The heme-binding protein Dap1 links iron homeostasis to azole resistance via the P450 protein Erg11 in Candida glabrata

Authors


Correspondence: Taiga Miyazaki, Department of Molecular Microbiology and Immunology, Nagasaki University School of Medicine, 1-7-1 Sakamoto, Nagasaki 852-8501, Japan. Tel.: +81 95 819 7273; fax: +81 95 849 7285; e-mail: taiga-m@nagasaki-u.ac.jp

Abstract

The pathogenic fungus Candida glabrata is relatively resistant to azole antifungals, which target lanosterol 14α-demethylase (Erg11p) in the ergosterol biosynthesis pathway. Our study revealed that C. glabrata exhibits increased azole susceptibility under low-iron conditions. To investigate the molecular basis of this phenomenon, we generated a strain lacking the heme (iron protoporphyrin IX)-binding protein Dap1 in C. glabrata. The Δdap1 mutant displayed growth defects under iron-limited conditions, decreased azole tolerance, decreased production of ergosterol, and increased accumulation of 14α-methylated sterols lanosterol and squalene. All the Δdap1 phenotypes were complemented by wild-type DAP1, but not by DAP1D91G, in which a heme-binding site is mutated. Furthermore, azole tolerance of the Δdap1 mutant was rescued by exogenous ergosterol but not by iron supplementation alone. These results suggest that heme binding by Dap1 is crucial for Erg11 activity and ergosterol biosynthesis, thereby being required for azole tolerance. A Dap1-GFP fusion protein predominantly localized to vacuolar membranes and endosomes, and the Δdap1 cells exhibited aberrant vacuole morphologies, suggesting that Dap1 is also involved in the regulation of vacuole structures that could be important for iron storage. Our study demonstrates that Dap1 mediates a functional link between iron homeostasis and azole resistance in C. glabrata.

Introduction

Azole antifungals inhibit ergosterol biosynthesis by targeting the cytochrome P450 enzyme, lanosterol 14α-demethylase, encoded by ERG11 (Odds et al., 2003). Candida glabrata is the second most common cause of candidemia after Candida albicans and is most notable for its reduced susceptibility to azole antifungals such as fluconazole (Pfaller & Diekema, 2007; Pfaller, 2012). As only limited antifungal agents are available in clinical settings, there is an urgent need to develop a therapeutic strategy effective for a broad range of fungal pathogens, including C. glabrata.

Iron is an essential element for all living organisms, serving as a co-factor for a variety of proteins. Host-iron availability and its intracellular metabolism are crucial for the virulence of pathogens (Sutak et al., 2008; Almeida et al., 2009). In addition, previous studies of C. albicans, Cryptococcus neoformans, and Aspergillus fumigatus have demonstrated that these pathogenic fungi display increased fluconazole susceptibilities under iron-limited conditions (Prasad et al., 2006; Zarember et al., 2009; Kim et al., 2012). Iron depletion leads to down-regulation of ERG11 and a partial defect in ergosterol biosynthesis, thereby enhancing membrane fluidity and passive drug diffusion in C. albicans cells (Prasad et al., 2006). The syntheses of heme (iron protoporphyrin IX) and ergosterol are closely related, because they share the same upstream precursors (Weinstein et al., 1986). However, the molecular base of the link between iron metabolism and azole resistance are poorly understood in pathogenic fungi.

In the model yeast Saccharomyces cerevisiae, the heme-binding protein damage-resistance protein 1 (Dap1), which is related to cytochrome b5 (Mifsud & Bateman, 2002), is required for cell growth under low-iron conditions (Craven et al., 2007). A genome-wide analysis has revealed that some members of the iron regulon involved in siderophore transport and high-affinity iron transport systems are down-regulated in the Δdap1 mutant, indicating a specific role for Dap1 in iron metabolism (Jo et al., 2009). Intriguingly, Dap1 is also involved in ergosterol biosynthesis through the activation of Erg11 in S. cerevisiae (Hand et al., 2003; Mallory et al., 2005; Craven et al., 2007). It has also been reported that Dap1 directly binds to and positively regulates Erg11 in Schizosaccharomyces pombe (Hughes et al., 2007).

