Association of SNP markers with agronomic and quality traits of field pea in Italy

Abstract

Ferrari B., Romani M., Aubert G., Boucherot K., Burstin J., Pecetti L., Huart-Naudet M., Klein A., Annicchiarico P. (2016): Association of SNP markers with agronomic and quality traits of field pea in Italy. Czech J. Genet. Plant Breed., 52: 83-93.Only a few studies on pea (Pisum sativum) investigated the association of single nucleotide polymorphisms (SNP) markers with key agronomic traits. This study aimed to explore the association of a standard set of 384 SNP with grain yield, seed protein content, seed weight, onset of flowering, plant height and lodging susceptibility, in three connected bi-parental recombinant inbred line (RIL) populations including 90 lines each. These RIL originated from crosses between three cultivars that displayed high and stable grain yield across Italian environments, namely, Attika (A), Isard (I), and Kaspa (K). The 270 lines were phenotyped in a spring-sown environment of Lodi (northern Italy; 45°19'N, 9°30'E). Variation among lines within the populations was significant (P < 0.01) in all cases except lodging susceptibility in one cross and, when expressed in terms of the genetic coefficient of variation, proved moderately large for most traits (including grain yield and seed protein content). Overall, we detected six quantitative trait loci (QTL) in the A × I linkage map, eight QTL in K × A, and nine QTL in K × I. Among them, there were three QTL in K × A and two QTL in K × I for grain yield, and one QTL in A × I and two QTL in both K × A and K × I for seed protein content. The consensus map, which included 130 markers (covering about 1094 cM), retained one QTL for grain yield and one for flowering time that co-located on LGII, and three for seed weight on LGIII, LGVI and LGVII. The QTL co-locating for yield and flowering time explained 8% and 31% of the overall phenotypic variation, respectively, for the two traits, and could be exploited in marker-assisted selection for adaptation to the target region.

Pea genomic selection for Italian environments

“…Additional file 5: Table S4 reports detailed information on marker ranking and association scores for each significant marker-trait association, as well as results of the linkage analysis of these markers with the Illumina array markers used for a consensus map in [42]. The absence of high marker-trait linkage in the presence of genetic variation confirmed indirectly the highly polygenic control of grain yield, lodging susceptibility and seed weight.…”

“…In agreement with earlier findings [34], most winter-type material combined cold tolerance with relatively early onset of flowering, because its main cold tolerance mechanism is not cold stress escape by a late phenology but the display of a rosette-like winter growth habit that features cold-tolerant germplasm in pea and other grain legume species [14, 15] and can contribute to cold tolerance by its relationships with lower relative plant water content [47]. Grain yield and onset of flowering exhibited a moderate positive correlation in the current environments, in contrast with the negative correlation that emerged for the same RIL populations grown under severe terminal drought in a phenotyping platform [13] or under spring sowing in northern Italy [42] (where early flowering was important to escape terminal drought and heat stress in the absence of winter cold stress).…”

“…The GWAS confirmed at the genetic level the moderate positive relationship between onset of flowering and grain yield that was observed phenotypically. Many of the markers associated with both traits were related to a genomic region in LG II that was reported as a major QTL for flowering time in earlier studies [42, 51]. The co-location of this QTL with grain yield emerged already in two earlier studies [13, 42] in which, however, the relationship of grain yield with onset of flowering was negative rather than positive, according to the adaptive value of early flowering in environments featuring severe terminal drought and heat stress and lack of winter cold stress.…”

“…A marker ranking per trait was obtained, setting an association score threshold of three (corresponding to P < 0.001 significance of the association). Linkage disequilibrium (LD) was measured between the significantly associated markers and those used for a consensus map reported in [42], where trait-marker associations were studied on a germplasm set that included 180 lines in common with the current study. The squared correlation coefficient ( r 2 ) was used as a measure of LD, considering as highly linked the markers showing r 2 > 0.8).…”

Meta-analysis of QTL reveals the genetic control of yield-related traits and seed protein content in pea

“…Seed yield and seed protein content are complex traits and highly quantitative. Numerous loci controlling seed yield and seed quality have been identified in pea [4][5][6][7][8][9][10][11][12][13][14] . The r and rb loci encoding Starch-Branching Enzyme 1 15 and ADP glucose-pyrophosphorylase 16 , respectively, and which control the wrinkled seed phenotype in pea have long been known to impact seed development, yield and seed protein content 17,18 .…”