To gain new insights into the complex evolutionary history of Brassica rapa, we investigated the genetic diversity of wild and domesticated populations from an unexplored area in its center of origin (Mediterranean basin). These populations were collected in Algeria, which presents a wide ecogeographic range and an exceptional bioclimatic gradient. We wanted to answer the following questions: (1) Were the local landraces domesticated from the local wild forms? (2) Do these populations offer a new diversity never described? (3) How is this diversity related to B. napus? Morphological traits and SSR markers were analysed to explore the genetic diversity among 18 Algerian B. rapa accessions that were compared to those previously analysed in this species and to B. napus varieties. Among worldwide B. rapa diversity, the wild and cultivated Algerian groups showed the highest allelic richness, suggesting that breeding has not significantly eroded genetic diversity in Algerian local landraces. Wild and cultivated B. rapa accessions from Algeria formed two clusters regardless of their local geographic origin and were distinct from all the B. rapa groups already described. Surprisingly, B. napus accessions clustered either with wild Algerian group or with the group containing mainly Asian accessions. This new diversity will be of high interest for B. rapa and B. napus breeding.

Aphanomyces euteiches is a major soilborne oomycete pathogen that infects various legume species, including pea and alfalfa. The model legume Medicago truncatula has recently emerged as a valuable genetic system for understanding the genetic basis of resistance to A. euteiches in leguminous crops. The objective of this study was to identify genetic determinants of resistance to a broad host-range pea-infecting strain of A. euteiches in M. truncatula. Two M. truncatula segregating populations of 178 F-5 recombinant inbred lines and 200 F-3 families from the cross F83005.5 (susceptible) x DZA045.5 (resistant) were screened for resistance to A. euteiches. Phenotypic distributions observed suggested a dominant monogenic control of resistance. A major locus associated with resistance to A. euteiches, namely AER1, was mapped by bulk segregant analysis to a terminal end of chromosome 3 in M. truncatula and explained 88% of the phenotypic variation. AER1 was identified in a resistance-gene-rich region, where resistance gene analogs and genes associated with disease resistance phenotypes have been identified. Discovery of AER1 opens up new prospects for improving resistance to A. euteiches in cultivated legumes using a comparative genomics approach.

Partial resistance to Mycosphaerella pinodes in pea is quantitatively inherited. Genomic regions involved in resistance (QTLs) have been previously identified in the pea genome, but the molecular basis of the resistance is still unknown. The objective of this study was to map resistance gene analogs (RGA) and defense-related (DR) genes in the JI296 x DP RIL population that has been used for mapping QTLs for resistance to M. pinodes, and identify co-localizations between candidate genes and QTLs. Using degenerate oligonucleotide primers designed on the conserved motifs P-loop and GLPL of cloned resistance genes, we isolated and cloned 16 NBS-LRR sequences, corresponding to five distinct classes of RGAs. Specific second-generation primers were designed for each class. RGAs from two classes were located on the linkage group (LG) VII. Another set of PCR-based markers was designed for four RGA sequences previously isolated in pea and 12 previously cloned DR gene sequences available in databases. Out of the 16 sequences studied, the two RGAs RGA-G3A and RGA2.97 were located on LG VII, PsPRP4A was located on LG II, Peachi21, PsMnSOD, DRR230-b and PsDof1 were mapped on LG III and peabetaglu and DRR49a were located on LG VI. Two co-localizations between candidate genes and QTLs for resistance to M. pinodes were observed on LG III, between the putative transcription factor PsDof1 and the QTL mpIII-1 and between the pea defensin DRR230-b gene and the QTL mpIII-4. Another co-localization was observed on LG VII between a cluster of RGAs and the QTL mpVII-1. The three co-localizations appear to be located in chromosomal regions containing other disease resistance or DR genes, suggesting an important role of these genomic regions in defense responses against pathogens in pea.

A collection of 148 Pisum accessions, mostly from Western Europe, and including both primitive germplasm and cultivated types, was structured using 121 protein- and PCR-based markers. This molecular marker-based classification allowed us to trace back major lineages of pea breeding in Western Europe over the last decades, and to follow the main breeding objectives: increase of seed weight, introduction of the afila foliage type and white flowers, and improvement of frost tolerance for winter-sown peas. The classification was largely consistent with the available pedigree data, and clearly resolved the different main varietal types according to their end-uses (fodder, food and feed peas) from exotic types and wild forms. Fodder types were further separated into two sub-groups. Feed peas, corresponding to either spring-sown or winter-sown types, were also separated, with two apparently different gene pools for winter-sown peas. The garden pea group was the most difficult to structure, probably due to a continuum in breeding of feed peas from garden types. The classification also stressed the paradox between the narrowness of the genetic basis of recent cultivars and the very large diversity available within P. sativum. A sub-collection of 43 accessions representing 96% of the whole allelic variability is proposed as a starting point for the construction of a core collection.

The inheritance of resistance to Ascochyta blight, an economically important foliar disease of field pea (Pisum sativum L.) worldwide, was investigated. Breeding resistant pea varieties to this disease, caused by Mycosphaerella pinodes, is difficult due to the availability of only partial resistance. We mapped and characterized quantitative trait loci (QTLs) for resistance to M. pinodes in pea. A population of 135 recombinant inbred lines (RILs), derived from the cross between DP (partially resistant) and JI296 (susceptible), was genotyped with morphological, RAPD, SSR and STS markers. A genetic map was elaborated, comprising 206 markers distributed over eight linkage groups and covering 1,061 cM. The RILs were assessed under growth chamber and field conditions at the seedling and adult plant stages, respectively. Six QTLs were detected at the seedling stage, which together explained up to 74% of the variance. Ten QTLs were identified at the adult plant stage in the field, and together these explained 56.6-67.1% of the variance, depending on the resistance criteria and the organ considered. Four QTLs were detected under both growth chamber and field conditions, suggesting they were not plant-stage dependent. Three QTLs for flowering date and three QTLs for plant height were also identified in the RIL population, some of which co-located with QTLs for resistance. The relationship between QTLs for resistance to M. pinodes, plant height and flowering date is discussed.

Pisum sativum sequences were retrieved from Genbank/EMBL databases and searched for all possible dinucleotide and trinucleotide tandem repeats. One-hundred and seventy-one simple sequence repeats (SSRs) were found among 663 sequences, The different dinucleotide or trinucleotide motifs occurred,It varying frequencies. CT/AG was the most frequent dinucleotide, and TCT/AGA the most frequent trinucleotide. Forty-three microsatellite markers were generated from these sequences and used to assess the genetic variability among 12 pea genotypes. Thirty-one were polymorphic among the genotypes and the average number of variants per marker was 3.6 when considering only polymorphic markers. Overall, the number of variants for a given SSR marker was correlated with the length of the SSR but some 12-bp long SSRs showed the same degree of polymorphism as longer ones. The groupings resulting from the SSR genotyping among the 12 genotypes gave an interesting insight into the possible origin of one recent cultivar. Database-derived SSR markers are highly variable. They can provide useful information on the genetic diversity among P. sativum cultivated types.