Physiological and Genetic Analysis of Water-Saving Traits in Wheat Reveal Links Between Shoot, Root Hydraulics and Development.

Drought-prone environments tend to exhibit high levels of atmospheric vapor pressure deficit (VPD), which is the main driver of crop water loss via transpiration (TR). Recently, it was hypothesized that “water-saving” cultivars displaying TR decreases in response to high VPD under well-watered conditions should lead to enhanced drought tolerance under terminal drought conditions by increasing the availability of stored soil moisture during the critical grain filling phase. On wheat, we have shown that such strategy explained the drought tolerance of an Australian breeding line (RAC875). Subsequent experiments indicated that the lower TR exhibited by RAC875 under high VPD was the result of a decreased root hydraulic conductivity relative to a check cultivar (Kukri), likely stemming from a lack of an aquaporin population and a reduced central metaxylem size. To unravel the genetic basis of such responses, we phenotyped TR response curves to VPD for 143 double haploid lines resulting from a cross between RAC875 and Kukri, along with other traits. All traits were controlled by 68 QTL, with heritabilities ranging from 0.61 to 0.91. Importantly, the QTL that explained individually the highest percentage of the genetic variance (>25%), controlled the slope (SLP) of TR responses to VPD. The peak region of that QTL was mapped to 72 genes, 12 of which were directly drought-related. An independent RNASeq experiment revealed that 9 of these genes were root-specific. Most of these genes were involved in root and xylem development, consistently with the physiological mechanism identified for parent line RAC875. In addition, phenology genes Ppd-D1 and Ppd-B1 strongly influenced SLP. Combined, these findings reveal a strong interplay between shoot and root hydraulic traits in determining drought tolerance while pointing to the key influence of crop phenology on such traits. Such considerations will be particularly useful to crop modeling, breeding and phenomics-based research programs.