a haloacid dehalogenase-like hydrolase domain (PFAM00702) and a cation transporter ATPase N domain (PFAM 00690). Additionally, phylogenetic analysis including Pmc1 sequences from distinct eukaryotic organisms revealed that the C. neoformans Pmc1 had the highest similarity to the Ustilago maydis Pmc1 ortholog (55% identity and 69% similarity; Fig. 1). Prediction of transmembrane regions revealed that C. neoformans Pmc1 has 10 transmembrane domains, which is in agreement with the properties of well-described Ca2+-ATPases [19].

Figure 1.

Phylogenetic analysis of C. neoformans Pmc1 orthologs. The analysis was conducted by applying the neighbor-joining method and included Pmc1 sequences from distinct eukaryotic organisms (S. cerevisiae, Saccharomyces Genome Database accession number YGL006W; C. albicans, Candida Genome Database accession number orf 19.1727; A. fumigatus, Aspergillus Genome Database, accession numbers Afu1g10880 (PmcA), Afu3g10690 (PmcB), Afu7g01030 (PmcC); U. maydis, Broad Institute accession number UM_03470.1; M. oryzae, Broad Institute accession numbers MGG_07971, MGG_04890, MGG_02487; C. elegans, NCBI accession number CAA09303.1; C. neoformans, Broad Institute accession number CNAG_01232). The bar marker indicates genetic distance, which is proportional to the number of amino acid substitutions. Bootstrap values obtained from 1000 resamplings are shown at the nodes.

The C. neoformans pmc1 mutant displayed sensitivity to high calcium concentrations

The Ca2+ calcineurin signaling network in C. neoformans, which is crucial for adaptation to the host [9], includes the vacuolar Ca2+ transporter Vcx1 [18]. C. neoformans vcx1 mutant cells have an altered calcineurin-dependent Ca2+ tolerance and a reduced ability to kill mice [18]. On the basis of these observations, we asked whether Pmc1 is involved in similar events. For this purpose, a pmc1 null mutant, a pmc1 vcx1 double mutant and complemented strains were constructed (Fig. S1). The role of C. neoformans Pmc1 in Ca2+ tolerance was evaluated by monitoring the growth of the mutant strains on yeast extract–peptone–dextrose medium (YPD) agar plates supplemented with increasing concentrations of CaCl2. The pmc1 mutant strain had a reduced growth rate under high Ca2+ conditions (Fig. 2A). This phenotype was accentuated in the pmc1 vcx1 double mutant, which manifested similar growth defects at lower Ca2+ concentrations. We also evaluated growth rates of the C. neoformans pmc1 mutant strains in the presence of other divalent cations (Cd2+ and Mn2+). In general, no significant differences in fungal growth were observed on agar plates containing 4 mm MnCl2. The only exception was the pmc1 vcx1 double mutant, which displayed impaired growth under these conditions. The pmc1 and pmc1 vcx1 mutant strains exhibited increased resistance to 50 μm CdCl2 in comparison to wild-type (WT) and complemented strains (Fig. 2B).

Figure 2.

C. neoformans Pmc1 is involved in Ca2+ tolerance. Ten-fold serial dilutions of WT, pmc1 mutant, pmc1 vcx1 mutant, vcx1 mutant and pmc1::PMC1 complemented cell suspensions were plated onto YPD agar containing 1–500 mm CaCl2 (A) or MnCl2 or CdCl2 (4 mm and 50 μm, respectively; B). The plates were incubated for 2 days at 30 °C. Control samples consisted of cells grown in YPD agar only.

Disruption of PMC1 led to defective growth and impaired capsule formation in DMEM

