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

J. Biol. Chem., Vol. 279, Issue 27, 28744-28755, July 2, 2004

This Article Abstract Full Text (PDF) Supplemental Data All Versions of this Article: 279/27/28744    most recent M403426200v1 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 Google Scholar Google Scholar Articles by Laliberté, J. Articles by Labbé, S. Search for Related Content PubMed PubMed Citation Articles by Laliberté, J. Articles by Labbé, S. Social Bookmarking                      What's this?

The Schizosaccharomyces pombe Pccs Protein Functions in Both Copper Trafficking and Metal Detoxification Pathways^*

Julie Laliberté, Lisa J. Whitson¶||, Jude Beaudoin, Stephen P. Holloway¶, P. John Hart¶, and Simon Labbé, A New Investigator Scholar from the Canadian Institutes of Health Research**

From the Département de Biochimie, Université de Sherbrooke, Sherbrooke, Québec J1H 5N4, Canada and the ^¶Department of Biochemistry and the X-ray Crystallography Core Laboratory, The University of Texas Health Science Center, San Antonio, Texas 78229-3900

Received for publication, March 29, 2004 , and in revised form, April 21, 2004.

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Because copper is both an essential cofactor and a toxic metal, different strategies have evolved to appropriately regulate its homeostasis as a function of changing environmental copper levels. In this report, we describe a metallochaperone-like protein from Schizosaccharomyces pombe that maintains the delicate balance between essentiality and toxicity. This protein, designated Pccs, has four distinct domains. SOD activity assays reveal that the first three domains of Pccs are necessary and sufficient to deliver copper to its target, copper-zinc superoxide dismutase (SOD1). Pccs domain IV, which is absent in Saccharomyces cerevisiae CCS1, contains seventeen cysteine residues, eight pairs of which are in a potential metal coordination arrangement, Cys-Cys. We show that S. cerevisiae ace1 mutant cells expressing the full-length Pccs molecule are resistant to copper toxicity. Furthermore, we demonstrate that the Pccs domain IV enhances copper resistance of the ace1 cells by an order of magnitude compared with that observed in the same strain expressing a pccs⁺I-II-III allele encoding Pccs domains I-III. We consistently found that S. pombe cells disrupted in the pccs⁺ gene exhibit an increased sensitivity to copper and cadmium. Furthermore, we demonstrate that overexpression of pccs⁺ is associated with increased copper resistance in fission yeast cells. Taken together, our findings suggest that Pccs activates apo-SOD1 under copper-limiting conditions through the use of its first three domains and protects cells against metal ion toxicity via its fourth domain.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Copper is both an essential and yet toxic cellular constituent (1). As a redox metal, it is used by critical enzymes as a catalytic cofactor (2). However, this same property can render copper cytotoxic because of its ability to participate in Fenton-like reactions that can generate hydroxyl radical, which may in turn cause cellular damage (3). In order to provide organisms with sufficient copper while at the same time preventing copper toxicity, specialized mechanisms have evolved for its uptake, intracellular trafficking, and sequestration within cells (4, 5).

In recent years, the use of yeast as a model organism has revealed a wealth of new information on how cells establish and maintain copper concentrations that are compatible with their needs (6–8). Under copper-limiting conditions and following the reduction of Cu²⁺ to Cu¹⁺ by the Fre plasma membrane reductases (9–13), the budding yeast Saccharomyces cerevisiae transports copper ions into yeast cells using two separate high affinity permeases encoded by the CTR1¹ and CTR3 genes (14–18). Copper-limiting conditions increase levels of mRNA synthesized from the FRE1/7, CTR1, and CTR3 genes, whereas their expression is repressed under copper-replete conditions, thereby preventing copper accumulation during acute copper exposure (12, 19). The copper-responsive regulation of FRE1/7, CTR1, and CTR3 gene expression is mediated by the copper-sensing transcription factor Mac1 (12, 19, 20). In addition to transcriptional regulation, copper uptake is further regulated through the copper-dependent degradation of the Ctr1 transporter. Ctr1 is stable under conditions of copper deprivation, but is rapidly degraded in the presence of excess copper (21). Unlike Ctr1, however, the Ctr3 transporter is not affected by increased exogenous copper concentrations (17).

