Pre-breeding of wheat (Triticum aestivum L.) for Biomass allocation and drought tolerance.

Abstract

Bread wheat (Triticum aestivum L., 2n=6x=42) is the third most important cereal crop globally after maize and rice. However, its production and productivity is affected by recurrent drought and declining soil fertility. Wheat cultivars with a well-balanced biomass allocation and improved root systems have better water- and nutrient-use efficiency and, hence, increased productivity under dry-land farming systems. The overall objective of this study was to develop breeding populations of wheat with enhanced drought tolerance and biomass allocation under water-limited conditions. The specific objectives of the study were: (i) to evaluate agronomic performance and quantify biomass production and allocation between roots and shoots in selected wheat genotypes in response to different soil water levels to select promising genotypes for breeding for drought tolerance and carbon (C) sequestration, (ii) to determine variance components and heritability of biomass allocation and grain yield related traits among 99 genotypes of bread wheat and triticale (Triticosecale Wittmack) to optimize biomass partitioning for drought tolerance, (iii) to deduce the population structure and genome-wide marker-trait association of yield and biomass allocation traits in wheat to facilitate marker-assisted selection for drought tolerance and C sequestration, and (iv) to estimate the combining ability of selected wheat genotypes and their progenies for agronomic traits, biomass allocation and yield under drought-stressed and non-stressed conditions for future breeding and genetic advancement for drought tolerance and C sequestration. To achieve these objectives, different experiments were conducted. In the first study, 99 wheat genotypes and one triticale accession were evaluated under drought-stressed and non-stressed conditions in the field and greenhouse using a 10×10 alpha lattice design with two replications. Data on the following phenotypic traits were collected: days to heading (DTH), number of productive tillers per plant (NPT), plant height (PH), days to maturity (DTM), spike length (SL), thousand kernel weight (TKW), root and shoot biomass (RB and SB), root to shoot ratio (RS) and grain yield (GY). There was significant (p<0.05) genotypic variation for grain yield and biomass production. The highest grain yield of 247.3 g m-2 was recorded in the genotype LM52 and the least was in genotype Sossognon with 30 g m-2. Shoot biomass ranged from 830g m-2 (genotype Arenza) to 437 g m-2 (LM57), whilst root biomass ranged between 140 g m-2 for LM15 and 603 g m-2 for triticale. Triticale also recorded the highest RS of 1.2, while the least was 0.30 for LM18. Water stress reduced total biomass production by 35% and RS by 14%. Genotypic

variation existed for all measured traits useful for improving drought tolerance, while the calculated RS values can improve accuracy in estimating C sequestration potential of wheat. The following genotypes: BW140, BW141, BW152, BW162, LM26, LM47, LM48, LM71, LM70 and LM75 were selected for further development based on their high grain and biomass production, low drought sensitivity and marked genetic diversity. In the second study, data obtained from the above experiment were subjected to analyses of variance to calculate variance components, heritability and genetic correlations. Significant (p≤0.05) genetic and environmental variation were observed for all the traits except for spike length. Drought stress decreased the heritability of RS from 47 to 28% and GY from 55 to 17%. The genetic correlations between RS with PH, NPT, SL, SB and GY were weaker under drought-stress (r ≤ - 0.50; p<0.05) compared to non-stressed condition, suggesting that lower root biomass under drought stress compromises wheat productivity. The negative genetic correlation between GY and RS (r = -0.41 under drought-stressed and r = -0.33 under non-stressed conditions; p<0.05), low heritability (<42%) and high environmental variance (>70%) for RS observed in this population constitute several bottlenecks for improving GY and RS simultaneously. However, indirect selection for DTH, PH, RB, and TKW, could help optimize RS and simultaneously improve drought tolerance and yield under drought-stressed condition. In the third study, the 99 wheat genotypes and one triticale accession were genotyped using 28,356 DArTseq derived single nucleotide polymorphism (SNPs) markers. Phenotypic and genomic data were subjected to genome wide association study (GWAS). Population structure analysis revealed seven clusters with a mean polymorphic information content of 0.42, showing a high degree of diversity. A total of 54 significant marker-trait associations (MTAs) were identified. Twenty-one of the MTAs were detected under drought-stressed condition and 11% were on the genomic loci where quantitative trait loci (QTLs) for GY and RB were previously identified, while the remainder are new events providing information on biomass allocation. There were four genetic markers, two under each water treatment, with pleiotropic genetic effect on RB and SB that may serve as a means for simultaneous selection. Significant MTAs observed in this study will be useful in devising strategies for marker-assisted breeding to improve drought tolerance and to enhance C sequestration capacity of wheat.

Lastly, 10 better performing and genetically diverse wheat genotypes selected during the first experiment were crossed using a half diallel mating design to generate F1 families. The parents and crosses were evaluated using a completely randomized block design with 2 replications under a controlled environment condition. Significant (p<0.05) genotype by water regime interaction effects were recorded for RB, SB, RS and GY. Root and shoot biomass were reduced by 48 and 37%, respectively, due to drought stress hindering biomass allocation patterns and hence C sequestration potential of the tested genotypes. Further, drought stress reduced RS and GY by 18 and 28%, respectively compared with the non-stressed treatment. Analysis of variance showed that both general combining ability (GCA) and specific combining ability (SCA) effects were significant (p<0.05) in conditioning the inheritance of grain yield and related traits and biomass allocation. Non-additive gene effects were more important in controlling the inheritance of the measured traits under drought-stressed and non-stressed conditions. Parental genotypes LM47 and BW140 had significant and positive GCA effects for root and shoot biomass and GY under drought-stressed conditions. These are recommended for recurrent selection programs to improve the respective traits. The crosses BW141×LM48 and LM47×LM75 were good specific combiners for biomass allocation and GY under drought stress, while BW141×LM48 and LM48×LM47 were good combiners under non-stressed condition. These families are selected for advanced breeding to develop pure line cultivars. The preliminary results suggest that simultaneous improvement of grain yield and root biomass can be realized to improve drought tolerance and C sequestration ability in wheat. Overall, the study detected marked phenotypic and genetic variation among diverse set of wheat genetic resources and candidate crosses for drought tolerance and biomass allocation through field and greenhouse based experiments and genomic analyses. The selected parents and novel crosses are useful for wheat breeding to enhance drought tolerance, yield and yield components and biomass allocation for C sequestration. This is the first study that evaluated biomass allocation in wheat as a strategy to improve drought tolerance and carbon sequestration.