Ultra-high density intra-specific genetic linkage maps accelerate identification of functionally relevant molecular tags governing important agronomic traits in chickpea

Abstract

We discovered 26785 and 16573 high-quality SNPs differentiating two parental genotypes of a RIL mapping population using reference desi and kabuli genome-based GBS assay. Of these, 3625 and 2177 SNPs have been integrated into eight desi and kabuli chromosomes, respectively in order to construct ultra-high density (0.20-0.37 cM) intra-specific chickpea genetic linkage maps. One of these constructed high-resolution genetic map has potential to identify 33 major genomic regions harbouring 35 robust QTLs (PVE: 17.9-39.7%) associated with three agronomic traits, which were mapped within <1 cM mean marker intervals on desi chromosomes. The extended LD (linkage disequilibrium) decay (~15 cM) in chromosomes of genetic maps have encouraged us to use a rapid integrated approach (comparative QTL mapping, QTL-region specific haplotype/LD-based trait association analysis, expression profiling and gene haplotype-based association mapping) rather than a traditional QTL map-based cloning method to narrow-down one major seed weight (SW) robust QTL region. It delineated favourable natural allelic variants and superior haplotype-containing one seed-specific candidate embryo defective gene regulating SW in chickpea. The ultra-high-resolution genetic maps, QTLs/genes and alleles/haplotypes-related genomic information generated and integrated strategy for rapid QTL/gene identification developed have potential to expedite genomics-assisted breeding applications in crop plants, including chickpea for their genetic enhancement.

Figure 1

(A) Relative distribution and frequency of 43358 SNPs discovered in eight chromosomes and scaffolds of desi and kabuli chickpea using reference genome-based GBS assay. Distribution of 22331 SNPs physically mapped on eight desi (B) and kabuli (C) chromosomes are depicted in the Circos circular ideogram. The innermost circles display the distribution of SNPs identified through GBS assay, while outermost circles represent the eight chickpea chromosomes coded with different colours. (D) Proportionate distribution of transition and transversion SNPs identified using the reference desi and kabuli genomes-based GBS assays.

Figure 2

(A) Proportionate distribution of 4448 SNPs identified in different coding and non-coding sequence components and intergenic regions of 635 genes annotated from desi chickpea genome. (B) Number of SNPs, including synonymous and non-synonymous (missense and nonsense) SNPs annotated in the coding (CDS) as well as non-coding (intronic and regulatory) sequences of genes and intergenic regions of desi genome. The URR (upstream regulatory region) and DRR (downstream regulatory region) of genes were defined as 2000-bp upstream sequences from the translation start codon (ATG) and 1000-bp downstream sequences from the stop codon, respectively.

Figure 3

Integration of desi (A) and kabuli (B) genetic maps with kabuli physical (C) map of one major genomic region harbouring robust SW QTL (qSW5.1) mapped target QTL on 713 kb sequence interval of chromosome 5. This QTL was further narrowed down to about 22.5 kb sequenced region on chromosome 5 by integrating QTL mapping with QTL region-specific haplotype (LD)-based high-resolution trait association analysis (D). The 22.5 kb genomic region between the markers CaSNP1 and CaSNP2 contained five candidate protein-coding genes, of which one of the regulatory SNP-carrying embryo defective kabuli gene (Ca07594) showed strong association with SW in chickpea (D). The genetic (cM) and physical (bp) distance and identity of the markers mapped on the LGs/chromosomes are indicated on the left and right side of the LGs/chromosomes, respectively.

Figure 4. The molecular haplotyping, LD mapping and gene haplotype-specific trait association analysis in an embryo defective gene validating its strong association potential for SW in chickpea.

The genotyping of 12 SNPs, including one regulatory SNP (C/T) in the URR of this gene (A) among 244 association panel constituted four haplotypes (B). The four SNP haplotype marker-based genotyping information produced higher LD estimates covering the entire 9286 bp sequenced region of gene (C). A significant correlation of parental genotypes and RIL mapping individuals representing high (haplotype I: TAC) and low (haplotype II: CGT) SW haplotypes [constituted by three SNPs (shaded with yellow colour in B)] in URR of gene with their seed-specific diverse transcript expression (D) during seed development was evident. A superior favorable high SW-regulating haplotype (TAC) with decreased transcript expression was identified in the URR of gene (D).