• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of eukcellPermissionsJournals.ASM.orgJournalEC ArticleJournal InfoAuthorsReviewers
Eukaryot Cell. May 2008; 7(5): 848–858.
Published online Mar 28, 2008. doi:  10.1128/EC.00456-07
PMCID: PMC2394967

The Asexual Yeast Candida glabrata Maintains Distinct a and α Haploid Mating Types[down-pointing small open triangle]


The genome of the type strain of Candida glabrata (CBS138, ATCC 2001) contains homologs of most of the genes involved in mating in Saccharomyces cerevisiae, starting with the mating pheromone and receptor genes. Only haploid cells are ever isolated, but C. glabrata strains of both mating types are commonly found, the type strain being MATα and most other strains, such as BG2, being MATa. No sexual cycle has been documented for this species. In order to understand which steps of the mating pathway are defective, we have analyzed the expression of homologs of some of the key genes involved as well as the production of mating pheromones and the organism's sensitivity to artificial pheromones. We show that cells of opposite mating types express both pheromone receptor genes and are insensitive to pheromones. Nonetheless, cells maintain specificity through regulation of the α1 and α2 genes and, more surprisingly, through differential splicing of the a1 transcript.

The comparative study of reproductive cycles in fungi reveals varied and often surprising situations. Even quite closely related species, such as Saccharomyces cerevisiae and Candida glabrata, exhibit very different modes of reproduction; the former is a self-fertile, fully sexual species, while the latter has no apparent sexual phase and reproduces clonally. “Candida” species that lack a teleomorph are distributed over the whole hemiascomycete phylogenetic tree (15), illustrating both the plasticity of the trait observed and the lack of simple relationships between taxonomic nomenclature and phylogenomics.

The type strain of Candida glabrata, CBS138, was sequenced in 2004 through the Genolevures 2 sequencing effort that included four hemiascomycetes of various degrees of relatedness to the model yeast S. cerevisiae (6). This confirmed that asexual C. glabrata is phylogenetically very distant from Candida albicans and close to S. cerevisiae. Because of the relatedness of C. glabrata to the extensively studied model yeast, we have examined whether known mating pathways from S. cerevisiae exist in C. glabrata.

In S. cerevisiae, the MAT locus encodes master transcription factors that regulate three categories of genes: a-specific genes, α-specific genes, and haploidy-specific genes, involved in pheromone expression and sensing, the transduction cascade, and the inhibition of meiosis, respectively (12, 13). The S. cerevisiae genome also contains two additional loci, HML and HMR, harboring, respectively, the α and a genetic information, which are transcriptionally silenced.

Many genes involved in mating are conserved between the two species (26, 31); in particular, the MAT cassettes and the MAT-like cassettes (HMLα and HMRa homologs) have the same overall structure and encode the same proteins in both yeasts. A notable difference is that in S. cerevisiae, the three cassettes are on the same chromosome, whereas in C. glabrata, MAT and the HML-like cassette are on chromosome II and the HMR-like cassette is on chromosome V (7). Cassettes are also inverted with respect to the telomeres between the two species, and examination of other yeasts has shown that the inversion took place in S. cerevisiae (20).

In this work, we show that there are several apparent defects in the mating pathway of C. glabrata compared to S. cerevisiae's. Cells, nonetheless, maintain some mating type identity through differential expression of α genes and, more surprisingly, differential splicing of the a1 transcript.


Yeast media.

YPD medium was prepared with 1% (wt/vol) Bacto peptone (Difco), 1% (wt/vol) Bacto yeast extract (Difco), 2% (wt/vol) glucose, and 2% (wt/vol) Bacto agar, when needed. Selective SC medium was prepared with 2% (wt/vol) glucose, 2% (wt/vol) Bacto agar, and 0.67% yeast nitrogen base without amino acids (Difco) but supplemented with a mix of all amino acids, uracil, and adenine, with omission of nutrients whose prototrophy was selected for, when needed. WO medium was prepared with 0.67% (wt/vol) yeast nitrogen base without amino acids (Difco), 2% (wt/vol) glucose, and 2% (wt/vol) Bacto agar. SPO medium contained 0.25% (wt/vol) Bacto yeast extract (Difco), 1% KO acetate, 0.1% (wt/vol) glucose, and 2% (wt/vol) Bacto agar. Gorodkowa medium was prepared with 1% (wt/vol) Bacto peptone (Difco), 0.5% NaCl, 0.1% (wt/vol) glucose, and 2% (wt/vol) Bacto agar. SLAD medium was prepared with 0.17% (wt/vol) yeast nitrogen base without amino acids (Difco), 2% (wt/vol) glucose, and 2% (wt/vol) Bacto agar. McClary's acetate medium was prepared with 0.25% (wt/vol) Bacto yeast extract (Difco), 0.98% (wt/vol) potassium acetate, 0.07% (wt/vol) MgSO4, 0.1% (wt/vol) glucose, and 1.5% (wt/vol) Bacto agar. V8 medium was prepared with 20% (vol/vol) V8 juice, 0.3% (wt/vol) CaCO3, and 1.5% (wt/vol) Bacto agar, and its pH was adjusted to 6.8 or 5.5 with NaOH. Bacto potato dextrose agar (Difco) medium contained 20% (vol/vol) potato infusion, 2% (wt/vol) glucose, and 1.5% Bacto agar. Malt (Difco) medium contained 3% (wt/vol) malt extract and 1.5% (wt/vol) Bacto agar.

