Mutation of the CgPDR16 gene attenuates azole tolerance and biofilm production in pathogenic Candida glabrata

Authors


Correspondence to: J. Subik, Comenius University in Bratislava, Department of Genetics, Mlynska dolina B-1, 842 15 Bratislava 4, Slovak Republic. E-mail: subik.julius@gmail.com or subik@fns.uniba.sk

Abstract

The PDR16 gene encodes the homologue of Sec14p, participating in protein secretion, regulation of lipid synthesis and turnover in vivo and acting as a phosphatidylinositol transfer protein in vitro. This gene is also involved in the regulation of multidrug resistance in Saccharomyces cerevisiae and pathogenic yeasts. Here we report the results of functional analysis of the CgPDR16 gene, whose mutation has been previously shown to enhance fluconazole sensitivity in Candida glabrata mutant cells. We have cloned the CgPDR16 gene, which was able to complement the pdr16Δ mutation in both C. glabrata and S. cerevisiae. Along with fluconazole, the pdr16Δ mutation resulted in increased susceptibility of mutant cells to several azole antifungals without changes in sensitivity to polyene antibiotics, cycloheximide, NQO, 5-fluorocytosine and oxidants inducing the intracellular formation of reactive oxygen species. The susceptibility of the pdr16Δ mutant strain to itraconazole and 5-fluorocytosine was enhanced by CTBT [7-chlorotetrazolo(5,1-c)benzo(1,2,4)triazine] inducing oxidative stress. The pdr16Δ mutation increased the accumulation of rhodamine 6G in mutant cells, decreased the level of itraconazole resistance caused by gain-of-function mutations in the CgPDR1 gene, and reduced cell surface hydrophobicity and biofilm production. These results point to the pleiotropic phenotype of the pdr16Δ mutant and support the role of the CgPDR16 gene in the control of drug susceptibility and virulence in the pathogenic C. glabrata. Copyright © 2013 John Wiley & Sons, Ltd.

Introduction

Candida glabrata is an opportunistic human pathogen responsible for candidaemia. Behind C. albicans, this haploid yeast is considered to be the second most commonly isolated Candida species from both bloodstream (Pfaller et al., 1999) and vaginal infections (Sojakova et al., 2004). It is evolutionarily more closely related to Saccharomyces cerevisiae than C. albicans. Compared with them, C. glabrata is naturally less susceptible to antimycotics (d'Enfert and Hube, 2007) and is particularly resistant to oxidative stress (Nikolaou et al., 2009; Roetzer et al., 2011).

The susceptibility of yeasts to antifungal agents is under the control of two networks of genes. In S. cerevisiae, the PDR and YAP networks are activated by the main multidrug resistance transcription factors Pdr1p/Pdr3p and Yap1p, respectively (Balzi et al., 1987; Delaveau et al., 1994; Moye-Rowley et al., 1989). Some genes in these networks are under the control of both transcription factors (Miyahara et al., 1996; Moye-Rowley, 2003). Yap1p also regulates the expression of a set of genes involved in oxidative stress response (Temple et al., 2005). The PDR and YAP gene networks are evolutionarily highly conserved in C. glabrata. However, in this species, instead of the ScPDR1 and ScPDR3 genes, only a single CgPDR1 homologue occurs (Dujon et al., 2004; Drobna et al., 2008).

In S. cerevisiae, the ScPDR16 gene (synonym SFH3) is a member of the PDR1 regulon (DeRisi et al., 2000). It affects lipid biosynthesis and resistance of cells to azole inhibitors of ergosterol metabolism (van den Hazel et al., 1999). Deletion of the PDR16 gene in C. glabrata and C. albicans also enhances fluconazole sensitivity (Kaur et al., 2004; Saidane et al., 2006). Studies in the yeast S. cerevisiae have shown that ScPDR16 encodes a phosphatidylinositol transfer protein which belongs to the Sec14 homologue (SFH) family (Li et al., 2000; Griac 2007) and localizes to lipid particles, microsomes and at the cell periphery (Schnabl et al., 2003). The recombinant Pdr16p purifies as a dimer (Ren et al., 2011) and its structure differs from other monomeric Sec14 homologues (Yuan et al., 2013). While the overexpression of SFH2 and SFH4 homologues suppressed the sec14 growth defect in a more efficient way and SFH1 in a less efficient one, the overexpression of PDR16 and SFH5 failed to complement the sec14 mutation (Schnabl et al., 2003). Sec14p is required for the transport of proteins through the Golgi complex (Bankaitis et al., 1990). It may also regulate formation of secretory vesicles from the Golgi by stimulating turnover of phospholipids (Patton-Vogt et al., 1997) and participate with other phosphatidylinositol transfer proteins in lipid signalling (Cockcroft et al., 1997; Ghosh and Bankaitis, 2011). Recently it has been shown that overexpression of the PDR17 gene complements the azole-susceptible phenotype of the S. cerevisiae pdr16Δ mutant strain and corrects its enhanced sterol alterations in the presence of azoles (Simová et al., 2013).

