HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Marvin, M. E. Articles by Cashmore, A. M. Search for Related Content PubMed PubMed Citation Articles by Marvin, M. E. Articles by Cashmore, A. M. Agricola Articles by Marvin, M. E. Articles by Cashmore, A. M.

The Candida albicans CTR1 gene encodes a functional copper transporter

¹ Department of Genetics, University of Leicester, Leicester LE1 7RH, UK
² Department of Microbiology and Immunology, University of Leicester, Leicester LE1 7RH, UK

Correspondence
Annette M. Cashmore
amc19{at}le.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Copper and iron uptake in Saccharomyces cerevisiae are linked through a high-affinity ferric/cupric-reductive uptake system. Evidence suggests that a similar system operates in Candida albicans. The authors have identified a C. albicans gene that is able to rescue a S. cerevisiae ctr1/ctr3-null mutant defective in high-affinity copper uptake. The 756 bp ORF, designated CaCTR1, encodes a 251 amino acid protein with a molecular mass of 27·8 kDa. Comparisons between the deduced amino acid sequence of the C. albicans Ctr1p and S. cerevisiae Ctr1p indicated that they share 39·6 % similarity and 33·0 % identity over their entire length. Within the predicted protein product of CaCTR1 there are putative transmembrane regions and sequences that resemble copper-binding motifs. The promoter region of CaCTR1 contains four sequences with significant identity to S. cerevisiae copper response elements. CaCTR1 is transcriptionally regulated in S. cerevisiae in response to copper availability by the copper-sensing transactivator Mac1p. Transcription of CaCTR1 in C. albicans is also regulated in a copper-responsive manner. This raises the possibility that CaCTR1 may be regulated in C. albicans by a Mac1p-like transactivator. A C. albicans ctr1-null mutant displays phenotypes consistent with the lack of copper uptake including growth defects in low-copper and low-iron conditions, a respiratory deficiency and sensitivity to oxidative stress. Furthermore, changes in morphology were observed in the C. albicans ctr1-null mutant. It is proposed that CaCTR1 facilitates transport of copper into the cell.

The GenBank accession number for the sequence reported in this paper is AJ277398.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Copper is an important cofactor for a wide variety of cellular enzymes that carry out essential biological processes such as respiration, iron acquisition and protection against oxidative stress. The problem, however, is that copper exists in the environment as insoluble complexes and is toxic in the presence of oxygen because of the formation of destructive hydroxyl free radicals. Therefore, organisms have evolved mechanisms for the transport of copper into the cell and for maintaining intracellular concentrations at non-toxic levels.

In the baker's yeast Saccharomyces cerevisiae, a well-documented system has been described where the acquisition of copper is linked to the acquisition of iron. Like copper, iron presents a similar problem to the cell as it is essential for numerous biological processes, is found in the environment as insoluble complexes and is toxic in the presence of molecular oxygen. In S. cerevisiae, copper and iron acquisition is achieved through a specific high-affinity reductive uptake system that is negatively regulated by both metals. Both copper and iron are rendered soluble by a cell-surface ferric/cupric-reductase complex encoded by the genes FRE1 and FRE2 (Dancis et al., 1990; Georgatsou & Alexandraki, 1994). Uptake of iron is facilitated by a specific iron transporter complex comprising an iron permease, Ftr1p, and a multicopper ferroxidase, Fet3p (Askwith et al., 1994; Stearman et al., 1996). Copper uptake requires two functionally redundant high-affinity copper transporters, Ctr1p and Ctr3p, localized in the plasma membrane (Dancis et al., 1994a, b; Knight et al., 1996). Cells that express either Ctr1p or Ctr3p are competent for high-affinity copper transport. However, ctr1/ctr3 mutants exhibit copper starvation phenotypes that include the inability to grow on non-fermentable carbon sources, lack of measurable high-affinity iron uptake, and defective copper/zinc-superoxide dismutase activity (Dancis et al., 1994a, b; Knight et al., 1996). Three copper chaperones that deliver copper to specific intracellular targets have also been identified, Lys7p to superoxide dismutase, Cox17p to the mitochondria and Atx1p to the gene product of CCC2 (Culotta et al., 1997; Glerum et al., 1996; Lin et al., 1997; Pufahl et al., 1997). Ccc2p is an intracellular copper transporter that in turn delivers copper to the lumen of the Golgi for incorporation into Fet3p (Yuan et al., 1995; Lin et al., 1997). Since Fet3p relies on copper for biological activity, inhibition of copper uptake has the secondary effect of significantly reducing iron uptake (Askwith et al., 1994; Yuan et al., 1995; De Silva et al., 1995).

A further five additional FRE-like genes (FRE3–7) have been identified by their sequence similarity to FRE1 and FRE2. It has been demonstrated that the proteins encoded by FRE1, FRE2, FRE3 and FRE4 can reduce iron bound to low-molecular-mass iron-binding compounds called siderophores (Yun et al., 2001). As S. cerevisiae cannot produce its own siderophores, these gene products may facilitate the scavenging of iron bound to siderophores produced by other species. The specific function of the additional three reductases (encoded by FRE5–7) remains unclear.