Yeast strains.

Strains are described in Table Table1.1. Deletion mutants of C. glabrata were obtained by cotransformation of the split-marker vectors (8) pKA, pAN (for the construction of the ura3 strain HM100a from wild-type CBS138), pUR, and pRA (for the construction of HMR and sst2 deletants of ura3 strains), in which appropriate PCR fragments upstream and downstream of the targeted gene were cloned into the BamHI/KpnI sites. Constructions were controlled by Southern blot analysis (data not shown).

Descriptions of strains used in this work

Mating assays.

Cells from strains of opposite mating types were grown as patches on complete medium plates, collected, and mixed in a patch on plates with various media. After 4 days at 30°C, cells were collected and streaked on WO medium plates and incubated at 30°C. Plates were examined regularly during 1 week, and potential diploid cells were streaked a second time on WO medium plates.

RNA extraction.

RNA from C. glabrata and S. cerevisiae was prepared as described previously (3), by using hot phenol extraction after glass bead cell lysis of mid-log-phase cultures.


Four micrograms of total RNA was used per reaction. To avoid any remaining DNA in the preparation, RNAs were first treated with DNase I (RQ1 RNase-free DNase; Promega) and extracted with phenol-chloroform before being subjected to reverse transcription. Reverse transcriptase (RT) Superscript II (catalog no. 18064-014; Invitrogen) was used according to the manufacturer's recommendations. RNasin (Promega) was added to all reaction mixtures to avoid RNA degradation.


Quantitative RT-PCR (qRT-PCR) experiments were performed using an Abgene ABsolute MAX 2-Step qRT-PCR Sybr green kit. The first step of DNase I and phenol-chloroform extraction was performed as described for the RT-PCR experiments.

DNase-free total RNA (0.8 μg) was used for reverse transcription to obtain cDNA. qPCR was then performed in triplicate on 10-fold dilutions of the cDNA solution. Standard curves were obtained by PCR with serial dilutions of DNA of known concentrations. In each well, 12.5 μl of Sybr green was added to 5 μl (~10 ng) of cDNA or DNA and 8 pmol of the two specific primers (Table (Table2)2) in a final volume of 25 μl. Specific primers for each gene were designed using the Beacon Designer software, v. 4.

Description of oligonucleotide primers used in this work

The PCR program was 14 min at 95°C for initial denaturation and enzyme activation followed by 40 cycles of denaturation (30 s at 95°C) and of hybridization/elongation (30 s at 55°C), and a final step of 1 min at 95°C. The melting curve started at 55°C with 0.5°C increments every 10 s for 80 cycles. qPCRs were run on an iQ5 real-time PCR detection system (Bio-Rad) and analyzed with the iCycler software.

Sensitivity to pheromones and pheromone expression.

Synthetic pheromones (Eurogentec SA, Seraing, Belgium, and NeoMPS SA, Strasbourg, France) of C. glabrata were synthesized as explained below and in the legend to Fig. Fig.4.4. For sensitivity assays, 50 μl of a 200-μg/ml solution of C. glabrata a-factor or α-factor or of S. cerevisiae α-factor (Sigma-Aldrich Inc., St. Louis, MO) was spotted on plates with SC medium without tryptophan (SC-W) or SC medium without adenine (SC-ade) and, when dry, covered with a cell lawn containing 5·104 cells.

FIG. 4.
Mating pheromones. (A) On the left, alignments of sequences of precursors of the a-factors of S. cerevisiae and C. glabrata are shown. The box indicates peptides resulting from precursor proteolysis. On the right, the synthesized form of the a-factor ...

For pheromone expression experiments, cells from C. glabrata or S. cerevisiae were grown in YPD overnight at 30°C until the end of log phase. Cultures were centrifuged, and supernatants were filtered on 0.22-μm nitrocellulose. Fifty microliters of filtered medium was then spotted directly on SC medium-W or SC medium-ade plates. After the spot was allowed to dry, 5·104 cells of the tester strain were spread on the plate.

Pictures were taken after 2 days of growth at 30°C.


There is no report of the mating of C. glabrata in the literature. We have, ourselves, tried mating MATa strains descending from BG2 with MATα strains descending from CBS138 with different combinations of auxotrophic markers in order to select prototrophic diploid cells, as shown in Table Table3.3. Mating experiments were performed as indicated in Materials and Methods on different media: SPO, V8 at two different pHs, Gorodkowa, McClary's acetate, SLAD, malt, Bacto potato dextrose agar, and WO. Subsequent streaking on minimal media never allowed the isolation of prototrophic cells. Since no difference was observed between results with different media and since C. glabrata can in fact be considered to be part of the Saccharomyces species complex (6), we decided to examine the functionality of genes involved in mating under standard conditions used for S. cerevisiae.

