**6. Drug resistance**

**5. Miltefosine effects on lipid metabolism**

448 Leishmaniasis - Trends in Epidemiology, Diagnosis and Treatment

smaller glycosylphosphatidylinositols (GIPLs) [88].

in fungi and *Leishmania*, was not changed [89].

APLs, including miltefosine, have been found to interact with membrane lipids and affect lipid metabolism [75-77], these actions having been suggested being involved in their mechanism of action. In lipid monolayer studies, miltefosine molecules were inserted into the monolayer of lipids as monomers until the critical micellar concentration, and a high condensation was found between miltefosine and sterols showing a high affinity between miltefosine and sterols [78]. However, miltefosine did not act as detergent disturbing membrane integrity [78]. *Leishmania* parasites have high levels of ether-phospholipids [79-83], and these are mainly found in the glycosylphosphatidylinositol-anchored glycolipids and glycoproteins present on the surface of the parasites [84-86]. Because edelfosine and other APLs are ether lipids, it could be suggested that the biosynthesis of ether lipids occurring in the glycosomes of *Leishmania* might be affected. Miltefosine and edelfosine did not affect enzymes involved in early steps in ether lipid biosynthesis in *L. mexicana* promastigotes, including dihydroxyacetonephos‐ phate acyltransferase, *sn*-l-acyl-2-*lyso*-glycero-3-phosphocholine and *sn*-l-alkyl-2-*lyso*-glyc‐ ero-3-phosphocholine acyltransferases activities [87]. However, both miltefosine and edelfosine affected the later metabolism of alkyl-phosphatidylcholine intermediates by inhibiting the glycosomal located alkyl-specific-acyl-CoA acyltransferase in a dose-dependent manner with an inhibitory concentration of 50 µM, thus suggesting these drugs can perturb ether-lipid remodelling [87]. However, the fact that inhibition of alkyl-specific-acyl-CoA acyltransferase required drug concentrations higher than those showing cytotoxicity to *L. mexicana* (IC50, 14 µM and 18 µM for miltefosine and edelfosine, respectively) [87] challenges the putative involvement of this pathway as the primary target of these drugs. In addition, the role of glycosylphosphatidylinositols and ether phospholipids in the survival of *Leishmania* amastigotes is questioned by the viability of *L. major* null mutants for alkyldihydroxyaceto‐ nephosphate synthase (ADS), the first committed step of ether lipid synthesis. These mutants lacked all ether phospholipids, including plasmalogens, lipophosphoglycan (LPG), and

Treatment of *L. donovani* promastigotes with 10 µM miltefosine significantly reduced the phosphatidylcholine content and enhanced the phosphatidylethanolamine content in parasite membranes, suggesting a partial inactivation of phosphatidylethanolamine-N-methyltrans‐ ferase [89]. Phospholipase D activity was not affected by miltefosine, whereas the enhancement of the lysophosphatidylcholine content could be ascribed to phospholipase A2 activation. No effect was observed in the fatty acid alkyl chain length or the fatty acid unsaturation rate upon miltefosine treatment, whereas a two-fold increase was detected in the amount of cholesterol within the membranes [89]. Because cholesterol is not biosynthesized by the *Leishmania* parasite, but is taken from the external medium, it might be envisaged that miltefosine promotes cholesterol uptake in promastigotes perhaps by the condensation effect between miltefosine and cholesterol [78]. In contrast, a strong reduction of about two times in the C24 alkylated sterol content was detected in miltefosine-treated membranes, even though the level of the final C24 sterol alkylating product, ergosterol, the predominant plasma membrane sterol

A major hurdle in successful leishmanial chemotherapy is emergence of drug resistance. Miltefosine, the first orally administrable anti-leishmanial drug, has shown the potential against drug-resistant strains of *Leishmania*. However, at the same time the readiness in generating miltefosine-resistant parasites *in vitro* in the laboratory [27-29] raises some concerns about the life span for an efficient use of miltefosine. The major underlying mechanisms involved in the generation of miltefosine-resistant parasites seem to reside in defective drug uptake into the parasite and increased drug efflux. In addition, its long half-life (approximately 150 hours) [91] might facilitate the emergence of resistance.

