Yield effects of flowering time genes in Mediterranean conditions

Abstract

Evidences of QTL studies and biogeography of landraces in the Western Mediterranean suggest a large role of vernalization and photoperiod genes on agronomic performance of barley in the region. Two hypotheses have been put to test, a possible agronomic advantage of the dominant allele of VrnH2 over the recessive (vrnH2), and a better adaptation of a haplotype combining a reduced vernalization requirement (VrnH2/VrnH1-4) and an active PpdH2 over a typical winter type (VrnH2/vrnH1/ppdH2). Sets of lines were selected from two biparental populations with the aid of molecular markers, according to their haplotypes of flowering time genes. Four sets of 25 lines were made in SBCC097/Plaisant, considering their haplotypes at VrnH1 and PpdH2. Eight sets of 8-16 lines were prepared in SBCC145/Beatrix, as per their haplotypes at VrnH2, VrnH3 and PpdH1 (also all the lines had the Beatrix, spring allele at VrnH1). Equal amounts of seeds of these 12 sets of lines were bulked, and sown in field trials for up to three years and two sowing dates per year, normal (November) and late (early February). There were significant differences in grain yield among the bulks. There was a significant yield advantage of the VrnH2 over vrnH2 bulks in SBCC145/Beatrix (16%) and of the PPdH2 bulks in SBCC097/Plaisant (6%). Although the differences were significant across sowings, they were even higher in the autumn one. Small differences in heading date could not account for the grain yield differences found. The relationship of these results with winter temperatures in the experimental locations will be studied and reported. These differences could be due to either a pleiotropic effect of the genes targeted by this study, or to other so far undetected linked genes with large agronomic effect. These hypotheses will be tested further over the next years.