The aim of this study was to investigate the role of the CgPDR16 gene in the control of cell properties related to the drug susceptibility and pathogenicity of C. glabrata. It was found that, unlike some other drugs, oxidants and polyene antibiotics, the CgPDR16 mutation sensitizes the cells to azole antimycotics, enhances rhodamine 6G accumulation in de-energized cells and reduces the virulence of this pathogenic yeast.

Materials and methods

Strains and culture conditions

Candida glabrata BG2 (Fidel et al., 1996), its isogenic BG14 (Cgura3::Tn903Neor) and pdr16Δ (Cgpdr16::Tn903Neor) mutant strains (Kaur et al., 2004) were obtained from B. Cormack (Johns Hopkins University, USA). C. glabrata ΔHTU (his3Δ trp1Δ ura3Δ) and its isogenic Cgyap1Δ, Cgskn7Δ and Cgyap1Δskn7Δ mutant strains (Roetzer et al., 2011) were provided by Ch. Schüller (Vienna University, Austria). S. cerevisiae FY1679-28C (MATa his3Δ200 leu2Δ1 trp1Δ63 ura3-52; EUROSCARF, Frankfurt, Germany) and its isogenic pdr16Δ mutant strain (van den Hazel et al., 1999) were used in heterologous complementation of the pdr16 deletion. Cells were grown on glucose-rich (YPD) medium (2% glucose, 1% yeast extract, 2% bactopeptone), glycerol-rich (YPG) medium (as YPD, but 2% glycerol was used instead of 2% glucose) or on minimal medium (YNB) containing 0.67% yeast nitrogen base without amino acids, 2% glucose (YNB-D) or 2% glycerol plus 2% ethanol (YNB-GE) and appropriate nutritional requirements. The media were solidified with 2% bactoagar. The media containing azole antifungals were buffered with 0.165 m MOPS, pH 7.0. For Ura selection of pdr16Δ, minimal medium containing glucose was supplemented with uracil (25 µg/ml) and 5-fluororotic acid (5-FOA) (1.4 mg/ml) (Castano et al., 2003). The Escherichia coli XL1-Blue strain was used as host for transformation, plasmid amplification and preparation. The bacterial strains were grown at 37 °C in Luria–Bertani medium (1% tryptone, 0.5% NaCl, 0.5% yeast extract, pH 7.5) supplemented with 100 µg/ml ampicillin for selection of transformants.

PCR amplification, plasmids, DNA sequencing and other recombinant DNA techniques

Genomic DNA from C. glabrata BG2 was extracted (Xu et al., 2000) and used as a template for amplification of the CgPDR16 gene. PCR was carried out with an Extensor Hi-Fidelity DNA polymerase (Thermo Scientific, UK) with the primers 5′-TCGTTATCAGTATCGAACACCT-3′ (forward) and 5′-TTCCCAGATAAACTAGAGGGAC-3′ (reverse). The overhanging A on the 3′ ends of the amplicon were synthesized using a GoTaq Hot Start DNA polymerase (Promega, Germany). The resulting amplicon was purified using a PCR Purification Kit (Qiagen, Hilden, Germany) and ligated into the pCR2.1–TOPO (pUC ori, lacZα, f1 ori, KanR, AmpR) vector (Invitrogen, USA). From the resulting pCR2.1–CgPDR16 plasmid, the BamHI–EcoRI DNA fragment containing CgPDR16 was isolated, purified and ligated into the BamHI- and EcoRI-restricted pCgACU-5 (CgARS, CgCEN, CgURA3, AmpR) vector obtained from D. Sanglard (Lausanne University, Switzerland). In the resulting pCgACU5–CgPDR16 plasmid, the nucleotide sequences for both strands of the CgPDR16 gene were determined by primer elongation with automated DNA sequencer (ABI Prism 3100, Applied Biosystems, Foster City, CA, USA). Sequence data were compared with the CgPDR16 sequence (http://cbi.labri.fr/Genolevures/elt/CAGL), using the BLAST program. From the pCR2.1–CgPDR16 plasmid, the SpeI–NotI genomic insert containing the CgPDR16 gene with its own promoter was subcloned into the pRS306K vector (2 µ, URA3, ARS1, KARS2, lacZα, f1(+), ori, AmpR) (Heus et al., 1994) and the resulting pRS306K–CgPDR16 plasmid was used to transform the S. cerevisiae pdr16Δ mutant strain. Standard protocols were used for generating recombinant DNAs, restriction enzyme analysis and gel electrophoresis (Sambrook et al., 1989). Plasmid DNA from E. coli was prepared by an alkaline lysis method. C. glabrata and S. cerevisiae transformations were carried out using modified lithium acetate protocols (Sanglard et al., 1996). The pCgPDR14672 plasmid containing the gain-of-function mutation in CgPDR1 has been described previously (Tsai et al., 2006; Berila et al., 2009). The centromeric YCp-lac111–ScPDR16 (LEU2, AmpR, CEN) and multicopy YEp-lac195–ScPDR16 (URA3, AmpR, 2 µm) plasmids were kindly provided by P. Griac (Slovak Academy of Sciences, Ivanka pri Dunaji, Slovakia).

