Role of an Expanded Inositol Transporter Repertoire in Cryptococcus neoformans Sexual Reproduction and Virulence

In previous studies, we discovered a defined medium, modified MS (Murashige and Skoog) medium, which supports the efficient mating of both C. neoformans and Cryptococcus gattii (26). By omission and readdition of components, myo-inositol was shown to be the active component stimulating mating. Standard MS medium contains 100 mg/liter myo-inositol. myo-Inositol was added at different concentrations to MS inositol dropout medium, and a concentration as low as 15 mg/liter myo-inositol was found to stimulate mating. A medium made from V8 juice is classically used for mating assays in C. neoformans. In related studies, the myo-inositol concentration in V8 juice has been measured (38). Based on this information, standard V8 mating medium contains 16.675 mg/liter inositol, and this level is sufficient to support mating in MS medium. The addition of more myo-inositol to V8 medium triggers increased mating filament and spore production (26). These observations suggest that myo-inositol is an active compound in V8 medium that stimulates mating. Compared with mating on V8 medium, mating on MS medium shows significantly increased sporulation efficiency.

Because myo-inositol is the main carbon source in MS medium, we replaced myo-inositol with other carbon sources, including glucose, sucrose, galactose, and glycerol. None of those sugars stimulated mating at the same concentration, indicating that myo-inositol elicits mating in addition to its known ability to serve as a carbon source (Fig. 1A).

We also tested several other inositol isomers, including scyllo-inositol, allo-inositol, muco-inositol, and chiro-inositol. None stimulated mating, indicating that mating stimulation is specific to myo-inositol (Fig. 1B). We have also investigated whether any of these inositol isoforms can compete to inhibit myo-inositol-stimulated mating. By adding 100 mg/liter of each isomer to MS medium containing 50 mg/liter myo-inositol, respectively (yielding a 2:1 ratio), a clear inhibition of mating by allo-, muco-, or chiro-inositol was observed (Fig. 1C). Lowering the myo-inositol concentration to 20 mg/liter in these experiments to yield a 5:1 ratio (at a constant concentration of 100 mg/liter of the inositol isomers) completely blocked mating filament production (Fig. 1D). These results suggest that allo-, muco-, and chiro-inositol antagonize mating by competing with myo-inositol. However, no antagonistic effect was observed by adding scyllo-inositol to MS medium with 20 mg/liter or 50 mg/liter myo-inositol (Fig. 1C and D). Because scyllo-inositol is an inactive form of inositol, these results may reflect an inability of scyllo-inositol to compete with myo-inositol for binding.

Fungal cells can acquire inositol from the following two main sources: internal inositol synthesis from glucose or transport from the environment via inositol transporters. We have identified a large myo-inositol transporter (ITR) gene family in Cryptococcus as encoding potential inositol permeases or sensors based on sequence identity shared with inositol transporter genes identified in S. cerevisiae and S. pombe (26). Because the ITR genes belong to the sugar transporter superfamily, here we identified all sugar transporter homologs from the genomes of the C. neoformans strain H99, S. cerevisiae, and C. albicans, and the phylogenetic relationships among them were compared (see Fig. S1 in the supplemental material). Homologs of Itrs in S. cerevisiae and C. albicans form a cluster in the phylogenetic tree generated. A total of 10 proteins in H99 were considered ITR candidates and included in this cluster. We then compared all of the Itr homologs identified from the following four complete Cryptococcus genome sequences: C. neoformans var. grubii H99 (serotype A, VNI), C. neoformans var. neoformans JEC21 (serotype D, VNIV), C. gattii WM276 (serotype B, VGI), and C. gattii R265 (serotype B, VGII) (Table 1). The Itr homologs can be divided into two distinct groups (Fig. 2A). The group 1 genes are homologs of Itr genes identified from S. cerevisiae and S. pombe and of the Itr1 gene from C. albicans. The group 2 genes are homologs of the HGT19 (orf19.5447) gene from C. albicans, which also exhibits myo-inositol transporter activity based on computational analysis presented in the Candida Genome Database (http://www.candidagenome.org/) but has not been extensively studied (21). The first group may have undergone recent duplication because the numbers of Itr paralogs differ among Cryptococcus strains, while the second group is well conserved among all of the strains that were compared. Based on the similarity among these two groups of Itrs in Cryptococcus, they can be further divided into six different subgroups (Itr1 to -6). Among these, group 1 Itrs can be further divided into subgroups 1, 2, and 3, while group 2 Itrs can be divided into subgroups 4, 5, and 6. Subgroups 4, 5, and 6 are well conserved, and each subgroup has one Itr (Itr4, Itr5, and Itr6) from each Cryptococcus strain. Interestingly, subgroup 3 has expanded in C. neoformans to include four homologs in H99 (ITR3, ITR3A, ITR3B, and ITR3C) and five homologs in JEC21 (ITR3, ITR3A, ITR3B1, ITR3B2, and ITR3C) but only one each for R265 and WM276 (Fig. 2A; Table 1).

