Extensive Variation in the Density and Distribution of DNA Polymorphism in Sorghum Genomes

Joseph Evans,John Mullet,Josie Hilley,Ryan F Mccormick,William Rooney,Daryl Morishige,Sara N Olson,Brock Weers,Patricia Klein,Giovanni G Vendramin

doi:10.1371/journal.pone.0079192

Abstract

Sorghum genotypes currently used for grain production in the United States were developed from African landraces that were imported starting in the mid-to-late 19th century. Farmers and plant breeders selected genotypes for grain production with reduced plant height, early flowering, increased grain yield, adaptation to drought, and improved resistance to lodging, diseases and pests. DNA polymorphisms that distinguish three historically important grain sorghum genotypes, BTx623, BTx642 and Tx7000, were characterized by genome sequencing, genotyping by sequencing, genetic mapping, and pedigree-based haplotype analysis. The distribution and density of DNA polymorphisms in the sequenced genomes varied widely, in part because the lines were derived through breeding and selection from diverse Kafir, Durra, and Caudatum race accessions. Genomic DNA spanning dw1 (SBI-09) and dw3 (SBI-07) had identical haplotypes due to selection for reduced height. Lower SNP density in genes located in pericentromeric regions compared with genes located in euchromatic regions is consistent with background selection in these regions of low recombination. SNP density was higher in euchromatic DNA and varied >100-fold in contiguous intervals that spanned up to 300 Kbp. The localized variation in DNA polymorphism density occurred throughout euchromatic regions where recombination is elevated, however, polymorphism density was not correlated with gene density or DNA methylation. Overall, sorghum chromosomes contain distal euchromatic regions characterized by extensive, localized variation in DNA polymorphism density, and large pericentromeric regions of low gene density, diversity, and recombination.

Highlights

Genome sequencing is being used to characterize genome-wide patterns of genetic variation and inheritance, to help identify regions of genomes under selection, and to aid in the discovery of single nucleotide polymorphisms (SNPs) and alleles associated with quantitative trait loci (QTL) [1,2,3,4,5]
A recombinant inbred line (RIL) population derived from BTx642 and Tx7000 is available and has been used to map QTL for the post-flowering drought tolerance stay-green trait and other traits [23,25]
Comparison of re-sequenced sweet sorghum genomes to BTx623 revealed high levels of genic polymorphism that could be associated with differences in traits in sweet sorghums compared with the grain sorghum BTx623 [38]

Summary

Introduction

Genome sequencing is being used to characterize genome-wide patterns of genetic variation and inheritance, to help identify regions of genomes under selection, and to aid in the discovery of SNPs and alleles associated with quantitative trait loci (QTL) [1,2,3,4,5]. Sequences from target genomes can be readily aligned to reference genome sequences, enabling the identification of sequence variants [6] This re-sequencing approach identified over a million sequence variants between closely related rice cultivars and between sorghum genotypes, including gene-coding variants that could affect function [5,7]. The genome sequences can be compared to help identify regions that have undergone selection [10,12,13] This is especially relevant for cultivated sorghums in the United States because the elite sorghums used in hybrid grain production are products of domestication in Africa ,4,000 to 6,000 years ago, followed by many subsequent generations of breeding and selection in different locations [14]. Sorghum races are prevalent in different regions of Africa, reflecting the diverse agro-eco-environments to which they are adapted: Kafir in southern Africa, Guinea in west Africa, Caudatum in northcentral Africa, Durra in Sudan/Ethiopia/Egypt and Bicolor, the most ancient race, in the northeast quadrant of Africa, the center of origin for sorghum [14]

Methods

Results

Conclusion