Phylogenetics, patterns of genetic variation and population dynamics of Trypanosoma terrestris support both coevolution and ecological host-fitting as processes driving trypanosome evolution

Abstract

Background: A considerable amount of evidence has favored ecological host-fitting, rather than coevolution, as the main mechanism responsible for trypanosome divergence. Nevertheless, beyond the study of human pathogenic trypanosomes, the genetic basis of host specificity among trypanosomes isolated from forest-inhabiting hosts remains largely unknown.

Methods: To test possible scenarios on ecological host-fitting and coevolution, we combined a host capture recapture strategy with parasite genetic data and studied the genetic variation, population dynamics and phylogenetic relationships of Trypanosoma terrestris, a recently described trypanosome species isolated from lowland tapirs in the Brazilian Pantanal and Atlantic Forest biomes.

Results: We made inferences of T. terrestris population structure at three possible sources of genetic variation: geography, tapir hosts and 'putative' vectors. We found evidence of a bottleneck affecting the contemporary patterns of parasite genetic structure, resulting in little genetic diversity and no evidence of genetic structure among hosts or biomes. Despite this, a strongly divergent haplotype was recorded at a microgeographical scale in the landscape of Nhecolândia in the Pantanal. However, although tapirs are promoting the dispersion of the parasites through the landscape, neither geographical barriers nor tapir hosts were involved in the isolation of this haplotype. Taken together, these findings suggest that either host-switching promoted by putative vectors or declining tapir population densities are influencing the current parasite population dynamics and genetic structure. Similarly, phylogenetic analyses revealed that T. terrestris is strongly linked to the evolutionary history of its perissodactyl hosts, suggesting a coevolving scenario between Perissodactyla and their trypanosomes. Additionally, T. terrestris and T. grayi are closely related, further indicating that host-switching is a common feature promoting trypanosome evolution.

Conclusions: This study provides two lines of evidence, both micro- and macroevolutionary, suggesting that both host-switching by ecological fitting and coevolution are two important and non-mutually-exclusive processes driving the evolution of trypanosomes. In line with other parasite systems, our results support that even in the face of host specialization and coevolution, host-switching may be common and is an important determinant of parasite diversification.

Keywords: Coevolution; Ecological fitting; Microgeographic divergence; Perissodactyla; Population bottleneck; Tapirus terrestris; Trypanosoma terrestris.

Fig. 1

Positive isolates of T. terrestris according to their biome of origin. Abbreviations: AM, Amazonia; CE, Cerrado; PA, Pantanal; AF, Atlantic Forest; P, Pampa; CA, Caatinga

Fig. 2

Patterns of genetic variation in T. terrestris. a Heat map of Nei’s genetic distances showing the average number of pairwise differences between ITS1 rDNA haplotypes. b ITS1 rDNA haplotype network inferred by minimum spanning network. Circle sizes correspond to the frequency of CBTs per haplotype and vertical lines connecting the network represent the number of mutations

Fig. 3

Results of the Bayesian clustering analysis in STRUCTURE. a Admixture model. b Non-admixture model. Isolates from Sumatra represent the trypanosomes of rhinos isolated from Southeast Asia (Trypanosoma vanstrieni)

Fig. 4

Results of the BARRIER test based on the bootstrapping of 100 K2P genetic distance matrices (Kimura 2-parameter) obtained from the random sampling of ITS1 haplotypes. Black and green lines represent the Voronoi/Delaunay tessellation/triangulation and the dots correspond to the geographical origin of ITS1 haplotypes sampled in Nhecolândia. Thickness of the red lines corresponds to the barrier robustness, identified by Monmonier’s maximum difference algorithm. In this case, sample 8 that has a bootstrap value of 86% is the genetic barrier belonging to H9

Fig. 5

Extended Bayesian skyline plot illustrating the entire posterior distribution of demographic trends for T. terrestris isolates employed in this study. Dashed lines are the median effective population sizes, whereas the solid ones belong to 95% HPD limits. The time is in units of million years before present and population uses a logarithmic scale (Log 4Neµ)

Fig. 6

Maximum clade credibility tree (MCC) inferred by gGAPDH and V7V8 SSU rDNA concatenated sequences, showing phylogenetic relationships and divergence times (tMRCA) of studied trypanosomes. Posterior probability at each node is indicated by values ranging from 0.0 to 1.0. Horizontal blue bars represent the posterior credibility limits (HPD) for divergence time estimates