When the group 1 and group 2 Itrs from Cryptococcus were compared with related Itr homologs from other fungi, group 1 was found to share a high level of identity with the Itrs in other yeasts, such as S. cerevisiae and S. pombe. There are two Itrs in S. cerevisiae and S. pombe, and both cluster within group 1. In most other ascomycetes with completed genome sequences, only one ITR gene was found in this group, including in hemiascomycetes (C. albicans and Ashbya gossypii) and euascomycetes (Coccidioides immitis and Neurospora crassa) (Fig. 3). Two basidiomycete genomes were used for comparison. The closely related species Tremella mesenterica has only one ITR homolog, while the corn smut fungus Ustilago maydis has two (Fig. 3). One exception in this group is the zygomycete Rhizopus oryzae, which has three Itr paralogs in group 1.

There are three Itrs in the four Cryptococcus strains that belong to group 2. Most fungi analyzed have only one Itr that belongs to group 2, including C. albicans, A. gossypii, C. immitis, N. crassa, U. maydis, and R. oryzae. T. mesenterica has two Itrs that belong to group 2, and interestingly, A. fumigatus has three group 2 representatives.

Based on the comparative genomic analysis, it is clear that the ITR gene family in Cryptococcus has expanded (Fig. 3). Because C. neoformans is an environmental microorganism that associates with a variety of niches, the expansion of the ITR gene family may reflect adaptation to nutrient-limited niches, such as survival on plants or in soil containing plant debris, where free inositol or phytic acid (IP6) is abundant. C. neoformans can utilize inositol as a sole carbon source, which may require this fungus to have more efficient inositol transport systems both to survive in natural habitats and to successfully infect the host and cause infection.

Most of the ITR genes in C. neoformans are located in telomeric regions of the chromosomes, which may have contributed to the expansion of this gene family (Fig. 2B) via recombination-mediated exchange of telomeric and subtelomeric regions between different chromosomes. Interestingly, in JEC21 chromosome 8 and its corresponding chromosome, 14, in H99, four ITR genes are located in telomeric regions, and three of these belong to subgroup 3, the least conserved transporter cluster. In addition, all four genes in subgroup 3 of H99 have very similar intron locations in the genes except ITR3C, which contains an extra exon (Fig. 2C). Comparison of 20-kb flanking sequence regions of the ITR genes in JEC21 and H99 revealed synteny between ITR3A (chromosome 4) and ITR3B2 (chromosome 10) in JEC21, an indication that this genomic region underwent duplication (see Fig. S2 in the supplemental material). These observations suggest that the subgroup 3 ITR genes on these chromosomes may have undergone recent duplications. In this study, we focused further effort on functional analyses of the seven group 1 Itr homologs in H99 to understand their roles in mating and virulence of C. neoformans.

