The ATP-binding cassette (ABC) superfamily is one of the largest evolutionarily-conserved families of protein transporters. ABC proteins play key roles in cellular detoxification of xeno- and endobiotics. Overexpression of certain ABC proteins, among them the multidrug resistance protein (MDR) and the multidrug resistance associated proteins (MRPs), contribute to drug resistance in a variety organisms ranging from parasitic protozoa to human neoplastic cells.

Membrane transporters, such as the Plasmodium falciparum chloroquine resistant transporter (pfcrt) and the Plasmodium falciparum mdr1, which is a member of the ABC superfamily, have been identified as key contributors in decreasing susceptibility to several anti-malarial drugs 1234. Research to identify additional potential contributors to Plasmodium drug resistance has lead to the identification of new candidate transporter genes, some of which belong to the ABC transporter superfamily 5678. The ABC transporter superfamily is comprised of eight subfamilies in eukaryotes: ABCA, ABCB, ABCC, ABCD, ABCE, ABCF, ABCG, and ABCH. Proteins within this superfamily were classified based on the sequence and organization of their conserved nucleotide binding domains (NBD). Characteristic motifs within these NBDs are found in the majority of adenine nucleotide hydrolases: the Walker A and Walker B boxes, ABC signature motif, H (histidine) loop, D (aspartate) loop, and Q (glutamine) loop 910111213. In general, functional ABC proteins contain two NBDs and two transmembrane domains (TMD) consisting of 6–11 transmembrane helices. Genes are organized either as full transporters containing two of each domain or half transporters with one of each (Figure 1c).

Members of the ABCC sub-family have been associated with drug resistance in organisms ranging from bacteria to man. This sub-family is comprised of a variety of proteins some of which have been designated as multidrug resistance associated proteins (MRPs). These proteins serve as primary active transporters of an array of structurally diverse compounds including organic anions such as glucuronide, glutathione (GSH), sulphate, drugs conjugated to GSH, and non-conjugated agents by GSH co-transport 1415. The human ABCC sub-family consists of 13 members, nine of which are transporters: MRP1, MRP2, MRP3, MRP4, MRP5, MRP6, MRP7, MRP8, and MRP9 1617. The human MRP1 has been the most studied among the MRP proteins because of its ability to transport a broad range of anticancer drugs through cellular membranes mediated by GSH co-transport or by the export of GSH-drug conjugates 1418. In addition, MRPs have been associated with drug resistance in other organisms such as the heavy metal resistance protein MRP-1 in Caenorhabditis elegans 19, the GSH conjugate transporters AtMRP1, AtMRP2, and AtMRP3 in Arabidopsis thaliana 202122, and yeast cadmium factor gene (ycf1) in Saccharomyces cerevisiae 23. Several mrp orthologs have been identified and linked to drug resistance in protozoa such as Trypanosoma brucei and Leishmania spp. In T. brucei, reduced sensitivity to melarsoprol has been linked to the overexpression of the MRP homologue, TbMRPA 24. In Leishmania, the pentamidine resistance protein 1(PRP1/ABCC7) and MRPA (formerly known as P-glycoprotein A/ABCC3) were shown to confer pentamidine resistance and antimony resistance, respectively 2526272829.

Homologues of these proteins have been identified in Plasmodium spp., but there is no clear evidence linking this transporter to anti-malarial resistance. Expression of MRP homologues was reported in P. berghei and P. yoelii 7 as well as in P. falciparum 30. The P. falciparum mrp1 gene is an intronless gene with highest mRNA expression in the late trophozoite and schizont stages. The P. falciparum mrp2, however, is mostly expressed in ring stages 31. Protein expression of these genes has only been reported for pfMRP1, where it has only been detected in the schizont stage, except in the CQ resistant P. falciparum strain, FAC8, where it was also detected in the trophozoite stage 30. Mutations within this gene have been associated with anti-malarial resistance in clinical isolates 5632.

In this study, the pbmrp gene was partially characterized using bioinformatics and molecular biology methods. We sequenced pbmrp and classified it as an ABCC sub-family member. Multiple sequence alignments with MRPs from P. y. yoelii, P. vivax, P. knowlesi and P. falciparum reveal a high degree of conservation within transmembrane and nucleotide binding domains. Bioinformatic analyses indicate that pbMRP is homologous to a Plasmodium spp. MRP2 gene. In addition, the gene copy number, structural organization, chromosomal localization and mRNA expression levels of the pbmrp gene were determined in drug sensitive (N clone) and the drug resistant derived line RC which was selected under CQ pressure and displays the multidrug resistance phenotype.

Energy-dependent transporters are responsible for maintaining the metabolic homeostasis in organisms. Understanding transporters involved in cellular detoxification and/or drug efflux in Plasmodium spp. can provide critical information about their function and their potential as new drug targets. Plasmodium berghei, a rodent malaria model, is a proven and valuable tool for studying Plasmodium biology as well as the development of resistance to anti-malarials which continues to be global health problem. In this study, a multidrug resistance associated protein gene in P. berghei was sequenced. This gene, pbmrp, was classified and characterized using bioinformatics. The pbmrp gene organization, copy number, and gene expression levels were compared between the drug sensitive N clone and the CQ selected (RC) P. berghei line.

