The use of CRISPR/LbCpf1 and CRISPR/xCas9 systems in wheat have not yet been reported. In this study, we compared the efficiencies of three CRISPR editing systems (SpCas9, LbCpf1, and xCas9), and three different promoters (OsU6a, TaU3, and TaU6) that drive single-guide (sg)RNA, which were introduced into wheat via Agrobacterium-mediated transformation. The results indicated that TaU3 was a better choice than OsU6a or TaU6. The editing efficiency was higher using two sgRNAs than one sgRNA, and mutants with a large fragment deletion between the two sgRNAs were produced. The LbCpf1 and xCas9 systems could both be used successfully. Two endogenous genes, TaWaxy and TaMTL, were edited with high efficiency by the optimized SpCas9 system, with the highest efficiency (80.5%) being achieved when using TaU3 and two sgRNAs to target TaWaxy. Rates of seed set in the TaMTL-edited T₀ transgenic plants were much lower than that of the wild-type. A haploid induction rate of 18.9% was found in the TaMTL-edited T₁ plants using the CRISPR/SpCas9 system. Mutants with reverse insertion of the deleted sequences of TaMTL and TaWaxy between the two sgRNAs were identified in the edited T₀ plants. In addition, wheat grains lacking embryos or endosperms were observed in the TaMTL-edited T₁ generation.

Summary: Genotype specificity is a big problem lagging the development of efficient hexaploid wheat transformation system. Increasingly, the biosecurity of genetically modified organisms is garnering public attention, so the generation of marker‐free transgenic plants is very important to the eventual potential commercial release of transgenic wheat. In this study, 15 commercial Chinese hexaploid wheat varieties were successfully transformed via an Agrobacterium‐mediated method, with efficiency of up to 37.7%, as confirmed by the use of Quickstix strips, histochemical staining, PCR analysis and Southern blotting. Of particular interest, marker‐free transgenic wheat plants from various commercial Chinese varieties and their F₁ hybrids were successfully obtained for the first time, with a frequency of 4.3%, using a plasmid harbouring two independent T‐DNA regions. The average co‐integration frequency of the gus and the bar genes located on the two independent T‐DNA regions was 49.0% in T₀ plants. We further found that the efficiency of generating marker‐free plants was related to the number of bar gene copies integrated in the genome. Marker‐free transgenic wheat plants were identified in the progeny of three transgenic lines that had only one or two bar gene copies. Moreover, silencing of the bar gene was detected in 30.7% of T₁ positive plants, but the gus gene was never found to be silenced in T₁ plants. Bisulphite genomic sequencing suggested that DNA methylation in the 35S promoter of the bar gene regulatory region might be the main reason for bar gene silencing in the transgenic plants.

Summary: Kernel size is an important trait determining cereal yields. In this study, we cloned and characterized TaDA1, a conserved negative regulator of kernel size in wheat (Triticum aestivum). The overexpression of TaDA1 decreased the size and weight of wheat kernels, while its down‐regulation using RNA interference (RNAi) had the opposite effect. Three TaDA1‐A haplotypes were identified in Chinese wheat core collections, and a haplotype association analysis showed that TaDA1‐A‐HapI was significantly correlated with the production of larger kernels and higher kernel weights in modern Chinese cultivars. The haplotype effect resulted from a difference in TaDA1‐A expression levels between genotypes, with TaDA1‐A‐HapI resulting in lower TaDA1‐A expression levels. This favourable haplotype was found having been positively selected during wheat breeding over the last century. Furthermore, we demonstrated that TaDA1‐A physically interacts with TaGW2‐B. The additive effects of TaDA1‐A and TaGW2‐B on kernel weight were confirmed not only by the phenotypic enhancement arising from the simultaneous down‐regulation of TaDA1 and TaGW2 expression, but also by the combinational haplotype effects estimated from multi‐environment field data from 348 wheat cultivars. A comparative proteome analysis of developing transgenic and wild‐type grains indicated that TaDA1 and TaGW2 are involved in partially overlapping but relatively independent protein regulatory networks. Thus, we have identified an important gene controlling kernel size in wheat and determined its interaction with other genes regulating kernel weight, which could have beneficial applications in wheat breeding.