Genetic diversity in European Pisum germplasm collections

Abstract

The distinctness of, and overlap between, pea genotypes held in several Pisum germplasm collections has been used to determine their relatedness and to test previous ideas about the genetic diversity of Pisum. Our characterisation of genetic diversity among 4,538 Pisum accessions held in 7 European Genebanks has identified sources of novel genetic variation, and both reinforces and refines previous interpretations of the overall structure of genetic diversity in Pisum. Molecular marker analysis was based upon the presence/absence of polymorphism of retrotransposon insertions scored by a high-throughput microarray and SSAP approaches. We conclude that the diversity of Pisum constitutes a broad continuum, with graded differentiation into sub-populations which display various degrees of distinctness. The most distinct genetic groups correspond to the named taxa while the cultivars and landraces of Pisum sativum can be divided into two broad types, one of which is strongly enriched for modern cultivars. The addition of germplasm sets from six European Genebanks, chosen to represent high diversity, to a single collection previously studied with these markers resulted in modest additions to the overall diversity observed, suggesting that the great majority of the total genetic diversity collected for the Pisum genus has now been described. Two interesting sources of novel genetic variation have been identified. Finally, we have proposed reference sets of core accessions with a range of sample sizes to represent Pisum diversity for the future study and exploitation by researchers and breeders.

Fig. 1

SSAP analysis of strongly assigned accessions. a Assignment of Q from Jing et al. (2010). b Neighbour joining (NJ) tree calculated using DARwin5 (Perrier et al. 2003). Allele frequency within the population sub-groups of Jing et al. (2010) was calculated in an Excel sheet and the majority consensus NJ tree with 100 bootstraps calculated. Bootstrap values higher than 80 % are indicated. c The corresponding NJ tree for individual accessions is shown

Fig. 2

Comparison of STRUCTURE analyses. The bottom horizontal ribbon (for both panels) shows the assignment to STRUCTURE sub-groups by Jing et al. (2010). The central vertical ribbon shows the assignment of Q values with the data set from this study (see Supplementary Figure 1c). The upper green group is referred to as Q _G and the red-brown and blue groups as Q _R and Q _B, respectively. Spots in the left panel graph indicate the locations of accessions in the two analyses. Observed versus expected numbers are indicated as a fraction for combinations of cells in the right panel, which are significantly different from expectation on the basis of a contingency test. Cells highlighted in yellow have significantly more accessions than expected and those highlighted in turquoise have fewer than expected

Fig. 3

Relationship between STRUCTURE groups containing exotic germplasm. Accessions assigned to the ‘exotic’ group (Q _G) of the K = 3 analysis of this data set (the green group of Supplementary Figure 1b) were analysed by STRUCTURE and the most consistent assignment into six sub-groups is presented. The locations of accessions analysed by Jing et al. (2010) (x axis) are marked by points (x, y) corresponding to their position in that and the analysis of the present data set (y axis). The diamond symbols at (x, y) are colour coded according to the scheme of Jing et al. (2010) as indicated on the right. Taxonomic groups strongly represented in sub-groups are indicated in brackets

Fig. 4

Multifactorial analysis of Pisum diversity. Distribution of accessions in MFA: these two dimensions explain 7.24 and 3.61 % of the variance, respectively (note that the variance is distributed over 4,532 dimensions). a Accessions previously analysed by Jing et al. (2010) b All accessions in the present analysis. c Accessions new to this analysis. d Accessions in a assigned to four main taxa other than P. sativum. e Region centred near (0.3, 0). f Outlying points

Fig. 5

Distribution of accessions by donor. This figure reproduces the MFA plot of Fig. 4, and shows the relative position of accessions obtained from various European germplasm collections. The codes are: All the complete data set, UK the John Innes Pisum collection, NLD the Dutch pea collection (Wageningen), POL the Polish pea collection (Wiatrowo), PRT the Portuguese pea collection (Elvas), FRA the French pea collection (Dijon), RUS the Russian pea collection (at the Vavilov Institute St. Petersburg), ESP the Spanish pea collection (Valladolid)

Fig. 6

Geographic and genotypic partitioning of Dutch accessions. a Multifactorial plot (from Fig. 5 NLD), colour coded. b Location of assigned collection points for accessions. The pale blue spot in a corresponding to a Mexican accession is not shown. The yellow spots in the Indian sub-continent are marked as ringed symbols. For reference, high points in the Himalayan range are marked in black. The cluster of accessions from northern Pakistan is boxed in red and shown expanded to the right, with the mountainous region shaded grey. For scale, the distance between Jalalabad and Islamabad is ~250 km. c The centre of the group of accessions corresponding to the solid green spots tightly clustered in northern Pakistan is at 35°47′N 72°36′E near Mahodand Lake south west of the Karakoram Mountains. The great circle distance from this point (in radians) for all accessions (y axis) is plotted against PC1 (x axis). Accessions are coloured yellow or green according to the two main groups in a. The yellow group is subdivided into ringed and solid colours according to the location shown in b. Three exceptions are: red an accession not clearly assigned to either yellow or green, blue a Mexican accession and the pale green ringed symbol that corresponds to an accession in the green group that was collected at a more southern location (indicated in b)

Fig. 7

Representative subsets of Pisum accessions. Multifactorial analysis (MFA) plots (a–c, l–o) and STRUCTURE assignments (d–k) of selected subsets of accessions are illustrated. The distribution of all accessions in the MFA space is shown in ‘a’ (identical to Fig. 4a). The assignments of accessions to STRUCTURE sub-groups of Jing et al. (2010) is shown in ‘d’, using their colour codes and accession order. e The assignment of accessions to three STRUCTURE groups identified here (blue Q _B, red Q _R and green Q _G). Those accessions common to a and b are in the order of Jing et al. (2010), but those on the right (new to this analysis) are in the order of Supplementary Figure 1c, with Q assigned as the averages of panel A1. b, f 141 accessions sampled on the basis of STRUCTURE assignments (black) and MFA (red). c, g Samples selected by Core Hunter at 5 % representation (black) with the seven accessions also in the minimum core highlighted in red. The 10, 20 and 30 % Core Hunter selections are in l–n (MFA) and h–j (STRUCTURE), respectively. o, k 10 % representation reselected from the 30 % selection