Drug susceptibility assays

Drug susceptibility was determined by spot assay. Suspensions (5 µl, 107 cells/ml and their 10-fold dilutions) of three independent clones, grown on solid glycerol-rich (YPG) medium to eliminate rho0 cells, were spotted onto rich and minimal media containing glucose or glycerol plus ethanol and supplemented with various concentrations of antifungals. Growth at 37 °C was scored after 2 days. The drug concentrations were as follows: fluconazole (FLC), 4, 8, 16, 32, 64 and 128 µg/ml; bifonazole (BIF), 0.0625, 0.125, 0.25, 0.5, 1 and 2 µg/ml; itraconazole (ITR), 0.125, 0.25, 0.5, 1, 2, 4, 8, 16, 32, 64 and 128 µg/ml; ketoconazole (KET), 0.5, 1, 2, 4, 8 and 16 µg/ml; clotrimazole (CLO), 0.25, 0.5, 1, 2, 4 and 8 µg/ml; miconazole (MIC), 0.05, 0.1, 0.2, 0.4 and 0.8 µg/ml; cycloheximide (CYH), 1, 2, 4, 6 and 8 µg/ml; (NQO), 1, 2, 4, 6 and 8 µg/ml; 5-fluorocytosine (5FC), 0.025, 0.05, 0.075, 0.1, 0.25, 0.5, 0.75 and 1.0 µg/ml. Susceptibility to polyene antibiotics and oxidants was assessed using zone-inhibition assays. Approximately 107 stationary phase cells were plated onto rich or minimal glucose agar. The filter discs (diameter 6 mm), soaked with an appropriate amount of antifungals, were placed on the plates, which were incubated at 37 °C for 2 days before determinating the diameter of the zone of growth inhibition.

Rhodamine 6G accumulation and efflux

The accumulation of rhodamine 6G (Sigma-Aldrich, Taukirchen, Germany) was measured by adapting the method described by Maesaki et al. (1999). Approximately 12.5 × 108 cells from an overnight culture were incubated in 50 ml YPD medium and grown for 3 h at 30 °C. The cells were pelleted and washed three times with 50 mm HEPES, pH 7.0. The cells were resuspended to a concentration of 108 cells/ml in 50 mm HEPES containing 5 mm 2-deoxyglucose and shaken for 1 h at 30 °C to exhaust the energy. Then 10 µm rhodamine 6G was added and, after incubation at 30 °C for the indicated times, samples were withdrawn and centrifuged at 9000 × g for 2 min. The supernatants were collected and added (100 µl) into Nunc 96-well fluoro-luminunc plates (Nalge Nunc International, Rochester, NY, USA). Rhodamine 6G fluorescences of the samples were measured at an excitation wavelength of 529 nm and an emission wavelength of 553 nm, using a Tecan Saphir II TM spectrofluorimeter (Tecan Austria GmbH, Grödig/Salzburg, Austria). Image analysis was also used to compare the amounts of accumulated rhodamine 6G in the wild-type and pdr16Δ mutant cells. After 30 min of incubation of the de-energized cells with 10 µm rhodamine 6G, samples (100 µl) were withdrawn and filtered through a nylon membrane (Boehringer Mannheim, Germany) using a 96-well Bio-Dot® Microfiltration Apparatus (Bio-Rad, USA). Filters with dots of cells were scanned and the images were analysed using GelQuant.NET software, provided by biochemlabsolutions.com. The active efflux of rhodamine 6G was measured as described previously (Borecka-Melkusova et al., 2008). De-energized cells (108 cells/ml) accumulating rhodamine 6G (10 µm) for 30 min were washed three times and resuspended in 50 mm HEPES, pH 7.0, to a concentration 109 cells/ml. At specific intervals after the addition of glucose (final concentration, 25 mm) to initiate rhodamine 6G efflux, the cells were centrifuged, 100 µl supernatants were added to Nunc 96-well fluoro-luminunc plates and rhodamine 6G fluorescences were measured as described above.