The interdependence of high-affinity copper and iron uptake in S. cerevisiae extends to the transcriptional activators Mac1p and Aft1p that operate in conditions of copper or iron depletion and recognize specific consensus sequences in the promoters of various genes (Radisky & Kaplan 1999; Yamaguchi-Iwai et al., 1995, 1996; Jungmann et al., 1993). The copper-responsive activator Mac1p regulates FRE1, CTR1, CTR3 and FRE7, the uncharacterized homologue of FRE1, while the iron-responsive activator Aft1p regulates FRE1, FRE2, FRE3-6, FTR1, FET3, ATX1, CCC2, and a family of intracellular siderophore transporter genes ARN1–4 (Dancis et al., 1992; Lin et al., 1997; Jungmann et al., 1993; Yamaguchi-Iwai et al., 1996; Labbe et al., 1997; Martins et al., 1998; Lesuisse et al., 1998; Heymann et al., 2000; Yun et al., 2000a, b). Thus FRE1 is regulated by both copper and iron through the activity of Mac1p and Aft1p, whereas the copper-trafficking genes ATX1 and CCC2 are regulated by iron through the activity of Aft1p. Two low-affinity transporters have also been described, Ctr2p for copper and Fet4p for iron and copper (Dix et al., 1994; Kampfenkel et al., 1995; Hassett et al., 2000).

Candida albicans is a dimorphic opportunistic pathogen that is a commensal of the human mouth and gastrointestinal tract in about 30–50 % of the population. However, in immunocompromised patients, C. albicans can cause both superficial and life-threatening diseases (Scherer & Magee 1990). Two genes have been identified in C. albicans (CaCUP1 and CaCRP1) that confer an unusually high resistance to copper (Weissman et al., 2000). The CaCUP1 gene encodes a metallothionein that sequesters internalized copper and CaCRP1 encodes a plasma-membrane P-type ATPase pump that transports copper out of the cell. Physiological concentrations of copper have been shown to be toxic to C. albicans under the acidic anaerobic conditions found in the gastrointestinal tract (Weissman et al., 2000). Therefore, by reducing intracellular free copper concentrations the products of these genes are believed to facilitate the survival of C. albicans in this environmental niche.

Iron has been implicated as an important factor for the growth, survival and virulence of C. albicans (Valenti et al., 1986; Chaffin et al., 1998; Fratti et al., 1998). The organism can acquire iron from haem and can bind to and lyse erythrocytes (Manns et al., 1994; Moors et al., 1992). Although C. albicans can produce siderophores (Ismail et al., 1985; Sweet & Douglas, 1991) it is still not clear how it obtains ferrous iron from these complexes or from other sources. Growing evidence suggests that a similar reductive system to that described for S. cerevisiae operates in C. albicans and several homologous genes have been identified. Our laboratory has previously reported that C. albicans has a cell-surface-associated ferric/cupric-reductase that is regulated similarly to the S. cerevisiae enzyme (Morrissey et al., 1996). We have also reported a C. albicans gene (CFL1) that is able to rescue mutant phenotypes associated with a S. cerevisiae fre1-null mutant and, like the S. cerevisiae FRE1 gene, is regulated in response to both copper and iron availability (Hammacott et al., 2000). Eck et al. (1999) described a C. albicans multicopper oxidase gene (CaFET3) that shares 55 % identity with the S. cerevisiae FET3 gene. A C. albicans fet3/fet3 mutant strain displayed slow growth in low-iron medium, suggesting a loss of high-affinity iron uptake. Ramanan & Wang (2000) described two C. albicans high-affinity iron permeases, encoded by CaFTR1 and CaFTR2. C. albicans mutants lacking CaFTR1 displayed a severe growth defect in low-iron conditions and, importantly, were unable to establish a systemic infection in mice. These observations suggest that, as in S. cerevisiae, copper and iron uptake in C. albicans are intimately linked such that reduced copper uptake may also affect the uptake of iron.

To further investigate copper uptake in this organism, we have isolated and characterized a C. albicans gene, CaCTR1. This gene is able to rescue the mutant phenotypes associated with a S. cerevisiae ctr1/ctr3-null mutant and we describe the sequence, transcriptional regulation and functional analyses of this gene. Our findings support the role of CaCTR1 as a copper transporter gene. This is believed to be the first report of a transporter facilitating the uptake of copper by C. albicans.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Strains.
The S. cerevisiae strain W303 (MAT; SUC2; ade2-1; can1-100; his3-11,115; leu2-3,112 trp1-1; ura3-1, R. J. Rothstein, Columbia University, New York, USA) is a derivative of strain S288C (Yeast Genetic Stock Center), which is known to harbour a ctr3-null allele due to a Ty2 insertion in the promoter separating the TATA box from the transcriptional start site by 6 kb (Knight et al., 1996). One-step gene replacement (Rothstein, 1983) was used to generate a ctr1-null allele in strain W303 (to generate S. cerevisiae strain W303ctr1 : : URA3) using a disruption cassette as follows. Primers containing XbaI and KpnI sites (ScCTR1-233 and ScCTR1+1525, Table 1) were used to amplify a 1758 bp fragment containing the entire CTR1 ORF. The resulting fragment was cloned into the KpnI and XbaI sites of the vector pUC19 to generate recombinant plasmid pCO1. Additional primers incorporating BamHI sites (ScCTR1+62 and ScCTR1+1131, Table 1) were then used in an inverse PCR reaction with pCO1 in order to delete 1068 bp from the cloned CTR1 ORF. The amplification product was digested with BamHI, and allowed to self-ligate, generating plasmid pCD1. A 1141 bp fragment of plasmid YDpU (Berben et al., 1991) carrying the S. cerevisiae URA3 gene was inserted into the BamHI site of pCD1. The resulting construct, designated pCD2, carries a 1798 bp CTR1 disruption cassette comprising the URA3 marker flanked by 280 bp and 377 bp of genomic targeting sequences. The disruption cassette was removed from pCD2 by digestion with KpnI and XbaI, and used to transform S. cerevisiae strain W303 selecting for uracil prototrophy. Homologous recombination leading to disruption of the CTR1 gene was confirmed by Southern blot analysis of genomic DNA digested with EcoRV and probed with a [-³²P]CTP-labelled CTR1 KpnI and XbaI fragment excised from plasmid pCOI (data not shown). The S. cerevisiae mac1-null strain BY4741mac1 : : KanMX4 is a derivative of the wild-type strain BY4741 (MATa; his31; leu20; met150; ura30). Both strains were obtained from the EUROSCARF deletion library (http://www.rz.uni-frankfurt.de/FB/fb16/mikro/euroscarf/index.html).

