Geologic events coupled with Pleistocene climatic oscillations drove genetic variation of Omei treefrog (Rhacophorus omeimontis) in southern China

Abstract

Background: Pleistocene climatic oscillations and historical geological events may both influence current patterns of genetic variation, and the species in southern China that faced unique climatic and topographical events have complex evolutionary histories. However, the relative contributions of climatic oscillations and geographical events to the genetic variation of these species remain undetermined. To investigate patterns of genetic variation and to test the hypotheses about the factors that shaped the distribution of this genetic variation in species of southern China, mitochondrial genes (cytochrome b and NADH dehydrogenase subunit 2) and nine microsatellite loci of the Omei tree frog (Rhacophorus omeimontis) were amplified in this study.

Results: The genetic diversity in the populations of R. omeimontis was high. The phylogenetic trees reconstructed from the mitochondrial DNA (mtDNA) haplotypes and the Bayesian genetic clustering analysis based on microsatellite data both revealed that all populations were divided into three lineages (SC, HG and YN). The two most recent splitting events among the lineages coincided with recent geological events (including the intense uplift of the Qinghai-Tibet Plateau, QTP and the subsequent movements of the Yun-Gui Plateau, YGP) and the Pleistocene glaciations. Significant expansion signals were not detected in mismatch analyses or neutrality tests. And the effective population size of each lineage was stable during the Pleistocene.

Conclusions: Based on the results of this study, complex geological events (the recent dramatic uplift of the QTP and the subsequent movements of the YGP) and the Pleistocene glaciations were apparent drivers of the rapid divergence of the R. omeimontis lineages. Each diverged lineages survived in situ with limited gene exchanges, and the stable demographics of lineages indicate that the Pleistocene climatic oscillations were inconsequential for this species. The analysis of genetic variation in populations of R. omeimontis contributes to the understanding of the effects of changes in climate and of geographical events on the dynamic development of contemporary patterns of genetic variation in the species of southern China.

Fig. 1

Sampling locations of the populations of R. omeimontis. The locality codes and coordinates are presented in Table 1. The three terrain steps in China and the primary mountain systems are marked. The yellow shadow areas represent the distribution range of R. omeimontis on the IUCN range map (http://www.iucnredlist.org/). The genetic lineages are labeled red, green and blue for the lineages SC, YN and HG, respectively

Fig. 2

Phylogenetic relationships and haplotype network based on the mtDNA haplotypes of R. omeimontis. (a) Phylogenetic tree from Bayesian inference analysis with Bayesian posterior probabilities/maximum parsimony bootstrap values near the branches; and (b) Median-joining network with node sizes proportional to the frequencies of haplotypes. The numbers of mutations separating the haplotypes are shown on the branches, except for the one-step mutations. Empty nodes indicate undetected haplotypes

Fig. 3

STRUCTURE clustering results deduced from microsatellite alleles within populations of R. omeimontis. (a) Ln P(X/K) and ΔK values as a function of the K values according to 20 run outputs; (b) STRUCTURE results at K = 3, with different colors representing different clusters; and (c) STRUCTURE results at K = 6, with different colors representing different sub-clusters

Fig. 4

Estimation of divergence time for R. omeimontis. The numbers near the primary branches are the estimated split times with 95 % highest posterior density. The three primary lineages are indicated by different colors: red, SC lineage; green, YN lineage; and blue, HG lineage

Fig. 5

Demographic history analyses for the three lineages according to mismatch distributions and Bayesian skyline plots. (a, b and c) Mismatch distributions for the SC, HG and YN lineages, respectively. The x coordinate represents the number of differences in each pair of sequence comparisons, and the y coordinate represents the frequencies of pairwise differences. (d, e and f) Bayesian skyline plots for the SC, HG and YN lineages, respectively. The x-axis indicates time in Ma BP, and the y-axis indicates the effective population size in units of Neτ (the product of effective population size and generation time in Ma). The blue areas represent 95 % highest posterior density