Phenotypic characterization and molecular Studies of Kenyan passion fruit (passiflora edulis (Sims))genotypes based on SSR markers

Abstract

Passion fruit (Passiflora edulis) was introduced to Kenya in the 1920s by European farmers. The fruit is a significant income earner for the country from both local and international markets. There are increased research activities on variety improvement and development of new varieties with the hall marker of this research being the KPFs 4, 11, 12 by KALRO. The KPFs are hybrids developed through natural crosses of the coastal yellow varieties and purple variety. Despite these gains in breeding, there is little information of both phenotypic and molecular variability of these new cultivars as well as the existing ones. The present study aimed at evaluating the molecular and phenotypic variation in the hybrid and parent varieties. Phenotypic characterization was done by targeting 7 morpho-agronomic descriptors with three biological replicates. Analysis was done using MINITAB 16 software, for description principle components and construction of a dendrogram using the Eucledian distance tool. The first principal component alone accounted for 51.3% variance, while the first two components accounted for cumulative variance of 74.3%. Molecular characterization was done using SSR markers. Eleven (11) markers were polymorphic while one marker; PE 20 was monomorphic. Calculation of genetic distance, PIC, allele number as well as gene diversity was done using Powermarker® version 3.25 while phylogeny reconstruction was done using DARWIN version 6. On the other hand, AMOV A was calculated using GenAlex 6.502. The PIC ranged from 0.2929 (PE 04) to 0.6179 (PE 90) with a mean of 0.3909 indicating that marker PE 90 was highly informative and other markers were reasonably informative. The mean PIC of 0.3909 indicated that generally, all the 11 polymorphic markers were informative and as such can be utilized for further analysis of Passiflora edulis genome. The dissimilarity matrix indicated that KRC-2 and KRP-2 had the highest level of dissimilarity, confirming the known molecular variation that exists between the yellow and purple varieties where the respective genotypes belong to. Several genotypes had a dissimilarity of 0% and indication that they shared closer ancestry compared to other genotypes. The phylogenetic tree discriminated the samples into three main clusters with several sub-clusters carrying different fruit populations, indicating genetic relatedness among members of the same population. From the phylogenetic tree, genetic variability was highest between genotype KRC-2 and KRP-4 which are coastal and purple varieties respectively. This corroborates the findings of dissimilarity matrix where they had a dissimilarity of 76%. The AMOVA indicated intra-population variation of 91% while inter-population variation was 9%. From the results, Molecular and phenotypic characterization proved to be useful tools for discrimination of various genotypes of Passiflora edulis. The results from this study will enable better detection and management of genetic variability that could hinder adoption of the KPF as an alternative crop for the semi-arid agro-ecologies.