Geographical barriers and climate influence demographic history in narrowleaf cottonwoods.

Studies of genetic variation can clarify the role of geography and spatio-temporal variation of climate in shaping demography, particularly in temperate zone tree species with large latitudinal ranges. Here, we examined genetic variation in narrowleaf cottonwood, Populus angustifolia, a dominant riparian tree. Using multi-locus surveys of polymorphism in 363 individuals across the species' 1800 km latitudinal range, we found that, first, P. angustifolia has stronger neutral genetic structure than many forest trees (simple sequence repeat (SSR) FST=0.21), with major genetic groups corresponding to large apparent geographical barriers to gene flow. Second, using SSRs and putatively neutral sequenced loci, coalescent simulations indicated that populations diverged before the last glacial maximum (LGM), suggesting the presence of population structure before the LGM. Third, the LGM and subsequent warming appear to have had different influences on each of these distinct populations, with effective population size reduction in the southern extent of the range but major expansion in the north. These results are consistent with the hypothesis that climate and geographic barriers have jointly affected the demographic history of P. angustifolia, and point the importance of both factors as being instrumental in shaping genetic variation and structure in widespread forest trees.

Population substructure in continuous and fragmented stands of Populus trichocarpa.

Population substructure has important implications for both basic and applied genetic research. We used 10 microsatellite markers to characterize population substructure in two ecologically and demographically contrasting populations of the model tree Populus trichocarpa. The Marchel site was a continuous stand growing in a mesic habitat in western Oregon, whereas the Vinson site consisted of three disjunct and isolated stands in the high desert of eastern Oregon. A previous study revealed that pollen-mediated gene flow is extensive in both populations. Surprisingly, model-based clustering, principal components analysis and analyses of molecular variance provided overwhelming support for the existence of at least two intermingled sub-populations within the continuous Marchel population (F(ST)=0.026, P<0.001), which occupied an area with a radius of only about 250 m. Genets in these two sub-populations appeared to have different relative clone ages and phenologies, leading us to hypothesize that they correspond to different seedling cohorts, each established from seeds produced by relatively few mothers. As expected, substructure was stronger in the fragmented Vinson population (F(ST)=0.071, P=0.001), and this difference appeared to result from the more extensive family structure in this population. Using group-likelihood methods, we reconstructed multiple interconnected half-sib families in the Vinson population, with some genets having as many as eight putative siblings. Researchers involved in ongoing and future association studies in P. trichocarpa should account for the likely presence of subtle but practically significant substructure in populations throughout the range of this species.

Extensive pollen flow in two ecologically contrasting populations of Populus trichocarpa.

Pollen-mediated gene flow was measured in two populations of black cottonwood using direct (paternity analysis) and indirect (correlated paternity) methods. The Marchel site was an area with an approximate radius of 250 m in a large continuous stand growing in a mesic habitat in western Oregon. In contrast, the Vinson site was an area with a radius of approximately 10 km and consisted of small, disjunct and isolated stands in the high desert of eastern Oregon. Pollen immigration was extensive in both populations, and was higher in the Marchel site (0.54 +/- 0.02) than in the substantially larger and more isolated Vinson site (0.32 +/- 0.02). Pollen pool differentiation among mothers was approximately five times stronger in the Vinson population (Phi FT = 0.253, N = 27 mothers) than in the Marchel population (Phi FT = 0.052, N = 5 mothers). Pollen dispersal was modelled using a mixed dispersal curve that incorporated pollen immigration. Predicted pollination frequencies generated based on this curve were substantially more accurate than those based on the widely used exponential power dispersal curve. Male neighbourhood sizes (sensu Wright 1946) estimated using paternity analysis and pollen pool differentiation were remarkably similar. They were three to five times smaller in the Vinson population, which reflected the substantial ecological and demographic differences between the two populations. When the same mathematical function was used, applying direct and indirect methods resulted in similar pollen dispersal curves, thus confirming the value of indirect methods as a viable lower-cost alternative to paternity analysis.