Growth at 37 °C, capsule formation and melanin production were evaluated in the pmc1 mutant strains because these attributes are determinants for C. neoformans virulence [21]. Disruption of PMC1 did not interfere with melanin production or with the ability to grow at human body temperature. The only exception was the pmc1 vcx1 double mutant, which displayed a slight growth defect at 37 °C (Fig. S2). Capsule formation and extracellular glucuronoxylomannan (GXM) release were affected in pmc1 mutant strains under specific conditions. When the cells were incubated in DMEM, a medium that has been reported to induce capsule enlargement, capsule formation and extracellular GXM release were impaired (Fig. 3A). Apparently, this observation is associated with other alterations in the fungal physiology, since fungal growth was also reduced when the mutant strains were cultivated in DMEM (Fig. 3C). The pmc1 vcx1 double mutant also showed a slightly increased sensitivity to 5% CO2 in vitro (Fig. S3). These phenotypic attributes were not observed in mutant cells that were incubated in minimal medium, another classical capsule-inducing condition (Fig. 3B). Capsular polysaccharides produced by WT, pmc1, pmc1 vcx1 and complemented cells were regularly recognized by monoclonal antibody 18B7 in both capsule-induction conditions, as demonstrated by immunofluorescence analysis (Fig. 3D).

Figure 3.

The C. neoformans pmc1 mutant presented growth defects and impaired capsule formation under specific growth conditions. (A) India ink counterstaining, capsule measurements and extracellular GXM determination in C. neoformans cultures incubated under capsule-inducing conditions (DMEM, 37 °C, 5% CO2): *< 0.05; ****< 0.0001. (B) India ink counterstaining, capsule measurements and extracellular GXM determination in C. neoformans cultures incubated in minimal medium at 30 °C. (C) Growth rates of WT, pmc1 mutant, pmc1 vcx1 mutant and complemented strains after incubation in DMEM: *< 0.05; **< 0.01 (24 h); ##< 0.01 (48 h). (D) Confocal microscopy of WT, pmc1 mutant, pmc1 vcx1 mutant and complemented cells incubated in DMEM or minimal medium (MM), as indicated. Cells were stained with calcofluor white (blue) and with the monoclonal antibody 18B7 (green) to visualize cell wall chitin and GXM, respectively. Bars, 5 μm.

Disruption of PMC1 influences the relative level of intracellular calcium concentration in C. neoformans cells

The relative concentration of free intracellular Ca2+ in WT, pmc1, pmc1 vcx1 and complemented strains was determined using the Ca2+sensitive dye Fura-2 acetoxymethylester (Fura-2 AM) (Invitrogen, Carlsbad, CA, USA) [18]. The pmc1 and pmc1 vcx1 mutants had increased relative intracellular calcium concentrations compared with WT and complemented strains (Fig. 4). This difference was more pronounced in the pmc1 vcx1 double mutant strain. These findings indicate that the loss of PMC1 and VCX1 could lead to a severe defect in intracellular calcium transport because both transporters regulate the cytosolic Ca2+ concentration and uptake into vacuoles [19].

Figure 4.

Disruption of PMC1 interferes with the relative intracellular Ca2+ levels in C. neoformans. The relative levels of intracellular Ca2+ in WT, pmc1 mutant, pmc1 vcx1 mutant and complemented cells were determined using Fura-2 AM. The relative Ca2+ concentration was determined based on the fluorescence ratio after dual-wavelength excitation. Data represent means ± standard deviations. *< 0.05; **< 0.01.

Loss of PMC1 affects the relative expression of other Ca2+ transporters in C. neoformans

Transcript levels of the Ca2+ calcineurin-related CCH1, ECAI, VCX1 and PMR1 Ca2+ transporters were evaluated in the WT, pmc1, pmc1 vcx1 and vcx1 mutant strains. Disruption of PMC1 led to a decrease in the expression of VCX1 and to an increase in the expression of ECA1 after a 10 min exposure to 100 mm CaCl2 (< 0.01) (Fig. 5). However, CCH1 and PMR1 transcript levels were not significantly affected in the pmc1 mutant strain. No statistically significant differences were observed in the transcript levels of calcium transporters in the control condition.

Figure 5.

Loss of PMC1 affects the relative expression of other Ca2+ transporters after exposure to CaCl2. The relative expression levels of various C. neoformans calcium transporters (CCH1, ECA1, PMR1 and PMC1) in WT, pmc1 mutant, pmc1 vcx1 mutant and vcx1 mutant cells were quantified by qRT-PCR. The data were normalized to actin cDNA levels in each set of PCR experiments. Data represent means ± standard deviations. *< 0.05; **< 0.01.