Consistent with the notion that there is little intracellular free copper available in the cytoplasm of the yeast cell (22), specialized metallochaperones have been identified, including Atx1 (23, 24), Cox17 (25–28), and CCS1 (29), that bind copper after it enters the cell. These chaperones subsequently distribute their copper cargo to specific intracellular proteins or compartments (30, 31). Atx1 is a 73 amino acid cytosolic protein with a predicted molecular mass of 8.2 kDa (23). This copper chaperone is known to coordinate a single metal ion via its Met-X-Cys-X₂-Cys motif (24). Atx1 shuttles copper from the cytosol to post-Golgi vesicles by specifically docking with the Ccc2 copper-transporting P-type ATPase (32). Once loaded, Ccc2 subsequently pumps copper into the lumen of the Golgi, to metallate the copper-dependent ferroxidase Fet3 (32, 33). Although ATX1 and CCC2 gene transcription is unaffected by intracellular copper status, these genes are transcriptionally activated in response to iron starvation, illustrating their importance to iron metabolism through the delivery of copper to Fet3 (23, 34). Cox17, a 8.05-kDa copper chaperone, delivers copper specifically to the mitochondria for the assembly of cytochrome c oxidase (25, 35). Cox17 binds three Cu¹⁺ ions through cysteine residues that are arranged in a Cys-Cys-X-Cys configuration (27). Copper-loaded Cox17 has been shown to shuttle in and out of the mitochondrial intermembrane space and is thought to dock to at least one inner mitochondrial membrane protein, Sco1, for subsequent copper donation to the Cu_A site of the cytochrome c oxidase (36–38). A third copper chaperone, CCS1, specifically activates SOD1 (29). CCS1 has a predicted molecular mass of 27.3-kDa and possesses three distinct domains (39). Domain I (residues 1–77) resembles the structure of Atx1 (40) and contains a Met-X-Cys-X₂-Cys motif (residues 15–20) within its N-terminal portion. This domain has been demonstrated to be important for insertion of copper into SOD1 in vivo under conditions of copper deprivation but is not needed for this function under copper-replete conditions (39). Domain II (residues 78–213) strongly resembles the overall structure of SOD1 (40). This domain participates in CCS1-SOD1 protein-protein interactions, and is absolutely required for donation of copper to SOD1 (41, 42). At the C terminus (residues 214–249) a short region, referred to as domain III, contains two conserved cysteine residues arranged in a Cys-X-Cys motif that binds copper in vitro (39). CCS proteins lacking the C-terminal domain III are unable to activate SOD1, suggesting that the Cys-X-Cys site binds copper (possibly with the aid of domain I) and facilitates its insertion into the catalytic site of SOD1 (39, 43). When environmental copper levels are in excess, intracellular buffering of copper is carried out by two metallothionein (MT) proteins, Cup1 and Crs5, as well as by SOD1 (44–48). The copper-dependent up-regulation of CUP1, CRS5, and SOD1 gene expression is mediated by the Ace1 copper-detoxifying transcription factor (49, 50).

In the fission yeast Schizosaccharomyces pombe, candidate molecules for sequestering excess metal ions have been reported, including the Zym1 MT and the phytochelatins (51, 52). Studies using an S. pombe mutant strain that is defective in phytochelatin biosynthesis revealed that phytochelatins play an important role in cadmium and arsenic detoxification but have no apparent clear function in copper detoxification (52). zym1⁺ encodes a MT that has a structure resembling the -domain of mammalian MTs (51). An S. pombe strain harboring a deletion of the zym1⁺ gene exhibits a reduction in zinc accumulation, a slight sensitivity to zinc, and a decrease in cadmium tolerance. Furthermore, zym1⁺ mRNA levels were shown to be induced by zinc and cadmium but not copper (51).

For growth under copper limiting conditions, the molecular mechanisms in the early steps of copper assimilation in S. pombe differ from those in S. cerevisiae (53). Two integral membrane proteins, Ctr4 and Ctr5, form a two-component copper transporting complex at the cell surface (54, 55). In the absence of Ctr5, Ctr4 is mislocalized within the secretory pathway. Similarly, it was found that in the absence of Ctr4, Ctr5 is retained in the endoplasmic reticulum (55). These results suggest that assembly of a Ctr4/Ctr5 complex is required for either protein to proceed through the secretory pathway to the plasma membrane. Within this complex, the exact function of each protein is currently unclear. Recently, we have identified a gene in fission yeast, ctr6⁺, which encodes an intracellular vacuolar copper transporter (56). In response to copper limitation, Ctr6 appears to mediate the efflux of usable copper from the vacuole into the cytosol (56). Like the ctr4⁺ and ctr5⁺ genes, ctr6⁺ expression is regulated by the Cuf1 transcription factor and is induced in copper-limited cells (54, 56, 57). Cuf1 plays an essential role in coordinating the copper-dependent transcriptional regulation of copper transporter gene expression in S. pombe. This regulation involves cis-acting copper-signaling elements (CuSEs) found in each of the ctr4⁺, ctr5⁺, and ctr6⁺ promoters (56, 57). Binding studies reveal that the Cuf1 N-terminal 174 amino acids are important for binding to the CuSE (58). A motif containing five clustered cysteine residues near its C terminus constitutes the minimal copper-sensing module of Cuf1 and serves to inactivate Cuf1 function when cells are grown under copper-replete conditions (58).