Mating experiments performeda

Expression of mating type-related genes.

In order to address the question of mating type expression by haploid C. glabrata cells, we examined the expression of CAGL0E00341g, CAGL0B01243g, and CAGL0B01265g, homologs of the key regulator genes a1, α1, and α2, respectively, at the MAT locus. We have not examined the a2 gene given the absence of any known role for a2 in either haploid or diploid cells in S. cerevisiae and the fact that the gene has no start codon in C. glabrata. It has been speculated that a2 could be a pseudogene of an ancestral gene common to the S. cerevisiae-S. glabrata branch (10). We also examined the homologs of the mating pheromone receptor-encoding genes that are involved in the initial steps of cellular fusion during mating, CAGL0K12430g, a homolog of STE2 (YFL026W) encoding the α-factor receptor, and CAGL0M08184g, a homolog of STE3 (YKL178C) encoding the a-factor receptor (7). Finally, we examined the expression of the homolog of HO (CAGL0G05423g). In S. cerevisiae, HO is necessary for the completion of the sexual cycle in clonal populations by inducing the formation of cells of opposite mating types within a clone. Since C. glabrata infections are usually monoclonal (1), sexual reproduction may depend on the ability of some cells to switch mating types and fuse with related cells.

We performed RT-PCR on a1 and α1 genes in three strains of C. glabrata of the α mating type, CBS138 (sequenced) and M4 and M5 (two isolates from patients), and in three strains of the a mating type, BG2 (a commonly used strain in laboratories) and M2 and M3 (two isolates from patients). For the a1 genes that contain two introns in S. cerevisiae and C. glabrata (20, 26), primers amplified a fragment containing the first intron in S. cerevisiae and both introns in C. glabrata. In the three independent C. glabrata strains of each mating type examined by RT-PCR (Fig. (Fig.1A),1A), the MATα gene α1 is expressed in all MATα strains and not in MATa strains. In contrast, the a1 gene is expressed in both MATa and MATα strains. In the latter, this transcript must arise from the expression of the HMRa locus, since the MAT and HML loci contain α-type information. We conclude that HMRa on chromosome V is not silenced in C. glabrata. Thus, the type strain expresses genes from both mating types. Reasoning that this might contribute to its apparent infertility, we constructed a CBS138 MATα strain with the HMRa locus deleted, strain HM102a. Further analyses include this strain. Control experiments with S. cerevisiae show that, as expected, there is clear-cut mating type-specific expression of the a1 and α1 genes (Fig. (Fig.1B)1B) (13). The size of the major PCR band on cDNA from the a1 transcript reveals the presence or absence of the splicing of its intron (5). In S. cerevisiae, the major band is smaller than the PCR band from genomic DNA, demonstrating splicing of the intron. In C. glabrata, despite the presence of multiple bands, the major PCR bands on a1 cDNAs have the same size as the PCR band from genomic DNA, suggesting a splicing defect. More-explicit experiments on a1 splicing are described below.

FIG. 1.
RT-PCR experiments with MATa and MATα cassettes of C. glabrata (A) and S. cerevisiae (B). Mating type is shown at the top, and the gene studied is shown above the gels; for C. glabrata, CgACT1 is CAGL0K12694g, Cga1 is CAGL0E00341g, and Cgα1 ...

In the qRT-PCR experiments (Fig. (Fig.2),2), we examined the expression of a1, α1, and α2 and of STE2, STE3, and HO in S. cerevisiae and their homologs in C. glabrata. As a standard, we used the actin transcript, estimated to occur at around 10 copies per cell in S. cerevisiae (21). Standard S. cerevisiae strains are ho mutants, so we included as a control the J5 strain, which contains a wild-type HO gene and α-type information at MAT, HML, and HMR.

FIG. 2.
qRT-PCR experiments with several genes related to mating. Histograms of transcript quantity per cell relative to the quantity of the actin transcript (fixed at 10 transcripts per cell [21]) in C. glabrata (A) and S. cerevisiae (B). Names of genes are ...

In C. glabrata, a1 is expressed at similar levels in both the MATa strain and the MATα strain, confirming the results of the above-described RT-PCR experiments. As expected, deletion of the HMR cassette in strain HM102a results in the absence of the a1 transcript (Fig. (Fig.2A).2A). Neither STE2 nor STE3 homologs display mating type specificity, as they are expressed at similar levels in MATa and MATα strains. In contrast, the expression of α1 and α2 is mating type specific in C. glabrata; they are expressed, respectively, 500-fold and 1,200-fold more in MATα cells than in MATa cells. Deletion of the HMR cassette does not interfere with the expression of these two genes. Control experiments show that a1, α1, α2, STE2, and STE3 exhibit mating type-specific expression in S. cerevisiae (Fig. (Fig.2B).2B). The HO gene is transcribed at similar levels in all strains of C. glabrata examined and at levels 20-fold lower than in haploid S. cerevisiae cells, where around one transcript per 10 cells is observed.

