Coalescent Simulation and Paleodistribution Modeling for Tabebuia rosealba Do Not Support South American Dry Forest Refugia Hypothesis

Abstract

Studies based on contemporary plant occurrences and pollen fossil records have proposed that the current disjunct distribution of seasonally dry tropical forests (SDTFs) across South America is the result of fragmentation of a formerly widespread and continuously distributed dry forest during the arid climatic conditions associated with the Last Glacial Maximum (LGM), which is known as the modern-day dry forest refugia hypothesis. We studied the demographic history of Tabebuia rosealba (Bignoniaceae) to understand the disjunct geographic distribution of South American SDTFs based on statistical phylogeography and ecological niche modeling (ENM). We specifically tested the dry forest refugia hypothesis; i.e., if the multiple and isolated patches of SDTFs are current climatic relicts of a widespread and continuously distributed dry forest during the LGM. We sampled 235 individuals across 18 populations in Central Brazil and analyzed the polymorphisms at chloroplast (trnS-trnG, psbA-trnH and ycf6-trnC intergenic spacers) and nuclear (ITS nrDNA) genomes. We performed coalescence simulations of alternative hypotheses under demographic expectations from two a priori biogeographic hypotheses (1. the Pleistocene Arc hypothesis and, 2. a range shift to Amazon Basin) and other two demographic expectances predicted by ENMs (3. expansion throughout the Neotropical South America, including Amazon Basin, and 4. retraction during the LGM). Phylogenetic analyses based on median-joining network showed haplotype sharing among populations with evidence of incomplete lineage sorting. Coalescent analyses showed smaller effective population sizes for T. roseoalba during the LGM compared to the present-day. Simulations and ENM also showed that its current spatial pattern of genetic diversity is most likely due to a scenario of range retraction during the LGM instead of the fragmentation from a once extensive and largely contiguous SDTF across South America, not supporting the South American dry forest refugia hypothesis.

Fig 1. Geographic distribution of haplotypes and Bayesian clustering for (a) ITS and (b) cpDNA, based on the sequencing of 235 individuals of Tabebuia roseoalba from 18 populations.

Different colors were assigned for each haplotype according to the figure legend. The circle size represents the sample size in each population and the circle sections represent the haplotype frequency in each sampled population. For details on population codes and localities see S1 Table. For BAPS clustering, each color represents an inferred cluster (6 clusters for ITS and 5 for cpDNA). Map modified from the IBGE shape, free available at http://www.ibge.gov.br/home/.

Fig 2. The demographic history scenarios simulated for T. roseoalba and their geographical representation.

The size and location of circle during the LGM indicate demographic population expansion or shrink, and geographic range shift at that time. LIG: last interglacial; LGM: last glacial maximum; Pres: present-day; N0: effective population size at time t0 (present); N1: effective population size at time t1750 (1,750 generations ago). The demographic scenarios correspond to: PLAH, Pleistocene Arc hypothesis; PPPH, the ‘Amazonian SDF’ hypothesis; Both (PLAH+PPPH), i.e., an expansion throughout the Central and Southwest Brazil and also westward toward the Amazonian Basin; Retraction, a retraction in geographic range in Central Brazil.

Fig 3. Variation in effective population size through time and relationships and TMRCA (time to most recent common ancestor) of Tabebuia roseoalba lineages based on concatenated sequences of cpDNA and ITS nrDNA, from 235 individuals.

(a) Extended Bayesian Skyline Plot showing effective population retraction at the LGM. (b) Coalescent tree showing that most lineage divergences occurred after the Lower Pleistocene. Tip color corresponds to haplotypes described in Fig 1; Gray bar corresponds to 95% Credibility Interval of the mean time to the common ancestor; numbers above the branches are the support to the node (posterior probability); numbers below the branches are the node dating (time to the common ancestor). Time scale is in millions of years (Ma) before present.

Fig 4. Maps of consensus expressing the ensemble of climatic suitability for Tabebuia roseoalba, hence its potential distribution across the Neotropics.

(A) LGM (21ka), (B) mid-Holocene (6 ka), and (C) present-day.

Fig 5. Spatial distribution of genetic diversity for Tabebuia roseoalba in relation to the historical refugium, i.e. areas climatically suitable throughout the time (in gray), based on the sequence of 235 individuals from 18 populations.

(A) mutation parameter theta (θ). (B) Effective population size (N_e). (C) Nucleotide diversity (π) for cpDNA. (D) Haplotype diversity (h) for cpDNA. (E) Nucleotide diversity (π) for ITS. (D) Haplotype diversity (h) for ITS. Circumference sizes are proportional to the value of genetic parameter, following the figure legends.