Genome-wide association mapping of partial resistance to Aphanomyces euteiches in pea

Research interest in polygenic resistance to plant diseases has increased worldwide in the past ten years to meet the challenge of sustainable agriculture. Recommendations to reduce chemical inputs and the frequent breakdown of major resistance genes in plants have encouraged the integration of polygenic resistance into cultivars of many crops. However, breeding strategies for polygenic resistance, which is controlled by many genes, have not been as widely developed as for monogenic or oligogenic resistance (controlled by one or few genes, respectively) [1]. Breeding schemes for polygenic traits are costly and time-consuming. The effects of Quantitative Trait Loci (QTL) controlling resistance are not always conserved in different genetic backgrounds and environments and markers tightly linked to resistance loci have also often been lacking. Further research is needed to validate QTL effects, reduce their confidence intervals (CIs) and identify their underlying causal genes [2, 3], to encourage and optimize QTL use in breeding [4].

The identification of plant resistance QTL has broadly been explored using linkage mapping populations derived from crosses between two parental genotypes [1]. With the decrease in genotyping costs and the massive development of markers in the recent past years, genome-wide association (GWA) studies, are becoming common approaches to detect natural variation underlying complex traits, especially polygenic resistance to major diseases, in a large range of crop species [5–7], including legumes [8, 9]. The advantages of GWA studies compared to bi-parental linkage mapping include access to wider genetic diversity, higher recombination rates due to the evolutionary history of the species, and thus substantially refined genomic regions associated with trait variations [5, 10]. Accurate marker density for GWA studies depends on the rate of linkage disequilibrium (LD) decay and should be higher in species with a rapid LD decay (a few kb, such as in maize) than in those with a slow LD decay (~100 kb, such as in rice) [5]. Segura et al. [11] recently proposed a Multi-Locus Mixed Model (MLMM) approach, in order to improve GWA studies precision and power of detection, and it was successfully applied [12]. The reduction of CIs of genomic regions associated with traits of interest, opened the possibility of identifying haplotypes for marker assisted selection (MAS) [8, 13] and pinpointing interesting candidate genes underlying QTL [14, 15]. However, the GWA approach has also been reported to have poor power to detect rare alleles associated with the trait of interest, leading to missing heritability [5, 16], and complementarity between linkage and GWA approaches has been underlined [7, 17]. Multi-parental designs, including Nested Association Mapping (NAM) [18, 19], Multi-parent Advanced Generation Inter-Cross (MAGIC) [20, 21] and breeding line populations [22, 23] were proven to efficiently increase power of GWA studies to detect rare variants, for which rates are increased by selection of rare-allele-carrier parental lines [7, 16].

Dry pea (Pisum sativum) is the third most important pulse crop worldwide [24], for which yield has been unstable for the past decades, mainly due to biotic and abiotic stresses. One of the most damaging biotic stresses of peas is Aphanomyces root rot due to Aphanomyces euteiches [25]. The soil-borne root pathogen, first described in 1925 [26], has been mainly reported as a yield limiting factor in the United States of America (USA) and Europe for more than twenty years [27–29], and more recently in Canada [30]. Two main pathotypes of A. euteiches were described by Wicker and Rouxel [28], including pathotype I predominant in France and pathotype III detected in some regions of the USA (Onfroy et al., personal communication). Both pathotypes cause honey brown necrosis symptoms on pea roots and epicotyls, resulting in dwarfism, foliage yellowing and then death of plants in highly infested fields. Increasing yield loss due to A. euteiches in dry and green pea production has been noted in Western Europe due to short crop rotations of susceptible pea varieties and the long lifespan of oospores [27]. The development of resistant cultivars has been considered as a major objective for the past two decades in France, as only prophylactic and cropping methods are available to manage the disease. Pea lines partially resistant to A. euteiches were identified from germplasm screening and breeding programs conducted in the USA [31–35], and more recently, from a French germplasm screening program of approximately 1900 Pisum lines [36]. The most resistant lines were integrated into crossing programs to develop breeding lines [37, 38], recombinant inbred lines (RILs) [39–43] and near-isogenic lines (NILs) [44]. Breeding lines with increased levels of resistance to A. euteiches were selected in a phenotypic recurrent selection-based breeding program developed by GSP (Groupement des Sélectionneurs de Pois Protéagineux, France) [37, 38]. RILs have also been used for discovery of Aphanomyces resistance QTL [39–41, 43]. A total of 27 meta-QTL were identified on a consensus genetic map from four RIL populations [43]. Eleven of them, corresponding to seven genomic regions, were detected on six of the seven pea linkage groups (LGs), with high consistency over locations, years, isolates and populations [43]. Marker assisted back-crossing was used to introgress each of the seven consistent genomic regions into one of the susceptible RIL parents and two main spring and winter pea varieties. The resulting NILs were used to validate individual or combined major resistance QTL effects [44]. Lavaud et al. [44] considered large QTL intervals for NIL creation, which brought undesirable morphological (coloured flowers, normal leaves) or developmental (late flowering) alleles linked to resistance alleles at several QTL.

Massive numbers of Single-Nucleotide Polymorphism (SNP) markers were recently developed from whole genome cDNA (coding deoxyribonucleic acid) [45–47] or genomic sequencing of pea lines [48, 49]. A GenoPea Infinium® BeadChip was developed by Tayeh et al. [49], containing 13,204 SNPs, all located in gene-context sequences. Pea diversity panels, especially the USDA (United States Department of Agriculture) core collection and the INRA (Institut National de la Recherche Agronomique) reference collection, were used to determine associations between low to medium density genetic markers (137–1233) and traits of interest [9, 50–52]. However, only a few sources of resistance to A. euteiches were identified in these collections (Pilet-Nayel et al., unpublished data), as was also found in larger Pisum screening programs for Aphanomyces resistance [34, 36].

The aim of this study was to validate and refine the CIs of previously reported Aphanomyces resistance QTL, as well as identifying new resistance loci, using a GWA approach. A novel panel, enriched in pea lines partially resistant to A. euteiches from gene pools previously studied, was designed including breeding and germplasm lines [37, 38, 43]. The recent GenoPea Infinium® BeadChip was used for high-density SNP genotyping of the collection [49]. The GWA study detected SNPs and LD blocks associated with Aphanomyces resistance from data collected in nine field environments and two strains under controlled conditions. The genomic positions and CIs of resistance loci detected by GWA study were compared to those previously identified by linkage analysis [41, 43]. The GWA study also identified loci associated with morphological and developmental traits, in order to analyse their linkages with Aphanomyces resistance loci. The GWA study allowed marker haplotypes and putative candidate genes to be identified, for future pyramiding of resistance alleles in breeding and to investigate the molecular basis of polygenic resistance.