COS 120-5
Demographic history of Western European populations of the trace elements hyperaccumulator Noccaea caerulescens (Brassicaceae)
Noccaea caerulescens (Brassicaceae) hyperaccumulates several trace elements (TE), namely cadmium, nickel and zinc, and is thus a common candidate for phytoremediation of contaminated soils. Significant variation in the hyperaccumulation ability between populations has been observed, mainly in France and its neighboring regions, representing the western part of the distribution area of the species. In this area, the geographic distribution of the species is characterized by high heterogeneity in edaphic types, with metalliferous and non-metalliferous soils.
To advance the understanding of factors behind variation among samples in the current distribution of hyperaccumulation ability, we used molecular genotyping to discern the historical and current genetic relationships among metalliferous and non-metalliferous samples.
Our sampling encompassed the majority of currently known sites in Western Europe, including most, if not all, M sites investigated to date. We analyzed the genetic structure of 62 sites in the French mountains using a combination of chloroplastic and nuclear genotyping markers. We applied classical analytical tools in phylogeography and population genetics. We also used recent methods developed from the Approximate Bayesian Computation (ABC) framework, inspired by Bayesian statistics, to track in more details the demographic scenarios that may be at the origin of detected genetic units.
Results/Conclusions
This study confirms that all samples in the western part of the distribution of N. caerulescens belong to one of two genetic units of the species. Microsatellite data revealed a strong genetic structure, enabling most samples to be assigned to one of three gene pools with high estimated membership coefficients. Noticeably, gene pool limits did not reflect soil types, but showed a good geographical coherence. Similarly, based on geographical proximity and the congruence between genetic data, samples were grouped into three genetic subunits. ABC suggests a first divergence separating samples from the southeastern France before the last glacial maximum and a second more recent divergence isolating more northern samples.
Neutral genetic structure observed here is not explained by edaphic types. That metalliferous samples belong to different genetic groups supports the hypothesis that the colonization of metalliferous sites occurred independently in distant geographic areas with convergence on TE tolerance and hyperaccumulation. Finally, it would be parsimonious to consider hyperaccumulation capabilities prior to the separation of the three subunits and before the colonization of metalliferous sites of recent anthropogenic origin.