Pmc1 is required for the progression of murine cryptococcal infection

Since all the previously characterized calcium transporters modulate virulence in C. neoformans [16-18], we asked whether Pmc1 could also have a role in cryptococcal pathogenesis. Mice infected with WT or complemented strains had similar mean survival periods of 22 and 26 days, respectively (= 0.0053). In contrast, pmc1 and pmc1 vcx1 mutant cells were strongly attenuated for virulence (< 0.0001) (Fig. 6A). The fungal burden in the lungs and brain of animals infected with WT, pmc1, pmc1 vcx1 and complemented strains was determined at 3, 7, 14 and 19 days post-infection. We found that pmc1 mutant strains displayed greatly reduced fungal burdens in lung tissues at all days analyzed compared with WT and complemented strains, indicating that Pmc1 is important for the progression of pulmonary infection in mice (Fig. 6B). Importantly, no viable fungal cells were recovered from the lungs of mice infected with the pmc1 vcx1 double mutant at days 14 and 19 post-infection, most probably due to infection clearance. Furthermore, animals infected with the pmc1 or pmc1 vcx1 mutant strains had no fungal burden in brain tissues, in contrast to what was observed for WT and complemented strains. Together, these observations demonstrate that Pmc1 is critical for C. neoformans virulence in a mice model of cryptococcosis.

Figure 6.

Pmc1 is essential for cryptococcal pathogenesis in mice. Mortality curves (A) and fungal loads (B) of mice infected with WT, pmc1 mutant, pmc1 vcx1 mutant and pmc1::PMC1 strains. (C) India ink counterstaining of fungal cells recovered from infected lungs at day 7 post-infection. Bars, 10 μm.

The hypocapsular phenotype observed after incubation of pmc1 and pmc1 vcx1 mutant cells in DMEM led us to ask whether these cells had the same phenotype during pulmonary infection. This question was addressed in supernatants of macerated lungs excised from mice infected with WT, pmc1, pmc1 vcx1 or complemented strains at day 7 post-infection. India ink counterstaining of these cells revealed normal capsules, indicating proper capsular assembly by the pmc1 mutant strains in vivo (Fig. 6C).

Discussion

The Ca2+-calcineurin signaling network is essential to many pathogenic mechanisms of C. neoformans, including mating, morphogenesis and growth at 37 °C [7, 9, 12, 13]. In the present study, we characterized a new component of the Ca2+-mediated signaling pathway in C. neoformans: the vacuolar Ca2+-ATPase Pmc1. Fungal Ca2+-ATPases are high affinity Ca2+ transporters that mediate the uptake of this ion into storage organelles in response to small alterations in cytosolic Ca2+ [19]. In particular, vacuolar Ca2+ATPases play an important role in calcium tolerance, as concluded from the observation that the PMC1 knockout in S. cerevisiae led to a Ca2+ hypersensitivity phenotype [20]. This observation corroborates our findings that C. neoformans Pmc1 is involved in Ca2+ tolerance in the presence of high concentrations of CaCl2. The vacuolar calcium exchanger Vcx1 partially contributes to this process, as inferred from the fact that the C. neoformans pmc1 vcx1 double mutant exhibited high sensitivity to low concentrations of CaCl2, indicating that vacuolar Ca2+ uptake is dramatically impaired in this mutant strain. Interestingly, a cadmium-resistance phenotype was detected for the pmc1 and pmc1 vcx1 mutant strains. Based on gene expression data, Mielniczki-Pereira et al. showed that S. cerevisiae Pmc1 could be involved in cadmium tolerance, but no significant sensitivity to Cd2+ was detected in S. cerevisiae pmc1 cells [22].

Our assays aiming at the characterization of additional regulators of the physiology of Ca2+ distribution in C. neoformans efficiently illustrate the need for different experimental conditions for phenotypic analysis of fungal mutants. For instance, Pmc1 was required for fungal growth and capsule formation when C. neoformans was incubated in DMEM but not in minimal medium or in vivo. This observation strongly suggests that PMC1 responds differently to specific conditions of the microenvironment to which the fungus is exposed. The components of DMEM that interfere with the physiological role of Pmc1 are still unknown. The ability to induce the formation of a polysaccharide capsule in response to a variety of host-specific environmental stimuli is one of the most important C. neoformans virulence attributes, which stimulates the investigation of this putative PMC1 regulator. Capsule formation is a complex biological process that is regulated at multiple levels, including the biosynthesis of basic components, the transfer of glycosyl units, polysaccharide transport, and its assembly and maintenance at the cell surface (reviewed in [23]). Divergence in capsular phenotypes was also described for C. neoformans rim20 and rim101 mutant strains when cultured in distinct capsule-inducing media [24]. In addition, deletion of GRASP, the gene coding an essential regulator of polysaccharide secretion in C. neoformans, resulted in smaller capsules in vitro, but the mutant produced regular amounts of GXM during lung infection [25].