Splicing of a1 introns.

Previous experiments led us to hypothesize that a1 may be nonfunctional in C. glabrata; we observed concomitant expression of a1, α1, and α2 in MATα cells, a situation that could not occur in S. cerevisiae because the a12 heterodimer represses the expression of α1 and the deletion of HMRa in MATα cells has no effect on the expression of α1, α2, STE2, and STE3. The fact that we observed unspliced forms of a1 prompted us to examine this in more detail. Indeed, in S. cerevisiae, when the a1 transcript is not spliced, the a1 protein is not produced (22). The a1 gene from C. glabrata contains two introns, as in S. cerevisiae, and four primer pairs were designed for each species in order to analyze the splicing of both introns (Fig. (Fig.3A3A).

FIG. 3.
Splicing of a1 introns in S. cerevisiae and C. glabrata. (A) In representations of genes, white boxes are exons (e1 is exon1, e2 is exon2, and e3 is exon3) and gray boxes are introns (i1 is intron1 and i2 is intron2); their sizes are indicated below the ...

As shown on Fig. Fig.3B,3B, in S. cerevisiae MATa cells (FY23), the size of the only cDNA band amplified around intron 1 corresponds to the spliced transcript; i.e., splicing is total, whereas for intron 2, some unspliced forms remain, although most transcripts are spliced. In C. glabrata cells, for both introns, the major cDNA band has the same size as the genomic DNA band. Minor cDNA bands corresponding to the predicted spliced forms of 84 and 70 bp are observed only in MATa cells (BG2, M2, M3). In C. glabrata MATα cells (CBS138, M4, M5), no spliced form from either intron can be detected. Amplification of the fragment of the transcript that contains both introns with the external primers (Fig. (Fig.3C)3C) shows that, in BG2 MATa cells, the doubly spliced form of the a1 transcript exists, although unspliced transcripts are more abundant and transcripts with only one intron spliced also exist. In the latter case, we cannot distinguish between transcripts spliced for intron 1 or for intron 2 because there is only 2 nucleotides' difference in size between the two introns and therefore between the two transcripts. The band of larger size than that of the unspliced a1 transcript is assumed to be a PCR artifact, as its presence is variable and as larger bands are also sometimes detected in the S. cerevisiae experiments.

In conclusion, in C. glabrata, the a1 transcript is partially spliced in the MATa strains examined, while it is not spliced at detectable levels in the MATα strains examined. Translation of the unspliced transcripts cannot give rise to a functional protein because there are in-frame stop codons in the first intron.

Response to pheromones.

Since C. glabrata MATa and MATα strains express both a- and α-factor receptor genes (Fig. (Fig.2),2), we wondered whether cells were sensitive to mating pheromones. The C. glabrata genome contains two genes predicted to encode pheromone precursors. The predicted translation product of CAGL0C01919g contains a 12-amino-acid long peptide similar to the a-factor of S. cerevisiae (Fig. (Fig.4A)4A) (7, 19). The predicted translation product of CAGL0H03135g contains three 13-amino-acid-long peptides similar to the α-factor of S. cerevisiae (Fig. (Fig.4B)4B) (7, 24). Two peptides are identical, but one differs by 2 amino acids. We used the two different sequences as putative α-factors for C. glabrata, form A and form B. For the synthesis of the artificial pheromones, we assumed that posttranslational modifications that occur in S. cerevisiae also occur in C. glabrata since the genes involved in these processes are conserved (7). These include precursor proteolysis of both pheromones and farnesylation and methylation at the C terminus of the a-factor (4).

We tested C. glabrata mating pheromones on S. cerevisiae cells, and we included sst1 and sst2 deletion mutants of S. cerevisiae because they exhibit greater sensitivity to pheromones (2). The SST1 gene, whose standard name is BAR1, encodes an aspartyl protease secreted into the periplasmic space of MATa cells that inactivates α factor. Thus, its action is mating type specific, as shown below. SST2 encodes a G protein regulator that is required to prevent receptor-independent signaling of the mating pathway. We constructed sst2 mutants (deletion of CAGL0H00374g) in C. glabrata cells of both mating types to check for increased sensitivity.

Null sst2 mutants of S. cerevisiae shmoo constitutively, even in absence of cells of the opposite mating type, and this results in a longer generation time compared to that of the wild type. We observed no constitutive shmooing of sst2 mutants of C. glabrata when grown in YPD or SC medium.

