Identiying traits and molecular markers for nitrogen-use efficiency in a wheat nested association mapping population

The global population is predicted to reach 9.1 billion in 2050 and the current rate of genetic progress in wheat yields is not sufficient to meet future demand. Nitrogen fertiliser is a key determinant of yield and represents a significant cost for the grower and has environmental impacts through nitrate leaching and N2O emissions. Therefore, there is need to develop wheat cultivars with improved Nitrogen-Use Efficiency (NUE; grain dry matter yield per unit of N availability). NUE is divided into two components, N-uptake efficiency, and N-utilization efficiency. Developing cultivars which have higher grain yield but use N efficiently may allow farmers to maximise income and minimise the pollution risk by reducing N fertilizer application. Due to wheat selection in plant breeding for high grain yield under high N fertility conditions, genetic diversity of modern wheat for NUE traits is reduced. Therefore, exploration for novel sources of genetic variation for NUE is needed. Landrace-derived lines which provide novel sources of genes could be used by plant breeders to improve NUE of modern bread wheat. The overall objective of the present project is to investigate favorable physiological traits for N-use efficiency in wheat landrace-derived lines and their genetic regulation. Two field experiments at the Bunny field station, two glasshouse experiments and one hydroponic experiment were conducted at the University of Nottingham. The current work can be divided into four main components. (i) quantifying the genetic variation for above-ground biomass, grain yield, NUE and its components in landrace-derived lines and modern bread wheat cultivar Paragon; (ii) Identifying novel physiological traits determining genetic variation in biomass, NUE and NUE components; (iii) exploration of root system architectural traits for improved N-Uptake efficiency (NUpE; above-ground N uptake / N available) of wheat landrace-derived lines compared to modern wheat; and (iv) identifying marker-trait association (MTAs) for the NUE traits. The experimental programme utilized a Nested Associated Mapping population of 286 genotypes comprising landrace-derived lines plus the elite bread wheat parent Paragon. These genotypes were assessed for biomass, grain yield, NUE traits and physiological traits under a moderate N stress condition in the field in two seasons. In subsets of germplasm, NUE traits including leaf photosynthesis traits were assessed in 14 landrace-derived lines and Paragon under HN and LN conditions in two glasshouse experiments; and root seedling architecture traits were assessed in 30 landrace-derived lines and Paragon under HN and LN under controlled-environment hydroponic conditions. Significant variation for most of the traits was identified among wheat genotypes where modern bread wheat Paragon had higher grain yield and generally shorter plant height compared to landrace-derived lines. Higher grain yield of modern elite bread wheat Paragon was due to improved harvest index compared to the landrace-derived lines. Lower NUE in landraces was therefore mainly due to lower harvest index as expected because the landraces were not bred for grain yield potential. Three lines (PxW396-62, PxW349-31 and PxCIM47-113) had higher NUpE than Paragon under moderate N stress conditions. In the glasshouse experiments, two lines (PxW223-89 and ParxPfau-59) under HN showed significantly later onset of flag-leaf senescence than Paragon associated with N-utilization efficiency (NUtE; grain yield DM/above-ground N uptake) and one line (PxW420-32) under LN conditions. However, under both, HN and LN conditions, generally, senescence traits were not well correlated with each of yield and biomass in the glasshouse experiments. This was also the case in the field experiments. These results indicated the grain yield of the NAM lines was mainly limited by sink rather than source in the experiments in the present study. In the glasshouse experiments, NUtE of modern wheat Paragon was higher than most landrace-derived lines mainly related to harvest index. Generally, NUpE explained more of the genetic variation in NUE than NUtE in the project experiments. Significant variation in root system architecture traits was identified in landrace-derived lines compared to Paragon, with landrace-derived lines showing higher expression for seminal root number, average primary root tip angle, total primary root length, average lateral root length and seminal root angle under N limitation compared to Paragon. A trend for a correlation was found between primary root tip angle and N-uptake efficiency in the field experiments (r=0.36, P=0.054). There was also a trend for a positive correlation between average primary root length and each of grain yield and NUE (r=0.33, P=0.07). Total lateral root length was also positively correlated with biomass (r=0.38, P=0.03) and there was a trend for a positive association between average lateral root length and biomass (r=0.33, P=0.07). Present results suggested that seedling root architectural traits could potentially be used as selection criteria for selecting genotypes for higher NUpE at maturity. Genome-Wide Association Study (GWAS) was carried out on the phenotypic data in the field experiments for the NAM population. In 2017, a total of 127 MTAs, including co-located MTAs, with a - log10 P-value of greater 4 were detected which explained from 5 to 15 % of the phenotypic variation. Major marker-trait associations (MTAs) were identified for anthesis date/ harvest index on chr 1A, grains m-2/1,000 grain weight on each of chr 1D, 2A, 2D and 5A, NUE/grains m-2 on chr 2B, NUE/AD/NDVI senescence rate/grain yield on chr 5A and AD/ NDVI senescence rate on chr 5A. In 2018, a total of 52 MTAs, including co-located MTAs, with a - log10 P-value of greater 4 were detected which explained from 4 to 10 % of the phenotypic variation. Major MTAs were identified for NDVI at anthesis/NDVI senescence rate on chr 2A and grain yield on chr 3B. Overall, the strongest co-localized genomic region was identified on chromosome 5A at 503137755 bp associated with four traits (grain yield, anthesis date, NUE and NDVI senescence rate). The genomic co-location of loci 503137755 bp with these four traits suggests a pleotropic effect of the 5A Vrn-A1 gene underlying these traits. A promising novel MTA for grain yield and NUE was identified on chr 3B. To our knowledge, the present work is the first published study in bread wheat that reports GWAS results on N-use efficiency in wheat crop using landrace-derived lines