Spatial and Temporal Genetic Structure in Landrace Cereals

Abstract

Genetic variation exists both on a geographical and a temporal scale. By tracing patterns in variation through time and space it is possible to explore the history of species. Since domestication, crop plants have diverged from their wild relatives due to selection for conditions created by farming practices. Many crop species have become dependent on humans for dispersal; patterns in genetic variation are thus strongly affected by mobility and economy in agrarian communities. In this thesis I have used phylogeographic approaches to study the genetic structure of two common, partially contrasting, crop species: barley (Hordeum vulgare) and rye (Secale cereale). The plant material studied in this thesis belonged to two different types of collections, extant landraces and historical landraces. Some extant landraces were collected recently; others have been maintained ex situ for over 100 years through cycles of rejuvenation. The historical landraces were harvested over 100 years ago “on farm”, have never been rejuvenated and are thus no longer viable. The thesis explores genetic structure on different levels. Landrace barley from Northern Europe was used to study genetic structure on an international scale and assess the link between SNP genotypes and local climate. The northern border of agriculture was explored in a study of barley where all accessions had origins north of the 65th parallel. Rye was studied on a continental scale, exploring genetic structure through Europe and neighboring territories. Genetic information for comparative analysis between individuals and accessions was primarily assessed with Single Nucleotide Polymorphism markers (SNPs). The SNP genotyping enabled the use of high throughput techniques, which allowed the generation of large quantities of data for each individual seed. Comparisons between genetic diversity in extant and historical barley landraces indicated that the levels of within-accession diversity were much more variable in landrace material that had been maintained in genebanks. Studies of landrace barley from Northern Europe showed that national borders and historical trade regulations had not been influential in determining the genetic structure. Instead, genetic structuring followed latitudinal clines. I also concluded that the diversity of the northernmost barley landraces was mostly found within accessions, and that the geographic structure is weak in that area. A significant correlation was found between genetic structure and harvest year, suggesting a change over time. In rye most genetic variation within rye landraces was found within accessions and clustering depended on origin to a higher extent than previous subspecies classifications. My work has shown that historical seed collections can be analyzed with high throughput molecular genetic techniques and that genetics can be used to study the history of crop landraces. I conclude that national borders did not affect genetic structuring of barley in northern Europe and hypothesize that the temporal structuring of barley in the far north is an effect of recurring crop failures and replacement of seed through trade. I also conclude that the genetic structure of rye depend more on geography than subspecies distinctions.