Genomic Changes in Response to 110 Cycles of Selection for Seed Protein and Oil Concentration in Maize

Identification of the genetic mechanisms underlying phenotypic variation between and within populations is essential to understanding natural selection and evolution. Changes in gene expression are an important source of phenotypic variation and may arise as a result of a combination of cisregulatory and trans-regulatory variation within complex genetic networks (Wittkopp, 2005). This complexity has hindered identification of genomic targets of both natural and artificial selection. Long-term phenotypic selection experiments in plants and animals provide an invaluable resource for studying evolutionary change and regulatory variation, with implications for enhancing the efficiency of genetic gain in plant and animal breeding (Dudley and Lambert, 2004). Plants have the additional unique advantage of regeneration, where replicate selection experiments may be conducted in parallel. The Illinois Long-Term Selection Experiment for grain composition is a classic in plant genetics; 800 cycles of selection have been conducted from the same source population during the past century (Figure 12.1). Previous quantitative trait loci (QTL) mapping studies suggest that changes in both protein and oil concentration in the Illinois experiment are governed by many genes (Laurie et al., 2004; Dudley et al., 2007) with small phenotypic effect. However, there is also evidence that variation at relatively few loci can be associated with significant phenotypic differences. Advances in high-density genotyping coupled with functional genomics approaches enable a more thorough understanding of the genetic basis for the dramatic phenotypic responses observed. In this chapter, we provide an update on more recent phenotypic responses within the Illinois Long-Term Selection Experiment and related populations, a review of past efforts to characterize the genetic architecture of selection response, and a preliminary view of gene expression changes resulting from divergent selection for grain protein, and we introduce the use of promoter-reporter genes as tools to understand how selection has altered the regulation of seed storage protein genes. Collectively, these approaches illustrate that long-term selection for kernel composition has targeted variation in both trans-acting regulatory factors and specific genes known to influence oil or protein concentrations.