Phylogeography and niche modelling of the relict plant Amborella trichopoda (Amborellaceae) reveal multiple Pleistocene refugia in New Caledonia
Amborella trichopoda Baill. (Amborellaceae, Amborellales), the sole living member of the sister group to all other extant angiosperms, is endemic to New Caledonia. We addressed the intraspecific phylogeography of Amborella by investigating whether its present population genetic structure could be re...
Saved in:
Published in | Molecular ecology Vol. 22; no. 24; pp. 6163 - 6178 |
---|---|
Main Authors | , , , , , , , , |
Format | Journal Article |
Language | English |
Published |
Oxford
Blackwell Publishing Ltd
01.12.2013
Blackwell |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Amborella trichopoda Baill. (Amborellaceae, Amborellales), the sole living member of the sister group to all other extant angiosperms, is endemic to New Caledonia. We addressed the intraspecific phylogeography of Amborella by investigating whether its present population genetic structure could be related to its current and past habitats. We found moderate range‐wide genetic diversity based on nuclear microsatellite data and detected four well‐differentiated, geographically distinct genetic groups using Bayesian clustering analyses. We modelled the ecological niche of Amborella based on the current climatic and environmental conditions. The predictive ability of the model was very good throughout the Central East mainland zone, but Amborella was predicted in the northern part of the island where this plant has not been reported. Furthermore, no significant barrier was detected based on habitat suitability that could explain the genetic differentiation across the area. Conversely, we found that the main genetic clusters could be related to the distribution of the suitable habitat at the last glacial maximum (LGM, c. 21 000 years BP), when Amborella experienced a dramatic 96.5% reduction in suitable area. At least two lineages survived in distinct putative refugia located in the Massif des Lèvres and in the vicinity of Mount Aoupinié. Our findings finally confirmed the importance of LGM rainforest refugia in shaping the current intra‐ and interspecific diversity in New Caledonian plants and revealed the possibility of an as yet unreported refugium. The combination of niche modelling and population genetics thereby offered novel insight into the biogeographical history of an emblematic taxon. |
---|---|
Bibliography: | http://dx.doi.org/10.1111/mec.12554 ArticleID:MEC12554 IRD Fig. S1 Unrooted Neighbor-joining tree of Amborella trichopoda individuals based on shared allele distances. Fig. S2 Estimated number of populations from a Geneland analysis of five replicates. Fig. S3 Individual Bayesian assignment probabilities for K = 2-18 for runs with the highest posterior probability (out of 20 runs per K value). Fig. S4 Response curves of the Maxent model created using each of the input variables only. Fig. S5 Model features. Fig. S6 Amborella populations (georeferenced vouchers or sites) and phytogeographical sectors as defined by Jaffré & Veillon () in red and Pintaud et al. () in yellow. Table S1 Georeferenced locations of Amborella. Localities are in decimal degrees (WGS84 Coordinate System). Table S2 Species level estimates of genetic diversity at each of the 10 microsatellite loci in Amborella trichopoda. Table S3 Detection of loci under selection using the hierarchical island model in Arlequin (No. of simulated groups: 12; No. of simulated demes per group: 100; No. of coalescent simulations performed: 1000). Table S4. Between Amborella trichopoda populations pairwise FST estimates (below the diagonal). Table S5. Pairwise FST values calculated between each site of sampled Amborella trichopoda. Table S6. Common thresholds and corresponding omission rates of the Maxent model. Table S7. Environmental variables used for ecological niche modelling with Maxent. istex:0F4A4A6BC57C9A8545B352A5B908F94D89CA4F76 ark:/67375/WNG-PR39PFV9-5 ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0962-1083 1365-294X |
DOI: | 10.1111/mec.12554 |