A Plasmodium falciparum copper-binding membrane protein with copper transport motifs
© Choveaux et al.; licensee BioMed Central Ltd. 2012
Received: 3 October 2012
Accepted: 20 November 2012
Published: 29 November 2012
Skip to main content
© Choveaux et al.; licensee BioMed Central Ltd. 2012
Received: 3 October 2012
Accepted: 20 November 2012
Published: 29 November 2012
Copper is an essential catalytic co-factor for metabolically important cellular enzymes, such as cytochrome-c oxidase. Eukaryotic cells acquire copper through a copper transport protein and distribute intracellular copper using molecular chaperones. The copper chelator, neocuproine, inhibits Plasmodium falciparum ring-to-trophozoite transition in vitro, indicating a copper requirement for malaria parasite development. How the malaria parasite acquires or secretes copper still remains to be fully elucidated.
PlasmoDB was searched for sequences corresponding to candidate P. falciparum copper-requiring proteins. The amino terminal domain of a putative P. falciparum copper transport protein was cloned and expressed as a maltose binding fusion protein. The copper binding ability of this protein was examined. Copper transport protein-specific anti-peptide antibodies were generated in chickens and used to establish native protein localization in P. falciparum parasites by immunofluorescence microscopy.
Six P. falciparum copper-requiring protein orthologs and a candidate P. falciparum copper transport protein (PF14_0369), containing characteristic copper transport protein features, were identified in PlasmoDB. The recombinant amino terminal domain of the transport protein bound reduced copper in vitro and within Escherichia coli cells during recombinant expression. Immunolocalization studies tracked the copper binding protein translocating from the erythrocyte plasma membrane in early ring stage to a parasite membrane as the parasites developed to schizonts. The protein appears to be a PEXEL-negative membrane protein.
Plasmodium falciparum parasites express a native protein with copper transporter characteristics that binds copper in vitro. Localization of the protein to the erythrocyte and parasite plasma membranes could provide a mechanism for the delivery of novel anti-malarial compounds.
Malaria is a serious, acute and chronic relapsing infection that kills close to one million people annually. More than 90% of these deaths are recorded in the sub-Saharan regions of Africa, with the majority being children under the age of five years . Human mortality, as a result of malaria infection, is predominantly caused by Plasmodium falciparum. Efforts to prevent and control the disease have, however, been hindered by an increased parasite resistance to currently available anti-malarial drugs, highlighting the need for novel anti-malarial drug development . A group of proteins gaining increased interest as potential therapeutic targets are the integral membrane proteins predicted to possess transport functions [3, 4]. Despite these proteins playing key roles in Plasmodium parasite growth and replication , they remain poorly understood and underexploited . Previous studies employing the intracellular copper chelator neocuproine established that copper is an essential micronutrient for in vitro parasite growth . Rasoloson et al. described the membrane-bound PfCuP-ATPase copper efflux protein and suggested this protein acts to reduce copper toxicity in P. falciparum. Studying P. falciparum copper metabolism may lead to the identification of novel anti-malarial drug targets.
Copper is an essential micronutrient that plays important catalytic and structural roles in numerous enzymes. Cuproenzymes harness the ability of copper to cycle between a stable oxidized Cu(II) and unstable reduced Cu(I) state for various redox reactions. However, this property also makes copper potentially toxic to cells since it can undergo free radical producing Fenton chemistry . Consequently, cells have evolved homeostatic mechanisms for the uptake, distribution, sequestration and secretion of copper to meet essential cellular requirements while reducing its toxic potential. From yeast to humans, copper acquisition is mediated by the high affinity copper transport protein, Ctr1 [7, 8]. Copper metallochaperones subsequently distribute intracellular copper to specific proteins or organelles. The copper chaperone for superoxide dismutase (CCS) distributes copper to cuprozinc superoxide dismutase (Cu/Zn SOD) in the cytosol and mitochondrion, antioxidant protein 1 (Atox1) transfers copper to the secretory pathway and nucleus and an ensemble of proteins deliver copper to cytochrome-c oxidase (CCO) in the mitochondrion [7, 8]. It has been suggested that Cu-ATPase-mediated secretion of excess copper is the main factor regulating copper homeostasis . Equally important to mammalian growth and development, however, is the copper transporter Ctr1. Cell-specific knock-out of Ctr1 in mouse intestinal epithelial cells caused marked growth retardation, coupled with cardiac hypertrophy and overall viability defects that caused postnatal lethality around three weeks of age . Similarly, knock-out of the Drosophila melanogaster Ctr1A gene resulted in developmental arrest at early larval stages . The C terminus of yeast copper transport protein, Ctr1, appears to have a role in copper regulation and stopping the build-up of copper to toxic concentrations .