To understand how the Itrs are involved in inositol sensing and transport, quantitative real-time PCR (qRT-PCR) was performed to detect the expression of ITR genes during growth under high-inositol conditions and during mating on MS medium containing inositol. In this study, we designed specific primers for seven ITR genes of H99 that belong to group 1 and that are closely related to well-studied ITR genes in S. cerevisiae and S. pombe. Cells of strain H99 and mating coculture of H99 and KN99a were inoculated on MS medium containing 100 mg/liter inositol. Cells were recovered from the medium surface at 24 h, 48 h, and 72 h postinoculation. Total RNAs were purified, and first-strand cDNAs were synthesized. Each cDNA sample was used as a template to amplify all seven ITR genes using gene-specific primers in a qRT-PCR. Our qRT-PCR results showed that these ITR genes are differentially expressed both during mating and during growth as a single strain in the presence of inositol (Fig. 4). Within the first 24 h of incubation, the mating mixture undergoes cell fusion, and all seven ITR genes were found to be expressed. Increased expression was observed when incubation was continued for 48 h or 72 h. The expression of ITR1 at all time points tested was significantly repressed in the single-strain culture but was not repressed during mating. Meanwhile, ITR1A and ITR2 were constitutively highly induced under both culture conditions, especially ITR1A, which represented the highest expression at all conditions; this suggests that these two may play important roles in inositol uptake. ITR3, ITR3B, and ITR3C are highly induced only during mating but not in the single-strain culture, possibly also consistent with a role in mating. ITR3A was expressed at a low level under both culture conditions, and thus, it may not play a significant role in inositol transport.

Overall, our qRT-PCR results suggest that there is a mating effect on the expression of ITR genes, since several ITR genes are expressed at a much higher level in mating culture than in single-strain culture. Our expression profiles of the inositol transporter gene family also provide an important indication that this gene family may play redundant but distinguishable roles in response to inositol, during both mating coculture and growth in monoculture. It will be of interest to investigate the expression profiles of these ITR genes at different sexual development stages or in other strain backgrounds, such as in ino1 and itr mutant strains.

To understand the potential role of the C. neoformans ITR gene family in inositol uptake, we generated S. cerevisiae itr1 itr2 double mutants lacking both inositol transporters. Such double mutants showed a growth defect in yeast extract-peptone-dextrose (YPD) at 37°C. Full-length cDNAs for all seven ITR genes from the C. neoformans H99 strain were cloned into the yeast expression vector pTH19 under the control of the ADH1 promoter and expressed in an S. cerevisiae itr1 itr2 double mutant strain. The heterologous expression of all seven ITR genes of H99 was confirmed by RT-PCR analyses (data not shown). Growth assays were performed for these S. cerevisiae strains expressing C. neoformans ITR genes, and the potential complementation of the S. cerevisiae itr1 itr2 growth defect phenotype was investigated. Expression of ITR1A and ITR3C fully complemented the growth defect of the S. cerevisiae itr1 itr2 mutant strain, while ITR3A and ITR3B partially rescued the slow growth on YPD medium (Fig. 5). Yeast strains expressing ITR1, ITR2, or ITR3, on the other hand, showed growth rates similar to that of the original S. cerevisiae itr1 itr2 mutant strain, suggesting no obvious complementation. These results suggest that Itr1A and Itr3C are important for inositol uptake, which is consistent with our qRT-PCR results suggesting that both ITR1A and ITR3C are highly expressed in response to the availability of environmental myo-inositol. The ITR genes that did not complement the growth defect of S. cerevisiae itr1 itr2 might be caused by either insufficient inositol uptake or expressed proteins that are unstable or not functional.

Gene deletion mutations were generated by biolistic transformation and homologous recombination for all seven ITR genes in C. neoformans var. grubii in both the α and a mating-type backgrounds. Each individual mutation was not sufficient to block sexual reproduction based on mating assays with both MS and V8 mating media. Normal mating hyphae and spores were observed in all bilateral mutants by mutant mating assays. The exception was that bilateral mating between itr1 mutants exhibited reduced mating hypha production (Fig. 6A). Cell fusion assays showed that itr1 mutants have cell fusion efficiency similar to that of wild-type strains, suggesting that the mating defect could be caused by post-cell fusion events, such as insufficient dikaryotic hypha production or a meiosis defect (Fig. 6C). Because expression of ITR1 in an S. cerevisiae itr1 itr2 mutant strain failed to rescue the growth defect of that yeast strain, Itr1 may not be important for inositol uptake but may play other roles, such as functioning as an inositol sensor. The fact that ITR1 expression is repressed by high concentrations of inositol also suggests that Itr1 may be functional as an inositol sensor. To further understand the potential role of Itr1 in the regulation of other Itrs, qRT-PCR analysis was performed to test the expression of the other six ITR genes in the itr1 mutant background, both as single cultures and during mating, using the same approach as that used for wild-type strains (Fig. 4C and D). Our results revealed that ITR3, ITR3B, and ITR3C are expressed at a much higher level in itr1 single cultures than in the H99 wild-type strain. During mating, most ITR genes were highly expressed, with similar expression patterns in both itr1 mating cultures and wild-type strains, except that ITR3A was highly expressed only in itr1 mating cultures (Fig. 4). These results indicate that the ITR3 subgroup genes may be regulated by Itr1, which further suggests that Itr1 could be an inositol sensor that regulates other Itrs.

