Dispersal patterns of Trypanosoma cruzi in Arequipa, Peru

Abstract

Anthropogenic environmental alterations such as urbanization can threaten native populations as well as create novel environments that allow human pests and pathogens to thrive. As the number and size of urban environments increase globally, it is more important than ever to understand the dispersal dynamics of hosts, vectors and pathogens of zoonotic disease systems. For example, a protozoan parasite and the causative agent of Chagas disease in humans, Trypanosoma cruzi, recently colonized and spread through the city of Arequipa, Peru. We used population genomic and phylogenomic tools to analyze whole genomes of 123 T. cruzi isolates derived from vectors and non-human mammals throughout Arequipa to determine patterns of T. cruzi dispersal. The data show significant population genetic structure within city blocks-parasites in the same block tend to be very closely related-but no population structure among blocks within districts-parasites in neighboring blocks are no more closely related to one another than to parasites in distant districts. These data suggest that T. cruzi dispersal within a block occurs regularly and that occasional long-range dispersal events allow the establishment of new T. cruzi populations in distant blocks. Movement of domestic animals may be the primary mechanism of inter-block and inter-district T. cruzi dispersal.

Fig 1

Map of sample collection locations in A) Arequipa, Peru and the B) Mariano Melgar district. A) All 123 T. cruzi samples used in this analysis were collected within 100km of Arequipa, Peru from 2008–2015. Sampled T. cruzi isolates are represented by red dots. Mariano Melgar—the city center and most densely-sampled district—is indicated with a yellow star. Inset shows the location of Arequipa in southern Peru. Map of Arequipa was obtained from https://landlook.usgs.gov, sample locations were plotted using QGIS v. 3.4 [8], and inset was modified from https://commons.wikimedia.org/wiki/Atlas_of_the_world [9]. B) Blocks from which T. cruzi were collected in the Mariano Melgar district (N = 56) are numbered 1–12 with the number of samples collected per block in parentheses. No parasites were collected in unlabeled blocks. Blocks are displayed in a grid to maintain privacy with black lines representing streets separating blocks.

Fig 2. T. cruzi isolates collected from the same block are genetically similar.

Violin plots represent the number of pairwise SNP differences among pairs of samples collected in different blocks from the same district (left) and among pairs of samples collected in the same block (right). The width of the violins represents the frequency that pairwise distances were detected. Samples found in different blocks within a district are significantly more different from one another than samples collected within the same city block (p<0.001). Genetically-similar samples tend to cluster within blocks despite each district containing much of the overall genetic diversity among all samples.

Fig 3. T. cruzi collected from the same block often forms a monophyletic group.

The maximum clade credibility phylogeny includes all 123 T. cruzi samples collected from Arequipa with samples from Mariano Melgar labeled blue and those from La Joya labeled green. Block numbers are labeled for the 48 samples collected from Blocks 1–7 in Mariano Melgar. Black branches represent samples collected from the other districts in the Arequipa region: Bustamante, Cayma, Huanca, Miraflores, Sachaca, Tiabaya, Uchumayo, and Vitor. T. cruzi from each district, for example, Mariano Melgar and La Joya, span the phylogenetic diversity in the region suggesting that most of the genetic variation in Arequipa can be found within each single district. The T. cruzi in many blocks within Mariano Melgar were likely introduced from another district as their closest relatives were detected in other districts and not in other blocks within Mariano Melgar. Jagged lines in most basal branch indicate that the branch was shortened for visualization. Posterior probabilities less than 0.80 are labeled for basal nodes. A modifiable phylogeny containing all tip labels and posterior probabilities can be found in S1 Appendix.