Although a P. falciparum P-type ATPase copper efflux protein has been described , an additional protein-mediated mechanism for copper acquisition has not. The current study describes a novel membrane-bound P. falciparum protein with copper transporter characteristics. A recombinant form of the protein's amino terminal region bound reduced copper in vitro and within E. coli host cells. The native P. falciparum copper transport protein was found to initially localize to the infected erythrocyte membrane, followed by relocation to a parasite membrane through the development of the parasite’s erythrocytic cycle.
Copper-dependent protein orthologs to Homo sapiens proteins identified in the P. falciparum proteome
P. falciparum ortholog identified
PlasmoDB protein identifier
Cu/Zn Superoxide dismutase
Cytochrome-c oxidase subunit I
Cytochrome-c oxidase subunit II
PF14_0288 (IIa) PF13_0327 (IIb)
Cytochrome-c oxidase subunit III
Cytochrome-c oxidase subunit Vb
Cytochrome-c oxidase subunit VIb
The amino acid sequence for a putative P. falciparum copper transport protein (PF14_0369) was retrieved from PlasmoDB and subjected to Predict7  analysis to identify immunogenic regions and the peptide CNLQKEEDTVVQLQD was selected for synthesis (GL Biochem Ltd. Shanghai, China). The selected peptide was coupled to rabbit albumin via the native amino terminal cysteine residue using m-maleimidobenzoyl-N-hydroxysuccinimide ester. Hy-Line Brown hens were used for antibody generation and antibodies were isolated from egg yolk as previously described [18, 19]. Animal ethics clearance was obtained from the University of KwaZulu-Natal animal ethics committee (003/10/Animal).
Plasmodium falciparum (D10) gDNA was isolated from infected red blood cells using the FermentasTM DNA purification kit. The full-length PF14_0369, open reading frame was amplified using the following specific primers: PfCtr369-fwd: 5'-at GAATTCGACAAAAGCGACAATAGTATTTG-3' and PfCtr369FL-rev: 5'-ACATCCACAACAAGCTGGATC. For functional copper binding studies, the coding domain for the protein's amino terminal domain minus signal peptide was amplified using the PfCtr369-fwd primer containing an engineered EcoRI cleavage site and PfCtr369Nt-rev: 5'-ca CTGCAGTTACGATTTGGTTTCCCATTTG-3' containing a PstI cleavage site (cleavage sites underlined). PCR products were analyzed on a 1% (w/v) agarose gel. The PCR amplicon for the PF14_0369 amino terminal domain (designated Nt) minus signal peptide (PfCtr369Nt-S) was cloned in frame with the maltose binding protein (MBP) gene in the pMal-p2x bacterial expression vector (New England BioLabs, USA), as previously described .The integrity of the resultant pMal-p2x/PfCtr369Nt-S plasmid insert was confirmed by DNA sequencing. Using the strategy employed for pMal-p2x/PfCtr369Nt-S, a plasmid containing pMal-p2X/PfCtr211Nt-S from the PF14_0211 gene was prepared.
A single E. coli JM103[pMal-p2x/PfCtr369Nt-S colony was grown overnight at 30°C in sterile LB medium supplemented with 100 μg.ml-1 ampicillin. The overnight culture was diluted in fresh broth, plus antibiotic, and incubated at 30°C until the A600 reached ~0.6. MBP-PfCtr369Nt-S expression was induced with 0.3 mM IPTG and the cultures grown for a further 16 h before harvesting of cells by centrifugation (4000 g, 15 min, 4°C). The E. coli periplasmic content was isolated following the method described by French et al. and the recombinant protein purified by amylose affinity chromatography. Recombinant MBP-PfCtr211Nt-S was expressed using the same procedure. Protein samples were analyzed on a 12.5% SDS-PAGE gel  followed by Western blotting and probed with chicken anti-PF14_0369 and anti-MBP antibodies  (see below). Protein concentration was determined with the Bradford dye-binding protein assay, using ovalbumin as the standard .