We have also tested the development of virulence factors and found that all mutants produced normal amounts of capsule and melanin and had normal growth at 37°C (data not shown). The absence of notable phenotypes may indicate that these transporters could be functionally redundant. Another possibility is that although one Itr could play a major role in inositol uptake, the inositol internal biosynthesis pathway might produce sufficient inositol to compensate for reduced inositol uptake from the environment in itr mutants.

The key enzyme for the myo-inositol biosynthetic pathway, inositol 1-phosphate synthase (Ino1), has been identified, and ino1 mutants have been generated, thereby blocking the inositol internal biosynthesis pathway. Importantly, ino1 mutants failed to grow on medium without myo-inositol (data not shown). No obvious mating defect was observed in an ino1 × ino1 bilateral mating assay using MS medium (Fig. 6A). To study the function of Itrs without the potential interference of the inositol biosynthesis pathway, ino1 with itr1, itr1a, itr2, itr3, itr3a, itr3b, or itr3c double mutants were generated in both of the mating-type backgrounds by genetic crossing. While none of the three major virulence factors were significantly altered in these double mutants, mating filament production and sporulation were clearly reduced in bilateral matings of ino1 itr1 × ino1 itr1 and ino1 itr1a × ino1 itr1a double mutants, suggesting that Itr1 and Itr1A are important for inositol transport or sensing (Fig. 6A). Compared with wild-type strains, the ino1 itr1 double mutant produced sporadic mating hyphae and reduced spore production, while the ino1 itr1a double mutant produced evenly distributed mating filaments but at much lower hyphal density and produced very few spores in MS medium. The sporulation defect in the ino1 itr1a double mutant is much more severe than in the ino1 itr1 double mutant (Fig. 6B). Bilateral mating of other double mutants produced normal mating structures, including dikaryotic hyphae and basidiospores. The identification of the ITR gene family and generation of itr and ino1 single and double mutants provide a valuable system to further study how C. neoformans senses and acquires inositol.

Although none of the itr mutants generated showed significant effects on several key virulence factors, such as capsule and melanin production and growth at 37°C (data not shown), their potential involvement in fungal virulence was assessed because virulence is a complex trait. All seven itr mutants (itr1, itr1a, itr2, itr3, itr3a, itr3b, and itr3c) were examined in a murine inhalation model of systemic C. neoformans infection. Groups of 10 female A/JCr mice were intranasally inoculated with 10⁵ yeast cells from each strain, and animals were monitored daily. As previously demonstrated, all mice infected with the wild-type strain H99α survived between 16 and 22 days (39, 40). Mice infected with the mutants exhibited mortality rates similar to those of mice infected with the wild-type H99 strain, indicating that none of these seven itr mutants are attenuated for virulence of C. neoformans under these conditions (Fig. 7). It is possible that the functional redundancy of this gene family or the existence of the inositol biosynthesis pathway is sufficient for the virulence of each single mutant.

Virulence studies were also conducted in the murine model for an ino1 mutant, an ino1 itr1 double mutant, and an ino1 itr1a double mutant. While the ino1 mutant was fully virulent, the ino1 itr1 double mutant was moderately attenuated for virulence, with statistical significance (P = 0.001) (Fig. 7). The median survival time was delayed from 18 days in H99 to 21.5 days in the ino1 itr1 double mutant. Interestingly, the ino1 itr1a double mutant exhibited an even more severe virulence attenuation (P < 0.0001), and the median survival time was prolonged to 24.5 days (Fig. 7). Evaluation of fungal burdens at the end point of the infection revealed that brains infected by the ino1 itr1a mutant contained 10-fold fewer yeast cells than those with H99 infection, which may indicate that the ino1 itr1a mutant either has a defect in central nervous system (CNS) entry or proliferates at a lower rate in the brain (Fig. 7C). These results indicate that both Itr1 and Itr1A are important for inositol acquisition in C. neoformans and that inositol acquisition is important for fungal virulence.