In the present study, our objectives were to evaluate the effects of iron starvation on azole susceptibility in C. glabrata and to explore molecular mechanisms linking iron homeostasis with azole resistance. To this end, we have characterized the functions of the Dap1 ortholog in C. glabrata.

Materials and methods

Media, drugs, and culture conditions

Candida glabrata cells were routinely propagated at 30 °C in synthetic defined (SD) medium, synthetic complete (SC) medium, or SC medium lacking tryptophan (SC-trp) (Kaiser et al., 1994), unless otherwise noted. Iron-depleted conditions were induced by adding the bacterial siderophore desferrioxamine (DFO; EMD Chemicals, San Diego, CA) or the Fe2+ chelator bathophenanthroline disulfonate (BPS; MP Biomedicals, Solon, OH) into media. Fluconazole was purchased from Sigma (St. Louis, MO), and voriconazole was kindly provided by Pfizer (New York, NY). Voriconazole was dissolved in dimethyl sulfoxide (DMSO), and other compounds were dissolved in distilled water. DMSO alone did not interfere with cell growth at the final concentrations used in this study. The AnaeroPack system (Mitsubishi Gas Chemical Company Inc., Tokyo, Japan) was used for cultures under conditions of low oxygen tension.

Strain and plasmid construction

Strains, primers, and plasmids used in this study are listed in Tables 1, 2, and 3, respectively. A C. glabrata DAP1 deletion construct was made using a one-step PCR-based technique, as described previously (Miyazaki et al., 2010a,b). Briefly, a 1-kb DNA fragment containing C. glabrata HIS3 was amplified from pBSK-HIS (Miyazaki et al., 2010a,b) using primers tagged with 100-bp sequences homologous to the flanking regions of the DAP1 open reading frame (ORF). Candida glabrata strain KUE200 (Ueno et al., 2007) was transformed with the deletion construct according to a lithium acetate protocol (Cormack & Falkow, 1999), and the resulting transformants (Δdap1 mutants) were selected by histidine prototrophy. PCR and Southern blotting were performed to verify that the desired homologous recombination occurred at the target locus without ectopic integration of the deletion construct.

Table 1. Strains used in this study
StrainGenotype or descriptionReference or source
CBS138Wild-typeDujon et al. (2004)
2001TΔtrp1Kitada et al. (1995)
KUE200Δhis3, Δtrp1Ueno et al. (2007)
TG122001T containing pCgACTMiyazaki et al. (2011)
TG152001T containing pCgACTP-ERG11This study
TG251Δdap1::HIS3, Δtrp1 (made from KUE200)This study
TG252TG251 containing pCgACTThis study
TG253TG251 containing pCgACT-DAP1This study
TG254TG251 containing pCgACT-DAP1D91GThis study
TG255TG251 containing pCgACTP-DAP1-GFPThis study
TG256TG251 containing pCgACTP-ERG11This study
Table 2. Primers used in this study
PrimeraTarget geneSequence (5′–3′)b
  1. a

    ‘F’ and ‘R’ indicate forward and reverse primers, respectively.

  2. b

    Sequences homologous to flanking regions of the target open reading frame are shown in italics. Sequences shown in boldface are present in pBSK-HIS. Restriction enzyme sites are underlined.