Determination of cell surface hydrophobicity

Cells grown to stationary phase in YNB medium containing 0.9% glucose at 28 °C were washed with phosphate-buffered saline (PBS) and suspended in YNB medium containing 0.9% glucose to obtain an absorbance of 0.6–0.8 at 570 nm. Cell surface hydrophobicity was measured by the water-octane two-phase assay (Klotz et al., 1985). A570 values of strains in YNB broth without octane overlay were used as negative controls. The percentage of exclusion of the cells from the aqueous phase (% change in A570) corresponding to relative cell surface hydrophobicity (CSH) was calculated as: [(A570 of the control – A570 after octane overlay)/A570 of the control] × 100. The experiments were repeated five times for each strain and mean values ± standard deviation (SD) were calculated.

Biofilm formation

A standardized method for biofilm formation, based on an untreated polystyrene 96-well plate, was used. Biofilm formation was quantified using the 2,3-bis(2-methoxy-4-nitro-5-sulphophenyl)-2H-tetrazolium-5-carboxanilide (XTT; Sigma-Aldrich, USA) reduction assay, or crystal violet (CV; Loba Feinchemie, Austria) staining, followed by measuring of absorbance using a microplate reader (A490 for XTT formazan and A570 for crystal violet), essentially as described (Li et al., 2003). Wells containing only YNB medium without yeast cells were used as negative controls. The experiments were repeated three times for each strain and mean values ± SD were calculated.

Results

Molecular cloning of the CgPDR16 gene and the susceptibility to antifungals conferred by its deletion

The C. glabrata mutant strain deleted in the PDR16 gene was isolated after genome-wide transposon mutagenesis of the wild-type strain BG14 using selection for enhanced sensitivity to fluconazole (Kaur et al., 2004). The ura3 derivative of the pdr16Δ strain was selected as a spontaneous mutant growing in the presence of 5-fluororotic acid, as described in Materials and methods. Along with reported fluconazole hypersensitivity (Kaur et al., 2004), the pdr16Δ mutant was found to be more sensitive to other azole antifungals, such as bifonazole, itraconazole, ketoconazole, clotrimazole and miconazole (Table 1). No hypersensitivity of the mutant strain was displayed to cycloheximide, 4-nitroquinoline-N-oxide, 5-fluorocytosine (Table 1) and polyene antibiotics (amphotericin B, nystatin and pimaricin) (Table 2).

Table 1. Susceptibility to antifungals of the wild type strain (wt), the pdr16Δ mutant and its transformant containing the CgPDR16 gene on plasmid
 Minimum inhibitory concentration (µg/ml)
C. glabrataYPDYNB-D
 FLCBIFITRKETCLOMICCYHNQO5FC
  1. Minimum inhibitory concentrations of azoles, cycloheximide, 4-nitroquinoline-N-oxide and 5-fluorocytosine were determined after growth of cells on the corresponding media.

wt + pCgACU-51281128440.2480.25
pdr16Δ + pCgACU-5320.125320.50.250.05480.25
pdr16Δ + pCgACU5–CgPDR161280.5128440.2480.25
Table 2. Susceptibility to polyene antibiotics and oxidants of the wild-type (wt) and pdr16Δ mutant strains
C. glabrataDiameter of the growth inhibition zone (mm)
AMBNYSPIMCTBTPARMENDIAH2O2
  1. The diameters of the growth inhibition zones were determined after growth of cells on YPD medium. Amounts of polyenes and oxidants per disc were: amphotericin B (AMB), 50 µg; nystatin (NYS), 50 µg; pimaricin (PIM), 50 µg; CTBT, 0.048 µmoles; paraquat (PAR), 2.5 µmoles; menadione (MEN), 1.5 µmoles; diamide (DIA), 5 µmoles; H2O2, 44 µmoles.

wt + pCgACU-5101612181314910
pdr16Δ + pCgACU-511171318131488
pdr16Δ + pCgACU5–CgPDR1611161318131498

