Identification of candidate genes for dissecting complex branch number trait in chickpea

Abstract

a b s t r a c tThe present study exploited integrated genomics-assisted breeding strategy for genetic dissection of complex branch number quantitative trait in chickpea. Candidate gene-based association analysis in a branch number association panel was performed by utilizing the genotyping data of 401 SNP allelic variants mined from 27 known cloned branch number gene orthologs of chickpea. The genome-wide association study (GWAS) integrating both genome-wide GBS-(4556 SNPs) and candidate gene-based genotyping information of 4957 SNPs in a structured population of 60 sequenced desi and kabuli accessions (with 350-400 kb LD decay), detected 11 significant genomic loci (genes) associated (41% combined PVE) with branch number in chickpea. Of these, seven branch number-associated genes were further validated successfully in two inter (ICC 4958 × ICC 17160)-and intra (ICC 12299 × ICC 8261)-specific mapping populations. The axillary meristem and shoot apical meristem-specific expression, including differential up-and down-regulation (4-5 fold) of the validated seven branch number-associated genes especially in high branch number as compared to the low branch number-containing parental accessions and homozygous individuals of two aforesaid mapping populations was apparent. Collectively, this combinatorial genomic approach delineated diverse naturally occurring novel functional SNP allelic variants in seven potential known/candidate genes [PIN1 (PIN-FORMED protein 1), TB1 (teosinte branched 1), BA1/LAX1 (BARREN STALK1/LIKE AUXIN1), GRAS8 (gibberellic acid insensitive/GAI, Repressor of ga13/RGA and Scarecrow8/SCR8), ERF (ethylene-responsive element-binding factor), MAX2 (more axillary growth 2) and lipase] governing chickpea branch number. The useful information generated from this study have potential to expedite marker-assisted genetic enhancement by developing high-yielding cultivars with more number of productive (pods and seeds) branches in chickpea.

Genetic dissection of plant growth habit in chickpea

“…For GWAS of PGH traits, the genetic diversity information including phylogenetic tree, PCA (principal component analysis) and population structure among 92 chickpea accessions were obtained from our previous study (Kujur et al 2015a, c). The population structure ancestry coefficient (Q), kinship matrix (K) and PCA (P) data along with genome-wide SNP genotyping and PGH phenotyping information of 92 accessions were analysed through mixed model (P + K, K and Q + K)-based CMLM (compressed mixed linear model) and P3D (population parameters previously determined, Zhang et al 2010;Kang et al 2011)/EMMAX (efficient mixed model association eXpedited) approaches of GAPIT (Lipka et al 2012) as described previously (Kujur et al 2015a;Bajaj et al 2016;Upadhyaya et al 2015Upadhyaya et al , 2016. To ascertain the accuracy and validity of SNP marker-trait association, the observed and expected -log 10 (P) value relative distribution estimated for each PGH-associated genomic locus was compared based on quantile-quantile plot.…”

“…To ascertain the accuracy and validity of SNP marker-trait association, the observed and expected -log 10 (P) value relative distribution estimated for each PGH-associated genomic locus was compared based on quantile-quantile plot. Subsequently, the correction of their adjusted P value threshold of significance for multiple comparison was performed by false discovery rate (FDR cut-off ≤0.05, Benjamini and Hochberg 1995) following Kujur et al (2015a), Upadhyaya et al (2015Upadhyaya et al ( , 2016 and Bajaj et al (2016). The degree of association of SNP loci with diverse PGH traits was measured by the R 2 using a model with the SNPs and adjusted P value adopting a FDR-controlling strategy.…”

“…The SNPs revealing significant association with diverse PGH traits at a cut-off P value ≤1 × 10 −8 are marked with a dotted line Differential expression profiling (Feuillet et al 2012;Thudi et al 2014;Kujur et al 2015a;Upadhyaya et al 2015Upadhyaya et al , 2016Bajaj et al 2016;Liu et al 2016). In this perspective, a high-resolution GWAS approach deployed by us to scale down the potential genomic loci/genes governing quantitative traits of PGH at a genome-wide scale appears much expedient for understanding the complex genetic architecture of target traits in order to drive genomicsassisted crop improvement of chickpea (Kujur et al 2015a;Bajaj et al 2016;Upadhyaya et al 2015Upadhyaya et al , 2016. Four non-synonymous SNPs and one intronic SNPcontaining candidate genes (major intrinsic protein, ankyrin repeat domain containing protein, ABC transporter, sucrose non-fermenting protein and B3 transcription factor) exhibiting significant association (P = 1.37 × 10 −8 ) with PGH traits (based on our GWAS) were validated in both intra-and interspecific mapping populations contrasting for diverse PGH traits by selective genotyping (Figs.…”

Regional Association Analysis of MetaQTLs Delineates Candidate Grain Size Genes in Rice

“…However, this strategy detects considerable number of false-positive marker trait associations due to presence of population structure among germplasm accessions used for association analysis (Breseghello and Sorrells, 2006; Sneller et al, 2009). This constraint of spurious marker-trait association can be addressed to certain extent by considering the significant effects of population structure in diverse statistical models as well-adopted by previous association mapping studies in crop plants (Kujur et al, 2015; Upadhyaya et al, 2015, 2016; Bajaj et al, 2016). Despite these efforts, association mapping result still suffers from significant degree of confounding due to strong population structure and cryptic relatedness among accessions used for association analysis.…”

Association analysis of biotic and abiotic stresses resistance in chickpea (Cicerspp.) using AFLP markers

“…Chickpea is well suited to association analysis for its selfpollination mating system [25]. Association analysis has been performed in chickpea [26][27][28][29][30][31] and QTLs controlling different traits, such as flowering time [32], branch number [33], pod and branch number/plant and plant hairiness [34], 100-seed weight and seed coat colour [35,36], seed protein content [37], resistance to Fusarium wilt and Ascochyta blight [38], as well as drought tolerance [26], have been identified.…”