Plasmodium falciparum 3D7 parasites were cultured in human O+ erythrocytes according to standard protocols, except cultures were incubated in gassed flasks . Synchronized asexual parasites were obtained by sorbitol treatment .
For Western blot analyses of P. falciparum lysates, protein samples were prepared from saponin-isolated, sorbitol-synchronized parasites. Equivalent amounts of 1 x 107 parasites were analysed by 12.5% SDS-PAGE and transferred to a nitrocellulose membrane. Primary chicken anti-PF14_0369 and chicken anti-P. falciparum lactate dehydrogenase (LDH) antibodies against specific peptides on each protein were employed and detected using rabbit anti-chicken horseradish peroxidase-conjugated secondary antibodies (DAKO, Santa Cruz). Immunoblots were developed using enhanced chemiluminescence. Recombinant MBP-PfCtr369Nt-S expression was analysed by Western blot using chicken anti-MBP and anti-PF14_0369 anti-peptide antibodies. These antibodies were detected using rabbit anti-chicken horseradish peroxidase-conjugated secondary antibodies (Jackson Immuno-Research Laboratories, USA) and visualized by incubating the membrane in 0.06% (w/v) 4-chloro-1-naphthol and 0.0015% (v/v) H2O2.
For in vitro copper binding studies, purified MBP-PfCtr369Nt-S (10 μM) was mixed with 200 μM CuCl2, with or without 10 mM ascorbic acid, and incubated at RT for 15 min. Unbound copper was removed by overnight dialysis against sodium phosphate buffer [0.1 M NaH2PO4, 0.01% (w/v) NaN3, pH 7.5] at 4°C. Copper binding to MBP-PfCtr369Nt-S and the oxidation state of bound copper was assessed using the bicinchoninic acid (BCA) assay , in which protein-copper complexes are disrupted by acid denaturation and copper detected upon release. BCA binds reduced copper (Cu+), forming a purple complex detectable at 354 nm. If copper is present in solution as Cu2+, ascorbate must be added to the sample following acid denaturation to yield a Cu+-BCA complex. However, if the addition of ascorbate does not influence Cu+-BCA complex formation, this indicates copper was released from the protein in the Cu+ state. Solutions of CuCl2 in sodium phosphate buffer were used as standards and affinity purified MBP served as a negative control. The ability of MBP-PfCtr369Nt-S to bind copper within E. coli cells was also examined. Cells containing the recombinant protein were grown in the presence or absence of 0.5 mM CuCl2, and the recombinant protein purified as described above. Prior to cell disruption and MBP-PfCtr369Nt-S isolation, pelleted cells were washed with a Tris buffer (0.2 M Tris–HCl, pH 7.5) to remove extraneous CuCl2. As with in vitro copper binding, copper bound to MBP-PfCtr369Nt-S within E. coli cells was detected using the BCA assay.
The ability of MBP-PfCtr369Nt-S to inhibit the copper-catalysed oxidation of ascorbic acid (H2Asc) was tested in vitro. In a typical experiment a mixture was prepared containing 5 μM MBP-PfCtr369Nt-S,, or 5 μM MBP (control), 8 μM CuCl2 and 120 μM H2Asc solution, adjusted to pH 4.5 with NaOH. The absorbance of this solution was monitored at 255 nm with readings taken every 5 s for 300 s. Reactions were carried out at RT.
Immunofluorescence assays were carried out following cell fixation with 4% paraformaldehyde/0.00075% (w/v) glutaraldehyde as previously described  except fixation was carried out at 37°C for 30 minutes and quenching was performed with 125 mM glycine/PBS. Chicken anti-peptide antibodies directed against P. falciparum lactate dehydrogenase and PF14_0369 and a rabbit antibody against Exp-1 were used. Primary antibodies were diluted in 3% (w/v) BSA/PBS and detected with either goat anti-chicken-Cy3 or goat anti-rabbit-Cy2 antibodies (DAKO, Santa Cruz). Hoechst 33258 (Molecular probes) was used at a concentration of 50 ng.ml-1 to stain parasite DNA. All images were acquired at room temperature on a Zeiss Cell Observer using appropriate filter sets.