CgDAP1 100-F CgDAP1 CTGAAAAGATGATTAATAAGGCCTGGATTGTTGGTGAAGGATAATATTGCAATTGATTTATTATTATATTTTTACCGAGTTGATAACAGGTTAACTAGG TAATACGACTCACTATAGGGC
CgDAP1 100-R CgDAP1 TTCATAAAAACTAAAAACAAACAATTCAGCAGGTAATATTTCCTAGACTATTTCCATGCAGAGTCACCTCAACAACTTGGGAGATATGTATCACTCTTA CGCTCTAGAACTAGTGGATCC
CgDAP1-F(-989)-Bam CgDAP1 CGAGGATCCTTGCGAACCAAGAGTGCTCC
CgDAP1-F1-Bam CgDAP1 CGGGATCCATGTCCTTCTTGAAGAATTTACT
CgDAP1-R(459)-Xho CgDAP1 CCGCTCGAGCTAAACATTGACACCTGGCT
CgDAP1-R456-HdIII CgDAP1 CCCAAGCTTTTAACATTGACACCTGGCTCAG
CgDAP1-mut-F1 CgDAP1 GGTGCCTCTCGTGGTCTTGCG
CgDAP1-mut-R1 CgDAP1 ATGACCTGCAAAATTAGAGTATGGT
EGFP1-F-HdIII yEGFP1 AAAAAGCTTTATTAAAATGTCTAAAGGTGAAG
EGFP1-R-Xho yEGFP1 CCGCTCGAGCTGCAGTTATTTGTACAATTCATCCATA
CgERG11-F1-Xho CgERG11 CCGCTCGAGATGTCCACTGAAAACACTTCTTTGG
CgERG11-R(190)-Xho CgERG11 CCGCTCGAGAGCAGCAAAGCCCTCTAAACG
Table 3. Plasmids used in this study
PlasmidDescriptionReference
pBSK-HISpBluescript II SK+ containing C. glabrata HIS3Miyazaki et al. ( 2010a, b)
pCgACTC. glabrata centromere-based plasmid containing autonomously replicating sequence and C. glabrata TRP1Kitada et al. (1996)
pCgACT-PpCgACT containing the S. cerevisiae PGK1 promoter, polylinker, and C. glabrata HIS3 3′UTRMiyazaki et al. (2010a, b)
pYGFP1pUC19 containing yEGFP1Cormack et al. (1997)
pCgACT-DAP1pCgACT containing the 1-kb upstream region and open reading frame of C. glabrata DAP1This study
pCgACT-DAP1D91GpCgACT containing a C. glabrata DAP1 variant, in which the residues D91 within the heme-binding domain was mutated to glycineThis study
pCgACTP-DAP1-GFPpCgACT-P containing the C. glabrata DAP1-yEGFP1 fusion geneThis study
pCgACTP-ERG11pCgACT-P containing C. glabrata ERG11This study

To generate complementation plasmids, C. glabrata genes were amplified from the genomic DNA of CBS138 (Dujon et al., 2004). A 1.5-kb PCR fragment containing the 1-kb upstream region and ORF of C. glabrata DAP1 was amplified with primers CgDAP1-F(-989)-Bam and CgDAP1-R(459)-Xho, digested with BamHI and XhoI, and then inserted into the BamHI–SalI site of pCgACT to generate pCgACT-DAP1. A Dap1-D91G mutation was introduced in pCgACT-DAP1 using the KOD-Plus-Mutagenesis kit (Toyobo, Osaka, Japan) and mutagenic primers CgDAP1-mut-F1 and CgDAP1-mut-R1, forming pCgACT-DAP1D91G. A 0.5-kb PCR fragment containing the C. glabrata DAP1 ORF without stop codon was amplified from pCgACT-DAP1 with primers CgDAP1-F1-Bam and CgDAP1-R456-HdIII, and then digested with BamHI and HindIII. A 0.7-kb PCR fragment containing yEGFP1 were obtained from pYGFP1 with primers EGFP1-F-HdIII and EGFP1-R-Xho, and then digested with HindIII and XhoI. These DNA fragments were concurrently inserted into the BamHI-SalI site of pCgACT-P to generate pCgACTP-DAP1-GFP. A 1.8-kb PCR fragment containing C. glabrata ERG11 was amplified with primers CgERG11-F1-Xho and CgERG11-R(190)-Xho, digested with XhoI, and then inserted into the SalI site of pCgACT-P to generate pCgACTP-ERG11. All of the plasmids constructed using PCR products were verified by sequencing before use.

Drug susceptibility assays

Broth microdilution tests were performed essentially according to CLSI M27-A3 protocols using SD medium in a checkerboard format. The fractional inhibitory concentration (FIC) index was calculated by the following formula based on the MICs of 2 drugs used alone as well as in combination: FIC index = (MIC80 of azole in combination/MIC80 of azole alone) + (MIC80 of DFO in combination/MIC80 of DFO alone). Drug interaction was classified as synergistic if its FIC index was ≤ 0.5, as described previously (Odds, 2003; Johnson et al., 2004).