Upon uptake into fission yeast cells, copper ions are presumably taken up by putative copper chaperones that are as yet uncharacterized at the molecular level. Examination of the S. pombe Genome data base suggests that the open reading frame SPAC22E12.04 encodes a putative ortholog of the S. cerevisiae CCS1. Although this putative ortholog bears 30% identity and 47% similarity to its bakers' yeast counterpart, notable differences exist between the two molecules. For instance, the N-terminal domain I of the S. pombe CCS ortholog (designated Pccs) lacks the copper-binding Met-X-Cys-X₂-Cys motif. In addition, Pccs harbors an extra domain at the C terminus that contains a series of cysteine residues, which are arranged in Cys-Cys configurations. Given these differences between the S. pombe and S. cerevisiae CCS proteins, we sought to dissect the functional features of Pccs. When the pccs⁺I-IV and pccs⁺I-III alleles were ectopically expressed in a S. cerevisae ccs1 strain, we found that cells producing a polypeptide spanning domains I-III displayed nearly wild-type levels of SOD1 activity. Under low basal copper conditions, S. pombe strains harboring a deletion of the pccs⁺ gene were defective in SOD1 activity. Transforming this strain with a plasmid expressing the first 222 amino acids of Pccs (domains I-III) restored SOD1 activity to the same level obtained with the full-length Pccs protein. When the full-length S. pombe pccs⁺ gene or a cDNA fragment that encodes only the Pccs domain IV was expressed in a S. cerevisiae ace1 strain, these cells exhibited a copper-resistant growth phenotype in the presence of exogenous copper. Consistently, pccs mutant cells were sensitive to copper and cadmium. Furthermore, overexpression of pccs⁺I-IV or pccs⁺IV alone conferred tolerance to elevated copper levels in fission yeast cells. Taken together, these results reveal that the S. pombe Pccs protein function in dual pathways to deliver copper to SOD1 during conditions of copper scarcity and to detoxify metal ions during conditions of metal excess.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Strains and Media—Six isogenic S. pombe strains, the wild-type FY254 (h^- can1-1 leu1-32 ade6-M210 ura4-D18) (59), cuf1 (h^- can1-1 leu1–32 ade6-M210 ura4-D18 cuf1::ura4⁺) (54), pccs (h^- can1-1 leu1-32 ade6-M210 ura4-D18 pccs::ura4⁺), sod1 (h^- can1-1 leu1-32 ade6-M210 ura4-D18 sod1::ura4⁺), zym1 (h^- can1-1 leu1-32 ade6-M210 ura4-D18 zym1::ura4⁺), and zym1 pccs (h^- can1-1 leu1–32 ade6-M210 ura4-D18 zym1::ura4⁺ pccs::KAN^r) were used in this study. To ascertain that the results seen were not specific to the S. pombe FY254 strain, identical experiments were conducted with the FY435 strain (h⁺ his7-366 leu1-32 ade6-M210 ura4-D18) (60). The pccs (h⁺ his7-366 leu1-32 ade6-M210 ura4-D18 pccs::KAN^r) and pccs zym1 ctr6 (h⁺ his7-366 leu1–32 ade6-M210 ura4-D18 ctr6::hisG zym1::ura4⁺ pccs::KAN^r) mutant strains are isogenic to the FY435 strain. S. pombe cells were grown in yeast extract plus supplements (YES) or in Edinburgh minimal medium with the necessary auxotrophic requirements (61). When pccs and sod1 mutants were grown, Edinburgh minimal medium was further supplemented with 225 mg/liter of adenine, histidine, leucine, methionine, lysine, and cysteine, unless otherwise stated. Under anaerobic growth conditions, fission yeast cells were grown in culture jars with BD GasPak EZ (BD Diagnostic System, Sparks, MD). For expression of the human CCS and S. pombe pccs⁺ genes in S. cerevisiae, the ccs1 (also named lys7) mutant strain, denoted EGy103lys7 (MAT leu2-3, 11 his3-1 trp1-289 ura3-52 lys7::LEU2) was utilized to ensure that within the cell, only the ectopically expressed hCCS or Pccs was the sole protein with the ability to deliver copper to SOD1. The EGy103lys7 strain was derived from the parent strain EGy103 (MAT leu2-3, 11 his3-1 trp1-289 ura3-52) (62) by gene deletion and replacement. S. cerevisiae strain DTY59 (MAT his6 leu2-3,-112 ura3-52 ace1-225 CUP1R-3) (63) was used to test the ability of Pccs domain IV to detoxify copper in bakers' yeast. DTY59 was derived from strain DTY7 (MAT his6 leu2-3,-112 ura3-52 CUP1R-3) as described previously (64). S. cerevisiae cells were grown in rich medium (1% yeast extract, 2% bactopeptone, 2% dextrose) or synthetic complete medium lacking histidine or uracil for strains transformed with plasmids harboring the HIS3 or URA3 gene, respectively.

Plasmids—The pccs⁺ gene was isolated by PCR using primers corresponding to the start and stop codons of the open reading frame from an S. pombe cDNA library (ATCC 87284, deposited by S. Elledge) (generous gift of Dennis J. Thiele, Duke University, Durham, NC). To clone the PCR product into the pRS313 vector (65), the EcoRI and BamHI sites found in pccs⁺ were modified by PCR mutagenesis, eliminating both sites without altering the amino acid sequence of the Pccs protein. The modified pccs⁺ allele was re-amplified by PCR using primers designed to generate EcoRI and XbaI sites at the upstream and downstream termini of the open reading frame. The PCR product obtained was digested with EcoRI and XbaI and cloned into the corresponding sites of the centromeric yeast plasmid pRS313 to generate pRSpccs⁺I-II-III-IV. Subsequently, the S. cerevisiae CCS1 promoter up to -395 from the start codon of the CCS1 gene was subcloned into pRSpccs⁺I-II-III-IV at the XhoI and EcoRI sites. Similarly, a 335-bp XbaI-BamHI DNA fragment containing the 3'-untranslated region of CCS1 was inserted into the same sites of pRSpccs⁺I-II-III-IV. The 666-bp EcoRI-XbaI fragment encoding the first three domains of Pccs was used to replace the EcoRI-XbaI fragment from plasmid pRSpccs⁺I-II-III-IV to produce the plasmid pRSpccs⁺I-II-III. The human and S. cerevisiae CCS genes were obtained by PCR amplification using primers that contained EcoRI and XbaI restriction sites using a human HeLa cell cDNA library or genomic DNA from S. cerevisiae strain DTY7 as templates, respectively. The purified DNA fragments were digested with EcoRI and XbaI and subsequently cloned into the corresponding sites in plasmid pRSpccs⁺I-II-III-IV to replace the pccs⁺I-II-III-IV gene.

