Inactivation of the FCY2 gene encoding purine-cytosine permease promotes cross-resistance to flucytosine and fluconazole in Candida lusitaniae

Abstract

In a previous work, we described the possible relationship between a defect of purine-cytosine permease and the acquisition of a cross-resistance to the antifungal combination flucytosine (5FC) and fluconazole (FLC) in Candida lusitaniae (T. Noël, F. François, P. Paumard, C. Chastin, D. Brethes, and J. Villard, Antimicrob. Agents Chemother. 47:1275-1284, 2003). Using degenerate PCR and chromosome walking, we cloned two FCY2-like genes in C. lusitaniae. Northern blot analysis revealed that only one gene was expressed; it was named FCY2. The other one behaved as a pseudogene and was named FCY21. In order to better characterize the possible role of FCY2 in cross-resistance to 5FC-FLC, disruption experiments with auxotrophic strain 6936 ura3(D95V) FCY2 with an integrative vector carrying the URA3 gene and a partial sequence of the C. lusitaniae FCY2 gene were undertaken. Southern blot analysis revealed that homologous recombination events occurred in all transformants analyzed at rates of 50% at resident locus FCY2 and 50% at resident locus URA3, resulting in the genotypes ura3 fcy2::URA3 and ura3::URA3 FCY2, respectively. It was then demonstrated that only transformants harboring a disrupted fcy2 gene were resistant to 5FC, susceptible to FLC, and resistant to the 5FC-FLC combination. Finally, complementation experiments with a functional FCY2 gene restored 5FC and FLC susceptibilities to the wild-type levels. The results of this study provide molecular evidence that inactivation of the sole FCY2 gene promotes cross-resistance to the antifungal association 5FC-FLC in C. lusitaniae.

FIG. 1.

Molecular characterization of C. lusitaniae FCY2 and FCY21 genes. (A) Southern blot analysis of EcoRV-digested genomic DNA from strain 6936 hybridized with the FCY2 DNA probe and then stripped and hybridized with the FCY21 DNA probe. (B) Northern blot analysis of total RNA from the strain 6936 hybridized with FCY2 and FCY21 RNA probes. The membranes were stripped and hybridized with the 18S RNA probe for RNA loading control.

FIG. 2.

Amino acid sequence alignments of two regions located in a transmembrane amphipathic domain (region 1) and in a hydrophilic loop (region 2) of C. lusitaniae Fcy2p and Fcy21p, C. albicans Fcy2p (Stanford database, contig6-1838; ORF6.1188) and Fcy22p (Stanford database, contig6-1777; ORF6.1000), and S. cerevisiae Fcy2p (GenBank accession no. X51751). The amino acid residues which are thought to be part of a hydrophilic pore involved in the base-H⁺ translocation process are in boldface. The region essential for the active three-dimensional structure of the transporter is framed. Absolutely conserved residues are marked with asterisks, highly conserved residues are marked with colons, and weakly conserved residues are marked with periods. The amino acid residue number for each protein is indicated at the beginning of each sequence.

FIG. 3.

Schematic representation of the genetic map of plasmids pUF2 and pFCY2 and Southern blot hybridization. (A) The 5.4-kb pUF2 plasmid, which contains the C. lusitaniae URA3 gene and a 944-bp cassette from the FCY2 gene (Fi cassette), was used to transform strain 6936 ura3(D95V) FCY2 to prototrophy. The 6.6-kb pUFCY2 plasmid, which contains the URA3 and the FCY2 genes of C. lusitaniae, was used to transform strain TR1 ura3(D95V) fcy2(S122P) to prototrophy. The locations of the probes used in this study are indicated in parentheses. (B) Hybridization patterns with URA3, Fi, and BLA DNA probes of EcoRV-digested genomic DNA from 6936 ura3(D95V) FCY2 (lane 1), a representative transformant of group 1 [ura3(D95V) fcy2::URA3] (lane 2), and transformant Rec2 ura3(D95V) fcy2::[URA3 FCY2] (lane 3). The hybridization patterns of EcoRV-digested genomic DNA from the strain TR1 ura3(D95V) fcy2(S122P) were identical to those of 6936 ura3(D95V) (lane 1). Signals revealed by the labeled probes correspond to those expected from the genomic restriction map. Hybridization of the genomic DNA of 6936 ura3(D95V) with the URA3 and Fi probes revealed fragments of 5.2 and 1.8 kb, respectively. For the transformant of group 1, integration of plasmid pUF2 at the FCY2 locus yielded two fragments of 5.2 and 2.2 kb when genomic DNA was hybridized with either the URA3 probe or the Fi probe. For transformant Rec2, integration of plasmid pUFCY2 at the fcy2 locus resulted in the detection of three fragments (5.2, 4.0, and 0.8 kb) when the genomic DNA was hybridized with the URA3 probe and of a single 1.8-kb fragment when the genomic DNA was hybridized with the Fi probe. Hybridization of the genomic DNA of the transformant of group 1 and transformant Rec2 with the BLA probe revealed fragments of 5.2 and 4.0 kb, respectively. DNA fragment sizes are indicated in kilobases.

FIG. 4.

Susceptibilities of control strains and of the strains genetically engineered this study to 5FC and FLC. (A) Typical antifungal susceptibility pattern of strain 6936 URA3 FCY2, of the transformants of group 2 (ura3::URA3 FCY2), and of transformant Rec2 ura3 fcy2::[URA3 FCY2]; (B) typical antifungal susceptibility pattern of clinical isolate CL42 URA3 fcy2; (C) typical antifungal susceptibility pattern of the transformants of group 1 (ura3 fcy2::URA3) and strain TR1 ura3 fcy2. Growth is given as a percentage of the growth level obtained for the drug-free control. ▵, 5FC concentration gradient; •, FLC concentration gradient; □, 5FC concentration gradient with a constant concentration of 16 μg/ml of FLC.

FIG. 4.

Susceptibilities of control strains and of the strains genetically engineered this study to 5FC and FLC. (A) Typical antifungal susceptibility pattern of strain 6936 URA3 FCY2, of the transformants of group 2 (ura3::URA3 FCY2), and of transformant Rec2 ura3 fcy2::[URA3 FCY2]; (B) typical antifungal susceptibility pattern of clinical isolate CL42 URA3 fcy2; (C) typical antifungal susceptibility pattern of the transformants of group 1 (ura3 fcy2::URA3) and strain TR1 ura3 fcy2. Growth is given as a percentage of the growth level obtained for the drug-free control. ▵, 5FC concentration gradient; •, FLC concentration gradient; □, 5FC concentration gradient with a constant concentration of 16 μg/ml of FLC.