The azole hypersensitivity of the pdr16Δ mutant strain was effectively complemented by the PCR-amplified and pCgACU-5 plasmid-borne CgPDR16 gene (Table 1). This gene, subcloned into pRS306K plasmid, was also able to complement fluconazole and miconazole hypersensitivity of the heterologous S. cerevisiae pdr16Δ mutant strain (Figure 1). Computer analysis revealed that the CgPDR16 gene (CAGL0J07436g; www.candidagenome.org) contains an open reading frame of 1035 bp and encodes a polypeptide of 344 amino acids. Sequence comparisons indicate that CgPDR16 has 23% identity at the amino acids level and 71% identity at the nucleotide level with the ScPDR16 gene. Inspection of the promoter region (1000 bp) of the CgPDR16 gene revealed the presence of one PDRE sequence (5′-TCCGTGGGA-3′) which has been found to be associated with genes whose expression in C. glabrata is under the control of CgPDR1 (Tsai et al., 2010; Ferrari et al., 2011), one YRE sequence (5′-TTACAAA-3′) known to be recognized by CgYap1p (Chen et al., 2007; Lelandais et al., 2008) and two sequences (5′-CGGTGTT-3′) probably recognized by CgStb5p (Noble et al., 2013).

Figure 1.

Heterologous complementation of the S. cerevisiae pdr16Δ mutation with the plasmid-borne ScPDR16 and CgPDR16 genes. The wild-type (wt) and pdr16Δ mutant strains contain an empty vector, pRS306K. ScPDR16(sc)–centromeric plasmid YCp-lac111-ScPDR16; ScPDR16(mc)–multicopy plasmid YEp-lac195-ScPDR16; CgPDR16–pRS306K-CgPDR16. Growth of yeast cells on YPD medium at 28 °C was scored after 3 days

These results expand the phenotype of the pdr16Δ mutant and support the assumption that azole hypersensitivity of the pdr16Δ mutant is caused by disruption of a single gene encoding CgPdr16p. This gene is a functional homologue of ScPDR16 and may be activated by the two main transcription factors (CgPdr1p and CgYap1p) involved in the control multidrug resistance in C. glabrata.

Susceptibility to oxidants

To evade macrophages and neutrophiles producing reactive oxygen species (ROS), fungal cells possess specific antioxidation mechanisms involving enzymes directly related to virulence (Nicola et al., 2008; Nikolaou et al., 2009; Roetzer et al., 2011). Moreover, endogenous reactive oxygen species were reported to be an important mediator of miconazole and fluconazole effects (Kobayashi et al., 2002). Therefore, we assessed the susceptibility of the wild-type and pdr16Δ mutant strains to different oxidants. As also shown in Table 2, the susceptibility of C. glabrata to CTBT, inducing superoxide generation in yeasts and filamentous fungi (Batova et al., 2010; Culakova et al., 2012), as well as to other oxidants inducing ROS formation, such as paraquat, menadione, diamide and hydrogen peroxide (Herrero et al., 2008), was not altered by the pdr16Δ mutation. While deletion of the YAP1 gene, but not SKN7, regulating the response to oxidative stress (Roetzer et al., 2011), slightly enhanced the susceptibility of yeast cells to CTBT (Figure 2), antioxidants such as glutathione, cysteine and ascorbate, but not quercetin and resveratrol, present in growth medium, decreased CTBT susceptibility in both the wild-type and pdr16Δ mutant strains (Table 3). These results indicate that the PDR16 gene is not involved in the oxidative stress response in C. glabrata.

Figure 2.

Susceptibility to CTBT (2 µg/disc) of the C. glabrata ΔHTU strain and its mutants deleted in the CgYAP1 and CgSKN7 genes. Cells were grown on YNB-GE medium at 37 °C for 2 days

Table 3. Effect of antioxidants on the growth inhibition induced by CTBT (5 µg/disc) in the wild-type (wt) and pdr16Δ mutant strains
AntioxidantConcentration (mm)Diameter of growth inhibition zone (mm)
wtpdr16Δ
  1. Cells were grown on YNB-D medium in the presence of the indicated antioxidants.

None2020
Glutathione11011
 599
Cysteine199
 5810
Ascorbate51312
 101213
Quercetin52018
 101920
Resveratrol51920

CTBT is a chemosensitizing agent enhancing fungal sensitivity to other antifungal agents, such as fluconazole, itraconazole and 5-fluorocytosine (Cernicka et al., 2007; Culakova et al., 2012). As shown in Figure 3, co-application of CTBT and subinhibitory concentrations of itraconazole or 5-fluorocytosine induced markedly larger growth inhibitory zones around CTBT (5 µg/disc) compared with media not containing these antifungals. This indicates that the drug-sensitizing effects of CTBT and the pdr16Δ mutation in C. glabrata are run by different mechanisms that may be additive when superimposed in the same cells.