To generate the pSP1sod1⁺ plasmid, a 1598-bp BamHI-NotI PCR-amplified DNA segment containing the S. pombe sod1⁺ locus starting at -860 from the translational start codon up to +273 after the stop codon was inserted into the BamHI and NotI sites of pSP1 (66). The pccs⁺ cDNA was isolated using the S. pombe cDNA library described above. The purified DNA fragment that contained flanking BglII and SmaI restriction sites was digested and cloned into the pBluescript SK vector (Stratagene, La Jolla, CA) at compatible BamHI and SmaI sites. To create a plasmid that has the pccs⁺ promoter driving the expression of the pccs⁺ gene, the S. pombe pccs⁺ regulatory region (positions -1105 to -1) was amplified by PCR and inserted just before the ATG codon of the pccs⁺ gene using the NotI and SpeI sites. Subsequently, a SmaI-PstI DNA fragment of the pccs⁺ terminator up to +542 from the stop codon was isolated by PCR from the S. pombe FY254 genomic DNA. Once generated and verified by DNA sequencing, the DNA fragment containing the pccs⁺ gene and its regulatory regions was isolated from the pSKpccs⁺I-II-III-IV plasmid using NotI and PstI and inserted into the corresponding sites of pSP1. The resulting plasmid was designated pSP1pccs⁺I-II-III-IV. To generate the pccs⁺I-II-III-IV-StuI-BspEI allele, a 12-bp StuI-BspEI linker was inserted in-frame and downstream of the last codon of the pccs⁺ gene by the overlap extension method (67). The insertion created four extra amino acid residues after the alanine at position 297 (Ala²⁹⁷-Arg-Pro-Ser-Gly-Stop) of Pccs. This allele was found to be functional because of its ability to fully restore SOD1 activity in vivo. We used the restriction sites StuI and BspEI created within pccs⁺ to insert a copy of the gfp gene (68) or four copies of the Haemophilus influenzae hemagglutinin epitope (69). The plasmid, denoted pSP1pccs⁺I-II-III-IV-GFP, was used to determine the localization of Pccs-GFP fusion protein in S. pombe by fluorescence microscopy. A similar strategy was utilized to generate the pccs⁺I-II-III-StuI-BspEI allele, except that the StuI-BspEI linker was placed in-frame at the end of a DNA fragment that encodes only the Pccs domains I, II, and III. Subsequently, the gfp or zym1⁺ gene with flanking StuI and BspEI restriction sites was cloned into pSP1pccs⁺I-II-III-StuI-BspEI to generate the pSP1pccs⁺I-II-III-GFP or pSP1pccs⁺I-II-III-zym1⁺ plasmid, respectively.

To ascertain if the expression of different versions of the pccs⁺ and zym1⁺ alleles contributed to the increased copper resistance of S. cerevisiae DTY59 cells, the p4XXGPD expression vectors were used as described previously (70). Using appropriate primers that contained SpeI and SmaI sites, the pccs⁺I-II-III-IV, pccs⁺I-II-III, pccs⁺IV, zym1⁺, and pccs⁺I-II-III-zym1⁺ alleles were isolated by PCR from the plasmids pSP1pccs⁺I-II-III-IV, pSP1pccs⁺I-II-III, pSP1pccs⁺IV, pSP1zym1⁺, and pSP1pccs⁺I-II-III-zym1⁺, respectively. The PCR products obtained were digested with SpeI and SmaI and cloned into the corresponding sites of p426GPD. To assess if overexpression of pccs⁺ and zym1⁺ alleles can rescue the copper hypersensitivity of a ctr6 strain overexpressing the wild-type ctr6⁺ gene, plasmids pREP3X-pccs⁺I-II-III-IV, pREP3X-pccs⁺I-II-III, pREP3X-pccs⁺-IV, pREP3X-zym1⁺, and pREP3X-pccs⁺I-II-III-zym1⁺ were constructed as follows. Five DNA fragments encompassing the pccs⁺I-II-III-IV, pccs⁺I-II-III, pccs⁺IV, zym1⁺, and pccs⁺I-II-III-zym1⁺ alleles were PCR amplified with flanking XhoI and SmaI sites from the pSP1pccs⁺ and pSP1zym1⁺ plasmids. The resulting PCR products were digested with XhoI and SmaI and cloned into the corresponding sites of pREP3X (71, 72). For ectopic expression of the ctr6⁺ gene, the thiamine-repressible promoter system was used as described previously (56).

Protein and Enzyme Assays—For Western blotting experiments, S. pombe and S. cerevisiae cells were grown to OD₆₀₀ of 1.0 in selective medium. Protein extracts were prepared from cells that were untreated or incubated for 3 h (S. cerevisiae) or 11 h (S. pombe) with either CuSO₄ (100 µM) or BCS (100 µM), and then quantitated as described previously (73). The extracts were resolved by SDS-polyacrylamide electrophoresis, transferred to polyvinylidene difluoride Hybond-P (Amersham Biosciences), and the immunoblots analyzed for steady-state levels of SOD1, PGK, and PCNA proteins using antiserum SOD-100 (Stressgen, Victoria, BC), 22C5-D8 (Molecular Probes, Eugene, OR), and PC10 (Sigma), respectively. After a 2-h incubation, the membranes were washed with TBS (10 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% bovine serum albumin), incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences), and visualized by chemiluminescence. SOD1 activity assays were performed using in-gel nitro blue tetrazolium staining as previously described (54). Spectrophotometric determination of SOD activity was also performed using the protein extracts by measuring the inhibition of the reduction rate of cytochrome c by SOD, which competes for reactive oxygen species produced from the xanthine-xanthine oxidase system (74) as described previously (56).

