Avaliação da variabilidade fenotípica e genética em genótipos de cana-de-açuçar utilizando marcadores moleculares RAPD e SSR

Abstract

The work had as aimed: (1) selecting progeny of sugarcane performancebased agronomic, index selection and genetic dissimilarity. (2) assess the genetic diversity in progenies of sugarcane by means of multivariate techniques based on eight characters agribusiness. (3) assess the genetic divergence in genotypes of sugarcane using molecular markers and microsatellites. The experiment was conducted in two stages. The first was held in the sugarcane zone of the North Coast of Pernambuco in the agricultural area of Usina Santa Teresa, Engenho Terra Rica, municipality of Goiana, with geographic coordinates (07º33’ S e 35º00’ W) and altitude of 13 m during the agricultural year 2006 to 2008 in red-yellow podzolic sandy texture. The experimental design was a randomized block design with five replications. Six progenies were evaluated, consisting of 200 individuals each, three standards considered, arising from clonal proliferation of commercial varieties RB943365, RB867515 and RB863129 and three of those self-pollinated varieties. For effective control of fertilization were used bells of TNT, measuring 0.50 cm in radius and 1.20 m long sealed. Each plot consisted of 5 rows of 8 m, spaced 1.20 m with 8 seedlings per line with 1 m between plants, thus totaling 40 seedlings per plot. The variables analyzed were: pol tons per hectare (PTH), sugarcane tons per hectere (STH), fiber (FB), correted pol% (CPP), purity (PTY), brix (BX),reducing sugar (RS), total retrievable sugar (TRS). Carried out the analysis of variance and estimation of genetic parameters, the standard Euclidean distance and the distance of Rogers were used as measures of dissimilarity matrices which were correlated by the Mantel test. The Mahalanobis distance was used to quantify the genetic divergence. Were used the method of hierarchical links averages (UPGMA) method and the optimization procedure. The second stage was performed at the Laboratory of Molecular Genetics, Department of Biology, Federal University of Lavras (UFLA) in Minas Gerais. For molecular analysis were selected 23 genotypes of the families surveyed in the first stage of the experiment. And three commercial varieties RB943365, RB867515 and RB863129 and 20 randomly selected families originating from selfing of commercial varieties(stratified mass selection in families). The genetic divergence was estimated based on the polymorphism generated by RAPD (Random Amplified Polymorphic DNA) and SSR (Simple Sequence Repeats). The arithmetic complement of the Simple Matching coefficient and Jaccard coefficient were used as measures of dissimilarity whose headquarters were correlated by the Mantel test. We applied the optimization method of Tocher and the hierarchical method of links averages (UPGMA). The methodology allowed the identification of progeny of higher genetic diversity in plant breeding by providing a safer choice of crosses to be made.