Physiological and environmental dissection of developmental drivers for predicting heading date in wheat based on Vrn1, Ppd1 and Eps-D1 allelic characterization

Abstract

Heading time is a critical feature determining wheat performance. Over the years, average global temperatures have increased, leading to adjustments in agronomic practices such as sowing dates and cultivar election. Bread wheat development is regulated by the effect of vernalization (Vrn), photoperiod sensitivity (Ps), and earliness per-se (Eps). The aim of this study was (i) to quantify the physiology behind wheat development for a wide range of contrasting commercial genotypes, and (ii) to design a simple wheat phenology prediction model that considers vernalization and photoperiod effects. Seventeen wheat genotypes that vary in VERNALIZATION-1 (Vrn1), PHOTOPERIOD-1 (Ppd1), and EARLINESS PER-SE D1 (Eps-D1) alleles were classified into four groups: Winter Sensitive (WS), Winter Insensitive (WI), Spring Sensitive (SS), and Spring Insensitive (SI). Field trials with vernalization treatments were carried out in different sites, years, and sowing dates. Genotypes with Vrn1 triple winter alleles exhibited the highest vernalization requirements, while cycle duration in spring genotypes remained constant. Photoperiod sensitivity was higher for those genotypes molecularly classified as Ppd1 sensitive but a differential response was evident when control and vernalization treatments were compared: after 6-week vernalization, Ps in WI decreased significantly without significant differences for WS, SS, and SI. For winter genotypes, the optimum photoperiod (Po) was the longest for the 6-week vernalized groups, and within that level of vernalization WI showed the longest Po values. The variability in Ps and Po values could be explained by the interaction between photoperiod and vernalization over Vrn1 upregulation and their consequent effect on wheat cycle duration. A wheat phenological model using photoperiod and vernalization correction factors was calibrated to predict heading dates in the Argentine wheat region. The model accurately predicted heading date with an error of 8.1, 6.5, 6.2, and 7.0 days for WI, WS, SI, and SS, respectively. The development of a model that uses simple universal process-based parameters is crucial for farmers, advisors, and researchers as it increases the precision in wheat cycle arrangement, but in a relatively parsimonious fashion.