Kernel size (KS) and kernel weight play a key role in wheat yield. Phenotypic data from six environments and a Wheat55K single-nucleotide polymorphism array–based constructed genetic linkage map from a recombinant inbred line population derived from the cross between the wheat line 20828 and the line SY95-71 were used to identify quantitative trait locus (QTL) for kernel length (KL), kernel width (KW), kernel thickness (KT), thousand-kernel weight (TKW), kernel length–width ratio (LWR), KS, and factor form density (FFD). The results showed that 65 QTLs associated with kernel traits were detected, of which the major QTLs QKL.sicau-2SY-1B, QKW.sicau-2SY-6D, QKT.sicau-2SY-2D, and QTKW.sicau-2SY-2D, QLWR.sicau-2SY-6D, QKS.sicau-2SY-1B/2D/6D, and QFFD.sicau-2SY-2D controlling KL, KW, KT, TKW, LWR, KS, and FFD, and identified in multiple environments, respectively. They were located on chromosomes 1BL, 2DL, and 6DS and formed three QTL clusters. Comparison of genetic and physical interval suggested that only QKL.sicau-2SY-1B located on chromosome 1BL was likely a novel QTL. A Kompetitive Allele Specific Polymerase chain reaction (KASP) marker, KASP-AX-109379070, closely linked to this novel QTL was developed and used to successfully confirm its effect in two different genetic populations and three variety panels consisting of 272 Chinese wheat landraces, 300 Chinese wheat cultivars most from the Yellow and Huai River Valley wheat region, and 165 Sichuan wheat cultivars. The relationships between kernel traits and other agronomic traits were detected and discussed. A few predicted genes involved in regulation of kernel growth and development were identified in the intervals of these identified major QTL. Taken together, these stable and major QTLs provide valuable information for understanding the genetic composition of kernel yield and provide the basis for molecular marker–assisted breeding.

Background High yield and quality are essential goals of wheat (Triticum aestivum L.) breeding. Kernel length (KL), as a main component of kernel size, can indirectly change kernel weight and then affects yield. Identification and utilization of excellent loci in wheat genetic resources is of great significance for cultivating high yield and quality wheat. Genetic identification of loci for KL has been performed mainly through genome-wide association study in natural populations or QTL mapping based on genetic linkage map in high generation populations. Results In this study, an F3 biparental population derived from the cross between an EMS mutant BLS1 selected from an EMS-induced wheat genotype LJ2135 (derived from the hybrid progeny of a spelt wheat (T. spelta L.) and a common wheat) mutant bank and a local breeding line 99E18 was used to rapidly identify loci controlling KL based on Bulked Segregant Analysis (BSA) and the wheat 660 K single-nucleotide polymorphism (SNP) array. The highest ratio of polymorphic SNPs was located on chromosome 4A. Linkage map analysis showed that 33 Kompetitive Allele Specific PCR markers were linked to the QTL for KL (Qkl.sicau-BLE18-4A) identified in three environments as well as the best linear unbiased prediction (BLUP) dataset. This QTL explained 10.87—19.30% of the phenotypic variation. Its effect was successfully confirmed in another F3 population with the two flanking markers KASP-AX-111536305 and KASP-AX-110174441. Compared with previous studies and given that the of BLS1 has the genetic background of spelt wheat, the major QTL was likely a new one. A few of predicted genes related to regulation of kernel development were identified in the interval of the detected QTL. Conclusion A major, novel and stable QTL (Qkl.sicau-BLE18-4A) for KL was identified and verified in two F3 biparental populations across three environments. Significant relationships among KL, kernel width (KW) and thousand kernel weight (TKW) were identified. Four predicted genes related to kernel growth regulation were detected in the interval of Qkl.sicau-BLE18-4A. Furthermore, this study laid foundation on subsequent fine mapping work and provided a possibility for breeding of elite wheat varieties.

Spikelet number per spike (SNS) is one of the crucial factors determining wheat yield. Thus, improving our understanding of the genes that regulate SNS could help develop higher-yielding wheat varieties. A genetic linkage map constructed using the GenoBaits Wheat 16K Panel and the 660K SNP array contained 5991 polymorphic SNP markers spanning 2813.26 cM. A total of twelve QTL for SNS were detected in the recombinant inbred line (RIL) populationmsf× Chuannong 16 (MC), and two of them, i.e.,QSns.sau-MC-3D.1andQSns.sau-MC-7A, were stably expressed.QSns.sau-MC-3D.1had high LOD values ranging from 4.99 to 11.06 and explained 9.71-16.75% of the phenotypic variation. Comparison ofQSns.sau-MC-3D.1with previously reported SNS QTL suggested that it is likely a novel one. A kompetitive allele-specific PCR (KASP) marker, KASP-10, tightly linked toQSns.sau-MC-3D.1was developed to successfully validate its effect in three segregated populations and a natural population. Genetic analysis indicated thatWHEAT ORTHOLOG OFAPO1(WAPO1) was a candidate gene forQSns.sau-MC-7A. The combined additive effect ofQSns.sau-MC-3D.1andWAP01had a great additive effect increasing SNS by 7.10%. In addition, our results suggested that SNS is not affected by 1BL/1RS translocations in the MC RIL population. Correlation analysis between two major QTL and other agronomic traits showed thatQSns.sau-MC-3D.1was likely independent of these agronomic traits. However, the H2 haplotype ofWAPO1may affect effective tiller number and plant height. This indicated that the breeding potential ofQSns.sau-MC-3D.1is better than that ofWAPO1. The geographical distribution ofQSns.nsau-MC-3D.1showed thatQSns.sau-MC-3D.1positive allele frequency was dominant in most wheat-producing regions of China and it has been positively selected among modern cultivars released in China since the 1940s. Two genes,TraesCS3D03G0222600andTraesCS3D03G0216800, associated with SNS development were predicted in the physical interval ofQSns.sau-MC-3D.1. qRT-PCR results of the two genes showed that only the expression level ofTraesCS3D03G0216800was significantly different between msf and CN16. These results enrich our understanding of the genetic basis of wheat SNS and will be useful for fine mapping and cloning of genes underlyingQSns.sau-MC-3D.1, and provide a basis for marker-assisted selection breeding.Author summaryIn this study, we identified two major QTL (QSns.sau-MC-3D.1andQSns.sau-MC-7A) in a RIL population.WAPO1was demonstrated to be the candidate gene forQSns.sau-MC-7A. QSns.sau-MC-3D.1was a novel and stably expressed QTL, and further confirmed in different genetic backgrounds. Our results further demonstrate thatQSns.sau-MC-3D.1has better breeding potential because of its no adverse effect on other agronomic traits thanWAPO1, and it has been positively selected during Chinese breeding programs since the 1940s. Taken together, the identification ofQSns.sau-MC-3D.1offers a promising resource to further increase wheat yields.