Cryptococcus neoformans Pmc1 also affected the expression of other calcium transporters. We previously demonstrated that PMC1 is upregulated in the vcx1 knockout strain, most probably due to a compensatory effect and functional redundancy [18]. However, in response to CaCl2 exposure, we found that VCX1 is downregulated in the pmc1 knockout strain, indicating the existence of a complex system that regulates calcium transport in C. neoformans. In S. cerevisiae, calcineurin reduces the Ca2+ tolerance of pmc1 mutants through inhibition of Vcx1 function [26]. This observation may be related to our current findings. Furthermore, there was an increase in ECA1 expression in the pmc1 mutant strain in response to a short time exposure to CaCl2. Interestingly, ECA1 expression was not affected by deletion of VCX1 in the pmc1 strain. A similar trend was observed when PMR1 was analyzed, but the differences in relative expression levels were not statistically significant.

Our findings demonstrated that the vacuolar calcium transporter Pmc1 is required for C. neoformans virulence in a mouse model of cryptococcosis. Reduced proliferation of pmc1 and pmc1 vcx1 mutant cells in the murine host was observed, as concluded from the reduced fungal burden in lung tissues and impaired brain colonization. The molecular mechanisms involved in dissemination are not yet completely understood; however, there are a number of C. neoformans mutants that fail to cause central nervous system disease (reviewed in [27]). The genes deleted in these mutants are involved in a variety of cellular functions, suggesting that deletion of PMC1 and VCX1 cause general defects in fungal physiology and consequently altered survival during interaction with the host. The impaired virulence of the mutant strains described in this study could be related to a defect in vacuolar Ca2+ uptake, which would result in excessive cytosolic Ca2+ concentrations. This perturbation of Ca2+ homeostasis may interfere with a wide variety of signaling pathways involved in C. neoformans pathogenesis. Importantly, the C. neoformans calcium transporters Cch1, Eca1 and Vcx1 also modulate virulence [16-18], supporting a crucial role of Ca2+ transport in cryptococcal pathogenesis. The function of Pmc1 in fungal virulence has been reported in other human and plant pathogens, including A. fumigatus and Magnaporthe oryzae [28, 29]. The Afumigatus pmcA mutant strain also displayed calcium sensitivity and attenuated virulence [28]. The knockdown of three M. oryzae PMC1 homologs resulted in significant reduction in mycelial melanization and conidial development. Additionally, the M. oryzae PMC1 homologs were essential for pathogenicity in plant hosts [29]. These observations reinforce the essential role for Pmc1 in fungal pathogenesis.

In conclusion, we have shown that Pmc1, a vacuolar Ca2+-ATPase, is required for essential cellular physiology and virulence events in C. neoformans. These events include Ca2+ tolerance, regulation of intracellular Ca2+ concentration and progression of cryptococcosis in mice. Although it is clear that further studies are required for a complete understanding of how Ca2+ homeostasis associated with virulence is regulated in C. neoformans, these results reveal previously unknown aspects of the mechanisms by which the fungus controls Ca2+ availability. Our data also reinforce the notion that Ca2+ regulation is absolutely essential to the pathogenic processes required for the development of cryptococcosis.

Materials and methods

Fungal strains, plasmids and media

The C. neoformans serotype A strain H99 was utilized for the construction of the pmc1 mutant strain. Fungal cells were maintained on YPD medium (1% yeast extract, 2% peptone, 2% dextrose and 1.5% agar). YPD plates containing nourseothricin (100 μg·mL−1) were used to select C. neoformans PMC1 knockout transformants (pmc1 and pmc1 vcx1 strains). YPD plates supplemented with hygromycin (200 μg·mL−1) were used to select C. neoformans PMC1 complementation transformants (pmc1::PMC1 strain). The pJAF15 [30] and pAI4 [31] plasmids were the source hygromycin and nourseothricin resistance cassettes, respectively. Plasmids were maintained in Escherichia coli grown at 37 °C in LB broth or on agar supplemented with 50 μg·mL−1 kanamycin.

