Functional Cloning of the Miltefosine Transporter

The antitumor drug miltefosine (hexadecylphosphocholine, MIL) has recently been approved as the first oral agent for the treatment of visceral leishmaniasis. Little is known about the mechanisms of action and uptake of MIL in either parasites or tumor cell lines. We have cloned a putative MIL transporter (LdMT) by functional rescue, using a Leishmania donovani-resistant line defective in the inward-directed translocation of both MIL and glycerophospholipids. LdMT is a novel P-type ATPase belonging to the partially characterized aminophospholipid translocase subfamily. Resistant parasites transfected with LdMT regain their sensitivity to MIL and edelfosine and the ability to normally take up [14C]MIL and fluorescent-labeled glycerophospholipids. Moreover, LdMT localizes to the plasma membrane, and its overexpression in Leishmania tarentolae, a species non-sensitive to MIL, significantly increases the uptake of [14C]MIL, strongly suggesting that this protein behaves as a true translocase. Finally, both LdMT-resistant alleles encompass single but distinct point mutations, each of which impairs transport function, explaining the resistant phenotype. These results demonstrate biochemically and genetically the direct involvement of LdMT in MIL and phospholipids translocation in Leishmania and describe for the first time a P-type ATPase involved in MIL uptake and potency in eukaryotic cells.

Leishmania donovani is a protozoan parasite that causes visceral leishmaniasis, an endemic disease of tropical and subtropical regions that is fatal if untreated (1). There are an estimated 500,000 cases per year in developing countries, half of them occurring in India (2). First-line treatment is based on pentavalent antimonials, but this is compromised by drug toxicity, high costs, and drug resistance (3). The growing incidence of visceral leishmaniasis among AIDS patients in the Mediterranean area emphasizes the need for optimal therapeutic man-agement (4). The use of miltefosine (MIL) 1 as the first effective oral drug against visceral leishmaniasis (5) and its recent approval for clinical use in India has provided some hope that we may now have an effective treatment (6). It is expected that MIL will play a key role in the control and possible eradication of visceral leishmaniasis (2), an Indian priority for the year 2010 (6). Thus, it is of particular importance to understand the mechanisms underlying the development of MIL resistance in Leishmania before its wide clinical use.
MIL was first developed as an antitumoral drug, together with the alkyl-glycerophospholipid edelfosine and other related drugs. A direct correlation between drug uptake and cellular sensitivity has been found in different eukaryotic cells (7)(8)(9)(10), but the mode of uptake of MIL and alkyl-glycerophospholipids is not well understood, and may vary depending on the cell type. We have described previously the biochemical characterization of the L. donovani MIL resistant line M-40 R. A 40-fold decrease in drug uptake was shown to be responsible for the resistant phenotype (10). The inability to normally take up MIL in the resistant line was the direct result of a defect in the inward translocation of the drug (the movement from the outer to the inner leaflet of the plasma membrane), because binding to the membrane, outward movement (efflux), drug metabolism, and bulk-phase endocytosis remained unaltered as compared with wild-type parasites. This resistant line is also defective in the inward-directed translocation of 7-nitrobenz-2oxa-1,3-diazol-4-yl)amino (NBD)-glycerophospholipid analogs across the plasma membrane, suggesting that MIL uptake and lipid translocation may share the same transporter (10). Reduced drug uptake has emerged as a common characteristic of drug-resistant trypanosomes, both in vitro and in vivo, enabling the molecular identification of drug transport systems (11,12). This paper reports the cloning of a new Leishmania P-type ATPase gene (LdMT, from Leishmania donovani putative Miltefosine Transporter) by functional rescue of the MIL-resistant line. LdMT belongs to the aminophospholipid translocase (APT) subfamily and localizes to the plasma membrane. LdMT mediates MIL and NBD-glycerophospholipids translocation across the plasma membrane in Leishmania parasites. New functional insights into LdMT activity strongly suggest that certain APTs behave as true phospholipid translocases. Further, we establish that single but distinct point mutations within the two alleles of the LdMT locus are responsible for the resistant phenotype through inactivation of the protein. These results provide the first genetic and biochemical evidence of a protein directly involved in MIL uptake in eukaryotic cells and point to LdMT as a gene possibly responsible for treatment failure in the future.