Analyses of Metal Ion Sensitivity—Cells were grown for 48 h without shaking, re-inoculated to OD₆₀₀ of 0.5, and grown to an OD₆₀₀ of 1.0 (1 x 10⁷) at 30 °C. Each cell culture was diluted (2 x 10⁴) and inoculated into 5 ml of Edinburgh minimal medium containing 50 µM CuSO₄ and further supplemented with increasing concentrations of metal ions specified in Fig. 8. After incubation for 7 days at 30 °C without shaking, total growth was measured at OD₆₀₀.

View larger version (18K): [in this window] [in a new window]   FIG. 8. Effects of pccs, zym1, and pccs zym1 gene deletions on tolerance to heavy metals. Growth of the indicated isogenic yeast strains was assessed in Edinburgh minimal medium that was either untreated (Control) or supplemented with increasing concentrations of CuSO4 (A) and CdCl2 (B). Exponentially growing cells (2 x 104) were inoculated into 5 ml of liquid cultures and were incubated at 30 °C for 7 days without shaking. Total growth relative to that obtained in the absence of the metal ion (% control growth) was determined turbidimetrically at OD600. Each point represents the mean of three separate experiments ± S.D.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
The Primary Sequence of Pccs—Analysis of genomic DNA sequences from the S. pombe Genome Project revealed an open reading frame (SPAC22E12.04) related to two previously identified groups of copper homeostasis proteins: the CCS and MT proteins, which are involved in delivering copper to SOD1 and in copper detoxification, respectively. The SPAC22E12.04-encoded protein, which we have denoted Pccs, has four distinct domains. The first three domains of the protein have extended homology to the CCS proteins, including S. cerevisiae CCS (CCS1) and human CCS (hCCS) (Fig. 1). Domain I of Pccs (residues 1–72) displays 24 and 18% identity to the same region of the S. cerevisiae and human CCS proteins, respectively. Interestingly, domain I of Pccs lacks the Met-X-Cys-X₂-Cys motif found in CCS1 and hCCS that is predicted to coordinate a single Cu¹⁺ atom under conditions of copper starvation (39). Instead, Pccs has a single cysteine residue at position 11 that aligns with the last cysteine of the Met-X-Cys-X₂-Cys sequence (Fig. 1). Domain II of Pccs (residues 73–186) exhibits 28 and 23% identity with the domain II of S. cerevisiae CCS1 and human CCS, respectively. Supporting the notion that domain II of Pccs has greater similarity to CCS1 than hCCS, this domain lacks the three histidine and one aspartic acid residues that bind zinc in SOD1 and are preserved in human CCS (76). Furthermore, three of the four histidine residues known to bind copper in SOD1 and are conserved in hCCS domain II are absent in Pccs domain II. However, consistent with the model that domain II is involved in recognizing SOD1 and the subsequent formation of the CCS-SOD1 heterodimer, the residues predicted to be involved in protein-protein interactions between CCS and SOD1 (40) are highly conserved in all three Pccs, CCS1, and hCCS molecules (Fig. 1). Domain III of Pccs (residues 187–222) shows 47% identity with the domain III of both the S. cerevisiae and human CCS. This is the most conserved region, with an invariant Cys-X-Cys motif that is capable, in the case of CCS1, of binding copper (39). As shown in Fig. 1, the Pccs protein has an extra domain (residues 223–297) at its C terminus, which is not found in CCS1 or hCCS. The sequence of Pccs domain IV is highly reminiscent of the metal binding protein, MT (77). This relatively short protein segment contains 17 cysteine residues, eight pairs of which are present in a Cys-Cys configuration. Furthermore, lysine and serine residues represent 19 and 28% of the total Pccs domain IV, respectively. Similar to MTs, domain IV lacks aromatic amino acids. Based on these observations and upon inspection of various sequence databases, the sequence of Pccs domain IV exhibits characteristics that are hallmarks of MTs. Therefore, Pccs bears strong similarity to domains found in the copper delivering CCS proteins, as well as the metal ion buffering MTs.

View larger version (46K): [in this window] [in a new window]   FIG. 1. Sequence alignment of S. pombe Pccs with S. cerevisiae and human CCS. Amino acid residues that are identical in at least two of the compared proteins are indicated above the alignment. The domains are numbered I-IV. Domains I, II, and III are boxed in blue, green, and brown, respectively. The fourth domain of Pccs is underlined. The MXCXXC sequence in domain I and the CXC motif in domain III are depicted in red and indicated with a double line above the sequence. The dots indicate residues that are predicted to be involved in CCS-SOD1 protein-protein interactions (40).

View larger version (34K): [in this window] [in a new window]   FIG. 2. Transcription of pccs+ mRNA is unregulated by copper, silver, cadmium, and zinc, and a functional cuf1+ gene is not required for its expression. A, the isogenic strains FY254 (cuf1+) and SPY1 (cuf1) were grown to mid-logarithmic phase in YES media. Cultures were incubated in the absence (-) or presence of CuSO4 (1 and 100 µM), or 100 µM BCS for 1 h at 30 °C. Total RNA was isolated and analyzed by Northern blot. pccs+ and act1+ mRNA steady-state levels are indicated by arrows. B, total RNA from the wild-type FY254 and the pccs disruption strain were analyzed by RNase protection assay. pccs+ and act1+ mRNA levels are indicated by arrows. Total RNA was isolated from control untreated cells (-), and cells treated with CuSO4 (100, 250, 500, and 1000 µM), AgNO3 (100, 250, 500, and 1000 µM), CdCl2 (100, 250, 500, and 1000 µM), or ZnSO4 (100, 250, 500, and 1000 µM). The results shown are representative of three independent experiments.