imm434

imm434

Escherichia coli strain DH5 [80lacZM15 recA1 endA1 gyrA96 thi-1 hsdR17() supE44 relA1 deoI (lacZYA–argF)U169] was used for propagation of library clones and plasmid constructs. For transposon mutagenesis the donor strain was streptomycin-sensitive E. coli DH1 [recA1 endA1 gyrA96 thi-1 hsdR17() supE44 relA1] carrying plasmid R388 transposed with modified Tn1000 containing the S. cerevisiae HIS3 gene as a selectable marker. The recipient strain was MH1578 [recA endA1 gyrA96 thi-1 hsdR17() supE44 relA1 rpsL], a streptomycin-resistant derivative of DH1 (Sedgwick & Morgan, 1994). Plasmid pCRC1 was used to transform E. coli strain DH1 [R388 : : Tn1000(HIS3)], which was then used as a donor strain for conjugal mating with recipient strain MH1578 with selection for resistance to ampicillin and streptomycin. Plasmid DNA was isolated from transconjugants and used to transform strain W303ctr1 : : URA3 with selection for leucine and histidine prototrophy.

Growth conditions.
S. cerevisiae and C. albicans cultures were grown at 30 °C unless otherwise stated. For growth in non-selective conditions yeast-extract peptone medium was used, with glucose added at a final concentration of 2 % w/v (YPD). To test the strains for the ability to grow on non-fermentable carbon sources, glucose was replaced with either 3 % (v/v) glycerol (YPG) or 3 % (v/v) ethanol (YPE). To test for sensitivity to oxytetracycline, overnight cultures were transferred into fresh yeast-extract peptone medium and grown to a cell density of 1x10⁷ cells ml^-1. The cells were then diluted back to 1x10³ cells ml^-1 and 100 µl was transferred to solid YPD medium containing 100 µg ml^-1 and 1 mg ml^-1 oxytetracycline, respectively. For yeast grown in selective conditions SD medium [0·67 % (w/v) yeast nitrogen base with ammonium sulphate (B101)] was used. Minimal defined medium (MD) was used to verify the ability of strains to grow in low-copper or low-iron conditions; the medium was based on the yeast nitrogen base recipe of Wickerham (1951) with the addition of 20 mM sodium citrate pH 4·2 (Eide et al., 1992) and omitting copper or iron as necessary. Bathocuproine disulphonic acid (BCS) or bathophenanthroline disulphonic acid (BPS) were added to media to reduce the availability of copper and iron, respectively. The final concentration of BCS or BPS was 50 µM unless stated otherwise. Amino acid supplements were added to all selective media at final concentrations described by Sherman et al. (1986) and glucose was added at a concentration of 2 % (w/v). Growth of all strains in liquid medium was monitored using a haemocytometer at 30 min intervals. For all solid media 2 % (w/v) agar was added. E. coli cultures were grown at 37 °C in Luria–Bertani medium [1 % (w/v) Oxoid Bacto-tryptone; 0·5 % (w/v) Oxoid Bacto-yeast extract; 0·5 % (w/v) sodium chloride pH 7·2].

Genetic techniques.
A C. albicans Sau3AI partial digest genomic library cloned into the BamHI site of vector YEp213 was obtained from P. Meacock, Department of Genetics, University of Leicester, UK. Transformation of S. cerevisiae and C. albicans was achieved by the lithium acetate method described by Geitz et al. (1992) and Braun & Johnson (1997), respectively. Isolation of plasmid DNA from S. cerevisiae was achieved using the procedure described by Holm et al. (1986). Plasmid DNA was isolated and purified from E. coli by alkaline lysis (Ish-Horowicz & Burke, 1981) and transformation of E. coli was performed using the calcium chloride method (Mandel & Higa, 1970).

Qualitative solid-phase ferric-reductase assay.
A modified version of the qualitative ferric-reductase assay described by Dancis et al. (1990) was used. Transformed strains were grown on SD agar plates for 3–5 days and placed at 4 °C overnight. Cells were transferred to nylon filters (Hybond-N) and incubated for 5 h on the surface of MD agar medium containing 300 µM ,'-dipyridyl and 2 mM ferric chloride. The filters were placed on Whatman 3MM paper soaked in assay buffer [50 mM trisodium citrate pH 6·5, 5 % (w/v) glucose] for 5 min and then transferred to fresh 3MM paper soaked in assay buffer containing 1 mM FeCl₃ and 1 mM BPS and incubated for a further 5 min. Ferric-reductase activity was determined by red staining of the filter around colonies caused by the formation of a BPS–[Fe²⁺] complex.

DNA manipulation and sequence analysis.
Restriction analysis, DNA cloning, agarose gel electrophoresis and Southern blot analysis were performed using standard procedures (Maniatis et al., 1982). Polymerase chain reactions were carried out in a MIR-D30 DNA amplifier (Sanyo), and DNA sequencing was performed using an ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin Elmer) and a DNA sequencer (ABI model 377XL).

Hyphal induction.
Overnight cultures were grown in YPD medium and transferred to fresh YPD medium at a cell density of 1x10⁶ cells ml^-1. Growth was allowed to continue at 37 °C with agitation. Following addition of 5 % (w/v) bovine calf serum, cells were examined every 15 min for germ tube formation.

