Genomic and transcriptomic alterations in Leishmania donovani lines experimentally resistant to antileishmanial drugs

Abstract

Leishmaniasis is a serious medical issue in many countries around the World, but it remains largely neglected in terms of research investment for developing new control and treatment measures. No vaccines exist for human use, and the chemotherapeutic agents currently used are scanty. Furthermore, for some drugs, resistance and treatment failure are increasing to alarming levels. The aim of this work was to identify genomic and trancriptomic alterations associated with experimental resistance against the common drugs used against VL: trivalent antimony (Sb^III, S line), amphotericin B (AmB, A line), miltefosine (MIL, M line) and paromomycin (PMM, P line). A total of 1006 differentially expressed transcripts were identified in the S line, 379 in the A line, 146 in the M line, and 129 in the P line. Also, changes in ploidy of chromosomes and amplification/deletion of particular regions were observed in the resistant lines regarding the parental one. A series of genes were identified as possible drivers of the resistance phenotype and were validated in both promastigotes and amastigotes from Leishmania donovani, Leishmania infantum and Leishmania major species. Remarkably, a deletion of the gene LinJ.36.2510 (coding for 24-sterol methyltransferase, SMT) was found to be associated with AmB-resistance in the A line. In the P line, a dramatic overexpression of the transcripts LinJ.27.T1940 and LinJ.27.T1950 that results from a massive amplification of the collinear genes was suggested as one of the mechanisms of PMM resistance. This conclusion was reinforced after transfection experiments in which significant PMM-resistance was generated in WT parasites over-expressing either gene LinJ.27.1940 (coding for a D-lactate dehydrogenase-like protein, D-LDH) or gene LinJ.27.1950 (coding for an aminotransferase of branched-chain amino acids, BCAT). This work allowed to identify new drivers, like SMT, the deletion of which being associated with resistance to AmB, and the tandem D-LDH-BCAT, the amplification of which being related to PMM resistance.

Keywords: 24-Sterol methyltransferase; Aminotransferase of branched-chain amino acids; Amphotericin B; D-lactate dehydrogenase; Genome; Leishmania donovani; Miltefosine; Paromomycin; Transcriptome; Trivalent antimony.

Graphical abstract

Fig. 1

General analysis of differentially expressed transcripts in the amphotericin B (A), miltefosine (M), paromomycin (P) and antimonial (S) lines. A) Venn diagram showing transcripts down- and up-regulated more than 2-fold in the resistant lines regarding the expression levels found in the parental WT line. The Venn diagram was constructed using the online software Venn diagrams, available at Web site of the Bioinformatics and Evolutionary Genomics group of the University of Gent and the VIB institute (http://bioinformatics.psb.ugent.be/webtools/Venn/). B) Heatmap of a hierarchical clustering analysis based on the data from three individual replicates. Euclidean distances were calculated over rlog transformed counts using DESeq2 and plotted with pHeatmap R package (https://cran.r-project.org). C) Volcano plots build using DESeq2 results. Dots in red represent differentially expressed genes with abs(log₂(FC)) > 1 and adjusted p-value (Q value) < 0.01. To avoid calculation errors, we have assigned the minimum adjusted p-value higher than 0 to those transcripts with adjusted p-value equal to 0. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Chromosome copy number variation in L. donovani parental line (WT) and the S, A, M and P resistant lines. The somy estimation was done using a 2-loop method (Mondelaers et al., 2016). The median coverage of the genome is shown by a solid line, and it was assigned as 2, taking into account that diploid is considered the major ploidy status in Leishmania. The dotted lines indicated the estimated values for other somies. Graphs were generated from the median coverage values by barplot function of R package (https://cran.r-project.org).

Fig. 3

Gene expression analysis of PAD, SMT, D-LDH, and BCAT genes in different L. donovani lines. Total RNA of L. donovani WT line, S, A and P resistant lines, and WT + Lexsy, A + Lexsy, WT + PAD, A + SMT, WT + D-LDH, WT + BCAT, WT + D-LDH + BCAT and WT + BCAT + D-LDH transfected lines, was extracted from promastigotes grown at log-phase as described in Materials and Methods. RT-qPCR expression values of the genes in each line were normalized with the expression of GAPDH. The relative expression of each gene was calculated as the fold-change between the resistant lines and the WT line (which was set to 1.0) (A) or transfected lines with PAD, SMT or D-LDH, BCAT, and WT + Lexsy, WT + Lexsy + pIR1SAT or A + Lexsy (set to 1.0) (B), see Material and Methods for further details. Results shown are the means ± SD from two independent experiments. In all cases, significant differences versus the controls were determined using Student's t-test (p < 0.01).