Cytochrome-c oxidase (CCO) is a key multi-subunit catabolic protein containing three functionally important copper ions in its binuclear CuA and mononuclear CuB sites . Metallation of the CuA and CuB sites requires the copper metallochaperones Cox17, Sco1 and Cox11 . Identification of P. falciparum orthologs of these metallochaperones and other related copper-requiring proteins supports a plasmodial requirement for copper, as previously suggested . A BlastP search of PlasmoDB, using mammalian and yeast proteins, identified six copper binding protein orthologs (Table 1). These were Cox11, Cox17, Cox19, Sco1, five CCO subunits and S-adenosyl-L-homocysteine hydrolase. The identification of CCO subunit I and II orthologs appears most important since these house the binuclear CuA and mononuclear CuB copper binding sites . Interestingly, the P. falciparum proteome appears to lack an identifiable class of metallothioneins and Cu/Zn SOD (Table 1). In yeast and mammalian cells, these proteins are essential for copper storage and free radical detoxification, respectively [7, 8].
A cladogram was generated for the complete Plasmodium copper transporters to compare their similarity to seven characterized copper transporters. The Plasmodium copper transporters formed a cluster related to, but separate from, the characterized copper transporters (Figure 1C). Of the characterized transporters, the sequence with the greatest similarity to that of Plasmodium parasites was from Theileria parva. This was to be expected since T. parva belongs to the same phylum as Plasmodium parasites, the Apicomplexans . Closer analysis of the Plasmodium sequences indicated that the three rodent-infecting malaria parasites cluster together, whilst P. vivax, P. knowlesi and P. falciparum form a separate but related cluster. Similar relationships have been observed from a phylogenetic analysis of the Plasmodium cytochrome-b and small subunit rRNA sequences . The separate cluster of the mammalian, yeast and plant reference sequences suggests a more distant similarity to the Plasmodium sequences.
Considering its clinical importance, P. falciparum was selected for further studies of the putative copper transporter. The gene encoding the putative copper transport protein (PF14_0369) is located on the positive strand of chromosome 14 as a two-exon gene with a 336 bp first exon and 372 bp second exon, translating to a 235 amino acid protein with a predicted mass of 27.15 kDa. The full length and amino terminal coding domain coding sequences for the putative P. falciparum copper transport protein (PF14_0369) were PCR amplified from genomic DNA, producing amplicons of approximately 890 bp and 290 bp, respectively (Figure 1D). An 820 bp product was amplified for the LDH control (Figure 1D). The identity of the amplicon for the putative transporter’s amino terminal domain (PfCtr369Nt-S) was confirmed by sequencing. This amplicon was cloned and expressed as a MBP fusion partner (MBP-PfCtr369Nt-S).
The ability of MBP-PfCtr369Nt-S to bind copper within a cellular environment was examined by expressing the protein in the presence of 0.5 mM CuCl2. This concentration of copper does not affect E. coli growth . MBP-PfCtr369Nt-S isolated and affinity purified from E. coli cells grown in copper-enriched media, bound copper (Figure 3B). Extraneous copper was removed by washing the cells prior to protein isolation. Protein isolated from cells grown without copper had negligible amounts of copper present (p < 0.01) (Figure 3B). The oxidation state of copper bound within the cell was inferred by including or excluding ascorbate from the BCA assay (Figure 3B). Like the in vitro analysis, Cu+ appeared to be the preferred oxidation state bound by the protein within the E. coli periplasm. An interesting consequence when copper was added to the E. coli growth medium was a four-fold increase in MBP-PfCtr369Nt-S yield from 1.53 mg to 5.98 mg per litre of culture (p < 0.01) (Figure 3B, inset). In all experiments MBP served as a negative control. Recombinant MBP did not bind copper and CuCl2 had no perceptible influence on MBP expression (Figure 3).