RNA isolation and Northern blot analysis.
Total RNA was extracted from mid-exponential-phase cultures of S. cerevisiae and C. albicans strains (1x10⁷ cells ml^-1) by the SDS-hot phenol method described by Schmidt et al. (1990). Purified RNA was separated by electrophoresis on a 1·7 % denaturing agarose gel and then transferred to a nylon filter (Hybond-N) and hybridized overnight at 65 °C in 0·5 M sodium phosphate pH 7·4, 7 % w/v SDS, 1 mM EDTA (Church & Gilbert, 1984) with [-³²P]CTP-labelled probes. The CTR1 probe was a 508 bp PCR product representing an internal fragment of the C. albicans CTR1 gene, generated using primers CaCTR1+60 and CaCTR1+568 (Table 1). To ensure equal loading of samples, total RNA was also probed with a 342 bp HindIII–EcoRI fragment of the S. cerevisiae ACT1 gene purified from plasmid pBS-Actin (Peter A. Meacock, personal communication), or a 1·3 kb XbaI–SalI fragment of the C. albicans URA3 gene purified from plasmid YpB1 (Goshorn et al., 1992). Stringent washes were carried out at 65 °C using 3x SSC/0·1 % (w/v) SDS, 1x SSC/0·1 % (w/v) SDS and 0·5x SSC/0·1 % (w/v) SDS. The filter was then exposed to Fuji Medical X-ray film at -80 °C to visualize hybridizing bands.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Functional rescue of S. cerevisiae mutant strain W303ctr1 : : URA3
High-affinity copper uptake in S. cerevisiae is facilitated by two genes, CTR1 and CTR3 (Dancis et al., 1994a, b; Knight et al., 1996). However, the parent of the S. cerevisiae strain used in this study (W303) is S288C, which is known to carry a defective allele of CTR3 due to a Ty2 insertion (Knight et al., 1996). Therefore, it was only necessary to disrupt CTR1 in strain W303 in order to construct a mutant that was defective in high-affinity copper uptake. A S. cerevisiae strain carrying a null allele of CTR1 was generated using one-step gene disruption (Rothstein, 1983). The resulting strain, W303ctr1 : : URA3, displayed phenotypes typical of a strain defective in high-affinity copper uptake, specifically a respiratory deficiency resulting in the inability to grow on yeast-extract peptone medium containing non-fermentable carbon sources such as glycerol and ethanol (YPG and YPE), high ferric-reductase activity on iron-rich medium (Dancis et al., 1994a) and the inability to grow on MD medium depleted of copper or iron (data not shown).

The respiratory-deficient phenotype of strain W303ctr1 : : URA3 was exploited to isolate a C. albicans gene functionally analogous with S. cerevisiae CTR1 or CTR3. Strain W303ctr1 : : URA3 was transformed with a YEp213 C. albicans Sau3AI partial digest genomic library (Peter A. Meacock, personal communication) and screened on YPE medium. Selected colonies were tested for ferric-reductase activity using a qualitative plate assay (Dancis et al., 1990). A single colony that exhibited wild-type low ferric-reductase activity on high-iron medium was picked for further study. Transformation of W303ctr1 : : URA3 with plasmid DNA (designated pCRC1) from this colony resulted in complementation of all the ctr1/ctr3-null-associated phenotypes (data not shown). Since these phenotypes reflect the lack of high-affinity copper transport, we propose that pCRC1 contains a C. albicans gene encoding a protein that facilitates copper transport into the cell.

Localization and sequence analysis of the rescuing gene in pCRC1
Restriction digest analysis indicated that plasmid pCRC1 carried a 9·3 kb genomic insert. To identify the rescuing ORF within this insert, a library of pCRC1 derivatives that carried random transposon insertions was assembled using transposon mutagenesis (Sedgewick & Morgan, 1994). Transformants were screened on YPE; three colonies unable to grow on this medium were picked and plasmid DNA used to retransform strain W303ctr1 : : URA3 in order to confirm the mutant phenotype. Restriction analysis showed that transposon insertion sites in these three plasmids were located within a 1·1 kb EcoRI fragment of insert DNA (Fig. 1).

View larger version (77K): [in this window] [in a new window]   Fig. 1. DNA and predicted protein sequence of C. albicans CTR1. The nucleotide sequence is numbered with respect to the transcriptional ATG start site with A of the start codon as +1. The putative TATA box is underlined with a double score. Transposon insertions mapped within a 1·1 kb EcoRI fragment are denoted as double-headed horizontal black arrows at nucleotides 122, 134 and 658. Promoter sequences resembling S. cerevisiae CuRE motifs are depicted as black boxes (nucleotides -398 to -90, -269 to -277, -238 to -230 and -146 to -138). Repeated amino acid sequences in the predicted C. albicans CTR1 protein are boxed in light grey for the two core sequences (residues 16–26 and 33–45) and as dark grey arrows for the four repeats between amino acids 11–19, 20–28, 29–37 and 38–46. The only example of a MXXM motif found is boxed in white, found between residues 52 and 55. High-scoring transmembrane segments (91–107, 194–210 and 216–232) identified using a PSORT II analysis are shown as dark grey blocks with white text.