Parasite Cell Culture and Cytotoxicity Assays-The wild-type (MHOM/ET/67/HU3) and the MIL resistant M-40 R (13) lines of L. donovani were cultured in M-199 medium (Invitrogen) supplemented with 40 mM HEPES (Sigma), 100 M adenosine (Sigma), hemin (0.2% of a 250 g/ml stock solution) (Sigma) and 10% heat-inactivated fetal bovine serum (Invitrogen) at 28°C. The Leishmania tarentolae wildtype line TarII described previously (14) was cultured in the same medium. For determination of parasite sensitivity to MIL or edelfosine, 1 ϫ 10 6 cells were incubated for 48 h at different drug concentrations before determining cell proliferation as described previously (15).
Transfection and Screening for MIL Sensitivity-To isolate cosmids containing the MIL transporter gene, eight independent transfections were performed as described previously (16,17). M-40 R stationary promastigotes were transfected with 10 g of DNA from a cosmid library of genomic DNA from the L. donovani infantum strain LEM1317 cloned into the shuttle vector cLHYG (16,18). One day after transfection, the medium was supplemented with 40 g/ml of hygromycin B, and 24 h later parasites were plated onto 1% agar plates containing culture medium plus 40 g/ml hygromycin B. After 10-days incubation at 28°C, isolated colonies were picked and transferred into 96-well microtiter plates containing culture medium plus 80 g/ml hygromycin B. Aliquots from each well were inoculated into replica plates containing culture medium plus 40 M MIL. Clones susceptible to 40 M MIL were expanded from the master microtiter plate, and the cosmids were rescued from the transfectants by alkaline lysis. Approximately 8,000 transfectants were screened, yielding one clone (19D10) that sensitized Leishmania M-40 R to MIL.
Isolation of LdMT, DNA Sequencing Analysis, and Construction of Plasmids-To localize LdMT within the 19D10 cosmid, restriction fragments were subcloned into the Leishmania expression shuttle vectors pX or cLHYG (16,19). The ϳ7-kb BglII fragment ( Fig. 1) containing the entire LdMT open reading frame (ORF) was completely sequenced on an ABI Prism 3100 DNA sequencer. For cloning and sequencing of the different LdMT alleles from the wild-type and M-40 R lines, PCR amplification was accomplished with two different sets of primers using the high fidelity Triple Master polymerase from Eppendorf. P1 (5Ј-GTAGATCTAAGAGCTGCGCAATTCATG) and P2 (5Ј-CATCTAGAC-CGTCATACAATCTTCACAGC) amplified only the coding region, whereas P3 (5Ј-GTAGATCTGTGCGTACGACACTTTG) and P4 (5Ј-GT-TCTAGATTGTGCACTTCGACGATTCG) amplified the entire ORF, flanked by 714 and 685 bp of the untranslated 5Ј and 3Ј regions, respectively. BglII and XbaI restriction sites were added (underlined in the sequence) for further cloning. PCR-amplified bands were subcloned into the pGEM-T vector (Promega) for sequencing. The selected alleles were also subcloned into the BamHI-XbaI sites of the pX expression vector.
To generate green fluorescent protein (GFP) fusions at the carboxyl terminus of LdMT (LdMT-GFP), the ORF of the wild-type LdMT gene without the STOP codon was amplified by PCR by using forward and reverse primers containing BglII and SmaI sites, respectively. After restriction digestion, the ORF was subcloned in frame into the BamHI-EcoRV sites of the Leishmania expression vector pXG-GFPϩЈ (20).
Inmunoblotting and Fluorescence Localization-Total cell lysates were prepared from wild-type and M-40 R parasites transfected with the LdMT-GFP construction and selected with increasing concentrations of G-418 as described (21). Immunoblots of comparable amounts of proteins were preformed with a polyclonal anti-GFP antibody (1:5000) (Molecular Probes) and a horseradish peroxidase-conjugated secondary goat anti-rabbit (1:5000) IgG (Sigma), as described previously (21).
For localization of the LdMT-GFP chimera, parasites were pelleted, washed three times in phosphate-buffered saline, attached to poly-Llysine-coated coverslips, and fixed with 3% (w/v) paraformaldehyde for 20 min. Coverslips were rinsed with phosphate-buffered saline and mounted on slides with Vectashield (Vecta Laboratoires Inc.). Images were acquired with a TCS-SP confocal microscope (Leica) and processed with the Leica Confocal Software and Adobe Photoshop.