Characterized copper transport proteins are defined as integral membrane proteins that, depending on cell type, localize to either the plasma membrane or intracellular vesicles . Analysis of the PF14_0369 protein sequence predicted an amino terminal signal peptide  and an apicoplast targeting signal . Further analysis indicated that PF14_0369 lacked an identifiable PEXEL/HT motif. This suggested the putative copper transporter would be targeted to the parasite’s apicoplast and not to the parasite or erythrocyte plasma membranes. To define the protein’s native sub-cellular location, the anti-peptide antibodies specific for the putative copper transporter were used in immunofluorescence assays. For this assay a modified form of the fixation protocol developed by Tonkin et al. was employed. Advantages of this technique, over the traditional methanol or methanol/acetone fixation methods, include decreased auto-fluorescence and improved preservation of cell morphology .
Export of the putative copper transporter to the erythrocyte surface was surprising considering the lack of an identifiable PEXEL/HT motif. Exported P. falciparum proteins lacking a PEXEL motif have previously been identified and are termed “PEXEL negative exported proteins” (PNEPs) . Partial PEXEL motifs have been identified in the amino terminal domain of the PNEPs REX2 (LAE)  and SBP1 (LAD)  that are essential for export. Inspection of the PF14_0369 amino terminal domain identified a partial PEXEL motif identical to the important LAD motif identified in SBP1. Analysis of the essential transmembrane domains from SBP1 and REX2 highlighted a relative abundance of phenylalanine residues  also found in the third transmembrane domain of PF14_0369 (Additional file 1).
Maintenance of copper homeostasis is essential for cell survival. Consequently sophisticated mechanisms have evolved for copper acquisition, distribution and excretion. In mammalian cells, copper is acquired by the copper transport protein Ctr1 and distributed to target proteins via dedicated copper metallochaperones such as CCS, Atox1 and Cox17 [7, 8]. To avoid the potentially toxic effects of copper, Cu-ATPase proteins mediate the excretion of excess copper  and a P. falciparum Cu-ATPase ortholog has been identified . One P. falciparum copper source appears to be from the ingestion of host erythrocyte Cu/Zn SOD . A screen of the PlasmoDB database, performed in the present study, identified six P. falciparum copper-requiring protein orthologs and a candidate copper transport protein. Native expression of the putative copper transporter was confirmed by Western blot and the protein’s subcellular location identified by immunofluorescence microscopy. A recombinant form of the transporter's amino terminal domain was shown to bind copper in vitro and within E. coli host cells, supporting the possibility that the full-length protein functions to acquire copper.
The identification of six P. falciparum copper-requiring protein orthologs suggested an important role for copper in P. falciparum metabolism. Treating P. falciparum parasites with the intracellular copper chelator, neocuproine, inhibited ring-to-trophozoite transition in vitro, highlighting the importance of copper for the parasite. During the course of its asexual cycle, P. falciparum appears to digest host erythrocyte Cu/Zn SOD to release copper . The presence of a membrane associated copper binding protein with copper transport protein characteristics in plasmodia suggests a mechanism for the transport of copper similar to that played by the Ctr1 copper transport protein in yeast and mammalian cells [7, 8]. Candidate copper transporter sequences were identified for eight species of the Plasmodium parasite and each sequence was shown to contain the essential and largely definitive Mx3M and Gx3G motifs (Figure 1B) [30, 45]. The predicted amino terminal domain of each putative transporter was found to contain one or more methionine motifs (MxM or MxxM). Although not limited to copper transporters, these motifs are important for Ctr1 protein function .
The presence of three putative membrane-spanning regions is considered definitive for copper transport proteins  and topological analysis of the putative P. falciparum copper transporter identified three such domains. Since three transmembrane domains cannot form a functional channel or pore for ion transport , a homotrimeric complex of copper transport protein monomers is formed for the characterized mammalian and yeast copper transport proteins . The P. falciparum copper transport protein may form a similar complex since the first transmembrane domain of the copper transport protein family serves as an adaptor allowing evolutionarily distant copper transporters to adopt a similar overall structure . An immunoblot of yeast and mammalian copper transport proteins identified monomeric and dimeric species of the trimeric complex . Similar monomeric and dimeric species were detected in an immunoblot of the putative P. falciparum copper transport protein in this study.