Analysis of the promoter region of CaCTR1 using the S. cerevisiae promoter regulatory sequence analysis website (http://rsat.ulb.ac.be/rsat/; Van Helden et al., 2000), revealed four motifs with the consensus sequence ^5'GCTCAT^3' (Fig. 1). Three of the four motifs were immediately preceded by ^5'TTT^3', making them identical to cis-acting copper response elements (CuREs) found in the promoters of CTR1, CTR3, FRE1 and FRE7 of S. cerevisiae (Labbe et al., 1997; Yamaguchi-Iwai et al., 1997; Martins et al., 1998). The core sequence of these transcription regulatory elements, ^5'TTTGC(T/G)C(A/G)^3', is essential for binding of the copper-sensing transactivator Mac1p (Jungmann et al., 1993; Georgatsou et al., 1997; Graden & Winge, 1997; Joshi et al., 1999). The fourth C. albicans motif differed from the S. cerevisiae consensus by the lack of two of the three T residues at the 5' end. Three of the potential C. albicans CuREs are preceded immediately by A, which has been shown to facilitate more efficient binding of Mac1p to the S. cerevisiae high-affinity copper transporter CTR1 (Joshi et al., 1999).

The predicted protein product of C. albicans CTR1
A BLASTp search (Altschul et al., 1997) using the predicted C. albicans Ctr1p sequence as a query against the SWISS-PROT database revealed significant similarity with the S. cerevisiae high-affinity copper transporter Ctr1p, as well as the Homo sapiens hCtr1p (Zhou & Gitschier, 1997) and the high-affinity copper transporter of Schizosaccharomyces pombe, Ctr4p (Zhou & Thiele, 2001). Comparison with the S. cerevisiae genome database (http://genome-www.stanford.edu/Saccharomyces) using the FastA program (Pearson & Lipman, 1988) revealed significant similarity to all three of the S. cerevisiae copper transporters Ctr1–3p. Direct comparison using the Needleman & Wunsch algorithm alignment (GAP) from the GCG package (Wisconsin package version 9.1, Genetics Computer Group, Madison, WI, USA) showed that the C. albicans Ctr1p had 39·6 % similarity and 33·0 % identity to the S. cerevisiae Ctr1p.

Analysis of the C. albicans CTR1 protein sequence using the SAPS analysis program (http://www.isrec.isb-sib.ch/software/SAPS_form.html; Brendel et al., 1992) revealed a predicted protein of 251 amino acids with a molecular mass of 27·8 kDa, lacking an amino-terminal leader sequence. Dancis et al. (1994a) previously reported that S. cerevisiae Ctr1p has a methionine- and serine-rich amino-terminal region that contains a putative copper-binding domain. This region includes three 19 amino acid repeats and 11 examples of the motif Met-XX-Met, which is also found in bacterial copper-binding proteins (Cha & Cooksey, 1991; Odermatt et al., 1993). Compositional analysis against all proteins in the SWISS-PROT database indicated that both methionine and serine are highly represented in the C. albicans predicted protein, predominantly at the amino end (residues 1–71). A search for repetitive sequences within this 71 amino acid region revealed four repeats of the motif M(A/X)M(S/A)(S/A/X)(S/A/T)(S/A/T)(S/A/T/X)(S/X) in and around two repeated core blocks of SAT(M/_)SMAM(S/_)(A/S)TS [where _=no residue and X=other uncharged or non-polar amino acids D/M/V with one exception, H], shown in Fig. 1. Only one example of a Met-XX-Met motif was found at residues 52 to 55 (MEGM), but there were eight examples of the motif Met-X-Met (residues 11–13, 20–22, 29–31, 36–38, 38–40, 50–52, 62–64 and 64–66). Distribution of all other amino acid types revealed two high-scoring transmembrane segments at residues 90–104 and 202–229, while pSORT type II analysis (http://psort.nibb.ac.jp/) indicated three transmembrane domains at residues 91–107, 194–210 and 216–232 (Fig. 1).

C. albicans CTR1 expression responds to copper availability in the growth medium
Northern blot analysis was carried out to examine expression of C. albicans CTR1 in low- and high-copper medium (Fig. 2). Total RNA was isolated from mid-exponential-phase cultures (1x10⁷ cells ml^-1) of C. albicans strain SC5314 grown in unsupplemented MD-BCS medium or in MD-BCS medium supplemented with 50 µM, 100 µM or 250 µM cupric chloride and probed with a 508 bp internal fragment of the CaCTR1 gene. After an exposure of 1 h the autoradiograph revealed a 1·1 kb band representing a highly transcribed gene in low-copper conditions.

View larger version (40K): [in this window] [in a new window]   Fig. 2. Northern blot analysis of the C. albicans CTR1 transcripts. Total RNA was isolated from cells growing exponentially in MD-BCS medium containing 0 µM (lane 1), 50 µM (lane 2), 100 µM (lane 3) and 250 µM (lane 4) cupric chloride. Following electrophoresis and transfer to a nylon membrane, duplicate sets of the four RNA samples were probed with either an [-32P]CTP-labelled 508 bp fragment of the C. albicans CTR1 gene or a 1·3 kb EcoRI fragment of the C. albicans URA3 gene as a loading control. Autoradiographs were then exposed to the labelled membrane for 1 h (a) or 48 h (b, c). After 48 h exposure further bands appeared, one fractionally greater than 1·1 kb observed in high-copper conditions and another much larger band (3·1 kb) in low-copper conditions.

