Figure 1.
Models for gene amplification in Leishmania.
(A) HR between DRs can lead to circular amplification (i) or to tandem duplication of the locus by nonequal crossing at sister chromatids (ii). The amplification can be nonconservative, leading to the deletion of the region amplified, or conservative (e.g., within replication forks), without genomic deletion. C1 and C2 are primers used to detect novel junctions formed by recombination between DRs. Arrows Δ1 and Δ2 are primers used to demonstrate a locus deletion between the DRs. (B) Role of IRs in the formation of linear amplicons. One model (i) suggests that IRs cause hairpin formation during replication and trigger the replication fork to stall and dissociate [7],[21]. The annealing of the repeats is used to prime DNA synthesis, leading to locus duplication up to the telomeric end [28]. Other mechanisms described in yeast and/or mammalian cells can also explain duplication events: (b) a single-strand break (SSB) during replication close to IRs or (ii) a double-strand break (DSB) initiating a linear duplication. The black bars represent the DNA segments that are amplified. L1 and L2 are primers used to detect novel junctions formed by rearrangements at IRs. The arrows in boxes indicate repeated sequences.
Figure 2.
Genomic distribution of repeated sequences in the L. major genome.
(A) The number of pairs of repeat sequences (RSs) is indicated for each of the 36 L. major chromosomes. The repeats are part of distinct RAGs. The repeats belonging to the SIDER retroposon family are the most abundant and are represented by yellow (SIDER1) or blue bars (SIDER2). Other repeated sequences are represented by orange bars. (B) Repeats on chromosomes 6 and 23 of L. major. Arrows indicate the orientation and the approximate location of the repeats. Repeats of the same RAG for each chromosome are displayed with the same color.
Figure 3.
DNA amplification/deletion occurs naturally in Leishmania parasites.
(A) Chromosome 23 with its three large polycistronic units is shown. Each vertical bar represents a gene. Intergenic regions of chromosome 23 were scanned for repeated sequences, and 15 RAGs (194–208) theoretically leading to 95 amplicons were detected (Table S1). RAG 203 (and its 29 putative amplicons) is not represented because it corresponds to complex duplicated regions difficult to investigate by PCR. We tested for the presence of 48 amplicons by PCR. PCR fragments of the expected size provided evidence for 25 amplicons formed at the level of DRs (upper map) and 15 amplicons formed at the level of IRs (lower map). PCR products of the right size detected (filled circle and filled square) or not detected (open circle and open square) in L. major LV39 (circles) and Friedlin (squares), respectively. (B) Selected examples of the detection of PCR products compatible with DNA rearrangements mediated at the level of either DRs or IRs in L. major strains LV39 or Friedlin. (C) Detection of PCR products diagnostic of deletion of three loci at the level of DRs using primers equivalent to Δ1 and Δ2 of Figure 1A.
Figure 4.
Adaptive gene amplification upon drug selection.
(A) Schematic representation of the DHFR-TS locus and the circular amplicon generated by HR between DRs (black arrows) on chromosome 6 in L. major. Genomic DNA was extracted from WT and MTX stressed promastigotes. Amplicon detection in the unstressed WT population (right panel, lane 1), after two passages with MTX (at EC50, 0.2 µM) (right panel, lane 2), and after removing MTX pressure (lanes 3–7). Ca and Cb are the primers used for detecting the rearrangements. (B) Schematic representation of the PTR1 and MRPA locus and the linear amplicon with inverted duplications generated after annealing of IRs (black arrows) on chromosome 23 in L. infantum (left panel). Detection of the PTR1 amplicons by PCR using primers La and Lb in the unstressed WT 263 population (right panel, lane 1), after four passages with MTX (at EC50, 0.2 µM) (lanes 2–4), and after removing MTX pressure (lanes 5–8). Lane 9 is a no DNA template control. (C) Kinetics of the selection and loss of the DHFR circular amplicon containing cells in LV39 (filled circle) and PTR1 linear amplicon containing cells in L. infantum (filled square) under MTX pressure (continuous line) and after drug removal (dashed lines). Average of three biological independent experiments is shown. (D) Adaptive MRPA gene amplification upon SbIII selection. Semiquantitative PCR was performed with primers La and Lb. Genomic DNA was extracted from L. infantum WT (lane 2) cells stressed with SbIII at 160 µM (4×EC50) (lanes 3–5) and from cultures after drug removal (lanes 6–8). Lane 1 is a no DNA template control. Densitometric ratios of PCR band intensities are indicated at the bottom. One representative experiment out of four is shown. Amplification of the chromosomal GAPDH gene was used as a reference to normalize the amount of template DNA loaded.
Figure 5.
Early events of gene amplification in Leishmania and expansion of clones with amplified DNA upon selection.