Signal peptides, transmembrane domains, rare codons, introns and genome AT-richness affect recombinant expression of P. falciparum proteins . The amino terminal signal peptide of the putative P. falciparum copper transporter was, therefore, excluded from the expression construct, resulting in periplasmic expression. The purified MBP-PfCtr369Nt-S recombinant protein bound reduced copper in vitro and within a cellular environment. A second copper transport protein, MBP-PfCtr211Nt-S, from the PF14_0211 gene was recombinantly expressed and isolated. Expression of the native P. falciparum protein encoded by this gene has not been ascertained at this stage.
In mammalian and yeast cells, the reduced cuprous ion (Cu+) is favoured for transfer and transport , since Cu+ is more exchange labile than Cu2+. Cu+ is highly reactive in an oxidizing environment and methionine motifs in the extracellular amino terminal domain of the copper transporter family are thought to sequester copper prior to its transport across the lipid bilayer . This was supported by a study demonstrating that methionine motif peptides inhibited copper-catalysed ascorbate oxidation through Cu+ chelation . MBP-PfCtr369Nt-S inhibited copper-catalysed ascorbate oxidation through copper chelation. Taken together with the data showing copper binding to MBP-PfCtr369Nt-S, it appeared that the amino terminal domain of the putative P. falciparum copper transporter preferably coordinates the Cu+ ion. Copper coordination was presumably a result of the methionine motif present in MBP-PfCtr369Nt-S.
The methionine motifs contained in the amino terminal domains of yeast and human copper transport proteins are essential for copper binding under copper limiting conditions . The yeast copper transport protein contains eight methionine motifs, but only the last methionine of the eighth motif was shown by Puig et al. to be essential for copper binding capability. The position of this methionine is conserved between copper transport proteins and is 20 amino acid residues N-terminal to the transporter’s first transmembrane domain . Interestingly, the last methionine of the methionine motif in the amino terminal domain of the putative P. falciparum copper transport protein is located in a similar position. However, the involvement of other amino acid residues, like cysteine and histidine residues that bind metal ions [47, 48], and are present in this protein cannot be excluded. Identification of the specific residues coordinating copper will be explored in future experiments.
An interesting consequence of expressing MBP-PfCtr369Nt-S in the presence of copper was a significant increase in recombinant protein yield. In the presence of excess copper it is a possibility that the copper bound to and stabilized the recombinant protein . The cytoplasm and periplasm of E. coli cells do, however, have different copper requirements since almost all bacterial copper proteins, such as multi-copper oxidases, amine oxidases and lysine oxidases are found in the periplasm or excreted extracellularly . Copper availability in E. coli is thought to be regulated by the actions of the DNA-binding metal sensor CueR, which controls the expression of genes encoding proteins involved in metal homeostasis . A copper-requiring protein is thought to gain access to copper only if the protein’s affinity for the metal ion is greater than the buffered cellular concentration of copper . Interaction of a recombinant P. falciparum protein with native E. coli proteins has been demonstrated for PfHsp70 , supporting the possibility that MBP-PfCtr369Nt-S interacts with native E. coli copper proteins for copper loading. These interactions could, in turn, influence the availability of copper for native E. coli copper-requiring proteins resulting in tightly regulated MBP-PfCtr369Nt-S expression under standard growth conditions. The addition of excess extracellular copper appears to have alleviated this growth stress producing the increase in the expression of MBP-PfCtr369Nt-S observed here.
During early stages of asexual development, the putative P. falciparum copper transporter appears to be targeted to the plasma membrane of the infected erythrocyte. As the parasite matures through its asexual cycle the protein was detected on a parasite membrane. This may represent a turnover of host cell membrane associated malarial proteins, followed by trafficking of the copper transporter to a parasite membrane. The precise localization of the copper transporter, when associated with the parasite, was not elucidated, due to limits to the fluorescence detection methods. The parasitophorous vacuolar membrane contains a “promiscuous” pore that permits the passage of solutes, nutrients and macromolecules [53, 54] and thus it is unlikely for there to be the need for a separate copper transporter. The parasite plasma membrane contains a variety of selective transporters (Reviewed by Martin et al.). The copper transporter is thus suggested to be associated with the parasite plasma membrane. PfCuP-ATPase was similarly localized to a parasite plasma membrane and the surface of the infected erythrocyte. However, unlike the putative copper transport protein, PfCuP-ATPase was associated with both the parasite plasma membrane and the erythrocyte membrane at the same time in trophozoites and schizonts, whilst no expression was detected in rings . These authors suggested that dual localization of PfCuP-ATPase represents a novel mechanism by which the parasite reduces copper toxicity through copper efflux . Reasons for the different localities of the copper transport protein are less apparent. There is the possibility that during ring stages, extracellular copper is required until the parasite digests endocytosed host cell Cu/Zn SOD and translocates the copper transporter protein to a parasite membrane.