Figure 3.

Enhanced growth inhibition zones of CTBT (5 µg/disc) in the presence of subinhibitory concentrations of itraconazole and 5-fluorocytosine in YNB-D medium. Growth of the wild-type (wt) and pdr16Δ mutant strains at 37 °C was scored after 2 days

Suppression of itraconazole resistance by the PDR16 deletion

The gain-of-function mutations in the CgPDR1 gene have been shown to result in overexpression of CgPDR1 (Tsai et al., 2006) and its target genes encoding the drug efflux membrane transporter genes (Vermitsky et al., 2006; Berila et al., 2009; Ferrari et al., 2009, 2011; Tsai et al., 2010). As expected, transformants of the wild-type and pdr16Δ mutant strains containing the gain-of-function PDR1(Cg4672) allele on pCgPDR14672 plasmid were resistant to itraconazole compared with their counterparts containing an empty vector (Figure 4). However, the pdr16Δ mutant cells containing the plasmid-borne gain-of-function PDR1(Cg4672) mutant allele were less resistant to itraconazole than the corresponding PDR1(Cg4672) transformant cells derived from the wild-type strain BG14. These results indicate that the absence of a functional Pdr16p also sensitizes the azole-resistant mutant strain bearing the gain-of-function mutation in the CgPDR1 gene.

Figure 4.

Effect of the pdr16Δ mutation on the susceptibility to itraconazole of the C. glabrata transformants containing the pCgPDR14672 plasmid. The wild-type (wt) and pdr16Δ mutant strains contain an empty cloning vector. Growth of yeast cells on YNB-D medium at 37 °C was scored after 2 days

Rhodamine 6G accumulations and efflux

The azole hypersensitivity of the pdr16Δ mutant can be caused by changes in passive diffusion across the membrane or in active efflux from cells. Therefore, rhodamine 6G was used to assess the drug accumulation and efflux in both wild-type and mutant strain. Rhodamine 6G is an acknowledged substrate of the multidrug resistance efflux pumps energized by ATP in both S. cerevisiae (Kolaczkowski et al., 1996) and C. glabrata (Izumikawa et al., 2003; Puri et al., 2011). The extent of its accumulation in yeast was determined by measuring the drug concentrations in supernatants of de-energized cells exposed to rhodamine 6G, as well as in cells separated by membrane filtration. Compared with the wild-type strain, the rhodamine 6G accumulation was higher in the pdr16Δ mutant cells (Figure 5). Rhodamine 6G uptake increased rapidly when de-energized cells of both strains were exposed to the drug and further rose slowly 5 min after drug addition (Figure 5A). Under used experimental conditions, after 20 min of incubation, the mutant cells accumulated 1.63 ± 0.27 times more drug than the wild-type cells. This is in good agreement with the image analyses of cells separated by filtration, accumulating rhodamine 6G for 30 min, revealing that the pdr16Δ mutant accumulates on average 1.54 ± 0.24 times more drug than its parental strain (Figure 5B). On the other hand, the energy-dependent rhodamine 6G efflux from the dye-preloaded cells was similar in wild-type and mutant cells (Figure 6), indicating that the increased susceptibility to azole antifungals in the pdr16Δ mutant strain could be caused at least by their enhanced passive diffusion into mutant cells and not by their reduced efflux from cells.

Figure 5.

Deletion of the CgPDR16 gene leads to an increased rhodamine 6G accumulation in de-energized cells. The wild-type (o) and pdr16Δ mutant (Δ) strains were depleted of energy and incubated in 10 µm rhodamine 6G, as described in Materials and methods. (A) Rhodamine 6G accumulation in cells was deduced from the decrease of the drug fluorescence in supernatants of centrifuged samples. (B) Image analyses of cells separated by filtration after 30 min of drug accumulation by de-energized cells. Dots of separated cells from a representative experiment are shown above the columns. Data are averages from five independent experiments

Figure 6.

