In vitro selection of miltefosine resistance in promastigotes of Leishmania donovani from Nepal: genomic and metabolomic characterization

Abstract

In this study, we followed the genomic, lipidomic and metabolomic changes associated with the selection of miltefosine (MIL) resistance in two clinically derived Leishmania donovani strains with different inherent resistance to antimonial drugs (antimony sensitive strain Sb-S; and antimony resistant Sb-R). MIL-R was easily induced in both strains using the promastigote-stage, but a significant increase in MIL-R in the intracellular amastigote compared to the corresponding wild-type did not occur until promastigotes had adapted to 12.2 μM MIL. A variety of common and strain-specific genetic changes were discovered in MIL-adapted parasites, including deletions at the LdMT transporter gene, single-base mutations and changes in somy. The most obvious lipid changes in MIL-R promastigotes occurred to phosphatidylcholines and lysophosphatidylcholines and results indicate that the Kennedy pathway is involved in MIL resistance. The inherent Sb resistance of the parasite had an impact on the changes that occurred in MIL-R parasites, with more genetic changes occurring in Sb-R compared with Sb-S parasites. Initial interpretation of the changes identified in this study does not support synergies with Sb-R in the mechanisms of MIL resistance, though this requires an enhanced understanding of the parasite's biochemical pathways and how they are genetically regulated to be verified fully.

Figure 1

The effect of drug treatment on the survival of promastigotes (A, B) or intracellular amastigotes (C, D) of L. donovani parasites treated with MIL (A, B, C) or sodium stibogluconate (SSG, D). Promastigotes of the Sb‐S or Sb‐S MIL‐R clone 8 strain were grown in the presence of medium alone (controls) or different concentrations of MIL (n = 6/treatment). Cytotoxicity was assessed by determining the mean suppression of drug treated samples compared to the relevant control (A, B). In amastigotes studies, macrophages were infected with the relevant parasite (C, Sb‐R WT or Sb‐R MIL‐R clone 9, D; Sb‐S WT or Sb‐S MIL‐R clone 8) and then incubated with medium alone (controls) or medium containing MIL (C) or SSG (D, n = 4/treatment). After 72 hours, the percentage of cells infected was assessed and the mean suppression in parasite numbers compared to the relevant control. Data is representative of a minimum of two experiments. *P < 0.05, ***P < 0.001 compared with relevant control.

Figure 2

The genetic mechanisms responsible for reducing the amount of LdMT transporter protein in parasites. A. Somy changes observed for Sb‐R WT (top) and Sb‐S WT (bottom) during MIL exposure (x‐axis) ranging from 0 (blue) to 74 µM (shades of red). The WT (0 uM) and fully resistant stages (74 µM) were completed with one and five replicates for Sb‐R, (respectively), and with three replicates for Sb‐S. For Sb‐R, a resistant sample was passaged with (74 + 74) and without the drug (74 + 0, with one replicate). For Sb‐S, a resistant sample was passaged without the drug too (74 + 0). The blue‐green‐beige shading indicates the ploidy state: 1 is disomic, 1.5 is trisomic, and 2 is tetrasomic. Chromosome 13 was downregulated in both strains, Chr33 was only in the Sb‐S one. B. The routes to reducing LdMT dosage for Sb‐S WT during MIL exposure (x‐axis) ranging from 0 to 74 µM: an initial decrease in dose though aneuploidy at 3 µM, a deletion also at 3 µM, and then A691P at 6 µM. The y‐axis indicates the alleles’ copy number within the Sb‐S population assuming one LdMT copy represents the expected disomic state. Black indicates the WT state (A691). At 3 µM MIL, the chromosome copy number decreased from 2.80 to 0.93, and a deletion occurred encompassing LdMT that continued to increase in frequency until 35 µM. At 12 µM MIL, P691 occurred and rose in level until the final step at 61 µM. C. Localisation of the LdMT deletion in the Sb‐R MIL‐R strain; upper part, sequencing read depth over the region; lower part, physical map of the LdMT locus: coordinates were given in a draft reference based on Pacbio sequencing, as this region contains gaps in the current reference and was not properly assembled. The two black arrows indicate the localization of the two 444 bp direct repeats at the deletion boundaries.

Figure 3

A comparison of MIL uptake in MIL‐R (Sb‐S MIL‐R clone 8, Sb‐R MIL‐R clone 9) and their corresponding WT strains (Sb‐S and Sb‐R) of L. donovani. Promastigotes were cultured in HOMEM medium with and without 7 µM MIL for 0‐120 min. Samples were quenched and lipid extraction carried out using promastigotes samples (4 × 10⁷, n = 4/treatment). The amount of MIL present (µg/ml) was determined using a calibration curve prepared using MIL standards.

Figure 4

Diagrammatic representation of the Kennedy pathway metabolites altered in the MIL‐R parasites of Sb‐S (A) and Sb‐R (B) compared with their corresponding WT. A blue circle denotes a significant down regulation (P < 0.05) of the metabolite in the MIL‐R strain compared to its WT. A red circle denotes a significant up regulation (P < 0.05) in the MIL‐R compared to WT. Key: a‐chol, acytelcholine; g‐3‐pchol, glycero‐3‐phosphocholine; chol, choline; CDP‐choline, cytidine diphosphate‐choline; PC, phosphatidylcholine; p‐d‐eth, phosphodimethylethanolamine; N‐meth‐eth‐p, N‐methylethanolamine phosphate; eth, ethanolamine; eth‐p, ethanolamine phosphate; CDP‐eth, cytidine diphosphate‐ethanolamine; PE, phosphodimethyethanolamine.