The concentration of copper in normal red blood cells has been reported as18 μM and copper is imported into red cells by a Cl--HCO3 - anion exchanger . Inside the red cell most of the copper is associated with Cu/Zn SOD and a 30 and a 40 kDa red cell protein . Rasoloson et al. reported a lower red cell copper concentration at 10 μM and further suggested that copper concentrations in parasitized cells was lower than uninfected cells. Copper in isolated parasites was lower than in red cells at 2 μM in rings and between 6 and 6.6 μM in trophozoites (depending on the isolation method). This data disagrees with recent data indicating that malaria infected red cell copper concentration was 30 μM, and double that of uninfected red cells in the same experiment . The different values for copper concentrations in infected red cells may be due to the different measuring methods employed and needs to be reconciled [5, 57]. Interestingly as the parasite develops from trophozoites to schizonts, the concentration of ingested Cu/Zn SOD decreases  and this coincides with data presented here showing the appearance of the putative copper transporter on parasite membranes, the highest level of expression of the protein, and mRNA levels described by Le Roch et al.. Thus as the source of copper in the form of Cu/Zn SOD decreases in the parasite , the parasite compensates with an increase in expression of the putative copper transport protein. Given the potential toxicity of copper and the need for the regulation of copper homeostasis, it is likely that the parasite employs a copper transport protein alongside the PfCuP-ATPase copper export protein described by Rasoloson et al.. It is suggested that, given the presence and membrane location of a copper binding protein with copper transporting motifs, and the increase in copper concentration in malaria infected red cells , that the protein described here has a role in copper transport. Data to support a transport role for the copper binding protein is being pursued.
Treatment of P. falciparum parasites with neocuproine inhibited ring-to-trophozoite transition , implicating copper metabolism as a potential target for novel anti-malarial drug development. Copper, when the concentration is above a critical level, has been shown to be toxic to the parasite . However, from in silico findings it seems unlikely that P. falciparum contains unique copper-dependent metabolic pathways. Copper-binding protein motifs also appear to be conserved from prokaryotes to eukaryotes , suggesting that designing a compound specific for P. falciparum copper-dependent proteins could prove difficult. An alternative approach would be to deliver anti-malarial compounds to the parasite by exploiting transport mechanisms, as suggested for the P. falciparum new permeation pathway and choline carrier . The human copper transport protein has been shown to transport the platinum containing anti-cancer drugs cisplatin and oxaliplatin . Cisplatin also exhibits anti-malarial activity through DNA damage [62, 63]. Given the structural similarities between the putative P. falciparum copper transport protein and the human copper transporter, it is possible that cisplatin is delivered to the parasite by the putative copper transport protein. This transport mechanism could perhaps be exploited for the delivery of novel platinum-based anti-malarial compounds.
Plasmodium falciparum actively remodels the erythrocyte during infection, leading to an increase in the permeability of the host cell membrane to low molecular weight solutes . This increase is mediated by ‘new permeability pathways’ that have also been shown to greatly facilitate the uptake of the antibiotics fosmidomycin and its derivative FR900098 . Parasitized erythrocytes also show a significant increase in the uptake of a copper-neocuproine complex when compared to uninfected erythrocytes . Association of the putative P. falciparum copper transport protein with the erythrocyte membrane, during early asexual development, therefore makes it tempting to speculate that the copper transport protein’s mechanism accounts for the increased rate of copper-neocuproine complex uptake by parasitized erythrocytes. Complexation of the anti-malarial buparvaquone to copper(II) significantly enhanced its anti-malarial activity . This was proposed to be a consequence of improved compound internalisation, which may be related to the transport mechanism of the putative copper transport protein. Localization of the P. falciparum copper transport protein to the erythrocyte and parasite membranes in late ring and early trophozoite stages may explain the increased susceptibility of these stages to cisplatin .