Glucose-induced efflux of rhodamine 6G from the wild-type and pdr16Δ mutant cells. Data are averages from four independent experiments

Cell surface hydrophobicity and biofilm formation

Along with the overexpression of CgPDR1 (and its target genes) due to its gain-of-function mutations (Vermitsky and Edlind, 2004; Tsai et al., 2006; Berila et al., 2009; Ferrari et al., 2009, 2011), cell surface hydrophobicity (CSH) and biofilm formation are well-recognized virulence factors in C. glabrata (Kaur et al., 2005; Bialkova and Subik, 2006; Berila et al., 2011). They influence adherence to biotic and abiotic surfaces, colonization and pathogenesis of fungal infections. As shown in Figure 7, the pdr16Δ mutant displayed significantly lower CSH and biofilm production compared to the wild-type strain or transformants containing the plasmid-borne CgPDR16 gene. This indicates that CgPDR16 may also contribute to the virulence of the C. glabrata.

Figure 7.

Reduced cell surface hydrophobicity (CSH) (A) and biofilm production (B) of the pdr16Δ mutant quantified by crystal violet staining (A570, black bars) and XTT reduction assay (A490, grey bars). The wild-type (wt) and pdr16Δ mutant strains contain an empty vector. The CgPDR16 gene was introduced into pdr16Δ mutant cells on the pCgACU5–CgPDR16 plasmid. Significance: *, +, p < 0.05; **, ++, p < 0.01. Values were calculated via Student's t-test: *, **, significant differences between wt and pdr16Δ mutant; +, ++, significant differences between pdr16Δ mutant and transformant containing a plasmid-borne CgPDR16 gene

Discussion

This study shows that CgPDR16 is a functional homologue of the ScPDR16 gene and its disruption confers a pleiotropic phenotype. This phenotype is manifested by increased sensitivity of cells to several azole antifungals, enhanced accumulation of rhodamine 6G, reduced cell surface hydrophobicity and biofilm formation. The CgPDR16 gene is synthenic with ScPDR16 and encodes a protein of 344 amino acids which is seven amino acids shorter than ScPdr16p. The CgPDR16 promoter displays the putative C. glabrata PDRE and YRE sequences, indicating that the expression of CgPDR16 may be regulated by the two multidrug resistance transcription factors CgPdr1p and CgYap1p. In S. cerevisiae and C. albicans the PDR16 gene is under the control of their respective counterparts ScPdr1p/ScPdr3p (DeRisi et al., 2000), ScYap1p (Lelandais et al., 2008), Tac1p (Znaidi et al., 2007) and Cap1p (Znaidi et al., 2009). In contrast to S. cerevisiae (DeRisi et al., 2000), however, no remarkable overexpression of CgPDR16 has been observed in transcriptome analyses using strains either containing gain-of-function CgPDR1 alleles (Vermitsky et al., 2006; Tsai et al., 2010; Caudle et al., 2011; Ferrari et al., 2011) or exposed to drugs (Lelandais et al., 2008).

CgPdr16p plays a significant role in C. glabrata drug resistance. The absence of this protein decreased the azole resistance not only in C. glabrata wild-type cells but also in transformants harbouring a gain-of-function mutation in the CgPDR1 gene. These results corroborate those observed in S. cerevisiae (van den Hazel et al., 1999) and C. albicans (Saidane et al., 2006), and point to the link between membrane lipid homeostasis and the drug susceptibility of yeast cells. In S. cerevisiae, Pdr16p exhibits phosphatidylinositol transport activity in vitro, controls the levels of various lipids, is localized to lipid particles and microsomes and its abundance increases in response to DNA replication stress (van den Hazel et al., 1999; Schnabl et al., 2003; Griac, 2007; Tkach et al., 2012).

As in S. cerevisiae (van den Hazel et al., 1999), despite azole hypersensitivity, the susceptibility of the pdr16Δ mutant cells to other drugs tested was not altered. The pdr16Δ mutation did not affect the sensitivity to oxidants, either. Apparently, CgPdr16p is not involved in the C. glabrata defence to ROS generated by CTBT, menadione, paraquat, diamide and hydrogen peroxide. These oxidants are known to activate Yap1p involved in response to both chemical and oxidative stresses (Moye-Rowley, 2003). The absence of this transcription factor resulted in a slightly increased susceptibility of cells to CTBT generating superoxide (Batova et al., 2010; Culakova et al., 2012). The susceptibility of cells was not enhanced further after deletion of the CgSKN7 gene, corroborating a main role of CgYap1p in the oxidative stress response (Roetzer et al., 2011). Whereas antioxidants such as glutathione, cysteine and ascorbate attenuated CTBT toxicity, the combination of oxidative stress induced by CTBT with chemical stress induced by itraconazole or 5-fluorocytosine rendered yeast cells more sensitive to drugs. This chemosensitizing effect of CTBT, however, was not abrogated by the pdr16Δ mutation.

