The genetic basis of barley (Hordeum vulgare L.) adaptation to Australian environment

Abstract

The main aim of this study was to identify genomic regions controlling adaptive traits and yield under Australian environment. Three doubled haploid mapping populations developed from inter-crossing three locally adapted elite genotypes (Commander, Fleet and WI4304) were used for the study. The parents were selected based on their long-term performances in southern Australia and are similar in maturity. Field trials were conducted at Minnipa (South Australia), Roseworthy (South Australia) and Swan Hill (Victoria) in 2012 and 2013 cropping seasons. Phenotypic evaluation comprised maturity (Zadoks), early vigour, leaf rolling, leaf waxiness, normalized difference vegetation index (NDVI), chlorophyll content (SPAD), grain plumpness, and grain yield. Three high-density genetic linkage maps were constructed using Genotyping by Sequencing (GBS) and single nucleotide polymorphisms (SNPs) for major phenology genes controlling photoperiod response and vernalization sensitivity. QTL mapping identified 13 maturity QTL, 18 QTL for other adaptive traits including three QTL for leaf rolling, six for leaf waxiness, three for early vigor, four for NDVI, and two QTL for SPAD. Seventeen QTL for grain plumpness and 18 yield QTL explaining from 1.2% to 25.0% of phenotypic variation were found across populations and environments. Significant QTL x environment interaction was observed for all maturity, grain plumpness and yield QTL except QMat.CF-5H.1, QPlum.FW-4H.1 and QYld.FW-2H.1. Seven of the 13 maturity QTL are coincident with known phenology genes. The major phenology genes Ppd and Vrn were not associated with variation in grain plumpness and yield in this study, and adjustment for maturity effect through co-variance analysis had no major effect on yield QTL. Adjustment for phenology genes confirmed six yield per se QTL that are independent of phenology genes. Six new yield QTL were identified in close proximity to phenology genes after phenology adjustment, with stable expression and major effects across environments, explaining up to 57.4% of phenotypic variance. Yield QTL common between two or all three populations were identified on chromosomes 2H and 6H. A yield QTL on chromosome 2H coincident with the HvCEN/EPS2 locus was identified in CW and FW populations. Controlled environment experiments was conducted under long day and short day light conditions using two contrasting recombinant lines selected from the yield QTL region on Chromosome 2H from each population. The result suggested that the yield QTL identified in CW and FW populations on 2H is independent of phenological variation. Further study is required to verify whether this yield QTL is related to HvCEN/EPS2 itself or whether a gene closely linked to the HvCEN locus is responsible for the observed yield xv variation. The three interlinked populations with high-density linkage maps described in this study are a significant resource for examining the genetic basis for barley adaptation in low to medium rainfall Mediterranean type environments. The identification of a QTL for increased yield that is not associated with maturity differences provides an opportunity to apply marker-based selection for grain yield.