No abstract

Biologic invasions can have important ecological, economic and social consequences, particularly when they involve the introduction and spread of plant invasive pathogens, as they can threaten natural ecosystems and jeopardize the production of human food. Examples include the grapevine downy mildew, caused by the oomycete Plasmopara viticola, an invasive species native to North America, introduced into Europe in the 1870s. We investigated the introduction and spread of this invasive pathogen, by analysing its genetic structure and diversity in a large sample from European vineyards. Populations of P. viticola across Europe displayed little genetic diversity, consistent with the occurrence of a bottleneck at the time of introduction. Bayesian coalescent analyses revealed a clear population expansion signal in the genetic data. We detected a weak, but significant, continental-wide population structure, with two geographically and genetically distinct clusters in Western and Eastern European vineyards. Approximate Bayesian computation, analyses of clines of genetic diversity and of isolation-by-distance patterns provided evidence for a wave of colonization moving in an easterly direction across Europe. This is consistent with historical reports, first mentioning the introduction of the disease in Bordeaux vineyards (France) and sub-sequently documenting its rapid spread across Europe. This initial introduction in the west was probably followed by a 'leap-frog' event into Eastern Europe, leading to the formation of the two genetic clusters we detected. This study shows that recent population genetics methods within the Bayesian and coalescence frameworks are extremely powerful for increasing our understanding of pathogen population dynamics and invasion histories.

The putative center of origin of Plasmopara viticola, the causal agent of grape downy mildew, is eastern North America, where it has been described on several members of the family Vitaceae (e.g., Vitis spp., Parthenocissus spp., and Ampelopsis spp.). We have completed the first large-scale sampling of P. viticola isolates across a range of wild and cultivated host species distributed throughout the above region. Sequencing results of four partial genes indicated the presence of a new P. viticola species on Vitis vulpina in Virginia, adding to the four cryptic species of P. viticola recently recorded. The phylogenetic analysis also indicated that the P. viticola species found on Parthenocissus quinquefolia in North America is identical to Plasmopara muralis in Europe. The geographic distribution and host range of five pathogen species was determined through analysis of the internal transcribed spacer polymorphism of 896 isolates of P. viticola. Among three P. viticola species found on cultivated grape, one was restricted to Vitis interspecific hybrids within the northern part of eastern North America. A second species was recovered from V. vinifera and V. labrusca, and was distributed across most of the sampled region. A third species, although less abundant, was distributed across a larger geographical range, including the southern part of eastern North America. P. viticola clade aestivalis predominated (83% of isolates) in vineyards of the European winegrape V. vinifera within the sampled area, indicating that a single pathogen species may represent the primary threat to the European host species within eastern North America.