The pbmrp gene has an intronless ORF of 5820 nucleotides and is predicted to encode a protein comprised of 1939 amino acids. NCBI's Conserved Domain Database classified pbMRP as a member of the ABCC transporter sub-family (Figure 1a). Using the predicted pbMRP sequence, six homologues were identified in four Plasmodium species (P. falciparum, P. vivax, P. knowlesi and P. y. yoelii) by sequence similarity searches in the PlasmoDB and in NCBI. We note that at the time the analyses were performed the annotation of the proteins at NCBI and PlasmoDB is only complete for the pfMRP2 and pfMRP1. For pvMRP1 and pvMRP2 the annotations are ABC transporter and multidrug-resistance associated protein, respectively. The classification as MRP1 and MRP2 is based on the results presented here from similarity searches and phylogenetic analyses. For pkMRP2, pyMRP2 and pbMRP2 the annotations are transporter/putative mrp, ABC transporter, and multidrug-resistance associated protein respectively. As before, classification as MRP2 is based on the results presented here from similarity searches and phylogenetic analyses. At this time, the presence of two mrp genes appears to be unique for the two human malaria species, P. falciparum and P. vivax (Table 2). An additional mrp gene may confer a biological advantage for P. falciparum or P. vivax with respect to fitness or the evolution of drug resistance but this has yet to be proven experimentally. BLAST results support that from an evolutionary standpoint pbMRP is more closely related to pyMRP, pkMRP, pfMRP2 and pvMRP2 rather than to pfMRP1 and pvMRP1 (Table 1). Interestingly, sequence similarity searches against the human genome show that pbMRP is more closely related to the MRP2. Taken together, the fact that pbMRP is more similar to pvMRP2 and pfMRP2 in addition to having the closest relatedness to the human MRP2 confirms that the P. berghei transporter is a member of the ABCC MRP2 sub-family. The phylogenetic analysis presented in Table 1 and Figure 4 supports the hypothesis that the MRP2 family is the ancestral family of MRPs in Plasmodium. It is noteworthy that at this time the MRP1 sub-family only appears in parasites that infect humans. Given the lack of knowledge about the function of the MRPs in Plasmodium, it is not possible to assess the significance of the finding at this time. It is also important to note that the MRP2 clade groups the gene from P. knowlesi with that of P. vivax, in agreement with published phylogenies [55 and references therein]. Human MRP2 is involved in the terminal phase of detoxification and excretion of endogenous and xenobiotic organic anions in a unidirectional manner. This protein actively transports glutathione, glucuronate, or sulphate conjugates in addition to a variety of non-conjugated anionic substances including glutathione and glutathione disulfide 56. Although MRP2 has broad substrate specificity, it has highest affinity for gluconarate and GSH conjugates of lipophilic substances 5758. Based on the sequence similarity to MRP2 and taking into account the biological function of this membrane transporter, it would be reasonable to suggest that the biological function of pbMRP might be related to cell detoxification and the secretion of conjugated and/or non conjugated endogenous and xenobiotic anions. In addition, human MRP2 has been shown to confer drug resistance to multiple chemotherapeutic agents in cell lines expressing the recombinant protein or antisense constructs 596061. Given that MRP2 confers resistance in human cells in addition to the fact that the overexpression of homologues in Leishmania and T. brucei 242629 is associated with drug resistance raises the possibility that pbmrp may also contribute to this phenomenon.

The pbMRP displays the typical core ABCC/MRP domain organization of two units of a TMD and a NBD (Figure 1c). All human MRPs, the best and most completely characterized members of this sub-family, possess this typical core structure with the exception of MRP1, MRP2, MRP3, MRP6 and MRP7 which have an additional N-terminal region composed of a third transmembrane domain (TMD0) 62631764. The pbMRP does not possess this additional TMD0 (Figure 1b) similar to the other members of human MRP family such as MRP4, MRP5, MRP8 and MRP9 63656667.

The NBD size and organization in the MRPs of the Plasmodium spp. studied were similar to those described in other organisms. A high level of similarity was observed in multiple sequence alignments, in particular within the NBDs with the exception of a large, non conserved insertion of approximately 103 amino acids in NBD1 (from position 961 to 1110 in the multiple sequence alignment) (Figure 3, Additional file 3). Characteristic NBD motifs such as the Walker A, Q loop, ABC motif, Walker B, D loop and the H loop were found within the NBDs (Figure 3). All of the characteristic motifs were identified with the exception of the H loop in NBD1. This NBD is unique in the sense that there is an insertion between the ABC motif and the Walker B motif. This insertion was present in all the Plasmodium MRP sequences studied. The Walker B motif and the D loop were identified after this large insertion in the multiple sequence alignment. The location and sequence of the Walker B motif in the region following the insertion is similar to that described as the Walker B in pfMRP1 30. The D loop could not be identified in the MRP NBD2 in all the Plasmodium species studied (Figures 2 and 3). The failure to identify the H loop in NBD1 and the D loop in NBD2 may be due to a lack of conservation in the sequences encompassing these motifs in Plasmodium. Both the H loop and the D loop have structural functions within the NBD although their precise role in the molecular mechanism is not clear. The H-loop histidine contacts the bound nucleotide and the D loop may form a hydrogen bond with the Walker A backbone 1368. Remarkably, the sequences characteristic of an ABC transporter NBD were highly conserved among the Plasmodium species studied but these are not strictly conserved when compared to the consensus sequences described for these motifs in vertebrates (Table 2).