In silico analysis of the C. neoformans PMC1 ortholog

The putative C. neoformans PMC1 gene sequence was identified by a blast search of the C. neoformans var. grubii strain H99 genomic database at the Broad Institute using the Pmc1 sequence of S. cerevisiae (S. cerevisiae Genome Database, accession number YGL006W). The amino acid sequences of Pmc1 orthologs from S. cerevisiae, Candida albicans, A. fumigatus, U. maydis, M. oryzae, Caenorhabditis elegans and C. neoformans were aligned using clustalx2. Mega4 was utilized for phylogenetic analysis using the neighbor-joining method, and the tree architecture was inferred from bootstrap values obtained from 1000 resamplings. Search for conserved domains in the C. neoformans Pmc1 amino acid sequence was performed using the Pfam database (http://pfam.sanger.ac.uk/). The tmhmm server was utilized for prediction of putative transmembrane segments (http://www.cbs.dtu.dk/services/TMHMM/).

Disruption and complementation of C. neoformans PMC1

Disruption of PMC1 was achieved using plasmid constructs generated by the DelsGate method [18]. The 5′ and 3′ PMC1 flanks (~ 700 bp each) were PCR-amplified and gel-purified using Illustra GFX PCR DNA and Gel Band Purification kit (GE Healthcare, Uppsala, Sweden). Approximately 300 ng of the pDONR-NAT vector [32] and 30 ng of each PCR product were used in the Gateway BP clonase reaction, according to the manufacturer's instructions (Invitrogen). The reaction product was transformed into E. coli OmniMAX 2-T1 cells. After confirmation of the correct deletion construct, the plasmid was linearized by I-SceI digestion prior to biolistic C. neoformans transformation [33]. Transformants were screened by colony PCR, and the deletion was confirmed by southern blot analysis. For complementation, a 6.6 kb genomic PCR fragment carrying the WT PMC1 gene and regulatory regions was cloned into the EcoRV site of pJAF15 vector. The resulting plasmid was used for transformation of the pmc1 mutant strain and hygromycin was used to select for positive transformants. Random genomic insertion of the complemented gene was confirmed by southern blot analysis. To construct the pmc1 vcx1 double mutant, the same plasmid used to construct the pmc1 mutant strain was used to transform vcx1 mutant cells. The primers used for these constructions are listed in Table S1.

Phenotypic characterization assays

For phenotypic characterization, WT, mutant (pmc1, vcx1 and pmc1 vcx1) and complemented strains were grown on YPD medium for 16 h, washed with NaCl/Pi and adjusted to a cell density of 107 cells·mL−1 in YPD. The cell suspensions were serially diluted 10-fold, and 3 μL of each dilution was spotted onto YPD agar supplemented with CaCl2 (1, 5, 10, 20, 100, 200, 300, 400 or 500 mm), MnCl2 (4 mm) or CdCl2 (50 μm). The plates were incubated for 2 days at 30 °C and photographed. Melanin production was examined on glucose-free asparagine medium agar plates (1 g·L−1 l-asparagine, 0.5 g·L−1 MgSO4.7H2O, 3 g·L−1 KH2PO4 and 1 mg·L−1 thiamine) containing 1 mm l-3,4-dihydroxyphenylalanine [34]. Capsule formation was evaluated in cells that were cultivated for 48 h in DMEM at 37 °C in 5% CO2 or for 48 h at 30 °C in a minimal medium composed of dextrose (15 mm), MgSO4 (10 mm), KH2PO4 (29.4 mm), glycine (13 mm) and thiamine-HCl (3 μm) (pH 5.5). Cell viability was monitored by colony forming unit (CFU) determination. Relative capsule sizes were defined as the distance between the cell wall and the capsule outer border divided by values of cell diameter. imagej software (http://rsbweb.nih.gov/ij/) was utilized for capsule measurements in at least 50 cells of each strain. Extracellular polysaccharide contents were evaluated by ELISA using antibodies to GXM, as described elsewhere [35]. Cell surface morphology was analyzed after incubation of yeast cells with calcofluor white and the monoclonal antibody 18B7, which recognizes GXM [36]. These probes were used to visualize cell wall chitin (calcofluor) and GXM (18B7) by confocal microscopy following a previously described protocol [37].

