Ability of alleles of PPD1 and VRN1 genes to predict flowering time in diverse Australian wheat (Triticum aestivum) cultivars in controlled environments

Abstract

Flowering time of wheat (Triticum aestivum L.) is a critical determinant of grain yield. Frost, drought and heat stresses from either overly early or overly late flowering can inflict significant yield penalties. The ability to predict time of flowering from different sowing dates for diverse cultivars across environments in Australia is important for maintaining yield as autumn rainfall events become less reliable. However, currently there are no models that can accurately do this when new cultivars are released. Two major Photoperiod1 and three Vernalisation1 development genes, with alleles identified by molecular markers, are known to be important in regulating phasic development and therefore time to anthesis, in response to the environmental factors of temperature and photoperiod. Allelic information from molecular markers has been used to parameterise models that could predict flowering time, but it is uncertain how much variation in flowering time can be explained by different alleles of the five major genes. This experiment used 13 elite commercial cultivars of wheat, selected for their variation in phenology and in turn allelic variation at the major development genes, and 13 near-isogenic lines (NILs) with matching multi-locus genotypes for the major development genes, to quantify how much response in time to flowering could be explained by alleles of the major genes. Genotypes were grown in four controlled environments at constant temperature of 22°C with factorial photoperiod (long or short day) and vernalisation (±) treatments applied. NILs were able to explain a large proportion of the variation of thermal time to flowering in elite cultivars in the long-day environment with no vernalisation (97%), a moderate amount in the short-day environment with no vernalisation (62%), and less in the short-day (51%) and long-day (47%) environments with vernalisation. Photoperiod was found to accelerate development, as observed in a reduction in phyllochron, thermal time to heading, thermal time to flowering, and decreased final leaf numbers. Vernalisation response was not as great, and rates of development in most genotypes were not significantly increased. The results indicate that the alleles of the five major development genes alone cannot explain enough variation in flowering time to be used to parameterise gene-based models that will be accurate in simulating flowering time under field conditions. Further understanding of the genetics of wheat development, particularly photoperiod response, is required before a model with genetically based parameter estimates can be deployed to assist growers to make sowing-time decisions for new cultivars.