INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Fluconazole, a widely used azole, selectively inhibits the cytochrome P-450-dependent C₁₄ lanosterol demethylase (P-450 14-DM), a key enzyme involved in ergosterol biosynthesis in fungi (19). The emerging resistance of Candida species to fluconazole is a matter of concern (20). A better understanding of the molecular responses of Candida species to fluconazole could enable physicians to make more effective use of this agent.

Saccharomyces cerevisiae, a genetically tractable fungus, is an attractive experimental system for the study of azole resistance. S. cerevisiae is closely related to the genetically intractable Candida albicans, and it has long served as a model system for studies of sterol biosynthesis (15). All of the reported mechanisms of fluconazole resistance in S. cerevisiae have also been described for C. albicans and involve the same gene products (11, 20). One well-characterized mechanism of azole resistance in Saccharomyces is conferred by loss-of-function mutations in sterol 5,6-desaturase, which is the product of the ERG3 gene (13, 20). Another mechanism of azole resistance is mediated by the pleiotropic drug resistance 5 gene (PDR5), an ATP-binding cassette efflux transporter, through decreased accumulation of fluconazole (2, 4). A complex regulatory network controls the PDR phenotype (4). Finally, loss of function of NADPH-dependent cytochrome P-450-oxidoreductase, which is encoded by the CPR1 gene, results in azole hypersensitivity (18). On the other hand, the importance of target-site (P-450 14-DM) alterations is unclear (11, 13, 20).

Despite advances in understanding the mechanisms of azole resistance in Saccharomyces, the components of the response pathways are not fully known. Recent methods such as DNA microarray technology (5) are precise and powerful tools for real-time analysis of the dynamic change of the host in response to signals, including drugs. However, these technologies will always rely on classical genetics and on the effects of gene disruption to explain gene function.

In the current work, I used an insertional Tn3::LEU2::lacZ transposon mutagenesis scheme to study the mechanisms of azole resistance in Saccharomyces. This strategy permits the generation and analysis of a large number of independent lacZ gene fusions (3). This approach offers several advantages over the traditional methods and oligonucleotide-based probe assays: (i) it frequently creates loss-of-function mutations by insertional mutagenesis, (ii) it facilitates easy identification of expression candidates by a simple colorimetric assay, and (iii) it allows for rapid sequencing of candidate genes (3). Using the transposon mutagenesis scheme, I found recessive mutations in genes that were previously found to be involved in azole resistance (e.g., PDR5 and CPR1) and sensitivity (e.g., ERG3). This approach also enabled me to identify recessive mutations in three genes previously not known to be involved in azole sensitivity. Two of the genes, ADA3 and SPT7, are general transcriptional activators or repressors; the third, YMR034c, is a putative sterol transporter. My results implicate the role of transposon mutagenesis as a promising tool for identifying novel drug targets and in identifying the molecular mechanisms involved in the resistance of S. cerevisiae to azoles.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Strains. I used standard methods for making the yeast growth media and standard techniques for yeast manipulation (12). All work was done in the 1278b genetic background. Fluconazole was a gift from Pfizer, Inc. (New York, N.Y.).

Transposon mutagenesis screening for fluconazole sensitivity and resistance. The S. cerevisiae haploid strain 10512-3C (MATa leu2::hisG his3::hisG; Fink laboratory, Whitehead Institute for Biomedical Research, Cambridge, Mass.) was transformed with the Tn3::LEU2::lacZ transposon genomic DNA library (Snyder laboratory, Yale University, New Haven, Conn.) (3). Approximately 50,000 Leu⁺ transformant colonies were pooled and replated at a density of ~600 colonies/plate on synthetic complete (SC)-leucine medium. To screen for sensitive mutants, colonies were replica plated on SC-leucine-8-µg/ml fluconazole plates. To screen for resistant mutants, the pooled transformants of 10512-3C were spread at approximately 10⁸ cells on SC-leucine-128-µg/ml fluconazole plates and incubated for 7 days at 30°C.

