Genome diversity ofLeishmania aethiopica

Abstract

Leishmania aethiopica is a zoonotic Old World parasite transmitted by Phlebotomine sand flies and causing cutaneous leishmaniasis in Ethiopia and Kenya. Despite a range of clinical manifestations and a high prevalence of treatment failure, L. aethiopica is the most neglected species of the Leishmania genus in terms of scientific attention. Here, we explored the genome diversity of L. aethiopica by analyzing the genomes of twenty isolates from Ethiopia. Phylogenomic analyses identified two strains as interspecific hybrids involving L. aethiopica as one parent and L. donovani and L. tropica respectively as the other parent. High levels of genome-wide heterozygosity suggest that these two hybrids are equivalent to F1 progeny that propagated mitotically since the initial hybridization event. Analyses of allelic read depths further revealed that the L. aethiopica - L. tropica hybrid was diploid and the L. aethiopica - L. donovani hybrid was triploid, as has been described for other interspecific Leishmania hybrids. When focusing on L. aethiopica, we show that this species is genetically highly diverse and consists of both asexually evolving strains and groups of recombining parasites. A remarkable observation is that some L. aethiopica strains showed an extensive loss of heterozygosity across large regions of the chromosomal genome, which likely arose from gene conversion/mitotic recombination. Hence, our prospection of L. aethiopica genomics revealed new insights into the genomic consequences of both meiotic and mitotic recombination in Leishmania.

Parasite genetic variation and systemic immune responses are not associated with different clinical presentations of cutaneous leishmaniasis caused by Leishmania aethiopica

“…We find limited genetic diversity across the set of isolates, although this is still very much more diverse than the largely clonal population responsible for most VL cases in the Indian subcontinent (44) but less so than in the L. braziliensis species complex (14). We find rapid decay in linkage disequilibrium within these isolates (as found across the species (17)). These isolates thus likely represent a single, interbreeding population of related L. aethiopica strains causing cutaneous leishmaniasis in Gayint: we find no evidence that other Leishmania species are involved in CL in Gayint as has been reported in Ethiopia (47) and elsewhere in Africa (48); Kenya (49) and in Yemen (50).…”

“…14, 16, 43, 44). The only previously available genomic data for L. aethiopica apart from the reference genome assembly (35) was from historical cryopreserved isolates (17), although microsatellite and RFLP data from small numbers of isolates has been published (45, 46). We find limited genetic diversity across the set of isolates, although this is still very much more diverse than the largely clonal population responsible for most VL cases in the Indian subcontinent (44) but less so than in the L. braziliensis species complex (14).…”

“…One exception is a recent genome-wide association study that found some evidence that clinical outcome is linked to parasite genotype in L. infantum (16). Research on L. aethiopica is particularly neglected, and the only genomic data available for this species is from 19 historical isolates and some hybrid forms collected from cryopreserved culture collections (17). For these kinds of samples accurate epidemiological or clinical data is typically not available, and there is often significant variation in sampling times and places which is likely to make detecting clinical associations much more challenging.…”

Evolutionary genomics of a zoonotic parasite across the Neotropical Realm

“…The clonal prevalences of the inferred populations were compared by means of a Chi-squared test using the stats R-package (68). Finally, we identified Loss-of-Heterozygosity (LOH) regions across the genome as regions in non-overlapping 10kb windows (69), for which the following parameters were used (70): minimum number of SNPs = 1; number of heterozygous SNPs = 0; minimum number of contiguous homozygous 10kb windows = 4; maximum number of 10kb gaps allowed within a LOH region = 1/3 of the windows; and maximum number of heterozygous SNPs allowed in a gap region = 2. Differences in the number and proportions of LOH regions among the inferred populations were tested by means of a Kruskal-Wallis test (stats R-package) (68) along with pairwise Dunn’s tests with BH corrected p-values (59).…”