Population genetic structure of Bromus tectorum in the mountains of western North America

Abstract

Premise of the study: Invasive species are often initially restricted to a narrow range and may then expand through any of multiple mechanisms including phenotypic plasticity, in situ evolution, or selection on traits preadapted for new habitats. Our study used population genetics to explore possible processes by which the highly selfing invasive annual grass Bromus tectorum has expanded into montane environments.

Methods: We used 69 single nucleotide polymorphic (SNP) markers to genotype ca. 20 individuals from each of 38 montane cheatgrass populations from throughout the Intermountain West and to identify characteristic SNP haplotypes and examine their distribution.

Key results: Five invariant SNP haplotypes were dominant in montane cheatgrass populations, making up 59% of genotyped individuals, with each haplotype present in 12 to 21 populations. Four of these were absent or present at low frequency in low elevation populations, while the fifth was also sometimes dominant at low elevation. Sixteen haplotypes made up 78% of all genotyped individuals. These haplotypes were distributed across several haplogroups within the clade that also includes most sagebrush steppe lineages.

Conclusions: The wide geographic distribution of several common haplotypes almost completely restricted to montane habitats suggests that dominant lineages in montane populations may possess adaptive syndromes that are preserved through reduced outcrossing rates or negative selection on outcrossed progeny. However, conclusive evidence of such local adaptation requires reciprocal seeding experiments and further characterization of adaptive traits and breeding system characteristics. Other lineages have likely risen to dominance in montane populations through selectively neutral processes.

Keywords: Poaceae; SNP genotyping; cheatgrass; downy brome; invasive species; local adaptation; pre-adaptation; range expansion; selfing; spatial sorting.