Transposon mutagenesis screening for fluconazole-responsive lacZ fusions. The S. cerevisiae diploid strain L5803 (MATa/ ura3-52/ura3-52 leu2::hisG/leu2::hisG; Fink laboratory) was transformed with the Tn3::LEU2::lacZ transposon-mutagenized yeast genomic DNA library. Approximately 280,000 transformants were obtained by selection on SC-leucine medium. The transformants were pooled and replated on SC medium at a density of ~600 colonies/plate. These colonies were printed to the surface of 8.26-cm filter circles (model no. 576; Schleicher & Schuell) on SC medium and on SC medium containing subinhibitory concentrations of fluconazole (8 µg/ml) and grown at 30°C for 18 to 24 h. The wild-type L5803 strain failed to grow at a concentration of 64 µg/ml. Filters were then processed with X-Gal (5-bromo-4-chloro-3-indolyl--D-galactopyranoside) as follows. The filters were first dipped into liquid nitrogen for 1 min and then thawed and placed on an 8.26-cm circular blotter (model no. 593; Schleicher & Schuell) saturated with 0.03% X-Gal in 2.5 ml of Z buffer. The mutant ypl110c::Tn3::LEU2::lacZ (which I found not to respond to fluconazole) was used as a blue-color control. The -galactosidase reaction was stopped by Na₂CO₃ after 4 h. Mutant colonies, which were more blue on the fluconazole plate filters, were patched onto SC medium, retested, purified, and retested again. Of the 18,000 blue colonies screened from a total plating of 180,000 colonies (only a subset of insertions express lacZ [3]), 62 mutants were found to be reproducibly induced. The pattern of -galactosidase induction for each mutant was studied by comparing the intensity of fluconazole-induced blue-color development with the color response to different classes of antifungals (amphotericin B, 10 mg/ml; nystatin, 25,000 U/ml; and 5-fluorouracil, 10 mg/ml) or other growth inhibitors (canavanine, 20 mg/ml; cycloheximide, 100 µg/ml; and 100% alcohol). To determine haploid phenotypes, the diploid expression mutants were sporulated and dissected on yeast extract-peptone-dextrose medium. Tetrads (10 per diploid) of ascospore colonies were replica plated to SC-leucine and filters on SC plates. The fluconazole growth phenotypes of haploid expression mutants were assayed by replica plating to SC plates containing different concentrations of fluconazole. The growth of reconstructed diploids formed by crossing mutants with wild-type sister spores was similarly tested to check for recessivity.

Molecular biology and biochemical methods. Genomic DNA immediately adjacent to Tn3::LEU2::lacZ in the mutants of interest was cloned as described earlier (3). DNA and protein homology searches were performed with the BLAST network (1).

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Identification of genes known to be involved in azole resistance and sensitivity. I found mutants with a Tn3::LEU2::lacZ disruption of PDR5 to be hypersensitive to fluconazole (2 µg/ml) (Fig. 1). The disruption phenotype of pdr5 mutants also included pleiotropic hypersensitivity to a variety of agents. I also found mutants with a Tn3::LEU2::lacZ disruption of CPR1 to be exquisitively sensitive specifically to fluconazole (0.1 µg/ml) (Fig. 1). Finally, I found several fluconazole-resistant mutants with a Tn3::LEU2::lacZ disruption of ERG3. The erg3 mutant alleles, despite their resistance to azoles, had a pleiotropic sensitivity to a variety of agents.

View larger version (82K): [in this window] [in a new window]   FIG. 1.   The Tn3::LEU2::lacZ transposon mutagenesis scheme identifies known genes and genes previously not known to be involved in azole resistance. One representative four-spore tetrad of ascospores of each of the diploids cpr1::Tn3::LEU2::lacZ/CPR1 leu2::hisG/leu2::hisG ura3-52/ura3-52, pdr5::Tn3::LEU2::lacZ/PDR5 leu2::hisG/leu2::hisG ura3-52/ura3-52, ada3::Tn3::LEU2::lacZ/ADA3 leu2::hisG/leu2::hisG ura3-52/ura3-52, spt7::Tn3::LEU2::lacZ::/SPT7 leu2::hisG/leu2::hisG ura3-52/ura3-52, and ymr034c::Tn3::LEU2::lacZ/YMR034c leu2::hisG/leu2::hisG ura3-52/ura3-52 shows that the cpr1::LEU2::lacZ, pdr5::Tn3::LEU2::lacZ, ada3::Tn3::LEU2::lacZ, spt7::Tn3::LEU2::lacZ, and ymr034c::Tn3::LEU2::lacZ disruptants segregate 2:2 (Leu+ and fluconazole hypersensitive). The spores were patched to SC plates and then printed to SC-leucine and SC-8-µg/ml fluconazole plates. Shown is the growth after 2 days at 30°C. The 10512-3C strain transformed with a cen LEU2 plasmid was also put in the plate as a haploid wild-type (Wt) control.