View larger version (23K): [in this window] [in a new window]   FIG. 3. Expression of full-length Pccs or Pccs domains I-III complements loss of CCS1 in S. cerevisiae. A, EGy103lys7 (also named EGy103ccs1) fails to grow on synthetic complete medium lacking lysine in the presence of oxygen. EGy103ccs1 cells were transformed with plasmids expressing human CCS (hCCS) or S. pombe Pccs (pccs+) domains I-IV or domains I-III under the control of the S. cerevisiae CCS1 gene promoter and assayed for growth in the presence or absence of oxygen. As a positive control, EGy103ccs1 cells were also transformed with a plasmid expressing S. cerevisiae CCS1 (CCS1) under the control of its own promoter and assayed for growth in the same conditions. B, EGy103ccs1 displays a deficiency in SOD1 activity when cells are grown under copper starvation conditions in the presence of the Cu+ chelator BCS (100 µM). These cells were transformed with vector only (-) or plasmids expressing the indicated domains of Pccs or CCS1 and hCCS as positive controls. Total extracts from transformed cells were assayed for SOD1 activity using an in-gel activity assay with nitro blue tetrazolium staining. C, SOD activity was determined from the cell lysates used in B using a spectrophotometric method with cytochrome c and xanthine oxidase. The SOD activities reported represent the means of three replicates experiments ± S.D. D, Western blot analysis of extracts used in B employing an antibody directed against human SOD1 that also recognized the S. cerevisiae SOD1 protein (89). Cellular levels of PGK was determined as a load control.

View larger version (29K): [in this window] [in a new window]   FIG. 4. An S. pombe pccs strain displays a deficiency in SOD1 activity under copper-limiting conditions. A, SOD1 activity was determined in pccs cells that were grown under conditions in which copper is required (100 µM BCS), in the presence of exogenous CuSO4 (100 µM) and after transformation with a wild-type copy of the pccs+ gene expressed from a plasmid. B, aliquots of whole cell extracts used in A were analyzed by immunoblotting. The positions of the SOD1 and PCNA proteins are indicated with arrows. C, SOD1 activity was determined in a wild-type strain of S. pombe (WT) and an S. pombe sod1 and pccs disruption strains in the presence or absence of exogenous CuSO4 (100 µM). D, a plasmid-borne copy of the sod1+ gene was tested for its ability to restore SOD1 activity to a S. pombe strain with a deletion of the sod1+ gene. E, the indicated strains were incubated in the absence (-) or presence of CuSO4 (100 µM) or BCS (100 µM). Protein extracts prepared from these strains were analyzed for steady-state protein levels of SOD1 by immunoblotting using either anti-SOD1 or anti-PCNA (as an internal control) antibody.

View larger version (39K): [in this window] [in a new window]   FIG. 5. pccs mutant cells expressing Pccs domains I-III exhibit similar levels of SOD1 activity to that observed for pccs cells expressing the full-length Pccs protein. A, exponentially growing cells were treated with CuSO4 (100 µM) or BCS (100 µM) for 11 h. Whole cell extracts were prepared from these cells and analyzed using an in-gel assay for SOD1 activity. pccs cells transformed with either pccs+I-IV or with a plasmid expressing pccs+I-III were examined for SOD1 activity. B, pccs cells were transformed with the indicated plasmids and superoxide scavenging activity was monitored by the nitro blue tetrazolium gel assay. C, the full-length Pccs and Pccs domains I-III fused to the green fluorescent protein fully complemented the defective SOD1 activity in a pccs strain in a manner indistinguishable from the unadulterated Pccs protein.

Based on previous studies that determined 95–99% of total S. cerevisiae CCS1 resides in the cytosol (81, 82), we ascertained the localization of the Pccs protein in S. pombe. To ensure that insertion of GFP at the Pccs C terminus did not interfere with its function, we transformed the pccs⁺I-IV-GFP and pccs⁺I-III-GFP fusion genes into a pccs mutant strain, and tested cell lysates from the transformants for the presence of SOD1 activity. As shown in Fig. 5C, pccs cells expressing plasmids containing the pccs⁺I-IV-GFP and pccs⁺I-III-GFP genes functionally complemented the pccs phenotypes similar to the full-length pccs⁺ and pccs⁺I-III genes, indicating that the Pccs I-IV-GFP and Pccs I-III-GFP fusion proteins were functional. The localization of each GFP fusion was determined by fluorescence microscopy. As shown in Fig. 6, pccs cells expressing the pccs⁺I-IV-GFP or pccs⁺I-III-GFP allele accumulated the full-length Pccs-GFP or Pccs domains I-III-GFP protein in the cytosol in the absence or presence of copper. The fluorescence was diffuse and distributed throughout the cytosol. The Pccs I-IV-GFP and Pccs I-III-GFP fusion proteins were predominantly localized in the cytosol, and in most cells, were largely excluded from the nucleus (Fig. 6 and data not shown). Our data do not allow us to establish whether or not a fraction of the total Pccs protein is localized in the mitochondrial intermembrane compartment as reported previously for CCS1 (81). Taken together, these observations suggest that the full-length Pccs and Pccs I-III proteins are primarily cytosolic components similar to SOD1.