View larger version (26K): [in this window] [in a new window]   Fig. 3. Northern blot analysis of the CaCTR1 transcripts in S. cerevisiae. Wild-type strain BY4741 (MAC1) and mutant strain BY4741mac1 : : KanMX4 (mac1) were transformed with plasmid YEp213 (lanes 1–2 and 7–8), pCRC1 (lanes 3–4 and 9–10) or pCaCO1 (lanes 5–6 and 11–12). The resulting transformed strains were grown in either unsupplemented MD-BCS medium (lanes 1, 3, 5, 7, 9 and 11) or MD-BCS medium supplemented with 100 µM cupric chloride (lanes 2, 4, 6, 8, 10 and 12). Total RNA was isolated from the transformed S. cerevisiae cells growing exponentially. Following electrophoresis and transfer to a nylon membrane the RNA samples were probed with an [-32P]CTP-labelled 508 bp fragment of the C. albicans CTR1 gene or a 342 bp fragment of the S. cerevisiae ACT1 gene as a loading control. Autoradiographs were then exposed to the labelled membrane for 30 min (CaCTR1) or 4 h (ACT1) to visualize hybridized bands.

When grown on YPD medium, strain BWP17ctr1 : : URA3/ctr1 : : ARG4 was slow-growing and formed small white colonies resembling S. cerevisiae petite mutants. It was also observed that with prolonged incubation (30 °C for 7 days), colonies of both strain BWP17ctr1 : : URA3/CTR1 and strain BWP17ctr1 : : URA3/ctr1 : : ARG4 produced invasive filaments that penetrated the growth medium (Fig. 4). In addition, strain BWP17ctr1 : : URA3/CTR1 always produced wrinkly colonies on YPD medium (Fig. 4b, e). To investigate whether copper available in the growth medium had an effect on morphology, overnight liquid cultures of BWP17ctr1 : : URA3/ctr1 : : ARG4, and wild-type BWP17 were set up in unsupplemented MD-BCS medium. The cells were then transferred at a concentration of 1x10⁶ cells ml^-1 into fresh pre-warmed unsupplemented MD-BCS or MD-BCS supplemented with 100 µM cupric chloride. Growth was continued with agitation and the cells were observed at 15 min intervals.

View larger version (84K): [in this window] [in a new window]   Fig. 4. Growth of strains BWP17 (a, d), BWP17ctr1 : : URA3/CTR1 (b, e) and BWP17ctr1 : : URA3/ctr1 : : ARG4 (c, f) on yeast-extract peptone medium. All three strains were streaked onto the medium and incubated at 30 °C for 7 days. Colony formation by each strain is shown in the top row (a–c) and a heavy inoculum of cells is shown in the bottom row (d–f).

View larger version (69K): [in this window] [in a new window]   Fig. 5. Changes in morphology displayed by strain BWP17ctr1 : : URA3/ctr1 : : ARG4 in unsupplemented liquid MD-BCS cultures. Cells growing in unsupplemented MD-BCS medium were transferred at a cell density of 1x106 cells ml-1 to fresh unsupplemented MD-BCS or MD-BCS supplemented with 100 µM cupric chloride. Growth of the cells was examined every 15 min and photographs were taken at 400x magnification. (a) The mutant BWP17ctr1 : : URA3/ctr1 : : ARG4 grown in unsupplemented MD-BCS medium; (b) wild-type BWP17 grown in unsupplemented MD-BCS medium; (c) strain BWP17ctr1 : : URA3/ctr1 : : ARG4 in MD-BCS medium supplemented with 100 µM cupric chloride. All the cells shown are at the mid-exponential phase of growth (5x106–2x107 cells ml-1).

View larger version (55K): [in this window] [in a new window]   Fig. 6. Strain BWP17ctr1 : : URA3/ctr1 : : ARG4 displays a growth defect on low-copper and low-iron media, and is respiratory-deficient and sensitive to hydrogen peroxide. (a, b) Cells grown in unsupplemented MD-BCS medium were harvested at mid-exponential phase, washed three times and suspended in distilled water at a concentration of 1x107 cells ml-1. Fivefold serial dilutions were made (1x107 to 5x104 cells ml-1, left to right) and 2 µl of each suspension was spotted in duplicate onto (a) MD medium containing 50 µM of the copper chelator BCS and supplemented with 100 µM cupric chloride or with no added copper, or (b) MD medium containing 50 µM of the iron chelator BPS and supplemented with 100 µM cupric chloride or with no added copper. (c) Single colonies were streaked onto unsupplemented YP medium containing either 2 % (w/v) glucose (YPD), or 3 % (w/v) glycerol (YPG) or ethanol (YPE) as a carbon source, or onto YPD, YPG and YPE medium containing 100 µM cupric chloride. (d) C. albicans strains were grown in unsupplemented MD-BCS medium to a cell density of 1x107 cells ml-1. The cells were then washed and 1 ml was suspended in 12 ml 1 % molten soft agarose equilibrated to 37 °C. This was used as a top layer to either unsupplemented MD-BCS medium or MD-BCS medium supplemented with 100 µM cupric chloride, and 10 µl hydrogen peroxide (8·8 M) was dispensed onto a nylon disc placed at the centre of the plate. The lack of growth forming a ‘halo’ around the disc was observed after 5 days incubation.

It has previously been reported that S. cerevisiae strains carrying null mutations of the SOD1 gene, which encodes Cu/Zn-superoxide dismutase, are sensitive to the antibiotic oxytetracycline (Avery et al., 2000). Levels of oxytetracycline as low as 20 µg ml^-1 in yeast-extract peptone medium have been shown to arrest growth of sod1 strains, whereas wild-type strains are unaffected at levels as high as 1 mg ml^-1. S. cerevisiae ctr1 and mac1 strains show similar sensitivities to oxytetracycline (Angrave et al., 2001) as a result of deficient levels of intracellular copper. This may be due to defective Cu/Zn superoxide dismutase activity, or through the direct antioxidant capability of copper (Liu & Culotta, 1994). Therefore we tested the growth of strains BWP17, BWP17 BWP17ctr1/CTR1 and BWP17ctr1 : : URA3/ctr1 : : ARG4 on yeast-extract peptone medium (YPD) containing oxytetracycline at concentrations of 100 µg ml^-1 and 1 mg ml^-1 (data not shown). Colony formation by the wild-type strain BWP17 and strain BWP17ctr1 : : URA3/CTR1 was unaffected by both concentrations of oxytetracycline, with colonies being formed on all the media after 2 days incubation. However, only very small colonies of BWP17ctr1 : : URA3/ctr1 : : ARG4 were apparent on YPD containing 1 mg oxytetracycline ml^-1 even after 7 days incubation. Colonies of BWP17ctr1 : : URA3/ctr1 : : ARG4 on YPD containing 100 µg oxytetracycline ml^-1 were observed after 3 days incubation although they were reduced in size in comparison to the other two strains.