A 34K SNP genotyping array for Populus trichocarpa: design, application to the study of natural populations and transferability to other Populus species.

Genetic mapping of quantitative traits requires genotypic data for large numbers of markers in many individuals. For such studies, the use of large single nucleotide polymorphism (SNP) genotyping arrays still offers the most cost-effective solution. Herein we report on the design and performance of a SNP genotyping array for Populus trichocarpa (black cottonwood). This genotyping array was designed with SNPs pre-ascertained in 34 wild accessions covering most of the species latitudinal range. We adopted a candidate gene approach to the array design that resulted in the selection of 34 131 SNPs, the majority of which are located in, or within 2 kb of, 3543 candidate genes. A subset of the SNPs on the array (539) was selected based on patterns of variation among the SNP discovery accessions. We show that more than 95% of the loci produce high quality genotypes and that the genotyping error rate for these is likely below 2%. We demonstrate that even among small numbers of samples (n = 10) from local populations over 84% of loci are polymorphic. We also tested the applicability of the array to other species in the genus and found that the number of polymorphic loci decreases rapidly with genetic distance, with the largest numbers detected in other species in section Tacamahaca. Finally, we provide evidence for the utility of the array to address evolutionary questions such as intraspecific studies of genetic differentiation, species assignment and the detection of natural hybrids.

Estimating pollen flow using SSR markers and paternity exclusion: accounting for mistyping.

Highly informative genetic markers, such as simple sequence repeats (SSRs), can be used to directly measure pollen flow by parentage analysis. However, mistyping (i.e. false inference of genotypes caused by the occurrence of null alleles, mutations, and detection errors) can lead to substantial biases in the estimates obtained. Using computer simulations, we evaluated a direct method for estimating pollen immigration using SSR markers and a paternity exclusion approach. This method accounts for mistyping and does not rely on assumptions about the distribution of male reproductive success. If ignored, even minor rates of mistyping (1.5%) resulted in overestimating pollen immigration by up to 150%. When we required at least two mismatching loci before excluding candidate fathers from paternity, the resulting pollen immigration estimates had small biases for rates of mistyping up to 4.5%. Requiring at least three mismatches for exclusion was needed to minimize the upward biases of pollen immigration caused by rates of mistyping up to 10.5%. The minimum number of highly variable SSR loci needed to minimize cryptic gene flow and obtain reliable estimates of pollen immigration varied from five to seven for a sampling scheme applicable to most conifers (i.e. when paternal haplotypes can be unambiguously determined). Between five and nine highly variable SSR loci were needed for a more general sampling scheme that is applicable to all diploid seed plants. With moderately variable SSR markers, consistently accurate estimates of pollen immigration could be obtained only for rates of mistyping up to 4.5%. We developed the POLLEN FLOW (PFL) computer program which can be used to obtain unbiased and precise estimates of pollen immigration under a wide range of conditions, including population sizes as large as 600 parents and mistyping rates as high as 10.5%.

Highly variable SSR markers in Douglas-fir: Mendelian inheritance and map locations.

Twenty-two highly variable SSR markers were developed in Douglas-fir [ Pseudotsuga menziesii (Mirb.) Franco] from five SSR-enriched genomic libraries. Fifteen PCR primer pairs amplified a single codominant locus, while seven primer pairs occasionally amplified two loci. The Mendelian inheritance of all 22 SSRs was confirmed via segregation analyses in several Douglas-fir families. The mean observed heterozygosity and the mean number of alleles per locus were 0.855 (SE=0.020) and 23 (SE=1.6), respectively. Twenty markers were used in genetic linkage analysis and mapped to ten known linkage groups. Because of their high polymorphism and unambiguous phenotypes, 15 single-locus markers were selected as the most suitable for DNA fingerprinting and parentage analysis. Only three SSRs were sufficient to achieve an average probability of exclusion from paternity of 0.998 in a Douglas-fir seed orchard block consisting of 59 parents.