View larger version (38K): [in this window] [in a new window]   FIG. 6. Localization of Pccs green fluorescent fusion proteins in S. pombe. A, cells expressing the full-length Pccs-GFP fusion protein were grown at 30 °C for 11 h in the presence of BCS (100 µM) or CuSO4 (100 µM) and visualized for GFP by fluorescence microscopy (top panel). The cells were also viewed by Nomarski microscopy for cell morphology (bottom panel). B, cells expressing a GFP protein fused to the C terminus of domain III of Pccs were examined for its cellular location as described in A.

View larger version (33K): [in this window] [in a new window]   FIG. 7. Functional comparison of Pccs domain IV with the Zym1 MT of S. pombe in an S. cerevisiae ace1 mutant strain. A, alignment of the amino acid sequences of Pccs domain IV, S. pombe Zym1, and mouse MT-I. The dots indicate amino acid residues that are identical in at least two of the compared proteins. Cysteine residues that are arranged in Cys-Cys, Cys-X-Cys, and Cys-X2-Cys configurations are boxed. The KXS motif is shown with a double line above or underneath. B, an ace1 mutant strain, transformed with plasmid p426GPD alone (-), plasmid p426GPDpccs+I-IV (I-II-III-IV), plasmid p426GPDpccs+I-III (I-II-III), plasmid p426GPDpccs+IV (IV), plasmid p426GPDzym1+ (zym1+), or plasmid p426GPDpccs+I-III-zym1+ (pccsI-II-III-zym1+), was spotted onto a synthetic complete medium containing the indicated CuSO4 concentrations (0, 50, 100, and 500 µM). The wild-type (ACE1) strain DTY7 transformed with p426GPD alone was used as a control.

Overexpression of Pccs Domain IV Suppresses the Copper Toxicity Phenotype Resulting from ctr6⁺ Overexpression—Based on the findings that deletion of the pccs⁺ gene lowered copper tolerance in S. pombe and that both Pccs and Pccs domain IV conferred copper resistance when expressed in S. cerevisiae ace1 cells, we ascertained the ability of these proteins to elicit copper resistance when overexpressed in an S. pombe strain that is hypersensitive to copper. Overexpression of ctr6⁺ from the thiamine-regulated nmt1⁺ promoter resulted in an increased sensitivity to copper toxicity when transformed cells were grown on medium containing elevated concentrations of CuSO₄ (56). Cotransformation of both the nmt1-ctr6⁺ and nmt1-pccs⁺I-IV genes into a S. pombe ctr6 pccs zym1 mutant strain conferred resistance to 15 µM copper on the transformed cells compared with the same cells that were co-transformed with nmt1-ctr6⁺ and an empty vector (pREP3X) (Fig. 9). Similar to the expression of the full-length pccs⁺ gene, cells expressing the pccs⁺IV allele displayed no copper sensitivity. On the other hand, cells expressing the pccs⁺I-II-III allele (without domain IV), were sensitive to copper. Consistent with its ability to sequester divalent metal ions, expression of zym1⁺ conferred copper tolerance to the transformed cells. Furthermore, coexpression of the chimeric ¹Pccs²²²-¹Zym1⁵⁰ and Ctr6 proteins in the ctr6 pccs zym1 mutant strain, allowed these cells to grow in the presence of exogenous copper at the same level as cells expressing the wild-type Pccs protein. This finding indicates that the first three domains of Pccs can be fused to a MT-like polypeptide to allow detoxification of excess copper ions. Together with the results from our studies on the ace1 yeast strain, these data strongly indicate that the role of Pccs in copper buffering is mediated through its fourth domain.

View larger version (41K): [in this window] [in a new window]   FIG. 9. Full-length Pccs, Pccs IV, Pccs I-III-Zym1, or Zym1 suppresses the copper toxicity phenotype resulting from ctr6+ overexpression. The isogenic wild type (WT) FY435 and ctr6 pccs zym1 DJY47 strains were co-transformed with the indicated plasmids. As controls, these strains were also co-transformed with the vectors pREP3Xhis7+ and pREP3XLEU2. Cells were spotted onto Edinburgh minimal medium supplemented with 0 and 15 µM CuSO4 and were grown at 30 °C for 4 days.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

With respect to the second question, the analysis of Pccs function in a S. cerevisiae ace1 strain revealed that expression of Pccs domain IV conferred copper tolerance to these cells, protecting them from lethal copper concentrations. Similarly, ace1 cells expressing the full-length Pccs or Zym1 MT protein acquired copper tolerance. Furthermore, consistent with the hypothesis that Pccs participates in intracellular copper buffering mainly through its fourth domain, we showed that ace1 cells expressing the first three domains of Pccs are more sensitive to copper by an order of magnitude compared with cells expressing full-length Pccs or Pccs domain IV. The involvement of Pccs in metal ion tolerance was further supported by three additional observations. First, the analysis of a S. pombe strain in which pccs⁺ has been insertionally inactivated clearly demonstrates a phenotype with marked sensitivity to copper and cadmium ions. Second, the primary sequence of Pccs domain IV exhibits extensive homology to MTs and specifically harbors numerous repeats of a putative copper-binding Cys-Cys motif. Third, in a S. pombe strain overexpressing the nmt1⁺-ctr6⁺ allele that generates a hypersensitivity to copper ions, overexpression of Pccs domain IV, as well as full-length Pccs, protects the cells from the toxic effects of copper ions. Taken together, these data suggest a model wherein the first three domains of Pccs are required to specifically deliver copper to SOD1 under low copper concentrations, while the fourth domain of Pccs functions to sequester metals in the presence of elevated copper concentrations.

