Wheat (Triticum aestivum L.) is one of the major food crops in the world, and is also the second-largest food crop in China. The development of new wheat varieties with high and stable yield is one of the important targets for wheat breeders worldwide. As grain weight is the key determinant for grain yield, cloning and characterization of the major genes controlling grain weight is the basis of genetic improvement of yield and the precondition for molecular design breeding in wheat. Previous studies from our lab have identified a major genomic region on chromosome 4A with stable loci for thousand grain weight (TGW) across different environments in the ND3338/JD6 doubled haploid (DH) population. Here, the QTgw-4A was studied in more depth by dissection, validation, fine mapping and candidate gene prediction.

Sixteen co-dominant SSR markers were developed to construct saturated genetic linkage map of QTgw-4A region, and five gene markers (TaARGOS-4A, TaSnRK2.10, TaCWI-4A, TaTGW6 and TaGS3-4A) associated with grain weight previously reported were anchored on chromosome 4A. Using the ND3338/JD6 DH population, QTgw-4A was validated and mapped into a 5.73 cM interval flanked by Xcau4A_11 and Xcau4A_13 by CIM and ICIM methods, suggesting that QTgw-4A is a novel QTL loci for TGW.

To fine map QTgw.4A, five BC3F3:4 segregating populations from BC3F3 plants covering different heterozygote fragments of the QTgw.4A region were generated. Thus, the QTgw.4A was finely mapped into a physical interval smaller than 10 Mb. The Wheat660K SNP array was used to analyze genetic similarities of five homozygous BC3F5 NIL pairs (4A-/4A+). Compared with NIL (4A-), NIL (4A+) showed an increase in TGW ranging from 5.73 to 6.74% and a decrease in GNS ranging from 7.73 to 10.17% in field trials. It was also found that NIL (4A+) showed significantly higher rate of grain filling than NIL (4A-) during the latter stage of grain filling. Moreover, it is predicted that a total of 49 candidate genes were identified within the finely mapped interval of QTgw-4A, combining with the transcriptome analysis of the 6-DAP grains of NILs (4A/4A+) and the dynamic expression profiles of spike and grain tissue-specific developmental time course in an expVIP expression database. Our results provide a basis for further fine-mapping and map-based cloning of the major QTL for TGW and unraveling of the mechanisms underlying TGW-GNS tradeoffs in wheat, which would help to fine-tune these two components and maximize grain yield for wheat breeders.