Genetic basis of heterosis in a common wheat cross withstrong-heterosis

Abstract

<p indent="0mm">Heterosis refers to the phenomenon that a heterozygous hybrid exhibits superior performances over its parents, which has been suggested as a solution to overcome yield stagnation of wheat. However, the molecular basis of wheat heterosis is still poorly understood. We have previously found that a cross derived from the winter common wheat Nongda3338 (ND3338) and Jingdong 6 (JD6) showed strong mid-parent heteroses on plant height (PH) and thousand grain weight (TGW). In this study, we further aimed to investigate the genetic underpinnings of PH and TGW using an “immortalized F₂” (IF₂) derived from the ND3338/JD6 doubled haploid (DH) population. The PH and TGW of DH and IF₂ populations were evaluated across two different environments, and the mid-parent heterosis value (HV) was used to measure heterosis. The direction and magnitude of HV greatly varied between PH and TGW across the environments. A two-way analysis of variance (ANOVA) was performed for PH and TGW in the IF₂ population, revealing that genetic variation was the main source, despite the significant variations found from the environments and genotype by environment interaction. The broad sense heritability of PH and TGW in IF₂ population was 94.05% and 88.14%, respectively. Pearson’s correlation coefficients of PH, TGW, HV_PH and HV_TGW in the IF₂ population were estimated based on the best linear unbiased prediction (BLUP) values of two environments, showing that PH was significantly positively correlated with TGW and HV_TGW, TGW was significantly positively correlated with HV_TGW and negatively correlated with HV_PH, while HV_PH had a strong positive correlation with HV_TGW. The genotypes of the IF₂ population were deduced from corresponding DH parents. Based on the inclusive composite interval mapping method, quantitative trait locus (QTL) mapping was separately conducted with PH, TGW, HV_PH and HV_TGW using the QTL IciMapping software. We identified a total of 32 and 25 QTL that exhibited additive, dominant and overdominant pattern associated with PH and TGW, which accounted for 0.54%–36.05% and 0.87%–25.78% of the variance, respectively. Consistent with the results found in the ND3338/JD6 DH population, major and stable QTL on chromosomes 2D (<italic>QPh</italic>.<italic>2D</italic>.<italic>1</italic> and <italic>QTgw</italic>.<italic>2D</italic>), 4B (<italic>QPh</italic>.<italic>4B</italic>.<italic>1</italic> and <italic>QTgw</italic>.<italic>4B</italic>.<italic>1</italic>) and 4D (<italic>QPh</italic>.<italic>4D </italic>and <italic>QTgw</italic>.<italic>4D</italic>) that were coincident with the dwarfing genes <italic>Rht1</italic>, <italic>Rht2</italic> and <italic>Rht8</italic> were detected, suggesting that the IF₂ population design is feasible and reliable for identifying the genetic components of heterosis. Furthermore, 13 QTL for HV_PH and 15 QTL for HV_TGW were identified, of which most of these loci exhibited overdominant pattern. However, only few of the QTL controlling PH and TGW were overlapped with those controlling HV_PH and HV_TGW, suggesting potentially different genetic mechanisms between phenotypic traits and heterosis. In addition, epistasis analysis for PH and TGW revealed that four types of digenic interactions, including additive by additive (AA), additive by dominance (AD), dominance by additive (DA) and dominance by dominance (DD), were involved in the genetic basis of hybrid performance and heterosis in the IF₂ population. Notably, the DD-oriented QTL with the greatest interactive effects possessed the highest frequency. Overall, this study provides strong evidence for supporting the classical genetic hypotheses of heterosis (dominance, overdominance, and epistasis) with an IF₂ population, establishing the foundation to develop and select crosses with strong heterosis.