In this study, we inactivated the sod1⁺ locus by deletion and replacement with the S. pombe ura4⁺ gene. Although the sod1 cells grow poorly in shaken and well-aerated cultures, these cells were viable on standard YES medium or in a modified Edinburgh medium containing supplements of cysteine, methionine, lysine, adenine, histidine, and leucine. It should be noted that recent studies of S. pombe with a disrupted sod1⁺ gene showed that the viability of sod1 mutants varied depending on the genetic background of the fission yeast strain (87, 88). Although the nature of this genetic variation is not clear, it is possible that the composition of the growth media may be a factor due to the amino acid auxotrophies of sod1 cells.

Our results thus far support a dual role for Pccs in copper homeostasis. It functions as a copper chaperone when copper concentrations are limiting and as a detoxifier when copper concentrations are in excess. Because its function is required in both high and low levels of copper, it is not surprising that the steady-state levels of pccs⁺ mRNA is constitutive and unaffected by changes in copper concentrations. Consistently, expression of pccs⁺ mRNA at steady-state levels, was independent of cuf1⁺, a gene encoding the nutritional copper sensing trans-inducer of the copper transport genes ctr4⁺, ctr5⁺, and ctr6⁺ in fission yeast. Interestingly, pccs⁺ mRNA was fairly abundant in wild-type S. pombe cells possibly to maintain adequate intracellular levels of Pccs for both copper distribution and detoxification pathways.

Two notable differences exist between the S. pombe and S. cerevisiae CCS chaperones. First, the N-terminal domain I of the S. pombe Pccs lacks the copper-binding Met-X-Cys-X₂-Cys motif, except for the last cysteine residue, and second, Pccs contains a fourth domain at its C terminus. Given that Pccs lacking the N-terminal domain I cannot activate the SOD1 under copper-limiting conditions, it is likely that other residues within domain I besides the Met-X-Cys-X₂-Cys site that is found in CCS1 but not Pccs, are important for its metallochaperone-like activity in vivo. Efforts are currently under way to identify the residues within domain I that may be important for this function. Based on computer algorithm analysis, the C-terminal 75 amino acids of Pccs exhibited sequence similarity to MTs. Thus far, a single MT encoded by the zym1⁺ locus has been described in S. pombe (51). In fission yeast cells, zym1⁺ is transcriptionally induced by zinc and cadmium, but not copper (51). Although Zym1 suppresses zinc toxicity (51), our findings showed that S. cerevisiae ace1 cells expressing Zym1 were protected against copper toxicity. Deletion of zym1⁺ (zym1) from fission yeast resulted in reduced in copper tolerance in the mutant strain. However, as reported previously (51), the steady-state levels of zym1⁺ mRNA were unaffected by cellular copper status.⁵ Interestingly, deletion in the pccs⁺ gene (pccs) in S. pombe cells also resulted in increased sensitivity to copper toxicity. In a heterologous strain, expression of pccs⁺I-IV or pccs⁺IV in the copper sensitive S. cerevisiae ace1 cells conferred significant levels of resistance to these cells. Using this strain, we further demonstrated that a chimeric Pccs protein harboring the S. pombe metallothionein Zym1 instead of Pccs domain IV allowed detoxification of excess copper at the same level as the wild-type protein. Thus, the fact that there is no known MT in S. pombe whose expression is induced by the presence of copper, fission yeast cells may circumvent this situation by the use of two genes, pccs⁺ and zym1⁺, to protect cells against copper poisoning.

	FOOTNOTES

 
^* This study was supported in part by Canadian Institutes of Health Research Grant MOP-36450 (to S. L.) and by grants from the NINDS, National Institutes of Health and Robert A. Welch Foundation (to P. J. H.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The on-line version of this article (available at http://www.jbc.org) contains supplementary material.

Supported in part by the Fondation de la Recherche sur les Maladies Infantiles du Québec and the Natural Sciences and Engineering Research Council of Canada.

^|| Supported by a predoctoral fellowship from the Ford Foundation.

^** To whom correspondence should be addressed: Département de Biochimie, Faculté de médecine, Université de Sherbrooke, 3001 12e Ave Nord, Sherbrooke, QC J1H 5N4, Canada. Tel.: 819-820-6868 (ext. 15460); Fax: 819-564-5340; E-mail: Simon.Labbe{at}USherbrooke.ca.

¹ The abbreviations used are: CTR1, copper transporter 1; BCS, bathocuproinedisulfonic acid; CCS, copper chaperone for SOD1; Cuf1, copper factor 1; CuSE, copper-signaling element; MT, metallothionein; Pccs, S. pombe copper chaperone for SOD1; PCNA, proliferating cell nuclear antigen; SOD1, copper-zinc superoxide dismutase; YES, yeast extract plus supplements; GFP, green fluorescent protein.

² Gel nitro blue tetrazolium assay showing the SOD1 activity of Pccs-HA₄ derivative polypeptides is shown; a Western blot analysis of each polypeptide using an anti-HA antibody.

³ Spot test for complementation of aerobic lysine auxotrophy is shown; an analysis of SOD1 activity by the nitro blue tetrazolium gel assay.

⁴ J. Laliberté and S. Labbé, unpublished data.

⁵ J. Beaudoin and S. Labbé, unpublished data.