L. major was either cultured in the presence of 0.5× MTX EC50 for eight continuous passages (P1–P8) or cultured for one passage at 0.5×EC50 and then shifted for four passages at 1× MTX EC50 (P41) (see Figure S4A for the selection scheme). (A) DNAs extracted from the population of parasites grown at 0.5×EC50 MTX for up to eight passages (P1–P8) and for four passages at 1×EC50 MTX (P41) were analyzed by Southern blot for the presence of DHFR-TS circles. The blot was hybridized to a probe corresponding to the first 1,000 bp of LinJ.06.0830 that discriminate the chromosomal loci at 10 kb from its amplified region at 4 kb (see Figure S4B). The blot was also hybridized with a probe covering the coding sequence of DHFR-TS (DHFR) and to a GAPDH probe to monitor the DNA loaded in each lane. Fold increases were normalized with the GAPDH signal. (B) DNAs extracted from the population of parasites at every 0.5×EC50 passages (P1–P8) were further analyzed by semiquantitative PCR for the presence of the new junction derived from HR between the direct repeats flanking the DHFR-TS locus (see Figure 4A). The number of PCR cycles for the DHFR-TS amplicon and the GAPDH control is indicated. For passages P6–P8, the PCR signal was saturated at cycle 35 and is thus not shown. (C) Parasites from five selected passages (boxed in A and B) were cloned on plates. For each condition, total DNA was extracted from 10 randomly selected colonies and tested for the presence of DHFR-TS amplification by PCR. The number of PCR cycles for the DHFR-TS amplicon and the GAPDH control is indicated on the left of each panel.
Figure 6.
CNV of a linear amplicon derived from chromosome 6 in the presence of fetal bovine serum.
(A) The small chromosomes of L. infantum were separated by pulse-field gel electrophoresis. Total DNA was isolated from cells cultured with FBS (lane 1) and from cells cultured in serum-free medium for 15 passages (lane 2), FBS was then added for 15 passages (lane 3), and removed again for 15 passages (lane 4). Lane M, marker of S. cerevisiae chromosomes. (B) Southern blot hybridized to a probe derived from LinJ06.1210. (C) PCR assay mapping the rearrangement leading to the linear amplicon shown in (D). The primers used are indicated by small arrows located between LinJ06.1150 and LinJ06.1160 and LinJ06.1240 and LinJ06.1250. A line indicates nonadjacent lanes that have been brought together for producing the figure. (D) Sequence coverage depth as determined by NGS of the linear amplicon. Below are the genomic organization of the subtelomeric region of chromosome 6 in L. infantum and the structure of the linear amplicon with its inverted duplication. Large arrows indicate the presence of repeated sequences where one of the rearrangements leading to extrachromosomal linear amplification occurred, whereas thin arrows indicate the location of PCR primers.
Figure 7.
Adaptive gene amplification in RAD51−/− parasites.
(A) Diagram of the RAD51 locus in L. infantum WT with the NEO and HYG disruption cassettes (left panel). Southern blot analysis (right panel) of WT (lane 1); RAD51/RAD51::NEO (lane 2) and RAD51::NEO/RAD51::HYG (lane 3) genomic DNAs digested with PvuII (P) and hybridized with the 5′ UTR RAD51 probe. (B and C) Circular amplicon selection in RAD51−/− parasites. L. infantum WT pSPαZEOα (1 and 4), RAD51−/− pSPαZEOα (2 and 5), and RAD51−/− pSPαZEOαRAD51 (3 and 6) were either cultured in the absence of drug (−) or in the presence of 0.2 µM MTX for five passages. Total DNA was extracted from cells, and semiquantitative PCRs were performed to detect DHFR-TS (B) or PTR1 (C) circular amplicons. Amplification of the chromosomal GAPDH gene was used as reference to normalize the amount of template DNA. The data shown are averages of three independent experiments (*p≤0.05; **p≤0.005; ***p≤0.0005, two-tailed Student's t test).
Figure 8.
MRPA locus amplification upon arsenite selection and the role of the RAD51 recombinase.
(A) The MRPA locus and the repeats allowing its amplifications are shown; DRs are A–A′, A–A″, B–B′, and C–C′ and IRs are D–D′. Primers 1–9 were used to determine the rearrangements that occurred in L. infantum WT cells (B), in RAD51−/− (C), and in RAD51 add-back revertant parasites (D) all selected for arsenite resistance (up to 250 µM = 5×EC50). The DNA of resistant clones was separated by pulse-field gel electrophoresis and hybridized to a MRPA probe. The rearrangement point for each amplicon was determined by PCR (Figure S7) and is indicated below for each lane. Lane 1, unselected population; lanes 2–10, independent arsenite-resistant clones.