Southern Analysis and Reverse Transcriptase-PCR Analysis-Genomic DNA was purified from wild-type and M-40 R lines with the DNAzol reagent (Invitrogen). Restriction enzyme-digested DNA was hybridized to the LdMT ORF following standard procedures (22). Total cell RNA was isolated with TRIzol reagent (Invitrogen). Subsequently, 5 g of total RNA was used to synthesize the first-strand cDNA with the SuperScript reverse transcriptase kit (Invitrogen) and the LdMT-specific primer P2 described above, following the manufacturer's indications. PCR amplification of an internal LdMT sequence was performed by using primers P5 (5Ј-GTTACGCAACACGGACTG) and P6 (5Ј-CTA-ACCAGAATGCAGCTG).
Functional Experiments-L. tarentolae TarII cells and L. donovani wild-type and M-40 R cells were transfected with either pX-LdMT (a construct containing P1-P2 amplified wild-type genomic DNA in the pX vector), pX-T420N (containing the P1-P2 T420N allele from M-40 R genomic DNA), pX-L856P (containing the L856P allele from M-40 R genomic DNA), or pX vector alone. The resulting transfectants were selected in medium containing G418 until a concentration of 200 g/ml was reached. At that point, the internalization of [ 14 C]MIL (1.33 MBq/ mmol) or fluorescent-labeled phospholipid analogs was measured as described previously (10). Briefly, 2 ϫ 10 7 promastigotes in culture medium were incubated with 0.09 Ci/ml [ 14 C]MIL (2.5 M) for 60 min at 28°C. After washing with phosphate-buffered saline containing bovine serum albumin to allow for the removal of the drug fraction bound to the outer leaflet of the plasma membrane, followed by a second phosphate-buffered saline wash, both protein concentration and counts per minute were determined. With respect to NBD-phospholipid accumulation, 2 ϫ 10 6 /ml cells were preincubated with 100 M phenylmethylsulfonyl fluoride before the direct addition of NBD analogs from an ethanol stock solution. After 30 min at 28°C, parasites were spun down, washed as described above (including back-exchange with bovine serum albumin), and the intracellular fluorescence of 10,000 gated cells was acquired using a FACScan flow cytometer (BD Biosciences).
Screening for Miltefosine Resistance in Mutagenized Parasites-1 ϫ 10 8 L. donovani wild-type promastigotes were treated with 1.2 g/ml of ethylmethane sulfonate (Sigma) during 4 h in culture medium at 28°C. The cells were then pelleted and washed twice in culture medium. This treatment resulted in less than 5% killing. Mutagenized cells were recovered for 24 h before being plated in semi-solid medium (1 ϫ 10 7 per plate) containing 50 M MIL. Resistant isolated colonies were picked after 6 days of incubation.

Cloning of the Putative MIL Transporter by Functional
Rescue of the MIL Sensitivity Phenotype-We isolated a putative MIL transporter by functional complementation in the M-40 R line using a cosmid genomic library from L. donovani infantum (18). A number of cosmids were isolated and, upon retransfection, one cosmid (19D10) was able to bestow MIL sensitivity upon the M-40 R line. A restriction map of the 19D10 cosmid is shown in Fig. 1. To pinpoint the ORF within the cosmid that resensitizes cells to MIL, restriction fragments of cosmid 19D10 were subcloned into the Leishmania expression shuttle vectors pX or cLHYG, transfected back into the M-40 R line, and tested for restoration of MIL susceptibility. The smallest fragment still sensitizing resistant cells was a ϳ7-kb BglII fragment. Its sequence revealed an ORF of 1097 amino acids and was given the name LdMT. A PCR product containing only the LdMT ORF was also able to rescue the MIL-sensitive phenotype in the M-40 R line ( Fig. 2A).
Deduced Amino Acid Sequence of LdMT-Sequence analysis of LdMT revealed 28% amino acid identity and 47% homology with the human ATPase II (23) and other members of the P-type ATPases APT subfamily, such as the yeast Drs2p (23% identity). LdMT is 96% identical to the Leishmania major putative APT from chromosome 13. LdMT shares all the P-type ATPase consensus sequences and the APT subfamily-specific motifs from both the yeast (24) and the mammalian homologues (25) (Fig. 3). It also contains 10 easily predictable hydrophobic transmembrane segments (TM) and a large cytosolic loop after the fourth TM containing the DKTGTLT phosphorylation motif (Fig. 3).