Spot dilution tests were performed as described previously (Miyazaki et al., 2010a, b). Logarithmic-phase cultures of each C. glabrata strain were adjusted to 2 × 107 cells per mL, and then 5 μL of serial 10-fold dilutions were spotted onto SC-trp plates containing each compound at the indicated concentrations. Plates were photographed after 48 h of incubation at 30 °C. All sensitivity tests were performed on at least three separate occasions to ensure reproducibility.

Sterol analysis

Candida glabrata cell pellets (10 OD600 units) were harvested, washed twice with 0.5% Tween 80, and then rewashed with and resuspended in 1% NaCl. Ten micrograms of sitosterol, dissolved in chloroform, was added as an internal control. Total lipids were extracted using the Bligh-Dyer method (Bligh & Dyer, 1959) and saponified with 0.4 M methanolic KOH (Lewis et al., 1987). The extracted lipids were dried under N2 gas and dissolved in 100 μL of chloroform/methanol (2 : 1). Ten microliters of the aliquot were injected onto a C-18 column (COSMOSIL 5C18-AR-II 4.6 × 150 mm) and analyzed by reverse-phase high-performance liquid chromatography (HPLC) using the Alliance HPLC system with the Waters 2695 separation module and 2489 UV/visible detector (Waters Corporation, Milford, MA). The mobile phase consisted of 99.5% methanol with 5 mM CH3COONH4. Elution of compounds was automatically monitored by absorption at 210 nm. For sterol quantification, standard curves were prepared using the commercially available sterols (ergosterol, lanosterol, sitosterol, and squalene; Sigma). The dynamic range of each sterol standard curve was 10 ng–6 μg per injection. All experiments were performed on 2 independent occasions to ensure the reproducibility of the results.

Fluorescence microscopy

Vacuolar staining with FM4-64 was carried out as described previously (Vida & Emr, 1995) with a few modifications. Briefly, live exponentially growing C. glabrata cells (OD600 0.8–1.0) were harvested, washed, and resuspended in fresh SC-trp broth. The fluorescent dye FM4-64 (Molecular Probes, Eugene, OR) was added at a final concentration of 10 μM, and the cells were incubated at 30 °C for 15 min. The cells were then washed, resuspended in fresh SC-trp broth, and incubated for 30–60 min for steady-state experiments. Images were collected using a Carl Zeiss LSM780 confocal laser-scanning microscope and processed with the ZEN 2011 software (Carl Zeiss, Jena, Germany). FM4-64 was excited by a diode-pumped solid-state (DPSS) laser at 561 nm and Dap1-GFP was excited by an argon laser at 488 nm. For quantification of cells with aberrant vacuole structures, approximately 500 cells were examined microscopically for each sample. All experiments were performed on three separate occasions.

Results and discussion

Candida glabrata displays increased azole susceptibility under iron-limited conditions

To evaluate the effects of iron starvation on azole susceptibility of the C. glabrata wild-type strain, we conducted checkerboard assays using twofold serial dilutions of DFO, fluconazole, and voriconazole. Because C. glabrata is unable to utilize the bacterial siderophore DFO, the addition of this substance to growth medium induces an iron-depleted condition for this fungus (Nevitt & Thiele, 2011). Increased fluconazole and voriconazole susceptibilities were observed in the presence of DFO even in C. glabrata (Fig. 1). FIC indexes of the combination of fluconazole and DFO and the combination of voriconazole and DFO were 0.5 and 0.375, respectively, indicating synergistic effects of these azoles and DFO against C. glabrata.

Figure 1.

Azole susceptibility of Candida glabrata wild-type strain under iron-depleted conditions. MICs were determined by a broth microdilution test in a checkerboard format essentially according to CLSI M27-A3 guidelines. Cells were grown in SD medium in the presence of the bacterial siderophore desferrioxamine and fluconazole (left panel) or voriconazole (right panel) at the indicated concentrations. Plates were incubated at 35 °C for 48 h and absorbance at 600 nm (OD600) was measured.