Identification of genes previously not known to be involved in azole resistance. Tn3::LEU2::lacZ insertions into three genes previously not known to be involved in sensitivity to azoles resulted in sensitivity to fluconazole.

Identification of genes known to be involved in response mechanisms to azoles. I found multiple pdr5::LEU2::lacZ fusions to be specifically induced by fluconazole (Fig. 2; Table 1). In addition, an erg3::LEU2::lacZ fusion was found to be induced specifically by fluconazole. Finally, a variety of previously known genes were identified for the first time to be induced by fluconazole (Table 1).

View larger version (60K): [in this window] [in a new window]   FIG. 2.   Fluconazole induces expression of transporters in yeast. Diploid strains of indicated genotype (PDR5/pdr5::Tn3::LEU2::lacZ, YOR1/yor1::Tn3::LEU2::lacZ, YDR091c/ydr091c::Tn3::LEU2::lacZ, and YLL031c/yll031c::Tn3::LEU2::lacZ are shown. The YDR091c/ydr091c::Tn3::LEU2::lacZ diploid appears to be induced the least. The YPL110c/ypl110c::Tn3::LEU2::lacZ mutant and strain L5803 were used as blue- and white-color controls, respectively. Strains were grown on filters in SC and SC-fluconazole (8 µg/ml) at 30°C for 24 h. Filters were then processed with X-Gal. The reaction was stopped after 6 h.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Using transposon mutagenesis, I found genes known to be involved in fluconazole resistance (PDR5 and CPR1) and fluconazole sensitivity (ERG3). In addition, this approach led to the identification of novel genes. Hence, I found that general transcriptional repressors-activators affect azole sensitivity. Of note, Spt7p, Spt13p, and Spt20p as well as Ada2p, Ada3p, and Gcn5p are located in the transcriptional complex called SAGA (Spt-Ada-Gcn5-acetyltransferase) (8). Our data suggest a GCN5-independent role of the SAGA complex in regulating gene expression, because, whereas spt7 and spt20 mutants were hypersensitive to fluconazole, gcn5 mutants were not. The mechanism of action of the aforementioned transcription factors is not known. It may involve regulation of the PDR network and thus, indirectly, the efficiency of drug efflux. Evidence of the interaction between PDR1, a positive regulator of PDR5 (2, 4), and ADA3 has been reported with a two-hybrid system (14). Further work is needed to define the role of the general transcriptional machinery in the various drug-specific responses. Finally, I found that a disruption in the YMR034c gene caused sensitivity to azoles. YMR034c is homologous with bacterial (arsenic resistance protein), plant, and mammalian (sodium-dependent bile acid cotransporter) transporters (1). Loss of function of YMR034c could affect efflux of fluconazole either directly through decreased transport of the drug or through alterations of lipid fluidity of the fungal membrane and thus indirectly through impaired extrusion of fluconazole by Pdr5p.

I also screened for fluconazole-responsive lacZ fusions in a defined medium (SC), which lacks the lipid extracts, in order to avoid the influence of exogenous lipids on the expression of genes such as ERG3 whose transcription is affected by feedback mechanisms by sterol levels (17). I found genes known to be involved in azole response mechanisms as well as genes that were previously unknown to be upregulated by azoles. Transporters appeared to constitute an important element of the response to fluconazole. PDR5, in particular, was found to be specifically induced by fluconazole in my study. Induction of CDR1, the PDR5 homologue in C. albicans, by fluconazole has been reported for both laboratory and fluconazole-resistant clinical isolates of C. albicans (10, 16, 20).

My work has multiple implications. First, transposon mutagenesis holds promise for enabling us to uncover novel molecular targets involved in azole sensitivity and could be applicable to other classes of antifungals. Second, elucidation of the role of transcription factors or sterol transporters in azole resistance could lead to the discovery of new, potentially therapeutic targets. Finally, combining mutants with fluconazole-responsive lacZ fusions to form double mutants could reveal additional unknown drug phenotypes and shed light on the regulatory mechanisms involved in azole response.