Get access

Genetic structure and population dynamics of a heteroecious plant pathogen Melampsora larici-epitea in short-rotation coppice willow plantations

Authors


  • C.B, C.R., T.H. and M.H.P. contributed working as a research team on energy crop pathology. A.K. is the Head of the Centre for Bioenergy and Climate Change and the Head of Genetic Diversity Group at Rothamsted Research. I.T. is a project leader at the Biometrics, Surveys & Statistics Division, Forest Research, Alice Holt, Surrey.

Ming H. Pei, Fax: +44(0)1582760981; E-mail: ming.pei@bbsrc.ac.uk

Abstract

Complex life strategies are common among plant pathogens belonging to rust fungi (Uredinales). The heteroecious willow rust Melampsora larici-epitea produces five spore stages and alternates on larch (Larix). To shed light on the epidemiology of this pathogen, amplified fragment length polymorphisms (AFLPs) were used to determine the genetic diversity and genetic structure of rust samples collected from coppice willow (Salix) plantations at three UK sites (LA, CA and MC) over three sampling dates (September 2000, July 2001 and September 2001). Of the total of 819 isolates, 465 were unique AFLP phenotypes and there was a shift in genotype diversity between the two seasons (0.67 in 2000 and 0.87–0.89 in 2001). No phenotypes were common between the two seasons within a site, suggesting that the rust did not overwinter as an asexual stage within plantations. A temporal analysis detected large amounts of genetic drift (FS = 0.15–0.26) between the two seasons and very small effective population sizes (Ne = 2–3) within sites. These results all point to a new colonization of the plantations by the rust in the second season (2001). The FST-analogue values were ΦCT = 0.121, Weir and Cockerham’s θ = 0.086 and the Bayesian estimate θB = 0.087–0.096. The results suggest that the sources of inoculum were somewhat localized and the same sources were mainly responsible for disease epidemics in LA and CA over the two seasons. The relatively low FST-values among sites (0.055–0.13) suggest the existence of significant gene flow among the three sampled sites.

Ancillary