Identification of genetic loci and candidate genes related to β-glucan content in barley grain by genome-wide association study in International Barley Core Selected Collection

Abstract

β-glucan is an important trait to be improved in barley breeding programs, as it greatly affects the quality of the end products when barley grains are used as raw material of feed or malt production or consumed as food for human. Although the genes associated with β-glucan synthesis have been identified, genetic regulation of β-glucan accumulation in barley grains is still completely unclear. In this study, 100 accessions from International Barley Core Selected Collection (BCS) were planted in two environments for two consecutive years to determine the genotypic variation of grain β-glucan content. A genome-wide association study (GWAS) identified 14 stable marker-trait associations (MTAs) (-Log10(P)> 4) for grain β-glucan content. Significantly positive correlation was found between grain β-glucan content and the number of favorable alleles of 14 stable MTAs. Seven putative candidate genes encoding some enzymes in glucose metabolism were found to be associated with β-glucan content. One of the putative genes, HORVU6Hr1G088380, could be an important gene controlling barely β-glucan content, with the SNPs being closely linked in all tested accessions and divided into two haplotypes. High resolution melting (HRM) analysis of the first SNP suggested that the HRM-SNP marker is valid for marker-assisted selection in barley breeding. This study provides useful information for the genes and markers related to grain β-glucan content in barley.

Supplementary information: The online version contains supplementary material available at 10.1007/s11032-020-01199-5.

Keywords: Barley; Genome-wide association study (GWAS); High resolution melting (HRM); Marker-trait associations (MTAs); β-Glucan.

Fig. 1

The distribution frequency of β-glucan content (%) of the examined 100 BCS accessions (a–c). β-glucan content (%) of Changxing-2018 (a), Changxing-2019 (b), and Cixi-2019 (c)

Fig. 2

Manhattan plots and QQ plots of GWAS mapping for β-glucan content in three different environments. Manhattan plots of Changxing-2018 (a), Changxing-2019 (b), and Cixi-2019 (c). QQ plots of Changxing-2018 (d), Changxing-2019 (e), and Cixi-2019 (f). The horizontal line depicts the 1E–04 threshold for significant association

Fig. 3

Linear regression analysis for number of favorable alleles and β-glucan content. Data shown as mean ± SE

Fig. 4

Heat map of transcript expression of seven putative candidate genes and annotations. For each period, the transcript expression values were calculated as LOG ranged from − 3.1 to 3.1. Red indicates high expression, blue indicates low expression, and gray means no expression. LEA, shoots from seedlings (10 cm shoot stage); INF1, young developing inflorescences (5 mm); INF2, developing inflorescences (1–1.5 cm); NOD, developing tillers, 3rd internode (42 DAP); CAR5, developing grain (5 DAP); CAR15, developing grain (15 DAP); ETI, etiolated seedling, dark cond. (10 DAP); LEM, inflorescences, lemma (42 DAP); LOD, inflorescences, lodicule (42 DAP); PAL, dissected inflorescences, palea (42 DAP); EPI, epidermal strips (28 DAP); RAC, inflorescences, rachis (35 DAP); SEN, senescing leaves (56 DAP)

Fig. 5

The polymorphism of putative gene HORVU6Hr1G08838 and haplotype analysis of the 5 SNPs. The position of putative gene HORVU6Hr1G08838 on chromosome 6H of Changxing-2019 (a) and Cixi-2019 (b). The putative gene structure of HORVU6Hr1G08838 and relative position of the 5 SNPs identified across 100 barley accessions from the exome capture dataset (c). The 5 SNPs from 1 to 5 were S6H_566411615 (C/T), S6H_566412497 (A/G), S6H_566412521 (C/G), S6H_566412689 (C/G), and S6H_566412731 (C/G), respectively. Box plots of mean barley grain β-glucan content (%) basing on the polymorphism of putative gene HORVU6Hr1G08838 (d). Coding regions are shown in black boxes, and introns are shown in black lines. Orange triangles represent SNPs. Boxes represent the median, first, and third quartile, respectively, and the thick horizontal lines correspond to the mean. All outliers are represented as circles

Fig. 6

HRM analysis and box plots of 80 barley accessions, shown as fluorescence difference △Fs curves (a). Barley grain β-glucan content (%) of different SNP type on the first SNP of the putative gene HORVU6Hr1G08838 (b)