Grafting versus seed propagated apricot populations: two main gene pools in Tunisia evidenced by SSR markers and model-based Bayesian clustering

Hedia Bourguiba 0 1 2 3 4 Bouchaib Khadari 0 1 2 3 4 Lamia Krichen 0 1 2 3 4 Neila Trifi-Farah 0 1 2 3 4 Sylvain Santoni 0 1 2 3 4 Jean-Marc Audergon 0 1 2 3 4 0 H. Bourguiba L. Krichen N. Trifi-Farah Laboratoire de Genetique Moleculaire , Immunologie et Biotechnologie, Faculte des Sciences de Tunis, Campus Universitaire , 2092 El Manar, Tunisia 1 H. Bourguiba B. Khadari INRA, UMR 1098 Developpement et Amelioration des Plantes, Campus CIRAD, TA A96/03 Avenue Agropolis, 34398 Montpellier cedex 5, France 2 H. Bourguiba J.-M. Audergon (&) INRA, UR1052 Genetique et Amelioration des Fruits et Legumes, Domaine Saint-Maurice, 84143 Montfavet, France 3 S. Santoni INRA, UMR 1097 Diversite et Adaptation des Plantes Cultivees, 2 Place Viala, 34060 Montpellier cedex 1, France 4 B. Khadari Conservatoire Botanique National Mediterraneen , UMR 1098, 34000 Montpellier cedex 1, France Apricot was introduced into the Mediterranean Basin from China and Asian mountains through the Middle-East and the Central Europe. Traditionally present in Tunisia, we were interested in accessing the origin of apricot species in the country, and in particular in the number and the location of its introductions. A set of 82 representative apricot accessions including 49 grafted cultivars and 33 seed propagated 'Bargougs' were genotyped using 24 microsatellite loci revealing a total of 135 alleles. The model-based Bayesian clustering analysis using both Structure and InStruct programs as well as the multivariate method revealed five distinct genetic clusters. The genetic differentiation among clusters showed that cluster 1, with only four cultivars, was the most differentiated from the four remaining genetic clusters, which constituted the largest part of the studied germplasm. According to their geographic origin, the five identified groups (north, centre, south, Gafsa oasis and other oases groups) enclosed a similar variation within group, with a low level of differentiation. Overall results highlighted the distinction of two apricot gene pools in Tunisia related to the different mode of propagation of the cultivars: grafted and seed propagated apricot, which enclosed a narrow genetic basis. Our findings support the assumption that grafting and seed propagated apricots shared the same origin. - Apricot (Prunus armeniaca L.) is a species of Prunus genus belonging to the family Rosaceae that is commercially grown world-wide. Apricot is diploid (2n = 16) and has a small genome size (5.9 Mbp/2n) (Arumuganathan and Earle 1991) compared to other fruit woody species as cherry (6.8 Mbp/2n) and apple (7.5 Mbp/2n) (Yuepeng and Korban 2007). Three centres of apricot origin were proposed by Vavilov (1992): north eastern, central and western China, central Asian mountains and near-eastern centre. Apricot was introduced into the Mediterranean Basin through two different ways. The first one was through the Middle-East allowing the identification of the Irano-Caucasian group (Kostina 1969) and the second was through the Central Europe (Faust et al. 1998). Four apricot cultivar groups named Diversification, Geographically Adaptable, Continental Europe and Mediterranean Basin were identified by Hagen et al. (2002). These groups displayed a gradient of decreasing genetic diversity of varieties from east to south-west. Tunisia, one of the extreme dissemination zone of apricot in the Mediterranean Basin, enclosed a strongly distinct gene pool of local apricot cultivars with a low genetic diversity compared to the four previously defined groups (Khadari et al. 2006). Apricot in Tunisia included traditional cultivars propagated by grafting, cultivated from the north to the south of the country, and accessions propagated by seeds, specific to oasis agrosystems, locally called Bargougs and characterized by their shadow contribution to the oasian ecosystem. Khadari et al. (2006) supported that seed propagation was more frequent than grafting propagation during apricot introduction in Tunisia. Thus, grafted cultivars could be the results of few introduced genotypes which have been firstly propagated by seeds. However, this hypothesis was not completely verified since the analysis was limited to only grafting propagated cultivars. Based on these results, a likely scenario could be proposed considering that grafting propagated cultivars shared a same gene pool with seed propagated accessions supporting a single apricot introduction in Tunisia. An alternative scenario based on historical events suggested two main apricot introductions (Valdeyron and Crossa-Raynaud 1950; Carraut and Crossa-raynaud 1974). The first one, for the grafting propagated cultivars, was located in the North of the country and originated from Andalusian germplasm, and the second one, for the seed propagated accessions, was situated in the South of Tunisia and originated from the Irano-Caucasian group. Recently, several studies focused on characterization and genetic variability assessment of Tunisian apricot cultivars in order to preserve the local genetic resources and to understand the evolution of apricot in south Mediterranean areas. In fact, using microsatellites markers (or simple sequence repeats; SSRs), Krichen et al. (2006) established an identification key for the discrimination of 54 cultivars on the basis of only five loci. While, using amplified fragment length polymorphism markers (AFLPs), Khadari et al. (2006) compared the genetic diversity among 31 grafting propagated Tunisian apricot cultivars and accessions from Europe, North America, Turkey, Iran and China, in order to give insights into the origin and historical selection process of local germplasm. Results revealed that Tunisian apricot constituted a distinct group with close genetic relationships among cultivars. Moreover, using the same set of molecular markers, Krichen et al. (2008) studied the genetic relationships among 31 different apricot cultivars and revealed a clustering closely related to their geographic origin with the distinction of two major groups suggesting probably at least the introduction of two independent gene pools in Tunisia. However, these studies considered only grafted propagated cultivars and surveys were limited to some areas of apricot culture. In this paper, our investigations implied mapped microsatellites markers covering the eight linkage groups of Prunus genome to study a larger set of Tunisian apricot germplasm including both grafting propagated cultivars and seed propagated accessions. We tested two working hypotheses related to the scenarios described above: (a) two independent introductions, a northern-central group including grafting propagated cultivars and a south-oasian group composed by seed propagated accessions; and (b) a single introduction followed by a local diversification. For that purpose, a set of 82 apricot accessions including 49 grafted cultivars and 33 seed propagated Bargougs was analyzed using 24 singlelocus SSR markers, poly (...truncated)