The effect of synthetic pheromones on C. glabrata and S. cerevisiae cells is shown in Fig. Fig.5.5. Drops of synthetic pheromones were put on minimal medium agar plates, and cells were spread after absorption. As shown, form A of the C. glabrata α-factor is active on wild-type, ss1, and sst2 MATa S. cerevisiae cells. Form B is active only on the hypersensitive S. cerevisiae sst2 cells. Neither form is active on either wild-type or sst2 C. glabrata MATa cells. The synthetic a-factor deduced from the genome of C. glabrata has no activity on wild-type S. cerevisiae MATα cells or on sst1 cells but has a very strong effect on sst2 cells. It has no observable activity on either wild-type or sst2 C. glabrata MATα cells. The strain of C. glabrata with HMRa deleted does not respond to pheromones any more than the original CBS138 strain. Thus, C. glabrata cells are potentially able to produce active pheromones to which S. cerevisiae cells are sensitive but are themselves insensitive to them. Interspecific sensitivity to α-factor between S. cerevisiae, Saccharomyces kluyveri, and Saccharomyces exiguus has been described previously (14, 18), but the peptidic sequences of α-factors from C. glabrata are more diverged from S. cerevisiae than are the α-factors from S. kluyveri and S. exiguus (5 to 6 amino acids conserved out of 13 for S. cerevisiae versus 7 to 9 amino acids conserved out of 13 for S. kluyveri and S. exiguus). Despite this divergence, the sensitivity of S. cerevisiae to α-factor from C. glabrata is mediated by the Ste2 receptor, as ste2 mutants of S. cerevisiae do not respond to it or to α-factor from S. cerevisiae (not shown). We then asked whether C. glabrata cells actually produce pheromones, using S. cerevisiae as a test.

FIG. 5.
Effect of synthetic pheromones on S. cerevisiae (A) and C. glabrata (B) cells. Cell type is indicated at the top, and pheromones are indicated on the left.

Pheromone production by C. glabrata.

To test pheromone production, C. glabrata and S. cerevisiae cell lawns were plated on top of drops of filtered supernatants from late-log-phase cultures of C. glabrata and S. cerevisiae strains, as described above. The results (Fig. (Fig.6A)6A) show that S. cerevisiae sst2 mutants exhibit sensitivity to supernatants of cultures from S. cerevisiae cells of opposite mating types. Due to the limited concentration of pheromones in the supernatant, no effect is observed on wild-type cells. In contrast, no supernatant from C. glabrata has a mating type-specific inhibitory effect on wild-type or sst2 mutants of S. cerevisiae. However, this assay reveals the inhibitory effect of the CBS138 supernatant on all cell types (see below).

FIG. 6.
Effect of culture supernatant on S. cerevisiae (A) and C. glabrata (B) cells. Cell type is indicated at the top, and the nature of the supernatant is indicated on the left.

Figure Figure6B6B shows that C. glabrata cells are insensitive to the culture supernatants of both S. cerevisiae MATa and MATα strains and C. glabrata MATa strains (BG2). The inhibition of the growth of MATa cells by the C. glabrata MATα CBS138 supernatant cannot be taken as a mating type-specific effect, as this supernatant inhibits the growth of mating type cells of C. glabrata of both types (not shown) (but not of mating type cells of itself or derived strains) and of mating type cells of S. cerevisiae of both types, as explained above. Thus, strain CBS138 produces an inhibitor of the growth of other strains as the “killer” strains of other yeast species (17). This effect is observed when supernatants of rich medium cultures but not minimal medium cultures are used (not shown).

We then extended this growth inhibition assay to detect pheromone production to a collection of C. glabrata clinical isolates that we characterized independently, of which 80% are MATa (C. Hennequin, H. Muller, B. Dujon, and C. Fairhead, unpublished data). This collection contains 182 isolates that were grown in 96-well plates. Supernatants were filtered and spotted before sst2 S. cerevisiae tester cells were plated. No isolate was found to produce pheromones specifically inhibiting the MATa or MATα sst2 strains of S. cerevisiae, but three unrelated strains (two MATa strains and one MATα strain) were found to express the same general inhibitor as CBS138 (not shown).


The reason why sexual reproduction is so common in living species remains debated, and many species in which clonal propagation is possible get by without it. This is particularly true in the fungal kingdom, where lack of an observed sexual stage is often associated with pathogenic interactions with humans or plants or with obligatory symbiosis with plant species. In this work we have examined the reasons why no teleomorph has been observed in the hemiascomycete C. glabrata despite the presence of genes homologous to those known to be involved in mating (7, 26, 31). We have observed several differences in the mating pathway from that of S. cerevisiae.

First, our results show that α1 and α2 are expressed in a mating type-specific manner, in contrast to a1, which is expressed in both mating types, because of a lack of silencing at HMRa, interestingly situated on a different chromosome than MAT and HMLα. Sir1p is responsible for silencing HML/HMR loci in S. cerevisiae; thus, perhaps the absence of a SIR1 homolog in C. glabrata (7) explains this phenomenon. Nonetheless, HMLα is silenced in MATa strains, in which we do not observe the expression of the α genes.