Finally, we reconstituted CaHIS1 alone, and CaHIS1 along with CaCTR1, in strain BWP17ctr1 : : URA3/ctr1 : : ARG4 in order to construct the control strains BWP17ctr1/HIS1 and BWP17ctr1/HIS1 : : CTR1 (see Methods). We subsequently tested these strains along with the wild-type control strain, DAY185 (Davis et al., 2000), for the mutant phenotypes associated with the lack of copper uptake (Fig. 6). Strain BWP17ctr1/HIS1 displayed phenotypes that were identical to the Cactr1-null strain, BWP17ctr1 : : URA3/ctr1 : : ARG4. Strain BWP17ctr1/HIS1 : : CTR1 displayed phenotypes that were similar to strain BWP17ctr1 : : URA3/CTR1, showing that reintegration of the wild-type CaCTR1 allele to the his1 : : hisG locus was sufficient for mutant rescue. Interestingly, strain BWP17ctr1/HIS1 : : CTR1 also grew as wrinkly colonies on YPD medium.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
We have isolated and characterized a C. albicans genomic clone that is able to complement phenotypes displayed by a S. cerevisiae ctr1/ctr3-null mutant defective in high-affinity copper transport. In the S. cerevisiae mutant, the lack of high-affinity copper transport results in defective high-affinity iron transport as a result of insufficient delivery of copper to Fet3p (Askwith et al., 1994; Yuan et al., 1995; De Silva et al., 1995). The S. cerevisiae ctr1/ctr3 mutant was therefore unable to grow on copper- or iron-restrictive medium and the respiratory-deficient phenotype can be attributed to defective incorporation of both copper and iron into mitochondrial enzymes that require the transition metals as prosthetic groups. Low copper and iron levels have also been shown to affect cell-surface reductase activity (Lesuisse & Labbe, 1989; Dancis et al., 1990, 1994a; Georgatsou & Alexandraki, 1994; Georgatsou et al., 1997). This resulted in high ferric-reductase activity in the ctr1/ctr3 mutant when grown on high-iron medium. Therefore we propose that the rescuing clone facilitates restoration of high-affinity copper transport in the S. cerevisiae ctr1/ctr3-null strain. Due to the ability of the C. albicans genomic clone to rescue a S. cerevisiae ctr1/ctr3-null mutant and the significant similarity of the predicted protein and promoter elements to the S. cerevisiae Ctr1p we have named the C. albicans copper transporter gene CaCTR1.

The ORF that complements the S. cerevisiae ctr1/ctr3-null mutant shares 99 % identity with an ORF designated as CTR1-like on the Stanford C. albicans sequencing project website, with seven nucleotide mismatches between the two sequences along the whole ORF. Four of these mismatches result in amino acid substitutions: alanine for valine, methionine for threonine, alanine for threonine, and lysine for arginine at residues 21, 22, 39 and 248 respectively. The predicted amino acid sequence of the CTR1-like ORF is 251 residues long with a molecular mass of 27·8 kDa. A BLAST search using the C. albicans predicted protein sequence against the SWISS-PROT database revealed significant similarity to three high-affinity copper transporters from different species, with the S. cerevisiae Ctr1p scoring highest. Direct comparison to the S. cerevisiae Ctr1p sequence gave 39·6 % similarity and 33·0 % identity, respectively.

Sequence analysis of the predicted C. albicans Ctr1 protein revealed that methionine and serine are significantly overrepresented within 71 residues at the amino terminus. Within this highly structured region are repeated motifs that may signify a copper-binding domain. Similar methionine- and serine-rich domains have previously been described in prokaryotic copper-binding proteins (Cha & Cooksey, 1991; Odermatt et al., 1993) and the S. cerevisiae high-affinity copper transporter Ctr1p (Dancis et al., 1994a, b; Knight et al., 1996) that include repeats of the motif Met-XX-Met. Only one example of the Met-XX-Met motif was found in the C. albicans predicted protein but there were eight examples where the two methionines were separated by one amino acid residue (Met-X-Met). Distribution of all other amino acid types revealed a possibility of two or three transmembrane domains situated in the middle and at the carboxy terminus of the protein. As the predicted protein lacks an amino-terminal leader sequence it may utilize one of these domains for membrane insertion. The S. cerevisiae Ctr1p also lacks a leader sequence, has two or three transmembrane domains and exists as a multimer in the plasma membrane (Dancis et al., 1994b). The lack of a significant number of transmembrane domains in the C. albicans predicted protein probably means that it also exists as a multimer in the plasma membrane.

