A mistletoe tale: postglacial invasion of Psittacanthus schiedeanus (Loranthaceae) to Mesoamerican cloud forests revealed by molecular data and species distribution modeling

Abstract

BackgroundEcological adaptation to host taxa is thought to result in mistletoe speciation via race formation. However, historical and ecological factors could also contribute to explain genetic structuring particularly when mistletoe host races are distributed allopatrically. Using sequence data from nuclear (ITS) and chloroplast (trnL-F) DNA, we investigate the genetic differentiation of 31 Psittacanthus schiedeanus (Loranthaceae) populations across the Mesoamerican species range. We conducted phylogenetic, population and spatial genetic analyses on 274 individuals of P. schiedeanus to gain insight of the evolutionary history of these populations. Species distribution modeling, isolation with migration and Bayesian inference methods were used to infer the evolutionary transition of mistletoe invasion, in which evolutionary scenarios were compared through posterior probabilities.ResultsOur analyses revealed shallow levels of population structure with three genetic groups present across the sample area. Nine haplotypes were identified after sequencing the trnL-F intergenic spacer. These haplotypes showed phylogeographic structure, with three groups with restricted gene flow corresponding to the distribution of individuals/populations separated by habitat (cloud forest localities from San Luis Potosí to northwestern Oaxaca and Chiapas, localities with xeric vegetation in central Oaxaca, and localities with tropical deciduous forests in Chiapas), with post-glacial population expansions and potentially corresponding to post-glacial invasion types. Similarly, 44 ITS ribotypes suggest phylogeographic structure, despite the fact that most frequent ribotypes are widespread indicating effective nuclear gene flow via pollen. Gene flow estimates, a significant genetic signal of demographic expansion, and range shifts under past climatic conditions predicted by species distribution modeling suggest post-glacial invasion of P. schiedeanus mistletoes to cloud forests. However, Approximate Bayesian Computation (ABC) analyses strongly supported a scenario of simultaneous divergence among the three groups isolated recently.ConclusionsOur results provide support for the predominant role of isolation and environmental factors in driving genetic differentiation of Mesoamerican parrot-flower mistletoes. The ABC results are consistent with a scenario of post-glacial mistletoe invasion, independent of host identity, and that habitat types recently isolated P. schiedeanus populations, accumulating slight phenotypic differences among genetic groups due to recent migration across habitats. Under this scenario, climatic fluctuations throughout the Pleistocene would have altered the distribution of suitable habitat for mistletoes throughout Mesoamerica leading to variation in population continuity and isolation. Our findings add to an understanding of the role of recent isolation and colonization in shaping cloud forest communities in the region.Electronic supplementary materialThe online version of this article (doi:10.1186/s12862-016-0648...

Historical reconstruction of climatic and elevation preferences and the evolution of cloud forest-adapted tree ferns in Mesoamerica

“…Sequence and time divergences based on plastid DNA suggest a recent Miocene to Pliocene evolution of Alsophila and Cyathea tree ferns in the cloud forests of Mesoamerica with respect to other members of Cyatheaceae ( Sphaeropteris ). Although there is phylogeographic evidence of genetic divergence and range shifts from lowland tropical forests to cloud forests for Mesoamerican cloud forest species (e.g., Ornelas et al, 2016), only few examples have documented repeated habitat shifts in relation to climate changes along the evolutionary history of a lineage. For instance, the diversification of Bursera , from the Miocene onwards, occurred during a period of enhanced aridity, with divergence times between B. microphylla, B. hindsiana, B. laxiflora and B. fagaroides and its corresponding closest relatives, and range shifts from seasonally dry tropical forest to xerophytic scrubs, estimated between 7.7 and 4.6 Ma (De-Nova et al, 2012).…”

“…However, the shared phylogeographical breaks of cloud forest-adapted taxa occurred as multiple vicariant events at different times (Ornelas et al, 2013). In contrast, plant species that presumably colonized cloud forest in the region from South America are little studied (Ornelas, Ruiz-Sanchez & Sosa, 2010; Gutiérrez-Rodríguez, Ornelas & Rodríguez-Gómez, 2011; Ramírez-Barahona & Eguiarte, 2014; Ornelas & Rodríguez-Gómez, 2015; Ornelas et al, 2016). Population expansion due to Pleistocene climatic cycling has been uncovered in some of these cloud forest-adapted species that presumably originated in South America (Gutiérrez-Rodríguez, Ornelas & Rodríguez-Gómez, 2011; Ramírez-Barahona & Eguiarte, 2014; Ornelas & Rodríguez-Gómez, 2015; Ornelas et al, 2016) but the times of interspecific and intraspecific lineage divergence are temporally incongruent (Ornelas et al, 2013; Ramírez-Barahona & Eguiarte, 2014).…”

“…Despite the contradictory paleoecological information for reconstruction of past climate changes in the Neotropics (e.g., Ramírez-Barahona & Eguiarte, 2013), recent phylogeographic data suggest that cloud forest species persisted in these forests (Haffer, 1969; Ornelas & González, 2014) or expanded to the lowlands during glacial cycles and its distribution then contracted allopatrically during the interglacials (e.g., Ramírez-Barahona & Eguiarte, 2014; Ornelas & Rodríguez-Gómez, 2015; Ornelas et al, 2016). Alternatively, the evolutionary history of tree ferns reflects patterns of long-term persistence in the Mesoamerican cloud forests.…”

The influence of climatic niche preferences on the population genetic structure of a mistletoe species complex

“…Total genomic DNA was extracted from 601 sampled individuals using a modified cetyl trimethyl ammonium bromide (CTAB) protocol (Doyle & Doyle, ) and a DNeasy Plant Mini kit (Qiagen) employing the manufacturer's protocol. DNA sequences for the chloroplast trn L‐ trn F IGS (295 bp) were obtained for 530 individuals across the sampled populations (Table S1) from previously published works (Díaz Infante et al ., ; Ornelas et al ., ; M. J. Pérez‐Crespo J. F. Ornelas, A. González‐Rodríguez, E. Ruiz‐Sanchez, A. A. Vásquez‐Aguilar & S. Ramírez‐Barahona, unpublished; GenBank accession numbers: –, –, –).…”

Genetic Differentiation and Fragmentation in Response to Climate Change of the Narrow EndemicPsittacanthus auriculatus

“…Sequences of noncoding spacer region that lies between the large subunit of ribulose-1, 5-bisphosphate-carboxyalse (rbcL) and the beta-subunit of the chloroplast ATP-synthase (atpB; hereafter atpBrbcL) in combination with sequences of the plastid trnL-trnF intergenic spacer region (hereafter trnL-F) from the chloroplast genome (cpDNA) were amplified by PCR and sequenced. These markers show the appropriate level of variation for mistletoe studies at the species and population levels (e.g., Amico, Vidal-Russell, Garcı´a, & Nickrent, 2012;Ornelas et al, 2016; Pe´rez-Crespo Table 1) and geographical distributions of 8 chloroplast haplotypes (atpB-rbcL/ trnL-F) found in six populations of Psittacanthus auriculatus in Oaxaca. The blue haplotypes correspond to the northern populations group and the green haplotypes to the southern populations group identified by AMOVA.…”