Geographic Origin and Vertical Transmission of Leishmania infantum Parasites in Hunting Hounds, United States

Abstract

Vertical transmission of leishmaniasis is common but is difficult to study against the background of pervasive vector transmission. We present genomic data from dogs in the United States infected with Leishmania infantum parasites; these infections have persisted in the apparent absence of vector transmission. We demonstrate that these parasites were introduced from the Old World separately and more recently than L. infantum from South America. The parasite population shows unusual genetics consistent with a lack of meiosis: a high level of heterozygous sites shared across all isolates and no decrease in linkage with genomic distance between variants. Our data confirm that this parasite population has been evolving with little or no sexual reproduction. This demonstration of vertical transmission has profound implications for the population genetics of Leishmania parasites. When investigating transmission in complex natural settings, considering vertical transmission alongside vector transmission is vital.

Keywords: Leishmania infantum; United States; clonal evolution; genomics; hunting hounds; leishmaniasis; parasites; vertical transmission.

Figure 1

Neighbor-joining tree based on pairwise Nei distances demonstrating geographic origin of US hound Leishmania isolates. Phylogenies were reconstructed on the basis of whole-genome genotype calls of 83 parasite samples representing the dominant L. infantum zymodeme MON-1 from the United States, Europe, South America, and the Middle East, which were the samples most relevant in the context of the origin of the US samples (Appendix 2 Figure 2). The 3 righthand columns indicate population grouping using admixture with best fitting total number of groups (Appendix 2 Figure 1, panel A).

Figure 2

Geographic origin of US hound Leishmania isolates. A) Cladogram of the neighbor-joining tree from Figure 1 showing monophyletic groups for better visibility of evolutionary relationships of the US hound parasites. B) Cladogram of the maximum-likelihood phylogeny (Appendix 2 Figure 2, panel B). Cladograms were reconstructed on the basis of whole-genome genotype calls of 83 parasite samples representing the dominant L. infantum zymodeme MON-1 from the United States, Europe, South America, and the Middle East, which were the samples most relevant in the context of the origin of the US samples (Appendix 2 Figure 2). Numbers at internal nodes show bootstrap values.

Figure 3

Molecular clock estimates of the date of the most recent common ancestor of US hound Leishmania samples. Shaded densities are normal kernel densities for the bootstrap estimates from PATHd8 analysis and from posterior samples for strict clock and relaxed clock with uncorrelated gamma-distributed rates in BEAST version 1.10.4 (https://beast.community). These distributions in each case represent the estimated uncertainty in the divergence date of Leishmania infantum isolates from US hounds and from Europe. Vertical lines in the same colors are at the means of each distribution.

Figure 4

Number and density of segregating SNPs in each group of Leishmania infantum isolates by geographic region and type of host. Values are shown as both the number (left y-axis) and density (right y-axis) of segregating SNP sites in each group. Because group sizes vary, groups were subsampled in triplicate for each group size from 4 up to their respective size; means and SDs are shown. SNP, single-nucleotide polymorphism.

Figure 5

Extreme excess of heterozygous sites in the US hound–derived Leishmania infantum isolates. The group-specific inbreeding coefficient F is shown for all polymorphic sites in the respective parasite population. F measures the deviation of the frequency of heterozygotes from Hardy-Weinberg equilibrium with negative values indicating an excess and positive values a deficiency of heterozygotes over homozygotes. Horizontal lines within boxes indicate medians; box top and bottom lines indicate 25 and 75 percentiles; and error bars indicate minimum and maximum values, excluding outliers.

Figure 6

Decay of linkage disequilibrium with genomic distance across geographically confined groups of Leishmania infantum isolates. A) US_d_S5, B) BR_d_A5, C) IS_d_A5, D) BR_RGN_VLh_A5, E) BR_RGN_VLh_Ah_S5, F) BR_MA_VLh_Ah_S5, G) BR_MG_VLh_S5. H) BR_PI_VLh_S5, I) CH_mix_S5, J) IP_mix_A5, K) IT_mix_A5, L) SP_mix_A5. Long-range linkage disequilibrium was measured as R² for pairs of SNPs up to 100 kb apart within chromosomes and located on different chromosomes. Symbols show mean R² across SNP-pairs on all chromosomes, and lines show 1 SD for variants in bins of 5kb distance starting at the indicated distance. For groups with >5 samples, 5 have been randomly chosen to calculate R² values, indicated in group names for each subplot (S6, subsampled 5; A5, all 5 samples of the group were used). Symbol shapes indicates the number of pairwise comparisons available for each distance bin. Statistical significance of comparisons between R² between 4 different 5 kb windows at 0–4999 bp, 50–54.999 kb, 100–104.999 kb between SNP pairs for all between-chromosome comparisons are shown. FDR was determined based on the Kruskal-Wallis test, followed by the Dunn post hoc test when significant. For the groups in which only data for 2 of the 4 windows was present, the Mann-Whitney-Wilcoxon test was used. FDR, false discovery rate; NS, not significant; SNP, single-nucleotide polymorphism.

Figure 7

Aneuploidy variation of Leishmania isolates from US hunting hounds. A) Aneuploidy profiles, shown as a heatmap of estimated somy for each isolate and chromosome. The sample phylogeny is extracted from Figure 1. B) Chromosome-specific variation in somy across US hound isolates. Variation in somy between isolates provides a conservative estimate of somy variation, as it ignores within-isolate variation.