*L. donovani* promastigotes resistant to up to 40 µM miltefosine have been generated *in vitro* in the laboratory by continuous step-wise drug pressure, resulting in being 15-fold more resistant to miltefosine than wild-type promastigotes [28]. A drastic reduction (>95%) in the ability of resistant *L. donovani* promastigotes to internalize [14C]miltefosine was detected, whereas binding of miltefosine to the plasma membrane and drug efflux from preloaded cells were similar in both drug-sensitive and -resistant cells, and no metabolism of [14C]miltefosine was evident in either sensitive and resistant cells [92]. This miltefosine uptake was temperature and energy dependent and sensitive to the thiol-reactive agent N-ethylmaleimide [92]. Resistant parasites were also unable to take up other short-chain phospholipid analogs, independently of their polar head group, even though endocytosis remained unaltered [92]. The underlying basis for the generation of drug-resistant *L. donovani* promastigotes turned out to be a decrease in the uptake of miltefosine mediated by the plasma membrane P-type transporter *L. donovani* miltefosine transporter (LdMT) from the aminophospholipid translo‐ case subfamily and by LdRos3 [93, 94]. LdMT is an inward-directed lipid translocase belonging to the P4 subfamily of P-type ATPases, which comprises lipid flippases that catalyze the translocation of phospholipids from the exoplasmic to the cytosolic leaflet of cell membranes; and LdRos3 is a non-catalytic subunit of LdMT that belongs to the CDC50/Lem3 family, which includes proteins proposed as likely beta subunits for P4-ATPases [93, 94]. LdMT and LdRos3 proteins are primarily localized to the *Leishmania* plasma membrane and required for the rapid intracellular uptake of miltefosine and additional related choline-bearing lipids. Likewise, in the budding yeast *S. cerevisiae*, members of the two protein families have been found to form stable transporter complexes that function in the translocation of phospholipids from the exoplasmic to the cytoplasmic leaflet of cellular membranes [95, 96]. Despite both LdMT and LdRos3 normally localize to the plasma membrane, they are retained at the endoplasmic reticulum in the absence of the other protein or in the presence of inactivating point mutations in LdMT [94]. Both LdMT and the Cdc50-like protein LdRos3 form a stable complex that plays an essential role in maintaining phospholipid asymmetry in the parasite plasma membrane [97], and constitute part of the same translocation machinery that determines flippase activity, responsible for miltefosine uptake, as well as miltefosine sensitivity in *Leishmania* [94]. Loss of either LdMT or LdRos3 blocks ATP-dependent transport of NBD-labeled phosphatidyletha‐ nolamine and phosphatidylcholine from the outer to the inner plasma membrane leaflet in *L. donovani* promastigotes and results in an increased cell surface exposure of endogenous phosphatidylethanolamine, whereas infectivity was not compromised [97]. A promastigote line, M-mutR, that shows mutations in LdMT, thus leading to defective miltefosine internali‐ zation, is infective to macrophages *in vitro* and in BALB/c mice *in vivo*, and displays a good correlation of *in vitro* resistance between promastigotes and intracellular amastigotes [29]. The fact that M-mutR parasites retain the resistant phenotype *in vivo* indicate that miltefosineresistant *L. donovani* promastigotes transform to miltefosine-resistant amastigotes [29]. It was also observed no cross-resistance to other antileishmanial drugs in M-mutR amastigotes [29]. Some clinical studies have suggested, as indicated above, that miltefosine shows significantly less efficiency against the cutaneous and mucocutaneous leishmaniasis caused by *L. brazilien‐ sis* parasites, mainly due to their inability to internalize the drug because of the low expression levels of the beta subunit LbRos3 [98]. Overexpression of LbRos3 induced increased miltefo‐ sine sensitivity in both *L. braziliensis* promastigotes and intracellular amastigotes, further supporting the notion that miltefosine uptake is a major event in determining miltefosine antileishmanial potency [98]. Miltefosine-resistant promastigotes, displaying cross-resistance to the ether lipid edelfosine, but not to the standard anti-leishmanial drugs, shows no ampli‐ fication of specific genes, including the multidrug resistance P-glycoprotein gene, and resistance has been found to be stable up to 12 weeks in drug-free culture medium [28].

The evidence gathered so far has shown that reduced miltefosine incorporation has always led to a resistant phenotype. This lower accumulation of miltefosine can be achieved by two processes: a) a decrease in drug uptake, rendered by inactivation of any one of the two proteins responsible for the miltefosine uptake, namely LdMT and its beta subunit LdRos3; b) an increase in drug efflux, mediated by the overexpression of the ABC transporter P-glycoprotein [27]. Thus, in addition to a flaw in the uptake of miltefosine as stated above, an increased efflux of miltefosine has also been implicated in miltefosine resistance. A multidrug resistance (MDR) *L. tropica* line overexpressing a P-glycoprotein-like transporter was found to display significant cross-resistance to the ALP miltefosine and edelfosine, with resistant indices of 9.2- and 7.1 fold, respectively [99]. This resistance was mediated through overexpression of an ABC transporter, namely the *Leishmania* P-glycoprotein-like transporter (*Leishmania* ABCB1 or LtrMDR1) [99, 100]. ATP-binding cassette (ABC) transporters constitute one of the largest and most conserved protein families and have been considered major players in drug resistance during the treatment of cancer and infectious diseases. Interestingly, sesquiterpene C-3 completely sensitizes MDR parasites to APLs, acting as an inhibitor of LtrMDR1 [99]. In addition, overexpression of two *Leishmania*-specific ABC subfamily G (ABCG)-like transport‐ ers localized at the plasma membrane of *Leishmania* protozoan parasites, LiABCG6 and LiABCG4 half-transporters, conferred resistance to APLs in *Leishmania* parasites and *S. cerevisiae* [101, 102]. Overexpression of LiABCG6 not only leads to miltefosine resistance *in vitro*, but also to the antileishmanial oral drug 8-aminoquinoline analog sitamaquine [102].

Additional genes associated with miltefosine resistance have been identified by generating *L. major* promastigote mutants highly resistant to miltefosine (80-100 µM) in a step-by-step manner and subsequent analysis of the short-read whole genome sequencing [103]. In addition to the previously described P-type ATPase involved in phospholipid translocation, another new gene coding for pyridoxal kinase, involved in the formation of pyridoxal-5'-phophate (active vitamin B6), has been implicated in miltefosine susceptibility [103]. Following this genetic approach, it was clear the polyclonal nature of a resistant population with varying susceptibilities and genotypes, indicating that miltefosine resistance can be genetically and phenotypically highly heterogeneous [103].