Even though the expression of the a1 gene is not subject to mating type specificity, experiments suggest that a1 is not functional in MATα haploid cells. Moreover, the expression of a functional a1 gene in MATα cells would logically lead them to behave as diploid cells, with the risk of their trying to undergo meiosis, a possibly fatal event for a haploid cell. Even though we do not know which conditions could induce meiosis in C. glabrata, it is likely that they would be met at some point by cells. This hypothesis is in accordance with the fact that only haploid cells are ever isolated, so that if diploid cells are formed, they must sporulate readily. In S. cerevisiae diploid cells, the coexpression of a1 and α2 leads to the repression of α1 in addition to other haploidy genes. The expression of the three transcription factors a1, α1, and α2 simultaneously in C. glabrata MATα cells thus leads to the hypothesis that a1 is not functional. Additionally, the observation that the HMRaΔ MATα strain still expresses α1, STE2, and STE3, like the wild-type MATα strain, strengthens this hypothesis. The fact that we found that the splicing of the a1 transcript is detectable only in MATa cells can explain these observations. In S. cerevisiae, it has been shown that the unspliced form of a1 is not functional in diploid cells (22). In haploid cells, there is no known role for a1, the MATa phenotype being the default one. Thus, a partial failure to splice the a1 transcript in MATa cells is not expected to have any effect. In MATα cells, the lack of splicing of the a1 transcript may be necessary to compensate for the leakage of expression from HMR by functionally inactivating the transcript from the rogue gene. Alternatively, complete silencing may not be necessary because a1 expression is regulated in some other way, such as splicing. Since there is no sequence difference between MATa1 and HMRa1 that would explain differences in splicing efficacy (6, 26), the differential splicing must originate either from a general splicing defect in MATα strains or from a mating type-specific mechanism, with both alternatives leading to the inactivation of a1.

The pheromone receptor genes STE2 and STE3 (CAGL0K12430g and CAGL0M08184g) are expressed in both cell types in C. glabrata, while in S. cerevisiae, STE3 is highly regulated and STE2 less tightly regulated. In S. cerevisiae MATα cells, STE3 is activated by α1. In C. glabrata, the basal level observed in MATa cells is higher than in S. cerevisiae (100-fold, if we assume that the actin transcript is expressed at similar levels in both species) and there is no activation by α1. In S. cerevisiae MATα cells, STE2 is repressed by the binding of Mcm1 and α2 on the a-specific gene operator. Putative binding sites for Mcm1 and α2 are found upstream of the homolog of STE2 in C. glabrata, but their spacing is different from S. cerevisiae's. This suggests that the expression of this gene could be mating type specific under some as-yet-undefined conditions (28; B. Tuch, personal communication).

The expression of both receptors in C. glabrata does not make the cells sensitive to both pheromones, as our tests with artificial pheromones show. Furthermore, there is no pheromone production detectable in standard laboratory culture in our primitive test. Nonetheless, pheromones synthesized using the genome sequences of C. glabrata are active on S. cerevisiae cells. The genes encoding factors that modify the pheromones are also found in the genome of C. glabrata. Thus, C. glabrata has retained its ability to encode active pheromones but does not respond to them, possibly because the signal cascade leading to the G1-S cell cycle arrest in S. cerevisiae does not properly operate in C. glabrata. This is consistent with the absence of the shmooing of sst2 mutants of C. glabrata. Alternative explanations of the absence of a pheromone response in C. glabrata are that pheromones are expressed under some unknown conditions and that some additional regulation of the signal cascade exists in C. glabrata, allowing for a mating type-specific response to the opposite mating pheromone.

We also show that the HO gene is transcribed in C. glabrata. In S. cerevisiae, the HO endonuclease drives mating type switching by initiating a double-strand break at the MAT locus. This mechanism allows the creation of cells of opposite mating types in clonal populations, thus facilitating mating under conditions of isolation. Partially degenerate HO sites in C. glabrata (7) that match a previously published consensus sequence for the endonuclease from S. cerevisiae (TNNNYGCG/ANC/AANT/G) can be identified (23). Indeed, HO from S. cerevisiae cuts the MATa site but not the MATα site of C. glabrata (our unpublished results). In vivo mating type switching events have been reported to occur in C. glabrata (1, 16), always from MATa to MATα. In Kluyveromyces lactis, where the loci are organized as in C. glabrata (7), switching is also more likely from a to α than the other way around (11). These observations lead to two hypotheses. The first is that HO from C. glabrata is active and has the same specificity as the one from S. cerevisiae, so that the endonuclease is able to recognize and cut only MATa sites, not MATα sites. In this case, activity must be infrequent to account for the three- to fourfold predominance of MATa strains (26; C. Hennequin, H. Muller, B. Dujon, and C. Fairhead, unpublished data). This predominance could also be explained by a better fitness of MATa strains than that of MATα strains, a possibility that needs to be tested. The second hypothesis is that HO is inactive in C. glabrata, and switching from MATa to MATα is an uncontrolled event that is more likely than MATα-to-MATa switching because of the chromosomal configuration of the cassettes. In the latter case, there would be no active control of switching and therefore potential mating, but with the first hypothesis, the possibility of a yet-unproved active control of switching is open, perhaps in the human body, as suggested previously (25).