Plasmodium falciparum protein export beyond the parasitophorous vacuole membrane was suggested to be mediated by the PEXEL/HT motif, but this motif was later shown not to be the sole determinant of protein export. Over 300 PEXEL/HT proteins have been predicted, whereas only a few PEXEL negative export proteins (PNEPs) have been identified and a common PNEP motif is yet to be identified . One important feature of PNEP proteins is the presence of a transmembrane domain, since removal of this domain from the PNEPs MAHRP1, SBP1 and REX2 inhibited protein export [43, 44]. Similarly, the transmembrane domain of SURFIN4.2 was essential for protein trafficking to the infected erythrocyte and Maurer’s clefts .
Analysis of the PNEPs REX2  and SBP1  identified partial PEXEL/HT motifs (LAE and LAD, respectively) as being essential for protein export. Analysis of the PF14_0369 sequence identified an identical LAD motif within the first 20 amino acids of its amino terminal domain, a similar position to the LAD motif in SBP1. Mutational analysis of the related LAE motif in REX2 established that the glutamate residue was essential for export . This suggests the related aspartate residue in the LAD motif of PF14_0369 may be important for export. Cleavage of the signal peptide in the PF14_0369 sequence would generate a new amino terminus similar to that generated following cleavage of the PEXEL motif in PEXEL/HT proteins. PEXEL motif cleavage generates an xD/E/Q amino terminus, with the presence of D/E/Q being essential for protein export . Another potential export signal in SBP1 is located N-terminal of its transmembrane domain, between amino acids 180 and 190 . A comparison of this short signal sequence (NEYEVES) with the PF14_0369 sequence identified a similar sequence upstream of the first predicted transmembrane domain (NKWETKS), thereby implicating this sequence as a potential contributor to successful protein export. The presence of motifs in PF14_0369 similar to those important for PNEP export suggests that these motifs play a role in the export of the putative copper transporter.
The amino terminal domain of the P. falciparum copper binding protein described here binds copper in vitro and in an in vivo expression system. The protein has copper transport motifs and has been shown to be expressed by malaria parasites and locate to two different membranes (the erythrocyte and the parasite membrane) as the parasite develops within the infected red cell. This evidence implicates a copper transport role for the protein in malaria infected erythrocytes and this implication is being explored. Whether the protein imports copper or has a copper export role alongside the Cu-ATPase protein described by others , remains to be elucidated.
A putative P. falciparum copper transport protein was identified and showed that the recombinant amino terminal domain binds copper in vitro and within E. coli host cells. This ability, combined with conserved copper transport protein sequence features, suggest an additional malaria copper transport mechanism to that previously described . Interestingly, the putative P. falciparum copper transport protein initially localized to the erythrocyte membrane and then relocated to a parasite membrane through the course of asexual development. The presence of a putative malaria copper transport protein has interesting implications for malaria copper metabolism and the protein may have a potential role as a novel anti-malarial drug delivery system.
Antioxidant protein 1
Copper chaperone for superoxide dismutase
Copper transport protein 1
Cuprozinc superoxide dismutase
Export protein 1
Glyceraldehyde 3-phosphate dehydrogenase
Heat shock protein 86/90
Membrane-associated histidine-rich protein 1
Maltose binding protein
Plasmodium export element/host targeting signal
Plasmodium falciparum copper transport protein PF14_0369 amino terminal domain minus signal peptide
PEXEL negative exported protein
Ring exported protein 2
Skeleton binding protein 1.
Our thanks to Peter Smith (University of Cape Town) for providing us with the P. falciparum D10 cultures and to Rob Skilton (International Livestock Research Institute) for supplying the T. parva gene sequence. We thank Prof Theresa Coetzer for critical reading of the manuscript.
We would like to thank the South African Department of Science and Technology S.A. Malaria Initiative, S.A .National Research Foundation, the S.A .Medical Research Council and the UKZN Research Incentive Fund for financial support. JMP is supported by the DFG German-Africa projects in Infectology programme.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.