Assessment of coconut tree genetic divergence by compound sample RAPD marker analysis
Rogério Figueiredo Daher; Messias Gonzaga Pereira; Evandro Almeida Tupinambá; Antônio Teixeira do Amaral Junior; Wilson Menezes Aragão; Francisco Elias Ribeiro; Luís Orlando Oliveira and Ney Sussumu Sakiyama
The coconut tree (Cocos nucifera L.) is a tropical species widely cultivated throughout the world, which is found in all intertropical regions. The species shows wide phenotypic variability, which, however, is little understood at the genetic level. This study of the variability among the various coconut tree populations is important to increase the efficiency of the development of superior cultivars adapted to different ecological conditions. It also helps the selection of divergent progenies that can maximize heterosis in hybridizations. Genetic divergence among 19 coconut tree populations available in the BAG – Coco at EMBRAPA/CPATC was estimated by RAPD. Leaf samples from 21 plants of each cultivar were squashed together (compound samples) in liquid nitrogen and the DNA extracted using the modified Doyle and Doyle (1990) protocol. Samples of these DNA were amplified with 24 primers of the OPERON Technologies series. One hundred and twenty-seven polymorphic and 61 monomorphic loci were obtained. Six different clusters, possibly heterotic groups, were formed by the Tocher optimized clustering analysis which used the matrix of the complement of the Jaccard index. Group 1 included the dwarf group cultivars. Giant accessions, abbreviated to GBR (Brazilian Giant), formed group 2, except GBRPF, which together with West African Giant (GOA) formed group 4. The most distant accession was the Tonga Giant cultivar (GTG) that did not group with the others and presents potential for hybridization with the six cultivars in the dwarf group cultivars and with the five in the GBR group. Group 3 consisted of GRL, GPY and GRT and Group 5 of GML and GVT. The dendrogram obtained by the nearest neighbor method was in line with the clustering obtained by the Tocher optimization method. The markers used permitted identification of each one of the populations showing that they were genetically different (absence of duplicity). The use of compound samples was effective to investigate the interpopulational genetic diversity. However, to understand the intrapopulation genetic variability, individual sampling should be used.