In conclusion, although our experiments to mate C. glabrata have failed, the facts that (i) so many genes of the mating pathway remain in the genome, (ii) the modes of splicing of the a1 transcript differ between mating types (this study), and (iii) this would have no cause to arise unless some mating/meiosis pathways are at least partially active indicate that it is possible that C. glabrata cells mate under some still-to-be-discovered conditions, such as in the human body. This could be followed by a diploid phase that may be transient, allowing for meiotic or pseudo-meiotic recombination to occur, as in C. albicans (25).


We thank the Génolevures consortium and members of the Unité de Génétique Moléculaire des Levures, in particular, Bertrand Llorente and Alain Jacquier, for stimulating discussions. We thank Brian Tuch for critical reading of the manuscript and sharing of unpublished information.

This work was supported in part by ACI grants MIC0314, GDR 2354, and ANR GENARISE. B.D. is a member of the Institut Universitaire de France. H.M. is a recipient of a doctoral fellowship of the Ministère de la Recherche through University Paris 6 and of a final-year fellowship from the Fondation pour la Recherche Médicale.


[down-pointing small open triangle]Published ahead of print on 28 March 2008.


1. Brockert, P. J., S. A. Lachke, T. Srikantha, C. Pujol, R. Galask, and D. R. Soll. 2003. Phenotypic switching and mating type switching of Candida glabrata at sites of colonization. Infect. Immun. 717109-7118. [PMC free article] [PubMed]
2. Chan, R. K., and C. A. Otte. 1982. Physiological characterization of Saccharomyces cerevisiae mutants supersensitive to G1 arrest by a factor and α factor pheromones. Mol. Cell. Biol. 221-29. [PMC free article] [PubMed]
3. Chanfreau, G., G. Rotondo, P. Legrain, and A. Jacquier. 1998. Processing of a dicistronic small nucleolar RNA precursor by the RNA endonuclease Rnt1. EMBO J. 173726-3737. [PMC free article] [PubMed]
4. Chen, P., S. K. Sapperstein, J. D. Choi, and S. Michaelis. 1997. Biogenesis of the Saccharomyces cerevisiae mating pheromone a-factor. J. Cell Biol. 136251-269. [PMC free article] [PubMed]
5. Cormack, B. P., and S. Falkow. 1999. Efficient homologous and illegitimate recombination in the opportunistic yeast pathogen Candida glabrata. Genetics 151979-987. [PMC free article] [PubMed]
6. Dujon, B., D. Sherman, G. Fischer, P. Durrens, S. Casaregola, I. Lafontaine, J. De Montigny, C. Marck, C. Neuveglise, E. Talla, N. Goffard, L. Frangeul, M. Aigle, V. Anthouard, A. Babour, V. Barbe, S. Barnay, S. Blanchin, J. M. Beckerich, E. Beyne, C. Bleykasten, A. Boisrame, J. Boyer, L. Cattolico, F. Confanioleri, A. De Daruvar, L. Despons, E. Fabre, C. Fairhead, H. Ferry-Dumazet, A. Groppi, F. Hantraye, C. Hennequin, N. Jauniaux, P. Joyet, R. Kachouri, A. Kerrest, R. Koszul, M. Lemaire, I. Lesur, L. Ma, H. Muller, J. M. Nicaud, M. Nikolski, S. Oztas, O. Ozier-Kalogeropoulos, S. Pellenz, S. Potier, G. F. Richard, M. L. Straub, A. Suleau, D. Swennen, F. Tekaia, M. Wesolowski-Louvel, E. Westhof, B. Wirth, M. Zeniou-Meyer, I. Zivanovic, M. Bolotin-Fukuhara, A. Thierry, C. Bouchier, B. Caudron, C. Scarpelli, C. Gaillardin, J. Weissenbach, P. Wincker, and J. L. Souciet. 2004. Genome evolution in yeasts. Nature 43035-44. [PubMed]
7. Fabre, E., H. Muller, P. Therizols, I. Lafontaine, B. Dujon, and C. Fairhead. 2005. Comparative genomics in hemiascomycete yeasts: evolution of sex, silencing, and subtelomeres. Mol. Biol. Evol. 22856-873. [PubMed]
8. Fairhead, C., B. Llorente, F. Denis, M. Soler, and B. Dujon. 1996. New vectors for combinatorial deletions in yeast chromosomes and for gap-repair cloning using ‘split-marker’ recombination. Yeast (Chichester) 121439-1457. [PubMed]
9. Fidel, P. L., Jr., J. L. Cutright, L. Tait, and J. D. Sobel. 1996. A murine model of Candida glabrata vaginitis. J. Infect. Dis. 173425-431. [PubMed]
10. Fraser, J. A., and J. Heitman. 2004. Evolution of fungal sex chromosomes. Mol. Microbiol. 51299-306. [PubMed]
11. Herman, A., and H. Roman. 1966. Allele specific determinants of homothallism in Saccharomyces lactis. Genetics 53727-740. [PMC free article] [PubMed]
12. Herskowitz, I. 1988. Life cycle of the budding yeast Saccharomyces cerevisiae. Microbiol. Rev. 52536-553. [PMC free article] [PubMed]