To investigate how inositol affects Cryptococcus mating, the effect of inositol on the global transcriptional profile was evaluated by conducting whole-genome microarray experiments. Global gene expression profiles during mating under inositol induction conditions were evaluated in both wild-type strains (H99 and KN99a) and ino1 mutants (α ino1 and a ino1). Mating assays were performed in MS medium without inositol and in MS medium with 100 mg/liter inositol for 48 h before cells were collected and RNA was purified. C-3- and C-5-labeled cDNAs were hybridized to JEC21 70-mer whole-genome arrays. When the transcriptomes of wild-type mating mixtures in MS medium were compared with that in MS medium without inositol, only a small set of genes (less than 50 genes) were upregulated over 2-fold by inositol. A large proportion of these genes encode putative proteins (see Fig. S3 in the supplemental material). Among these, two inositol oxygenase genes (180.m00186 and 177.m03138) were upregulated over 10-fold. An α-glucoside transporter gene was also significantly upregulated (Table 2). When the transcriptomes of ino1 mutant mating mixtures in MS medium with and without inositol were compared, over 300 genes were found to be upregulated over 2-fold by inositol. Inositol oxygenase and glucoside transporter genes were also among the genes with the highest induction levels. The pheromone receptor Ste3 was found to be upregulated in both of the experiments. In the ino1 mutant results, more genes involved in inositol sensing and metabolism were found to be upregulated, including a phosphatidylinositol 3-kinase gene and several inositol transporter genes, including ITR1, ITR2, ITR3A, and ITR3B2, indicating that ino1 mutant cells are more responsive to environmental inositol than wild-type cells. Results obtained from such systemic analyses provide rich information for further understanding of inositol acquisition and cellular function in C. neoformans, especially with respect to sexual reproduction and fungal virulence.

The C. neoformans strains used in this study are listed in Table 3. Strains were grown at 30°C in yeast extract-peptone-dextrose (YPD) agar medium and synthetic defined (SD) medium. V8 medium (pH 5.0) was used for mating assays. Modified MS medium was used for mating and sporulation assays and prepared as previously described, with modification (26). Niger seed medium was used to test for melanin production. Dulbecco modified Eagle’s (DME) medium for assessing capsule production was prepared as previously described (54). All other media were prepared as described previously (55–57).

All fungal ITR gene sequences and other sequences were obtained from a variety of databases. All sequences for C. neoformans var. grubii and C. gattii were obtained from the Broad Institute (http://www.broadinstitute.org/), and sequences for C. neoformans var. neoformans were obtained from the TIGR database (http://www.tigr.org/). Related ITR sequences in S. cerevisiae and C. albicans were obtained from the Saccharomyces genome database and Candida genome database. The well-defined ITR genes in S. cerevisiae, S. pombe, and C. albicans were used as queries to identify the initial ITR homologs in the H99 genome via the BLAST algorithm with the default setting, and the output was sorted with top hits ranked by BLASTp scores (58). Phylogeny trees were generated using ClustalX 2.0 (59) and viewed via TreeView X (60).

To identify ITR gene homologs/paralogs in different Cryptococcus species, the identified ITR gene homologs in H99 were used as queries to conduct BLASTn and BLASTx searches in the genome database for C. neoformans strains H99 and JEC21 and C. gattii strains R265 and WM276. BLAST results were parsed using an in-house Perl script. Alignment of the deduced amino acid sequences of all ITR genes was performed with ClustalW (61). The aligned amino acid sequences were imported to PhyML to construct their phylogenetic organization using the maximum likelihood method (62). The generated phylogenetic tree was viewed and edited with FigTree software (http://tree.bio.ed.ac.uk/software/figtree/).

To test how the ITR genes respond to the presence of environmental myo-inositol, we measured the mRNA levels for all seven ITR genes under inositol induction conditions via quantitative real-time PCR (qRT-PCR). Mating assays were performed by mixing H99 and KN99a and coculturing on MS medium containing 100 mg/liter myo-inositol. A single-strain H99 culture was also grown in MS medium. Plates were incubated at room temperature in the dark. Mating mixtures and H99 cultures were collected from agar surfaces at 24 h, 48 h, or 72 h after plating using cell scrapers. Collected cells were washed with distilled water (dH₂O), and pellets were used for total RNA extraction. Total RNAs were extracted using Trizol reagents (Invitrogen) and purified with the Qiagen RNeasy cleanup kit (Qiagen) by following the manufacturer’s instructions. Purified RNAs were quantified using a NanoDrop instrument (Thermo Scientific). The same approach was used to prepare RNAs from itr1 mutant strains.