Similarities between the C. albicans CTR1 and the S. cerevisiae CTR1 extend to the promoter of each respective gene. Analysis of the CaCTR1 promoter sequence revealed four sequences resembling the S. cerevisiae copper response elements (CuREs). These motifs facilitate binding of the copper-sensing transactivator Mac1p (Dancis et al., 1992; Jungmann et al., 1993; Yamaguchi-Iwai et al., 1997; Labbe et al., 1997; Martins et al., 1998). The presence of CuRE-like elements in the promoter of CaCTR1 may explain how expression of the gene is controlled in response to copper availability in S. cerevisiae by the copper-sensing transactivator Mac1p. However, it also raises the possibility of the existence of a similar copper-sensing transactivator in C. albicans. A putative metal-binding transcriptional regulator is present on the Institut Pasteur C. albicans genomic database (http://genolist.pasteur.fr/CandidaDB/genome.cgi; accession number CA5628) that displays significant sequence homology to Mac1p of S. cerevisiae. Future studies on this ORF may reveal a similar copper-responsive transactivator in C. albicans. Three of the identified C. albicans Mac1p-like binding motifs found in the promoter of CaCTR1 are identical to the S. cerevisiae CuREs and the fourth is similar due to two mismatches. Electrophoretic mobility shift assays have previously shown that Mac1p binding is stronger to ^5'TATTTGCTC^3' than to ^5'TTTTTGCTC^3' and the transactivator makes specific and favourable contact to an adenine residue immediately 5' to the core sequence compared to a thymine (Joshi et al., 1999). Three of the C. albicans putative core sequences have adenine rather than thymine at this position and one has an identical sequence of ^5'TATTTGCTC^3' that may facilitate more efficient binding of a putative transactivator to these sequences. In S. cerevisiae the distance between the motifs has also been shown to have a limited effect on transcription, with greater spacing attenuating expression (Martins et al., 1998). Therefore the two C. albicans putative CuRE motifs with the least distance between them (-268 to -78 and -239 to -229) are the stronger candidates to facilitate the most effective transcriptional control. In S. cerevisiae the CTR1, CTR3 and FRE7 CuREs are inverted repeats whilst the FRE1 CuRE is a direct repeat. The two C. albicans putative CuREs with the least distance between them make up an inverted repeat. However, to date an inverted CuRE repeat rather than a direct CuRE repeat has not been shown to convey more effective transcription in S. cerevisiae. Northern blot analysis shows that CaCTR1 is negatively regulated by copper in C. albicans and is highly expressed in copper depleted conditions. Extra bands present after a 48 h exposure of the autoradiograph may reveal other copper-regulated C. albicans genes that contain significant sequence identity to CaCTR1.

A C. albicans ctr1-null mutant displays phenotypes consistent with the lack of copper transport, including the inability to grow on low-copper or low-iron medium, the inability to grow on non-fermentable carbon sources and an increased sensitivity to hydrogen peroxide and oxytetracycline. These phenotypes are directly comparable to those of S. cerevisiae mutants defective in high-affinity copper transport (Dancis et al., 1994b; Knight et al., 1996; Angrave et al., 2001). The C. albicans ctr1-null strain did not grow on low-copper or low-iron medium when the cells had previously been starved of copper. However, a much higher concentration of the copper chelator, BCS (2 mM), was required to arrest growth completely when compared to a corresponding S. cerevisiae ctr1/ctr3-null mutant (50 µM). Addition of copper rescued the Cactr1-null mutant when it was grown on low-copper or low-iron medium. Strain BWP17ctr1 : : URA3/ctr1 : : ARG4 also displayed an increased sensitivity to the oxidative-stress-generating agents hydrogen peroxide and oxytetracycline. This may be attributed to deficient intracellular copper conditions that may lead to defective Cu/Zn superoxide dismutase activity, or to lowered copper antioxidant capability (Liu & Culotta, 1994). However, a decrease in Cu/Zn superoxide dismutase activity has previously been described in a S. cerevisiae ctr1/ctr3-null mutant (Knight et al., 1996). The inability of the C. albicans ctr1-null strain to grow on non-fermentable carbon sources such as glycerol or ethanol can be attributed to defective mitochondria. This is presumably a secondary effect resulting from the deficient delivery of copper and possibly iron to the respiratory enzymes that provide electron transport.

The lack of CaCTR1 activity, and therefore deficient intracellular copper concentrations, may contribute to the change in morphology observed in strains BWP17ctr1 : : URA3/CTR1 and BWP17ctr1 : : URA3/ctr1 : : ARG4 when grown on yeast-extract peptone medium. In support of this is the observation that strain BWP17ctr1 : : URA3/ctr1 : : ARG4 changed morphology in medium depleted of copper and the addition of 100 µM cupric chloride to the growth medium rescued this. The production of pseudohyphae by strain BWP17ctr1 : : URA3/ctr1 : : ARG4 in low-copper medium may result from the cell responding to copper-starved conditions. However, it has previously been observed that low-iron conditions affect hyphal growth in C. albicans (Sweet & Douglas, 1991) and so the lack of copper transport affecting iron uptake due to inactive CaFet3p activity may also be a contributing factor.

Copper is believed to play a detrimental role in oxidative stress for C. albicans when the organism is in its environmental niche of the gastrointestinal tract. Both CaCUP1 and CaCRP1 may play an important role for the protection of C. albicans in this environment (Weissman et al., 2000). We have now isolated and characterized a C. albicans gene that may also play a role in this protection. The presence of CuRE-like elements in the promoter of CaCTR1 may facilitate tight control of copper uptake by a Mac1p-like transactivator, leading to copper homeostasis. Iron availability has been shown to affect C. albicans growth, hyphal production, adherence to host cells and the ability to set up an infection in mice (Sweet & Douglas, 1991; Valenti et al., 1986; Moors et al., 1992; Fratti et al., 1998; Eck et al., 1999; Ramanan & Wang, 2000). In S. cerevisiae the dependence of iron acquisiton on the uptake and delivery of copper to Fet3p is well documented and evidenc