13. Herskowitz, I., J. Rine, and J. N. Strathern. 1992. Mating-type determination and mating-type interconversion in Saccharomyces cerevisiae, p. 583-656. In E. W. Jones, J. R. Pringle, and J. R. Broach (ed.), The molecular and cellular biology of the yeast Saccharomyces: gene expression, vol. II. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
14. Hisatomi, T., N. Yanagishima, A. Sakurai, and H. Kobayashi. 1988. Interspecific actions of alpha mating pheromones on the a mating-type cells of three Saccharomyces yeasts. Curr. Genet. 1325-27. [PubMed]
15. Kurtzman, C. P., and J. W. Fell (ed.). 2000. The yeasts: a taxonomic study, 4th ed. Elsevier Science, Amsterdam, The Netherlands.
16. Lin, C. Y., Y. C. Chen, H. J. Lo, K. W. Chen, and S. Y. Li. 2007. Assessment of Candida glabrata strain relatedness by pulsed-field gel electrophoresis and multilocus sequence typing. J. Clin. Microbiol. 452452-2459. [PMC free article] [PubMed]
17. Marquina, D., A. Santos, and J. M. Peinado. 2002. Biology of killer yeasts. Int. Microbiol. 565-71. [PubMed]
18. McCullough, J., and I. Herskowitz. 1979. Mating pheromones of Saccharomyces kluyveri: pheromone interactions between Saccharomyces kluyveri and Saccharomyces cerevisiae. J. Bacteriol. 138146-154. [PMC free article] [PubMed]
19. Michaelis, S., and I. Herskowitz. 1988. The a-factor pheromone of Saccharomyces cerevisiae is essential for mating. Mol. Cell. Biol. 81309-1318. [PMC free article] [PubMed]
20. Muller, H., C. Hennequin, B. Dujon, and C. Fairhead. 2007. Ascomycetes: the Candida MAT locus: comparing MAT in the genomes of hemiascomycetous yeasts, p. 247-266. In J. Heitman, J. W. Kronstad, J. W. Taylor, and L. A. Casselton (ed.), Sex in fungi: molecular determination and evolutionary implications. ASM Press, Washington, DC.
21. Needleman, R. B., D. B. Kaback, R. A. Dubin, E. L. Perkins, N. G. Rosenberg, K. A. Sutherland, D. B. Forrest, and C. A. Michels. 1984. MAL6 of Saccharomyces: a complex genetic locus containing three genes required for maltose fermentation. Proc. Natl. Acad. Sci. USA 812811-2815. [PMC free article] [PubMed]
22. Ner, S. S., and M. Smith. 1989. Role of intron splicing in the function of the MATa1 gene of Saccharomyces cerevisiae. Mol. Cell. Biol. 94613-4620. [PMC free article] [PubMed]
23. Nickoloff, J. A., J. D. Singer, and F. Heffron. 1990. In vivo analysis of the Saccharomyces cerevisiae HO nuclease recognition site by site-directed mutagenesis. Mol. Cell. Biol. 101174-1179. [PMC free article] [PubMed]
24. Singh, A., E. Y. Chen, J. M. Lugovoy, C. N. Chang, R. A. Hitzeman, and P. H. Seeburg. 1983. Saccharomyces cerevisiae contains two discrete genes coding for the alpha-factor pheromone. Nucleic Acids Res. 114049-4063. [PMC free article] [PubMed]
25. Soll, D. R., S. R. Lockhart, and R. Zhao. 2003. Relationship between switching and mating in Candida albicans. Eukaryot. Cell 2390-397. [PMC free article] [PubMed]
26. Srikantha, T., S. A. Lachke, and D. R. Soll. 2003. Three mating type-like loci in Candida glabrata. Eukaryot. Cell 2328-340. [PMC free article] [PubMed]
27. Thierry, A., C. Fairhead, and B. Dujon. 1990. The complete sequence of the 8.2 kb segment left of MAT on chromosome III reveals five ORFs, including a gene for a yeast ribokinase. Yeast (Chichester) 6521-534. [PubMed]
28. Tsong, A. E., B. B. Tuch, H. Li, and A. D. Johnson. 2006. Evolution of alternative transcriptional circuits with identical logic. Nature 443415-420. [PubMed]
29. Weig, M., K. Haynes, T. R. Rogers, O. Kurzai, M. Frosch, and F. A. Muhlschlegel. 2001. A GAS-like gene family in the pathogenic fungus Candida glabrata. Microbiology (Reading) 1472007-2019. [PubMed]
30. Winston, F., C. Dollard, and S. L. Ricupero-Hovasse. 1995. Construction of a set of convenient Saccharomyces cerevisiae strains that are isogenic to S288C. Yeast (Chichester) 1153-55. [PubMed]
31. Wong, S., M. A. Fares, W. Zimmermann, G. Butler, and K. H. Wolfe. 2003. Evidence from comparative genomics for a complete sexual cycle in the ‘asexual’ pathogenic yeast Candida glabrata. Genome Biol. 4R10. [PMC free article] [PubMed]

Articles from Eukaryotic Cell are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...