Identification of the DAP1 ortholog in C. glabrata

To determine the molecular basis of the relationship between iron homeostasis and azole resistance in C. glabrata, we addressed functional characterization of the C. glabrata Dap1 ortholog. A single Dap1 ortholog (NCBI accession No.: XP_446124, Genolevures ID: CAGL0F03553g) was identified in the C. glabrata genome by BLASTp search of the NCBI (http://www.ncbi.nlm.nih.gov/BLAST/) and Genolevures (http://genolevures.org/blast.html#) databases, using the entire amino acid sequence of S. cerevisiae Dap1 (YPL170W) as a query. The deduced amino acid sequence of C. glabrata Dap1 displayed 85.6% similarity and 73.9% identity with that of S. cerevisiae Dap1. The NCBI BLASTp search also identified Dap1 orthologs in other Candida and Aspergillus species with high sequence homology, especially in the cytochrome b5-like heme-binding domain (Fig. 2). Based on the significant sequence homology, we named the corresponding C. glabrata gene DAP1.

Figure 2.

Amino acid sequence alignment of fungal Dap1 orthologs. The cytochrome b5-like heme-binding domain is underlined. An asterisk indicates the conserved heme-binding site. GenBank accession numbers of the Dap1 orthologs listed in this figure: Saccharomyces cerevisiae, NP_015155; Candida glabrata, XP_446124; Candida albicans, XP_717441; Candida parapsilosis, CCE42585; Candida tropicalis, XP_002545589; Candida dubliniensis, XP_002421969; Aspergillus fumigatus, XP_754570; Aspergillus niger, EHA22036; Aspergillus terreus, XP_001213759; and Aspergillus nidulans, XP_662543.

Candida glabrata cells require Dap1 for growth under iron-limited conditions and in the presence of azole antifungals

To investigate Dap1 functions, a Δdap1 mutant lacking the entire DAP1 ORF was constructed and then transformed with an empty vector and a plasmid containing a C. glabrata gene of interest. In S. cerevisiae Dap1, mutation of residue D91 in the heme-binding domain to glycine (D91G) abolished its heme-binding activity (Mallory et al., 2005). As this residue is highly conserved among fungi (indicated by an asterisk in Fig. 2), we introduced the corresponding mutation in C. glabrata DAP1 to generate a DAP1D91G variant. The Δdap1 mutant displayed growth defects on iron-deficient media containing 100 μM DFO or 25 μM BPS; moreover, these phenotypes could be complemented by wild-type DAP1, but not by DAP1D91G (Fig. 3a). As expected, supplementation of the media with 1 mM ferric chloride (FeCl3) completely rescued growth of all the strains tested. These results suggested that Dap1 plays a role in iron metabolism and that its heme-binding activity is necessary for this process in C. glabrata, consistent with previous findings in S. cerevisiae (Craven et al., 2007). On the other hand, the Δdap1 and DAP1D91G mutants exhibited increased azole susceptibility that was not fully rescued by addition of 1 mM FeCl3 into the media (Fig. 3a), indicating that both iron and Dap1 are required for azole resistance in C. glabrata.

Figure 3.

Spot dilution assay. (a) Serial 10-fold dilutions of Candida glabrata log-phase cells were spotted onto SC-trp plates containing each compound at the indicated concentrations. Plates were incubated at 30 °C for 48 h. C. glabrata strains: Wild-type control, TG12; Δdap1 + vector, TG252; Δdap1 + DAP1, TG253; Δdap1 + DAP1D91G, TG254; and Δdap1 + DAP1-EGFP1, TG255. FeCl3, ferric chloride; DFO, desferrioxamine; BPS, bathophenanthroline disulfonate; FLC, fluconazole; VRC, voriconazole; and FEN, fenpropimorph. (b) Effects of ergosterol supplementation on azole susceptibility of the C. glabrata DAP1 mutants. The spot dilution assay was performed as in part (a) except that C. glabrata cells were grown under low oxygen tension to facilitate ergosterol uptake. Plates were incubated at 30 °C for 3 days.

