The role of Central American barriers in shaping the evolutionary history of the northernmost glassfrog, Hyalinobatrachium fleischmanni (Anura: Centrolenidae)

Abstract

The complex geological history of Central America has been useful for understanding the processes influencing the distribution and diversity of multiple groups of organisms. Anurans are an excellent choice for such studies because they typically exhibit site fidelity and reduced movement. The objective of this work was to identify the impact of recognized geographic barriers on the genetic structure, phylogeographic patterns and divergence times of a wide-ranging amphibian species, Hyalinobatrachium fleischmanni. We amplified three mitochondrial regions, two coding (COI and ND1) and one ribosomal (16S), in samples collected from the coasts of Veracruz and Guerrero in Mexico to the humid forests of Chocó in Ecuador. We examined the biogeographic history of the species through spatial clustering analyses (Geneland and sPCA), Bayesian and maximum likelihood reconstructions, and spatiotemporal diffusion analysis. Our data suggest a Central American origin of H. fleischmanni and two posterior independent dispersals towards North and South American regions. The first clade comprises individuals from Colombia, Ecuador, Panama and the sister species Hyalinobatrachium tatayoi; this clade shows little structure, despite the presence of the Andes mountain range and the long distances between sampling sites. The second clade consists of individuals from Costa Rica, Nicaragua, and eastern Honduras with no apparent structure. The third clade includes individuals from western Honduras, Guatemala, and Mexico and displays deep population structure. Herein, we synthesize the impact of known geographic areas that act as barriers to glassfrog dispersal and demonstrated their effect of differentiating H. fleischmanni into three markedly isolated clades. The observed genetic structure is associated with an initial dispersal event from Central America followed by vicariance that likely occurred during the Pliocene. The southern samples are characterized by a very recent population expansion, likely related to sea-level and climatic oscillations during the Pleistocene, whereas the structure of the northern clade has probably been driven by dispersal through the Isthmus of Tehuantepec and isolation by the Motagua–Polochic–Jocotán fault system and the Mexican highlands.