| Maize is the main crop in China,which plays an important role in food security.Improving the heterosis of hybrid combinations is an effective way to increase maize yield.Heterosis can be measured using a variety of traits,such as plant height,yield,leaf area and so on.Hundred grain weight(HGW)is one of the important components of maize yield traits,and the HGW of hybrids is significantly higher than that of their parents.However,at present,there are relatively few studies on using HGW to measure heterosis and to explore the genetic mechanism of HGW heterosis.In this study,384 test cross species were constructed using 4 testers as male parents and 96 temperate inbred lines as female parents following NCII design.Through the identification of HGW phenotype and parent genotype in multiple environments,the HGW heterosis of maize was analyzed from three aspects: hybrid performance,combining ability and heterosis by using the methods of genome-wide association analysis(GWAS)and whole-genome selection(GS),and the genetic basis of HGW heterosis was analyzed.The main results are as follows:1.Through GWAS analysis of maize inbred lines with HGW,63 significant markers were identified.F1 hybrids were analyzed by GWAS using four models(additive model,dominant model,overdominant model and recessive model),and25,51,456,71 significant markers were located respectively.The phenotypic variation rates explained by significant markers in inbred lines and hybrids were calculated by multiple regression.The results showed that 32 significant markers located in inbred lines explained 70.3% phenotypic variation,and 16 significant markers located in F1 hybrids explained 52.6% phenotypic variation by using additive model,while dominant model,overdominant model and recessive model were used to explain only 36.8%,39.8% and 18.5% phenotypic variation.By calculating the relationship between the number of accumulated superior genotypes and phenotype of inbred lines and F1 hybrids,it was found that there was a significant positive correlation(the correlation coefficients were 0.64,0.46respectively).The above results proved the accuracy of GWAS mapping results.The results of genetic variance analysis using HGW phenotype of hybrids showed that the additive variance accounted for 75.5%,the dominant variance accounted for 15.7%,and the dominant × dominant interaction variance accounted for 0.8%.There is no additive × additive interaction variance and additive × dominant interaction variance,indicating that HGW traits are mainly controlled by geneadditive effects.The GS analysis results show that the prediction accuracy of the GS of the hybrid population is 74%,which is greater than the prediction accuracy of the molecular marker-assisted selection(MAS)using significant markers71.2%.The accuracy of MAS prediction using significant markers in inbred lines and hybrid populations(76.8% and 71.2%,respectively)was higher than that in MAS prediction using random selection markers(17% and 44.5%,respectively).The accuracy of GWAS analysis results of parent inbred lines and F1 hybrids was verified again.2.Through the GWAS analysis of the general combining ability(GCA)of maize inbred lines with HGW,the results showed that 82 markers were significantly associated with GCA,and GCA was mainly affected by gene additive effect.Through the GWAS of the special combining ability(SCA)of F1 hybrids,the results showed that the additive model did not locate any significant markers,and the dominant model,super-dominant model and recessive model were located at 25,543 and 62 significant markers.It was found that the additive model performed worst and the dominant model performed the best among the four models,indicating that SCA was mainly affected by the non-additive effects of genes,especially the dominant effects.3.By performing GWAS on the HGW heterosis(mid-parent heterosis),the results showed that the dominant model did not locate any significant markers,and only 14 dominant interactions were located in the dominant interaction model.there were no significant interactions in other epistatic interaction models.The variance of different genetic effect models was calculated,and the results showed that dominant variance and additive × additive interaction variance accounted for a large proportion in each model.In the prediction accuracy analysis of heterosis GS,the prediction accuracy of additive × additive interaction effect was 63.5%higher than that of dominant effect alone was 33.2%,the prediction accuracy of dominance and additive × additive interaction effect was 58.6%,the prediction accuracy of dominance,additive × additive interaction and additive × dominance interaction effect was 59.1%,and the prediction accuracy of dominance,additive× additive interaction,additive × dominance interaction and dominance ×dominance interaction effect was 57.5%.The above results proved that the heterosis of HGW was mainly controlled by gene interaction.In this study,NCII design was used to analyze the hybrid performance,heterosis,general combining ability(GCA)and special combining ability(SCA)of HGW maize by GWAS.Some QTL loci controlling the above traits were found,and it was proved that using these significant loci for MAS can improve theprediction accuracy.In this study,the genetic mechanism of HGW heterosis in maize was studied from different angles,which lays a foundation for the study of maize heterosis theory and molecular mechanism,and provides a theoretical basis for using GS technology to improve the breeding efficiency of strong heterosis. |
PDF Full Text Request