Identification Of QTL For Grain Amylose And Amylopectin Content In Barley(Hordeum Vulgare L.)

Barley?Hordeum vulgare L.?is the fourth largest cereal crop planted worldwide with the extensive utilization in various fields.Barley grain has been mainly used as livestock feed and raw material for malting and brewing,while it is also used as the major food source in some regions around the world.With the improvement of people's living level,the health benefits of barley have also being getting more attention.Starch,of which amylose and amylopectin are the two components,is the main constituent in barley grain like other cereal grains.Hence,the composition and structure of the starch in a certain extent determine the barley grain quality.The contents of starch compositions are quantitative traits controlled by multiple genes or quantitative trait loci?QTLs?.Traditional breeding methods,such as selected based the phenotypes,are difficult to improve these traits.Markers associated with phenotype may be integrated into regular breeding schemes,namely marker-assisted selection?MAS?,to improve the breeding efficiency.To date,very few studies identifying QTLs controlling starch components in barley grain have been performed.In this study,QTL linkage analysis together with genome-wide association studies?GWAS?were performed to identify markers linked to grain amylose and amylopectin content respectively using a doubled haploid?DH?population and an association panel of 185 diverse barley germplasms both genotyped with SNPs via genotyping-by-sequencing?GBS?.The study is hoped to provide help for quality improvement in barley breeding programs.Following are the main results achieved in this study.1.In the DH population,Naso Nijo showed significantly higher amylose and amylopectin content than TX9425 in two environments.Amylose and amylopectin contents of DH lines both showed a continuous and relatively normal or normal distribution in Yangzhou?2013-2014?and Yancheng?2014-2015?.In the association panel,amylose and amylopectin contents of 185 barley germplasms both showed a continuous and normal distribution in two years.Furthermore,the distribution of the average values in DH population and the values from BLUPs in association panel of amylose and amylopectin content in two environments were similarly continuous and relatively normal distribution.Significant differences in amylose and amylopectin contents existed among genotypes for both DH lines and varieties even though the effects of the environment and genotype x environment were also significant.2.A genetic linkage map of the DH population was generated from 1551 SNP markers covering a total map distance of 957.09 cM in seven linkage groups.The average distance between two positions across the whole map was 0.61 cM.The QTLs for grain amylose and amylopectin content were analyzed by using Windows QTL IciMapping v3.2 software with inclusive composite interval mapping?ICIM?.In two environments,three QTLs for amylose content were detected respectively on 3H,6H,7H,namely,qAC-3-1,qAC-6-1 and qAC-7-1,in addition,four QTLs for amylopectin content were detected respectively on 3H,4H,5H,7H,namely,qAPC-3-1,qAPC-4-1,qAPC-5-1,and qAPC-7-1.Among them,qAC-7-1 and qAPC-7-1 were detected in both two environments,explaining the genetic variation of 8.20 and 10.33%,9.83 and 9.49%,respectively.When using the average values of amylose and amylopectin contents from two environments for QTL analysis,the LOD value as well as the phenotypic variation of qAC-7-1 and qAPC-7-1 could significantly increase.3.According to the results of population structure and familial relationship analyses,two subgroups and about 0.5 frequency of kinship were evaluated for the association panel.The MLM model incorporating Q and K was used for GWAS for grain amylose and amylopectin content with 3826 SNPs evenly distributed on seven chromosomes.Nine SNP markers?located on chromosomes 1,2,5 7H,accounting for 8.2-11.3%of the phenotypic variation?responsible for amylose content and eleven SNP markers?located on chromosomes 1,2,3,4,7H,accounting for 7.4-12.4%of the phenotypic variation?responsible for amylopectin content were detected by the threshold of-Log?P?>3?P<0.001?.The SNP2310 associated with amylose content and the SNP3210 associated with amylopectin content were in the same or similar position respectively to qAC-7-1 and qAPC-7-1 identified by linkage mapping from the DH population.4.The candidate gene was predicted by the annotation according to synteny on rice in the co-localized region of QTL linkage mapping analysis and GWAS.The barley Waxy?GBSSI?gene?MLOC₆4657?and barley SSII-3?SSIIa?gene?MLOC₆9670?was considered to be the putative candidate gene of amylose content and amylopectin content,respectively.After sequencing and alignment,a 191-bp insertion/deletion in the region from the transcription start site to the start codon of Waxy gene and a 3 3-bp insertion/deletion in exon 2 of SSII-3 gene was identified between two parents?Naso Nijo and TX9425?of the DH population.The 191-bp and 33-bp nucleotide difference could be detected in both DNA and cDNA of two parents of the DH population.Amplified by primer Wx-191,185 germplasms could be grouped into two types?107 type-TX9425 and 74 type-Naso nijo?and a significant difference in amylose content was found between two types.Amplified by primer SSIIa-DNA,185 germplasms could be grouped into two types?58 type-TX9425 and 124 type-Naso nijo?and the difference between the two types was significant in the grain amylopectin content.The expression of the Waxy and SSII-3 gene during different stages of seed development was estimated via qRT-PCR.The Waxy expression levels were low at the beginning of seed development,increased rapidly from 10 DAF,peaked at 20 DAF,and then decreased.In addition,the Waxy expression level of Naso Nijo in each stage was significantly higher compared with TX9425.The expression levels of the SSII-3 gene showed the tendency of low at early seed development,increasing from about 10 DAF,highest at about 15-20 DAF and then decreasing rapidly.Moreover,the expression level of the SSII-3 allele in Naso Nijo was significant higher than that in TX9425 at the first four DAF stages.