Functional Characterization of LdMT-To establish the functional role of LdMT, the pX-LdMT construct was transfected into M-40 R L. donovani cells. We have established previously that NBD-glycerophospholipids and MIL are internalized in Leishmania parasites mainly by rapid transbilayer movement, a protein-mediated and energy-dependent process, independent of endocytosis (10,28). Uptake assays using radiolabeled MIL or fluorescent-labeled NBD-phospholipid analogs demonstrated that LdMT was able to rescue inward translocation and thus to increase uptake of MIL (78-fold) (Fig. 4A), NBD-PC (35-fold), NBD-PE (14-fold), and NBD-PS (12-fold), but not NBD-SM in the M-40 R line (Fig. 4B). It also restored the normal sensitivity to MIL and edelfosine (Fig. 2). These results indicate that LdMT participates in the inward translocation machinery, which distinguishes sphingolipids from glycerolipids and recognizes only the latter as substrates for translocation across the plasma membrane, together with the non-glycerol based alkylphosphocholine MIL. This feature makes LdMT the phospholipid translocase with the broadest substrate specificity so far studied.
One may still wonder whether LdMT is really a true translocase able to transport lipid analogs or just another member of the translocation machinery required for this activity. To gain further insight into this issue, we transfected wild-type L. donovani parasites with LdMT. Overexpression of LdMT increased the internalization of these phospholipid analogs (except NBD-SM) 2-5-fold as compared with control cells transfected with the empty vector (Fig. 4). A concomitant 2-3-fold increase was seen in sensitivity to MIL and edelfosine (Fig. 2). The increase in phospholipid uptake, both in wild-type and M-40 R parasites, correlated with the G418 selective pressure that controls plasmid copy number and thus the level of LdMT overexpression, as shown by Western blotting with a functional LdMT-GFP chimera (Fig. 5 and data not shown). These data further suggest that LdMT is not only involved in the translocation machinery, but that it is likely to be the direct MIL and glycerophospholipid transporter.
If LdMT were a true translocase, it should confer its biochemical activity to any other cell type lacking this function, as far as the system contains the elements required for proper functioning of the pump. The lizard parasite Leishmania tarentolae shows a natural resistance to alkylphosphocholines, with sensitivities one order of magnitude higher than those of the human parasites L. donovani, Leishmania tropica, Leishmania mexicana, or Leishmania panamensis (personal observations). We then tested MIL accumulation in L. tarentolae and found that this was 16-fold lower than that of L. donovani wild-type cells (Fig. 4A). Therefore, L. tarentolae parasites constitute an optimal system to study LdMT function. Expression of LdMT into L. tarentolae increased MIL uptake 20-fold (Fig. 4A), with the concomitant gain of MIL sensitivity (data not shown). These results demonstrate that LdMT, at least inside the Leishmania genetic background, behaves as a true translocase of lipid analogs, being the key factor modulating MIL uptake and potency.
Cellular Localization of LdMT-GFP-To localize LdMT inside the cells, we constructed LdMT versions with a GFP tag at the carboxyl terminus. Tagging did not interfere with LdMT function, because M-40 R parasites expressing the chimera were able to rescue the transport-deficient phenotype (Fig. 5). Subcellular location of LdMT-GFP was monitored by using GFP fluorescence. As shown in Fig. 6, a significant portion of LdMT-GFP was localized to the plasma membrane but excluded from the flagellum. There was some labeling of the flagellar pocket area, which may reflect a secondary location or just protein trafficking.
Analysis of the LdMT Locus, Transcripts, and Allele Sequences in the M-40 R Line-Genomic Southern blots (Fig. 7A) probed with the LdMT ORF revealed hybridizing fragments consistent with the map given for the cosmid in Fig. 1, indicating that LdMT is likely a single copy gene. The signal intensity and the restriction pattern in wild-type and M-40 R blots were similar, demonstrating that neither deletions nor rearrangement of the LdMT locus were responsible for the translocationdeficient phenotype on M-40 R cells. Reverse transcriptase-PCR experiments showed unique ethidium bromide-stained bands of the same intensity and expected size in both cell lines (Fig. 7B), indicating that loss of LdMT function in M-40 R cells is not caused by an alteration in RNA levels. To determine whether mutations within LdMT were responsible for the translocation-deficient and MIL-resistant phenotype, LdMT from the M-40 R cell line was cloned and sequenced, after its amplification from genomic DNA by PCR. M-40 R cells were compound heterozygotes at the LdMT locus containing two mutant alleles that encompassed single but distinct point mutations. One of the two LdMT-resistant alleles contained a Cys-1259 3 A mutation, which resulted in a T420N substitution inside the conserved, specific P-type phosphorylation motif DKT 420 GTLT (Fig. 3). The other allele contained a different single point mutation, Thr-2567 3 Cys, resulting in an L856P substitution in the large cytosolic loop before TM 5 (Fig. 3). These point mutations were confirmed by two additional independent PCRs from genomic DNA. Further, the presence of two independent alleles was confirmed by polymorphism analysis in the 5ЈUTR region of the gene (data not shown). M-40 R cells transfected with either the T420N or the L856P mutant alleles showed greatly reduced MIL uptake capabilities, similar to those observed in the M-40 R line, as opposed to the wild-type  (26). Asterisks indicate amino acid identity. Location of transmembrane segments (TM) as defined in human ATPase II (27) are highlighted from TM1 to TM10. Consensus sequences characteristics of P-type ATPases are boxed in black. Aminophospholipid translocase subfamily specific motifs are boxed in gray (24,25). The position of Thr-420 and Leu-856 in LdMT are marked by the lower arrowheads.