First-strand cDNAs were synthesized using a SuperScript III cDNA synthesis kit (Invitrogen) by following the instructions provided by the manufacturer. Expression of the ITR genes and the GAPDH (glyceraldehyde-3-phosphate dehydrogenase) gene was analyzed using Brilliant SYBR green quantitative PCR reagents (Stratagene). The gene expression level was normalized using the endogenous control GAPDH gene, and the relative levels were determined using the comparative threshold cycle (C_T) method. Real-time PCRs were performed using a Mx4000 quantitative PCR system (Stratagene), with an initial denaturation step at 95°C for 10 min, followed by 40 PCR cycles in which each cycle consisted of 95°C for 30 s, 60°C for 30 s, and 72°C for 30 s. SYBR green fluorescence emissions were monitored after each cycle. The primers for each gene are listed in Table S1 in the supplemental material. Amplification of specific transcripts was confirmed by melting curve profiles (cooling the sample to 60°C and heating it slowly to 95°C, with measurement of fluorescence) at the end of each PCR. The specificity of the PCR was further verified by subjecting the amplification products to agarose gel electrophoresis and sequencing.

S. cerevisiae itr1 and itr2 single mutants were obtained from Open Biosystems. A genetic cross between S. cerevisiae α itr1 and S. cerevisiae a itr2 was conducted in YPD medium, and the mating mixture was colony purified in yeast nitrogen base (YNB) supplemented with histidine, leucine, and uracil and incubated at 30°C for 2 days. Single colonies were inoculated in sporulation medium and incubated for 5 days, and meiotic products were then isolated by tetrad dissection. Five tetrads were analyzed, and two double mutants (S. cerevisiae itr1 itr2) were confirmed by PCR using gene-specific primers (see Table S1 in the supplemental material) and selected for further study. The growth rate of these double mutants was compared with those of wild-type yeast strain BY4742 and S. cerevisiae itr1 and itr2 single mutants at both 30°C and 37°C.

Seven ITR genes from C. neoformans were cloned into the yeast expression vector pTH19 (63), under the control of the ADH1 promoter. Each of the seven ITR expression plasmids was introduced into an S. cerevisiae itr1 itr2 double mutant background, and colonies were selected on SD-Ura medium. The expression of all of these ITR genes in this yeast heterologous system was verified with RT-PCR using gene-specific primers (see Table S1 in the supplemental material). Yeast strains were tested for growth in YPD at 30°C and 37°C.

Seven inositol transporters have been identified in the genome of H99 that share significant identity with the Itrs of S. cerevisiae (Itr1 and Itr2), S. pombe (Itr1 and Itr2), and C. albicans (Itr1). itr mutants were generated in the congenic C. neoformans serotype A α (H99) and a (KN99a) strains by overlap PCR as previously described (64). The 5′ and 3′ regions of each ITR gene were amplified with primers F1 and R1 and primers F2 and R2, respectively (see Table S1 in the supplemental material for primer sequences), from H99 or KN99a genomic DNA. The dominant selectable markers (Nat^r or Neo^r) were amplified with the M13 primers (M13F/M13R) from plasmid pNATSTM#122 or pJAF1 (65), respectively. Each target gene replacement cassette was generated by overlap PCR with primers F1 and R2 (see Table S1 in the supplemental material). Purified overlap PCR products were precipitated onto 600-µg gold microcarrier beads (0.8 µm [Bioworld Inc.] or 0.6 µm [Bio-Rad]), and strains H99 and KN99a were biolistically transformed as described previously (66). Stable transformants were selected in YPD medium containing nourseothricin (100 mg/liter) or G418 (200 mg/liter). To screen for mutants of each ITR gene, diagnostic PCR was performed by analyzing the 5′ junction of the disrupted mutant alleles with primers F4 and JH8994 (see Table S1 in the supplemental material). Positive transformants identified by PCR screening were further confirmed by Southern blot analysis. The ino1 deletion mutants in the H99 background were generated using the same strategy used to generate itr mutants, as described above. Because the ino1 mutant is myo-inositol autotrophic, all mutants were tested for growth in YNB medium without myo-inositol to verify the deletion of the INO1 gene.