Dap1 is involved in ergosterol biosynthesis via Erg11 in C. glabrata

The impaired growth of the Δdap1 and DAP1D91G mutants in the presence of azole antifungals was completely rescued by supplementation of the media with exogenous ergosterol (Fig. 3b). In contrast to the results with fluconazole and voriconazole, the loss of DAP1 did not affect susceptibility to fenpropimorph (Fig. 3a), which targets C-8 sterol isomerase (Erg2p) and C-14 sterol reductase (Erg24p) (Marcireau et al., 1990; Jia et al., 2002). These results suggest a possible involvement of Dap1 in ergosterol biosynthesis via Erg11. Thus, we next performed sterol assays using reverse-phase HPLC. Compared with the wild-type and DAP1-complemented strains, the Δdap1 and DAP1D91G mutants exhibited slightly reduced amounts of ergosterol and more than twofold higher accumulation of lanosterol and squalene (Fig. 4), which are 14α-methylated sterol intermediates produced upstream of Erg11 in the ergosterol biosynthetic pathway (Onyewu et al., 2003; Martel et al., 2010; Hull et al., 2012) (Fig. 5); this indicates a partial defect of Erg11 activity in the mutants. In S. cerevisiae, Dap1 does not affect ERG11 transcription but regulates the stability of the Erg11 protein (Mallory et al., 2005). Our qRT-PCR assay also revealed that the expression level of ERG11 in the Δdap1 mutant was comparable to that in the wild-type strain in C. glabrata (data not shown). Intriguingly, the overexpression of ERG11 resulted in increased azole tolerance in the wild-type background, while this effect was modest in the Δdap1 background in C. glabrata (Fig. 6), in contrast to previous findings in S. cerevisiae where ERG11 overexpression confers itraconazole resistance in both wild-type and Δdap1 backgrounds (Mallory et al., 2005). Taken together, these results suggest that Dap1 acts on Erg11 post-transcriptionally depending on its heme-binding activity but may have an additional role in azole resistance in C. glabrata. It is worth mentioning that the changes in sterol compositions in the C. glabrata Δdap1 mutant were similar to those previously observed in S. cerevisiae wild-type cells grown under low-iron conditions (Shakoury-Elizeh et al., 2010). In addition, another P450 enzyme, C-22 sterol desaturase (Erg5p), acts downstream of Erg11 in the ergosterol biosynthesis pathway (Lamb et al., 1999). A previous study in S. cerevisiae has suggested that Erg5 is not a target for Dap1 in methyl methanesulfonate resistance, but the Erg5p substrate ergosta-5,7-dienol accumulated in a Δdap1 mutant under normal growth conditions (Mallory et al., 2005). Future studies are needed to determine whether Erg5 is a target for Dap1 in azole resistance in S. cerevisiae and C. glabrata.

Figure 4.

Sterol analysis. Candida glabrata cells were grown under normal growth conditions using SD medium. Sterol contents were analyzed by reverse-phase HPLC, as described in Materials and Methods. The data shown are the results of a single experiment that was repeated on two independent occasions with similar results. C. glabrata strains: Wild-type control, TG12; Δdap1 + vector, TG252; Δdap1 + DAP1, TG253; and Δdap1 + DAP1D91G, TG254.

Figure 5.

Ergosterol biosynthesis pathway. A linear model of the pathway was adapted from references (Onyewu et al., 2003; Martel et al., 2010; Hull et al., 2012), and only selected intermediates and enzymes are shown. Solid and broken arrows indicate single and multiple enzymatic steps, respectively. The known targets of fluconazole, voriconazole, and fenpropimorph are indicated.

Figure 6.

Effects of ERG11 overexpression on azole susceptibility in wild-type and Δdap1 backgrounds. Serial 10-fold dilutions of Candida glabrata cells containing either an empty vector or an ERG11-overexpression (ERG11-OE) plasmid were spotted onto SC-trp plates containing fluconazole (FLC) and voriconazole (VRC) at the indicated concentrations. Plates were incubated at 30 °C for 48 h. C. glabrata strains: Wild-type + vector, TG12; Wild-type + ERG11-OE, TG15; Δdap1 + vector, TG252; and Δdap1 + ERG11-OE, TG256.