protein which is able to increase MIL uptake ϳ80-fold on the M-40 R line (Fig. 4A and data not shown). Moreover, the mutant alleles were unable to rescue the defective translocation of glycerophospholipids in the M-40 R line (data not shown). These latter results confirm that both independent mutations are sufficient to inactivate the putative transporter and that the molecular basis for the transport-deficient and drug-resistant phenotype consists of the acquisition of single but distinct point mutations within each of the two LdMT alleles present in the M-40 R line.
Phenotypic and Genotypic Analysis of Mutagenized Resistant Clones-After mild mutagenesis treatment, resistant clones were rapidly selected in semisolid medium. Six clones were selected for further analysis, and all of them were found to be defective in the internalization of glycerophospholipids and MIL (data not shown). Transfection with pX-LdMT was able to rescue this defect in all of the clones, suggesting that mutagenized clones bear the loss of LdMT function. Sequence analysis of one of those clones identified two different alleles, encompassing three (F414S ϩ F430S ϩ G824D) and four (L366P ϩ L780P ϩ G824D ϩ F987L) point mutations each. M-40 R parasites transfected with these alleles also failed to recover the lipid transport defect, further demonstrating the genetic basis for this MIL resistance phenotype.
These experiments point to LdMT as the protein inactivated in Leishmania during the generation of MIL resistance in vitro, either by stepwise drug pressure selection (M-40 R) or after mutagenesis treatment, suggesting that LdMT is likely to suffer high selective pressure in the field. translocation across the plasma membrane of both MIL and glycerophospholipids, independently of their polar head group (10). The present work identifies LdMT as a new member of the P-type ATPase APT subfamily expressed at the plasma membrane. LdMT mediates the translocation of MIL and glycerophospholipids in Leishmania parasites, thus modulating the uptake and potency of alkylphosphocholine drugs. It also shows that the molecular basis for the drug-resistant phenotype consists of the acquisition of single but distinct point mutations within each of the two LdMT alleles present in the M-40 R line, leading to a loss of functionality.
LdMT is clearly a member of the third subfamily of P-type ATPases, enzymes proposed to function as APTs (29) and currently represented by over a dozen genes in the human genome (25), 5 genes in Saccharomyces cerevisiae (24), 12 genes in Arabidopsis (30), and many others identified from protozoa to mammals. However, the biological functions remain elusive for the majority of them. ATPase II has been proposed as the protein likely responsible for the translocation of aminophospholipids in mammalian cromaffin granules (27,31). The five yeast members have been characterized recently (32,33), and Dnf1p and Dnf2p were found to be involved in phospholipid translocation and maintenance of the plasma membrane asymmetry (33).
All mammalian APT activities described so far have shown specificity for aminophospholipids, phosphatidylserine being a better substrate than phosphatidylethanolamine (29). Thus, a striking difference between previously described mammalian APT and the LdMT-dependent translocation activity is that the latter has an increased range of substrates. Indeed, LdMT is able to promote the translocation of not only aminophospholipids but also glycerophosphocholine molecules such as NBD-PC or edelfosine and the even more simple, non-glycerol based, alkylphosphocholine MIL. Thus, assuming a role for these proteins as the direct translocases and despite sequence homology (Fig. 3), the mammalian and parasitic APTs differ in their substrate specificity, which may explain in part the exquisite activity of MIL against Leishmania. Nevertheless, certain mammalian cell types and tumoral lines internalize glycerophosphocholine analogs and/or alkylphosphocholines (7,9,34,35). It is tempting to suggest that some other members of the P-type ATPase APT subfamily differentially expressed in these cell types may be responsible for the phosphatidylcholine/MIL accumulation phenotype and thus be implicated in MIL potency in tumor cells. It will be challenging to identify those P-type ATPases and the pattern of tissue distribution in human beings.