The enhanced susceptibility to azole antimycotics in the C. glabrata pdr16Δ mutant strain could be explained at least by their increased passive uptake. The latter probably does not play a significant role in other drugs exhibiting unaltered antifungal efficiency. Although de-energized pdr16Δ mutant cells displayed a higher rhodamine 6G accumulation compared with the wild-type strain, the glucose-activated rhodamine 6G efflux from the drug-loaded cells was similar in both wild-type and pdr16Δ mutant cells. A slightly higher range of its efflux in the mutant strain probably reflects the enhanced accumulation of rhodamine 6G in pdr16Δ cells compared with the wild-type strain. Increased passive rhodamine 6G and fluconazole accumulation has been previously observed in the azole-hypersensitive S. cerevisiae erg2, erg3, erg4 and erg6 mutants, lacking activities of enzymes involved in ergosterol biosynthesis (Mukhopadhyay et al., 2002). In the erg6 mutant, enhanced rhodamine 6G accumulation in de-energized cells has been demonstrated without affecting the Pdr5p-mediated drug efflux (Emter et al., 2002), supporting the assumption that the altered lipid composition of yeast membranes can increase the sensitivity of cells to azole antimycotics, merely enhancing their passive diffusion across the membrane. In fact, in S. cerevisiae the ScPDR16 gene has been shown to control the lipid composition of the plasma membrane, which modifies the passive diffusion of hydrophobic drugs across the plasma membrane, thereby modulating multidrug resistance (van den Hazel et al., 1999). The complementation of azole hypersensitivity and altered sterol composition of the S. cerevisiae pdr16Δ mutant by the overexpressed PDR17 gene (Simová et al., 2013), known to be essential for intermembrane transport of phosphatidylserine, along with the activity and the subcellular distribution pattern of ScPdr16p, seems to be compatible with a proposed role of Pdr16p in regulating sterol and phospholipid biosynthesis (Schnabl et al., 2003; Griac, 2007) and in phosphoinositide signalling during lipid transfer and membrane transport in yeast cells (Ghosh and Bankaitis, 2011; Schuh and Audhya, 2012).

The participation of Pdr16p in the dynamics of yeast membranes during the biogenesis of transport vesicles on the trans-Golgi network and endosomal membranes (Bankaitis et al., 2010) could also explain the decreased production of biofilm and cell surface hydrophobicity displayed in the C. glabrata pdr16Δ mutant cells. These observations point to certain differences in the cell wall structure that may be associated with the presence of both phosphatidylinositol-linked aspartyl proteases induced during cell wall remodelling and a family of cell wall-localized adhesins required for the adherence and virulence of C. glabrata (Bialkova and Subik, 2006; Kaur et al., 2007). This yeast species exhibits a significant positive correlation between a relative cell surface hydrophobicity and adhesion to abiotic surfaces being essential for the colonization (Luo and Samaranayke, 2002), and between the degree of in vitro biofilm formation and the virulence of the pathogen (Hasan et al., 2009).

Since the two virulence factors studied are associated with pathogenic functions that are required to cause diseases in vivo, CgPdr16p could serve as a potential target for novel drugs inhibiting its function. Along with the suppression of C. glabrata virulence, they could also augment the antifungal activity of azole antimycotics and decrease the frequency of appearance of multidrug-resistant cells. In S. cerevisiae, the deletion of ScPDR16 significantly reduces the capacity of the yeast population to evolve resistance to fluconazole (Anderson et al., 2009).

In summary, the pdr16Δ mutation in C. glabrata results in: (a) increased sensitivity to azole antimycotics; (b) unaltered sensitivity to oxidants that further increase the sensitivity of cells to antifungals, inhibiting sterol and nucleic acid biosynthesis; (c) a higher accumulation of rhodamine 6G while maintaining the activity of membrane efflux pumps; and (d) decreased cell surface hydrophobicity and suppressed biofilm formation associated with the pathogen virulence. These results point to a relationship among intracellular membrane trafficking, phosphatidylinositide metabolism, resistance to azole antifungals and virulence in pathogenic C. glabrata.

Acknowledgements

We thank B. Cormack, A. Goffeau, P. Griac, R. Kaur, D. Sanglard, Ch. Schuller and H. F. Tsai for the strains and plasmids used in this study. This work was supported by the Slovak Grant Agency of Science (Grant No. VEGA 1/0867/12), the Slovak Research and Developmental Agency (Grant No. APVV-0282-10) and Comenius University (Grant No. UK 281/2013).

Ancillary