Understanding the genetic and physiological basis of transpiration efficiency in Australian wheat

Water availability is one of the prominent limitations in wheat productivity. In the face of a rapidly changing climate with predictions for greater variability of rainfall across Australia’s wheat growing regions there is a clear and growing need to increase the efficiency with which water is used. Wheat is a staple crop for the majority of the world, and Australia is the 3rd largest exporter, making it a crop not only of importance for feeding the world, but also of significant economic importance. Through conventional breeding programs and technological advances in crop production over the last century, and particularly in the last 3-4 decades, grain yields have been able to increase proportionately in response to increasing demand. In the last decade annual yield increases have started to plateau, while population growth and demand for food has continued to grow. By the year 2050 global population is predicted to reach nearly 10 billion people, requiring an increase in food production of up to 60%. In order to maintain global food security, greater yields will need to be achieved in environments with more variable water availability.To address both the increased demand for food and increasing scarcity of water for crop production, Transpiration efficiency (TE) has been proposed as a trait which could increase yields using similar or reduced amounts of water. Although TE has been explored in the past using proxy traits such as carbon isotope discrimination (CID) and stomatal conductance, rapid and accurate measurement of direct water-use at the plant level for large populations has not been undertaken on a large scale, and therefore neither have genetic studies of the underlying genes controlling TE. Here we propose a method for rapidly screening large populations of wheat for water-use and TE, as well as utilising modern genome wide association study (GWAS) techniques to identify and validate multiple quantitative trait loci (QTL) for TE in wheat.A high-throughput methodology for screening large populations of wheat was developed by combining the ‘Pot-In-Bucket’ platform for measurement of water use at the plant level with the new approach of reducing trial duration in order to run multiple experiments within a single season. By running two experiments consecutively in the same cropping season, throughput of the platform was effectively doubled while the results for TE remained highly correlated.This methodology allowed for a large genetically structured mapping population to be grown and screened for TE rapidly before using GWAS to identify multiple QTL. In total, 22 genetic markers representing 20 QTL where identified for TE in a population of 11 families, developed using Suntop as a reference parent. The effect size of these QTL ranged from 0.145 g kg-1 to -0.142 g kg-1 (>6% effect). In a subsequent experiment, 6 of these QTL were validated in a previously untested population from a different genetic background. The parental origins of the donor alleles which conferred either a positive or deleterious effect on TE were also investigated in order to present the families where genes of interest may be obtained and bred into existing material. This genetic information can be used by breeders or researchers to further investigate the genetics of TE in wheat and the potential to rapidly introduce novel genes using modern breeding techniques.Further investigation into these QTL demonstrated that most of these QTL could be detected in a population with a different genetic background, validating the discovery of those QTL and the methodology used to detect them. Another study demonstrated that in a population with common parents it was possible to use genomic markers to predict phenotypic performance. This methodology was not able to successfully predict phenotypic performance in populations from different genetic backgrounds or environmental conditions which differed considerably from the original testing environment