Interestingly, the recently described Drs2p homologues, Dnf1p and Dnf2p, also localize to the plasma membrane and have been shown to be required for the rapid internalization of not only aminophospholipids but NBD-PC as well (33). Considering the amino acid identity between Dnf1p/Dnf2p and LdMT (21 and 20%, respectively), their cellular localization, and the phenotype exhibited when they are either deleted (33) or mutated (this work), it is possible to suggest that they are homologous proteins with similar functions. Moreover, the presence of some other distantly related P-type ATPases having a redundant function in the parasite, as occurs with Dnf1p and Dnf2p in yeast, cannot be ruled out. If this were the case, one would expect those proteins to be also inactivated in the M-40 R line. Another yeast protein, ROS3p/Lem3p, is required for the uptake of alkylphosphocholines NBD-PC and -PE but not -PS (36,37). Considering the requirement of Dnf1p/Dnf2p in yeast or LdMT in Leishmania parasites for the uptake of all glycerophospholipids and alkylphosphocholines, ROS3p must be a protein somehow modulating the proper function of these P-type ATPases, but never a translocase.
Final proof of P-type APTs acting as direct phospholipid pumps will require reconstitution of purified proteins into chemically defined liposomes, an issue that we are currently approaching. Nevertheless, we propose that LdMT behaves as a true lipid translocase. Apart from localizing to the plasma membrane and mediating the translocation of glycerophospholipids and alkyl-phosphocholine in L. donovani, LdMT expression in L. tarentolae, a species naturally resistant to MIL, shifts cells sensitive to MIL and able to accumulate the drug in levels similar to those of other sensitive Leishmania species (Fig. 4A). This result constitutes the strongest evidence to date of a member of the P-type APT subfamily behaving as a real transporter.
The absence of detectable deletions or rearrangements at the LdMT locus in M-40 R cells and the presence of similar levels of LdMT mRNA in both mutant and wild-type lines suggested the presence of stable mutations of the gene in the M-40 R line leading to the resistance phenotype. The present PCR-based approach identified single but distinct point mutations in both alleles that are able to inactivate LdMT function. Thus, the gain of missense point mutations, and not loss of heterozygosity, had to occur during drug pressure selection. The T420N mutation is located within the phosphorylation motif DKT 420 GTLT. All P-type ATPases form an aspartyl-phosphate intermediate, the Asp residue being essential for their functionality (38). Interestingly, site-directed mutagenesis in the corresponding Thr residue of the Ca 2ϩ -ATPase of sarcoplasmic reticulum (another P-type ATPase) has shown it to be involved in the catalysis of ATP and the coordination of catalytic Mg 2ϩ (39). Thus, the Thr-420 position seems to be essential for LdMT activity, further highlighting the central role of this Thr residue in the phosphorylation site of P-type ATPases. The L856P substitution within the second LdMT allele is a position only partially conserved in other putative APTs. Considering its proximity to TM 5, misfolding of the protein could not be discarded. These inactivating point mutations highlight the convenience of this approach as a genetic tool for identifying residues within LdMT (both inside or outside conserved motifs) that are critical for its function and, ultimately, for identifying the residues responsible for substrate specificity.
To answer whether the MIL resistance phenotype found in the M-40 R line could be generalized to other resistant cases, we created Leishmania resistant lines by mutagenesis. These lines showed not only a similar phenotype, but also a similar genetic basis for the acquisition of resistance: inactivating point mutations within LdMT. This result further validates the previous findings, strongly suggesting that the generation of MIL-resistant clinical isolates bearing specific point mutations in LdMT is likely to occur.
In conclusion, we have found that the novel P-type ATPase LdMT mediates the translocation of MIL and glycerophospholipid analogs in Leishmania. The new functional insights into LdMT shown here strongly suggest that this subfamily of proteins behaves as true phospholipid translocases. Loss of function point mutations within LdMT confer MIL resistance in vitro, which may have a bearing on the future of MIL as a successful drug against leishmaniasis. Knowledge of the molecular basis of MIL resistance in vitro could anticipate strategies to prevent its appearance in vivo and to design second line drugs for drug-resistant cases.