A genome wide association study to dissect the genetic architecture of agronomic traits in Andean lupin (Lupinus mutabilis)

Abstract

Establishing Lupinus mutabilis as a protein and oil crop requires improved varieties adapted to EU climates. The genetic regulation of strategic breeding traits, including plant architecture, growing cycle length and yield, is unknown. This study aimed to identify associations between 16 669 single nucleotide polymorphisms (SNPs) and 9 agronomic traits on a panel of 223 L. mutabilis accessions, grown in four environments, by applying a genome wide association study (GWAS). Seven environment-specific QTLs linked to vegetative yield, plant height, pods number and flowering time, were identified as major effect QTLs, being able to capture 6 to 20% of the phenotypic variation observed in these traits. Furthermore, two QTLs across environments were identified for flowering time on chromosome 8. The genes FAF, GAMYB and LNK, regulating major pathways involved in flowering and growth habit, as well as GA30X1, BIM1, Dr1, HDA15, HAT3, interacting with these pathways in response to hormonal and environmental cues, were prosed as candidate genes. These results are pivotal to accelerate the development of L. mutabilis varieties adapted to European cropping conditions by using marker-assisted selection (MAS), as well as to provide a framework for further functional studies on plant development and phenology in this species.

Keywords: Lupinus mutabilis; SNP; association mapping; flowering time; molecular markers; plant architecture.

Figure 1

Boxplots showing the variation of the plant architecture and yield related traits (BLUES data) under study in a collection of 223 L. mutabilis accessions. The upper and lower end of the boxes indicate respectively the 75th and 25th percentiles, while the line in the middle represents the 50th percentile (median). The whiskers represent the highest and lowest values, while circles represent outliers outside the 5th-95th percentile interval.

Figure 2

Markers distribution on the 20 chromosomes of L. angustifolius genome. In red, significant markers detected in association with agronomic traits using Bonferroni threshold (LOD 5.52, experiment wise error α = 0.05).

Figure 3

Manhattan plots displaying SNP markers-trait associations identified for Flowering time in GWAS using 223 accessions of L. mutabilis. The red line indicates the Bonferroni threshold (LOD = 5.52). Common SNPs showed significant association with variation in flowering time across locations, even if not all of them were detected as significant.

Figure 4

Manhattan plots suggestive of a SNP markers-trait association for plant height on chromosome LG16 and LG12, respectively for L. mutabilis growing in Mediterranean (Portugal) and North-Central European cropping conditions (NL-Sc and NL-Wi). The red line indicates the Bonferroni threshold (LOD = 5.52). The black dotted line indicates a suggestive threshold of LOD= 4. The rectangles highlight interesting piling up of SNPs below the threshold of significance, harboring interesting candidate genes related to plant height.