Determination of relative intracellular Ca2+ levels

The relative intracellular Ca2+ concentration was determined using Fura-2 AM according to a previously described protocol [18]. Briefly, WT, pmc1, pmc1 vcx1 and complemented cells were cultured in YPD medium overnight with shaking at 30 °C. Subsequently, 107 cells of each strain were incubated for 1 h in fresh YPD supplemented with 100 mm CaCl2. The cells were washed with NaCl/Pi and incubated with 10 μm Fura-2 AM for 30 min at 37 °C. Fura-2 fluorescence was measured at excitation wavelengths of 340 and 380 nm and an emission wavelength of 505 nm. The relative intracellular calcium concentration was expressed as the ratio of fluorescence intensities obtained by excitation at wavelengths of 340 and 380 nm. All data presented are representative of three independent experiments.

Quantitative real-time RT-PCR analysis

For RNA extraction, cultures of WT, pmc1, pmc1 vcx1 and vcx1 mutant cells were grown overnight in YPD medium at 37 °C with shaking. Subsequently, 107 cells of each strain were incubated for 10 min in fresh YPD with or without 100 mm CaCl2. Three independent RNA samples were prepared using TRIzol reagent (Invitrogen) according to the manufacturer's protocol. After DNase treatment, reverse transcriptase reactions were performed. Real-time PCR reactions were performed using StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). PCR thermal cycling conditions included an initial step at 95 °C for 5 min followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. Platinum SYBR green qPCR Supermix (Invitrogen) supplemented with 5 pmol of each primer and 2 μL of the cDNA template in a final volume of 20 μL was used as the reaction mix. Each cDNA sample was analyzed in triplicate with each primer pair. Melting curve analysis was performed at the end of the reaction to confirm the amplification of a single PCR product. Data were normalized to actin cDNA amplified in each set of PCR experiments. The 2−ΔCT method was used to determine the relative expression [38]. The primers utilized in these experiments are listed in Table S1.

Virulence assay

Virulence studies were conducted according to a previously described intranasal inhalation infection model [18] using seven female BALB/c mice (~ 5 weeks old) for each strain. Fungal cells were cultured overnight in 50 mL of YPD medium at 30 °C with shaking, washed twice with NaCl/Pi and suspended in the same buffer. Mice were infected with 105 yeast cells suspended in 50 μL NaCl/Pi and monitored daily for survival. The Mantel–Cox test for survival analysis was performed using prism graphpad software. For determination of fungal burden, mice (= 4) were infected as described above. At days 3, 7, 14 and 19 post-infection, the animals were euthanized, and the lungs and brain were aseptically excised. These tissues were macerated in NaCl/Pi, and after removal of host cell debris the resulting suspensions were plated on YPD for CFU determination. Student's t test was used to determine the statistical significance of differences in CFU counts. The use of animals in this work was performed with approval of the Universidade Federal do Rio Grande do Sul Ethics Committee for Use of Animals. Mice were housed in groups of four kept in filtered top ventilated cages with food and water ad libitum. The animals were cared for according to the Brazilian National Council for Animal Experimentation Control (CONCEA) and Brazilian College of Animal Experimentation (COBEA) guidelines. All efforts to minimize animal suffering were made. Before infection assays, mice were intraperitoneally anesthetized with 100 mg·kg−1 ketamine and 16 mg·kg−1 xylazine. Mice were analyzed twice daily for any signs of suffering, defined by weight loss, weakness or inability to obtain feed or water. At the first signs of suffering, mice were humanely sacrificed.

Acknowledgements

This work was supported by grants from the Brazilian agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and Fundação de Amparo a Pesquisa no Estado do Rio Grande do Sul (FAPERGS). We thank Joseph Heitman and Alex Idnurm for providing pJAF15 and pAI4 and Arturo Casadevall for providing the monoclonal antibody 18B7. We also thank the Electron Microscopy Center of the Federal University of Rio Grande do Sul (CME, UFRGS) for the confocal microscopy analysis and Henrique Biehl for technical assistance.

Ancillary