To generate complemented strains of itr1 mutants, a genomic DNA fragment that contains a 1.5-kb upstream region of the ITR1 open reading frame (ORF) and its 500 bp downstream region was amplified with a PCR using primers JH19630 and JH19631. This PCR fragment was fused with the Neo^r selective marker gene at its C-terminal end with an overlap PCR using primers JH19630 and M13R. The overlap PCR product was biolistically transformed in both α itr1 and a itr1 mutant strains. Mating assays were performed to identify the transformants that complemented the itr1 phenotype.

In a mating assay, C. neoformans cells of opposite mating types were mixed and cocultured in V8 or MS agar medium at 25°C in the dark for 10 days, and filamentation was examined by light microscopy. Spore production was also visualized by microscopy and photographed. Basidiospores were also dissected from mating performed in MS medium.

To generate ino1 itr1 double mutants, a mating assay between α ino1::NEO and a itr1::NAT mutants was conducted, and spores were visualized and isolated with an MSM system (Singer Instruments, England). All progeny that grew in YPD with both nourseothricin and G418 were tested in YNB without inositol, and those that failed to grow were subjected to genomic DNA extraction. PCR was used to screen for ITR gene deletion with primers F4/JH8994 and F3/R3 (see Table S1 in the supplemental material). INO1 gene mutations for all potential double mutants were also tested with primers JH18517/JH18518 and confirmed by Southern blotting. The same approach was used to generate double mutants between ino1 and other itr mutants.

To understand the effect of inositol on gene expression profiles during cell development, microarray experiments were performed to monitor the genes regulated during mating by inositol. Overnight cultures of H99 and KN99a were mixed in equal cell amounts and washed with dH₂O once, and cell mixtures were inoculated on MS medium with or without inositol. Cells were collected from MS plates 48 h postinoculation and washed with dH₂O, and total RNA was purified. Total RNAs were extracted using Trizol reagents (Invitrogen) and purified using the Qiagen RNeasy cleanup kit (Qiagen). Cy3- and Cy5-labeled cDNA were generated by incorporating amino-allyl-dUTP during reverse transcription of 5 µg of total RNA as described previously (67) and competitively hybridized to a JEC21 whole-genome array generated previously at Washington University in Saint Louis, MO. After hybridization, arrays were scanned with a GenePix 4000B scanner (Axon Instruments) and analyzed by using GenePix Pro version 4.0 and BRB array tools (developed by Richard Simon and Amy Peng Lam at the National Cancer Institute [http://linus.nci.nih.gov/BRB-ArrayTools.html]) as described previously (68).

Melanin production was assayed by inoculating C. neoformans strains into 2 ml of YPD liquid medium, incubating the culture overnight at 30°C, and spotting 5 µl of each culture on Niger seed agar medium. The agar plates were incubated at 30°C or 37°C for 2 days, and pigmentation of fungal colonies was assessed and photographed. To examine capsule production, 5 µl of overnight cultures was inoculated on DME agar medium and incubated at 30°C for 3 days. The capsule was visualized with India Ink negative staining and observed with a 100× Olympus CX41 equipped with an Infinity digital camera (Olympus).

Yeast strains were grown at 30°C overnight, and cultures were washed twice with 1× phosphate-buffered saline (PBS) and resuspended at a final concentration of 2 × 10⁶ CFU/ml. Groups of 10 female A/JCr mice (NCI-Frederick, MD) were intranasally infected with 10⁵ yeast cells of each strain as previously described (39). Animals that appeared to be moribund or in pain were sacrificed by CO₂ inhalation. Survival data obtained from the murine experiments were statistically analyzed between paired groups using the log rank test and PRISM program 4.0 (GraphPad Software) (P values of <0.01 were considered significant). At the end point of the infection, infected lungs, brains, and spleens were also isolated and homogenized using a homogenizer in 1× PBS buffer. Resuspensions were diluted, 50 µl of each dilution was spread on YPD medium with antibiotics, and colonies were counted after 3 days of incubation at 30°C.