Dap1 plays a role in the regulation of vacuole structures

We next investigated intracellular localization of Dap1 using a Dap1-GFP fusion protein in C. glabrata. The Δdap1 phenotypes were complemented by the introduction of DAP1-EGFP1 into the mutant (Fig. 3a), indicating that this fusion protein is biologically functional. Dap1-GFP predominantly localized to FM4-64-stained vacuole membranes and endosomes in C. glabrata (Fig. 7a). In S. cerevisiae, Dap1 localizes mainly to endosomes (Craven et al., 2007), while Erg11 localizes to the ER (Huh et al., 2003), cytoplasm (Natter et al., 2005), and granular sites (Wiwatwattana et al., 2007). Although Dap1 may participate in intracellular heme trafficking to activate the P450 hemoprotein Erg11, a stable physical interaction between Dap1 and Erg11 has not been detected in S. cerevisiae (Craven et al., 2007). These results suggest that interaction between Dap1 and Erg11 may be transient or indirect in S. cerevisiae and C. glabrata.

Figure 7.

Fluorescence microscopy. (a) Colocalization of GFP-tagged Dap1 with FM4-64-stained vacuole membranes and prevacuolar endosomal compartments. Live exponentially growing Candida glabrata wild-type cells expressing Dap1-GFP were stained with 10 μM FM4-64 at 30 °C for 15 min, and images were collected using a confocal laser-scanning microscope, as described in Materials and Methods. Arrows indicate Dap1-GFP located at endosomal fractions. Bar, 2 μm. (b) Comparison of vacuole morphology between the wild-type and Δdap1 cells in C. glabrata. FM4-64-staining was performed as described in part (a). Bar, 2 μm. (c) Percentage of cells with aberrant vacuolar structures in the C. glabrata wild-type and Δdap1 strains. Data were obtained by microscopic examination of approximately 500 cells for each sample. The experiments were repeated 3 times on independent occasions, and the means with standard deviations are shown.

The vast majority of vacuoles were stained with FM4-64 as circular structures in the C. glabrata wild-type cells, whereas a significant fraction of vacuoles displayed abnormal structures, such as multiple small constructions with irregular edges, in the Δdap1 mutant (Fig. 7b and c). It is likely that vacuolar iron storage is impaired in cells with aberrant vacuole structures (Raguzzi et al., 1988; Craven et al., 2007); thus, the results may, at least in part, account for the decreased tolerance of the Δdap1 mutant to iron starvation.

Concluding remarks

To our knowledge, this is the first report in pathogenic fungi, characterizing Dap1 as a molecule linking iron homeostasis and azole resistance. Dap1 interacted genetically with Erg11, thereby indirectly participating in ergosterol biosynthesis in C. glabrata. Iron depletion or dysfunction of Dap1 impairs Erg11 activity, leading to increased azole susceptibility in C. glabrata. Previous studies in pathogenic fungi, such as C. albicans, A. fumigatus, C. neoformans, and Rhizopus oryzae, suggest the potential use of iron chelators in antifungal therapy (Ibrahim et al., 2006, 2007; Prasad et al., 2006; Zarember et al., 2009; Fiori & Van Dijck, 2012; Kim et al., 2012), although the adverse effects of this therapy on mammalian cells remain an issue. Our results suggest that inhibition of Dap1 could be an alternative approach. As Dap1 is highly conserved at the sequence level from yeast to filamentous fungi, but not in humans, further studies on Dap1 orthologs in other pathogenic fungi are warranted in the quest for novel antifungal strategies.

Acknowledgements

We thank Dr. Brendan Cormack for providing pYGFP1, Dr. Hironobu Nakayama for pCgACT, and Dr. Hiroji Chibana for KUE200. This research was supported by a grant from the Global Centers of Excellence Program, Nagasaki University, Grant-in-Aid for Scientific Research (JSPS KAKENHI no. 24791027 to T.M. and no. 21390305 to Y.M. and S.K.) from the Japanese Ministry of Education, Culture, Sports, Science and Technology, grants from the Ministry of Health, Labour and Welfare, Japan (H23-shinko-ippan-007, H23-shinko-ippan-018, and H24-shinko-ippan-013 to Y.M.), and an Ueda award from the Japanese Society of Chemotherapy (T.M.).

Authors' contribution

N. Hosogaya and T. Miyazaki contributed equally to this work.

Ancillary