The structure and function of MRP2 genes in Plasmodium has not been described. In neoplastic cells as well as in Leishmania, gene duplication and/or overexpression correlate with the drug resistant phenotype 6970. In this study, the pbmrp gene was shown to have has similar gene organization and the same gene copy number in both the drug sensitive N clone and the CQ selected RC line (Figure 5). In addition, this gene was not differentially expressed at an mRNA level between these two lines (Figure 7). However, amplification or increased expression might have been lost upon removal of drug pressure given that RC line was not under continuous CQ pressure in these experiments. In primary cultures of rat brain endothelial cells treated with dexamethasone, increased expression of mrp2 at the mRNA and protein level was detected in a concentration dependent manner. This increase in expression was reversible in the absence of drug pressure 71. In a CQ selected P. falciparum clone, an increase in the size of chromosome 3 was observed with increased CQ pressure as a result of DNA amplification whereas removal of CQ selection resulted in the return of chromosome 3 to its original size 72. Single nucleotide polymorphisms within pfmrp1 have been associated with anti-malarial resistance in field isolates 5632. Nevertheless, mutations within pbmrp may also contribute to the CQ drug resistance phenotype.

The pbmrp locus was mapped to chromosome 13/14 in the drug sensitive N clone. Similarly, synteny analysis between the P. falciparum and the P. y. yoelii genome indicate that pymrp is located in chromosome 14 73. However, a striking finding in our study was that unlike N clone, the CQ selected line RC exhibited a chromosomal translocation of this gene to chromosome 8 (Figure 6). In a CQ resistant P. falciparum isolates subjected to mefloquine selection and constant drug pressure, pfmdr1, a member of the ABCB transporter sub-family, was amplified resulting in an increase in size of chromosome 5 74. A partial chromosomal duplication of the region containing the mdr1 gene was described in a Plasmodium chabaudi line selected for a stable mefloquine-resistance, where a second copy was translocated from chromosome 12 to chromosome 4 75. Pyrimethamine selection of a P. chabaudi line resulted in the duplication of the DHFR gene and a rearrangement of chromosome 7 76. The detection of pbmrp solely in chromosome 8 in the RC line and the fact that this chromosome appears to be of similar size as its counterpart in the drug sensitive N clone support that our observation is a translocation event without gene duplication. Previous work mapped the pbggcs gene to chromosome 8 in the P. berghei lines studied 54. Co-transfection experiments in Leishmania with the genes for PGPA and GGCS demonstrated that they work synergistically to confer resistance to antimony 77.

Therefore, the relocation of the pbmrp gene to the same chromosome of a potential linked gene, pbggcs, is intriguing and deserves further examination. It remains to be established whether pbMRP transports anions such as GSH, and/or GSH S-conjugates and whether it confers resistance to any anti-malarials.

The P. berghei mrp gene was sequenced and compared to homologues of this gene in other Plasmodium species and organisms. Using bioinformatics the pbmrp gene was found to have the highly conserved ABC motifs and was classified as an ABCC/MRP2 type membrane transporter.

There were no differences in gene organization, copy number, or level of expression when comparing a drug sensitive and a CQ selected drug resistant P. berghei line. A chromosomal translocation of the pbmrp gene from chromosome 13/14 to chromosome 8 was observed when comparing the drug sensitive N clone and the CQ resistant line RC.

The fact that some of these proteins are involved in the drug resistance phenomenon in other species justifies further investigations on the potential contribution of ABC transporters to anti-malarial resistance in Plasmodium. The results of this study justify the examination of the role of the Plasmodium berghei multidrug resistance associated protein in Plasmodium detoxification pathways. Future research on membrane transport mechanisms and intracellular biochemical processes mediated by these transporters could result in the identification of novel drug targets or multidrug treatment strategies to combat malaria.

The authors declare that they have no competing interests.

MGP participated in the work for the biological assays, performed the bioinformatic analyses, and drafted the manuscript and figures. ACS contributed to the study design, participated in the work for the biological assays, performed the sequencing experiments, carried out some bioinformatic analyses, and helped draft the manuscript. ACS submitted part of this work in partial fulfillment of her doctoral dissertation. RGM designed, supervised, and carried out part of the bioinformatics analyses. RGM carried out the phylogenetic analysis. RGM also helped draft the manuscript and figures. AES contributed to the design of the study, was responsible for the supervision of the work, and helped draft the manuscript. All authors read and approved the final manuscript