Molecular biodiversity and population structure in common ash (Fraxinus excelsior L.) in Britain: implications for conservation

Abstract

Current forestry policy promotes the use of local seed for new plantings, on the assumption that local material may be better adapted to local conditions. However, landscape-scale genetic studies which are necessary to underpin conservation and breeding strategies are often lacking. We investigated molecular diversity in common ash (Fraxinus excelsior L.) sampled from 42 British and six French sites with microsatellites. Chloroplast haplotype H04 was the most common and widespread in Britain, although rare and localized individuals with H02 and H09 were also detected. In addition, three new chloroplast haplotypes were identified, and these were rare and highly localized. In terms of nuclear microsatellite markers, allelic richness differed between sites and decreased in an east to west direction. Differentiation between sites was often very low (mean F(ST) 0.025), indicating few differences between the majority of sites. There was a clear excess of homozygotes (mean H(O) 0.669, mean H(E) 0.818) and a relatively high F(IS) (mean 0.182), suggests a consistent level of inbreeding or a widespread Wahlund effect in many F. excelsior sites. Gene pool ancestry analysis suggested that the majority of British F. excelsior belongs to a single meta-population which covers mainland western and central Europe. Three northern and western sites diverged markedly from the dominant population, and may represent remnants of two late potential Ice Age refugia in northern Britain. The data provide new information which will aid development of appropriate conservation policies for ash and other wind pollinated tree species.

Lack of genetic structure and evidence for long-distance dispersal in ash (Fraxinus excelsior) populations under threat from an emergent fungal pathogen: implications for restorative planting

“…The mean F IS calculated for ash populations across Northern Ireland (0.067) is significantly lower than that reported by Sutherland et al (2010), who used six of the eight loci analysed in the present study to examine populations throughout Great Britain (mean F IS = 0.182; Mann-Whitney test, z = 6.07, P < 0.0001). We designed new primers to amplify previously characterized microsatellite loci with the aim of circumventing any potential problems with null alleles, and our estimated null allele frequencies were generally much lower than those reported in Sutherland et al (2010).…”

“…The mean F IS calculated for ash populations across Northern Ireland (0.067) is significantly lower than that reported by Sutherland et al (2010), who used six of the eight loci analysed in the present study to examine populations throughout Great Britain (mean F IS = 0.182; Mann-Whitney test, z = 6.07, P < 0.0001). We designed new primers to amplify previously characterized microsatellite loci with the aim of circumventing any potential problems with null alleles, and our estimated null allele frequencies were generally much lower than those reported in Sutherland et al (2010). However, on genotyping a subset of our samples using the same primers as Sutherland et al (2010), we did not observe an increase in F IS or estimated null allele frequencies, suggesting that the previously observed homozygote deficiencies were not due to null alleles as a result of non-amplification.…”

“…Furthermore, the Irish samples of ash exhibited similar levels of nuclear genetic diversity to those in Great Britain and continental Europe, including populations from putative refugial areas that should harbour the highest levels of variation (Heuertz et al 2001;Morand et al 2002;Heuertz et al 2004a;Heuertz et al 2004b;Ferrazzini et al 2007;Sutherland et al 2010;Gérard et al 2013).…”

“…Several previous population genetic studies on ash have reported significant, and often high, levels of F IS , which have been attributed to various factors including inbreeding, null alleles, large allele dropout and Wahlund effect (Morand et al 2002;Ferrazzini et al 2007;Sutherland et al 2010). The mean F IS calculated for ash populations across Northern Ireland (0.067) is significantly lower than that reported by Sutherland et al (2010), who used six of the eight loci analysed in the present study to examine populations throughout Great Britain (mean F IS = 0.182; Mann-Whitney test, z = 6.07, P < 0.0001).…”

“…For nuclear microsatellite genotyping, we used six previously reported loci which have been widely used in population genetic studies on ash: Femsatl-4, Femsatl-8, Femsatl-11, Femsatl-16 andFemsatl-19 (Lefort et al 1999) andM230 (Brachet et al 1999), as well as two loci developed for the present study (FR639485 and FR646655). As previous studies highlighted the possibility of null alleles using the Lefort et al (1999) and Brachet et al (1999) primers (Morand et al 2002Ferrazzini et al 2007;Sutherland et al 2010), we designed new primers for all loci (Table 2) using the Primer3 program (v 0.4.0; http://primer3.ut.ee). The F IS values calculated in the present broad-scale study were lower Data analysis -broad-scale GENEPOP (V3.4; Raymond and Rousset, 1995) was used to test for linkage disequilibrium between nuclear microsatellite loci.…”