Fig. 4

Read coverage along the chromosome 29 in WT and A, S and P resistant lines. The coverage values obtained with genomeCoverageBed (BEDTools) were smoothed by an in-house Perl script that calculates the mean value within overlapping 5-Kb windows. Graphs were generated from the coverage values by plotly utility of R package (https://cran.r-project.org). Panel A shows the coverage along the complete chromosome 29, whereas panel B shows the region comprised between the coordinates 322,447 and 390,000 of this chromosome. Green-colored boxes denote ORF (the gene names are indicated), and red-boxes correspond to repeated sequences of the SIDER2 family (Requena et al., 2017).

Fig. 5

Deletion of a SMT gene in the AmB resistant line. (A) Read depth distribution of Illumina reads from WT and A in the region 948,788–962,094 of chromosome 36. An in-house Perl script was used for representing the smooth coverage with a 200 nucleotide-window and a normalization by the total number of reads from each sample. In the bottom, a graph showing the ORF of the genes located in this region is included. The ORFs of genes LinJ.36.2510 and LinJ.36.2520 are identical and code for SMT. The location and sizes of transcripts LinJ.36.T2510 and LinJ.36.T2520 are indicated by arrows with rhomboid heads. The slim arrows indicate the position of the oligonucleotides used for PCR amplification of the region. (B) PCR amplification with oligonucleotides gSMT-Fw and gSMT-Rv using as template genomic DNA from WT strain or A line. The green arrow points to the expected amplicon in the WT line, whereas the red one indicates the amplicon observed in the A line. HindIII-digested DNA of Φ29 phage was used as molecular weight marker (M). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

Amplification in the P line of the genomic region containing the genes LinJ.27.1940 (D-LDH) and LinJ.27.1950 (BCAT). (A) The reads coverage along a region (coordinates 853,000–872,000) of the chromosome 27 in both WT and P lines is shown. The coverage values obtained with genomeCoverageBed (BEDTools) were smoothed by an in-house Perl script that calculates the mean value within overlapping 200-nucleotide windows. Note that a logarithmic scale is used to represent the coverage values. (B) Schematic drawing of the amplified genomic region, showing the position of genes LinJ.27.1940 (D-LDH) and LinJ.27.1950 (BCAT). The red boxes correspond to two identical sequences of 424-nucleotides in length. The dotted arrow at the bottom denotes the transcriptional direction of the genes. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 7

Stability of the resistance to PMM and levels of expression of D-LDH and BCAT in L. donovani P line. (A) The stability of resistance with respect to the initial EC₅₀ of P line (month 0) was determined at 2, 3, and 4 months after removal of the drug pressure, using an MTT-based assay, as described in Materials and Methods. The results shown are the mean ± SD from three independent experiments. Asterisks denotes statistically significant differences (p < 0.05; Wilcoxon–Mann–Whitney test) between the P line before and after drug removal. (B) Relative expression levels of D-LDH and BCAT transcripts was determined by RT-qPCR before and after 2, 3 and 4 months after removal of drug pressure. Total RNA from different lines was extracted as described in Materials and Methods. Also, RT-qPCR expression values of the genes were normalized with the expression level of GAPDH transcript. The expression levels of the transcripts in the P line was set at 100%.The results shown are the mean ± SD from three independent experiments. Significant differences versus the control (P line) were determined using Student's t-test (*p < 0.01; **p < 0.005; ***p < 0.001).

Fig. 8

Mitochondrial membrane potential in different Leishmania lines. The different L. donovani lines were left untreated (black columns) or exposed to PBS (white columns) for 2 h. Promastigotes were labeled for 15 min with 0.5 μM of Rh123 as described in Materials and Methods. Measurements by flow-cytometry are expressed as mean of arbitrary fluorescence units ± SD from three independent experiments. Asterisks denotes statistically significant differences (p < 0.05; Wilcoxon–Mann–Whitney test) between the control line (WT + Lexsy) and both the P line and the BCAT- transfected line (WT + BCAT).

Fig. 9

PBS-induced ROS generation in L. donovani lines. The L. donovani lines were left untreated (black columns) or exposed to PBS (white columns) for 2 h. The parasites were incubated with 4 μM H₂DCFDA (A) or 5 μM MitoSOX (B) for 30 min at 28 °C. The fluorescence intensity was determined by flow-cytometry analysis and expressed as arbitrary fluorescence units (a.u). The data are the mean ± SD values from three independent experiments. Significant differences (p < 0.05; Wilcoxon–Mann–Whitney test) when comparing the different untreated lines with the untreated control line (WT + Lexsy) are denoted by asterisks, whereas significant differences between untreated and PBS-treated parasites for each line are marked by an ‘a’.