A Gene Expression Biomarker Reflects Drought Stress Intensity in Controlled and Natural Environments

To understand crop responses to drought or drought-interacting stresses, plant molecular physiologists, eco-physiologists, agronomists and breeders need to estimate the water available to plants in the soil during their growth. Commonly, they use soil- and plant-based measurements such as pre-dawn leaf water potential. Because transcriptomic responses to abiotic environment such as water deficit reflect the plant water status, they could be used as a molecular biomarker.

Using sunflower (Helianthus annuus) as a model species to study transcriptomic response to water deficit both in glasshouse and field conditions, we specifically identified three genes highly correlated to plant water status estimated with the pre-dawn leaf water potential, fraction of transpirable soil water, soil water content or fraction of total soil water in controlled conditions. We developed a generalized linear model to estimate these classical water status indicators from the expression levels of these three genes in controlled conditions. This estimation was independent of the genotype (XRQ, PSC8, Inedi, Melody), and the stage (pre- or post-flowering) of the plant. We further validated this gene expression biomarker for plant water status in field conditions on the four genotypes in three different trials and a large range of water status. Finally, we studied the circadian variation of the biomarker genes and were able to correct their expressions values for a large diurnal sampling period.

Together, we developed a gene expression biomarker able to estimate the plant water status in glasshouse and field conditions, independent of the tested genotypes and the flowering stage.