Academic journal article Genetics

Understanding the Evolution of Defense Metabolites in Arabidopsis Thaliana Using Genome-Wide Association Mapping

Academic journal article Genetics

Understanding the Evolution of Defense Metabolites in Arabidopsis Thaliana Using Genome-Wide Association Mapping

Article excerpt

ABSTRACT

With the improvement and decline in cost of high-throughput genotyping and phenotyping technologies, genome-wide association (GWA) studies are fast becoming a preferred approach for dissecting complex quantitative traits. Glucosinolate (GSL) secondary metabolites within Arabidopsis spp. can serve as a model system to understand the genomic architecture of quantitative traits. GSLs are key defenses against insects in the wild and the relatively large number of cloned quantitative trait locus (QTL) controlling GSL traits allows comparison of GWA to previous QTL analyses. To better understand the specieswide genomic architecture controlling plant-insect interactions and the relative strengths of GWA and QTL studies, we conducted a GWA mapping study using 96 A. thaliana accessions, 43 GSL phenotypes, and ~230,000 SNPs. Our GWA analysis identified the two major polymorphic loci controlling GSL variation (AOP and MAM) in natural populations within large blocks of positive associations encompassing dozens of genes. These blocks of positive associations showed extended linkage disequilibrium (LD) that we hypothesize to have arisen from balancing or fluctuating selective sweeps at both the AOP and MAM loci. These potential sweep blocks are likely linked with the formation of new defensive chemistries that alter plant fitness in natural environments. Interestingly, this GWA analysis did not identify the majority of previously identified QTL even though these polymorphisms were present in the GWA population. This may be partly explained by a nonrandom distribution of phenotypic variation across population subgroups that links population structure and GSL variation, suggesting that natural selection can hinder the detection of phenotype-genotype associations in natural populations.

NATURAL phenotypic variation within a species or population is largely quantitative, polygenic, and controlled by the interaction of environmental and genetic factors (Fisher 1930; Falconer and Mackay 1996; Lynch and Walsh 1998). Advances in both highthroughput genotyping and phenotyping has enabled the use of intraspecific natural variation to identify the molecular and genetic bases of complex traits such as disease resistance, growth and development and correspondingly provide a preliminary view of the ecological and evolutionary consequences of this variation. While quantitative trait locus (QTL) mapping has been the standard approach to studying complex traits in the past, its application to self-incompatible and longgeneration species has been limited by the labor and time required to generate and genotype mapping populations (Liu 1998; Lynch and Walsh 1998; Mauricio 2001). As such, the molecular basis of most quantitative traits remains unknown.

Genome-wide association (GWA) mapping has become a popular alternative to QTL mapping in recent years for studying natural genetic variation. GWA identifies association between phenotypes and genotypes, at a genome-wide level, using "unrelated" individuals that have been simultaneously genotyped and phenotyped (Hirschhorn andDaly 2005;Weigel and Nordborg 2005; Nordborg and Weigel 2008). Genetic recombination across generations leads to a decay of linkage disequilibrium (LD), or apparent genetic linkage, between neighboring polymorphisms such that polymorphisms separated by hundreds to thousands of bases are effectively inherited independently (Kim et al. 2007; Nordborg and Weigel 2008). The goal of GWA mapping is to identify polymorphisms associated with the quantitative traits of interest. This potential has been demonstrated, typically within candidate genes previously identified from molecular or QTL data (Begovich et al. 2004; Palaisa et al. 2004; Szalma et al. 2005; Brock et al. 2007; Easton et al. 2007; Harjes et al. 2008) but increasingly using genome-wide analyses (Easton et al. 2007; Zhao et al. 2007; Ghazalpour et al. 2008).

While the potential of GWA studies has been experimentally supported, one's ability to identify phenotype- genotype associations may be obscured by many factors, including (1) population structure, which can lead to a high level of false significant associations (de Bakker et al. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.