Caracterização agronômica, molecular e fitoquímica de Eplingiella Harley & J.F.B. Pastore

Abstract

Eplingiella fruticosa (Salzm. Ex Benth.) Harley & JFB Pastore is an aromatic species, native, occurring in six states in northeastern Brazil (Bahia, Sergipe, Pernambuco, Paraiba, Rio Grande do Norte and Ceará). Popularly known as "alecrim de vaqueiro", is commonly found in street markets of the region and used to combat pain and seizures. Reviews in mice and in vitro studies show analgesic activity, vasodilating, cardioprotetiva, anti-inflammatory and larvicidal of its essential oil and of different types of the leaves extract. Recent studies show great variability in essential oil chemical composition of E. fruticosa, related to soil and climatic conditions and different plant organs. Thus, the species has great potential for exploration both agronomic, and by pharmaceutical companies. The aim of this study was to evaluate the vegetative propagation capacity and characterize previously Eplingiella genotypes, through morphological, agronomic, phytochemicals and molecular data. In Chapter I, two experiments were conducted: the first tested the effect of three substrates and the second evaluated five concentrations of IBA and three periods of cultivation. The design was a randomized block design with four replications. We evaluated survival percentage (% S), percentage of rooted cuttings (% EE), root length (CRE), number of shoots (NBE), dry mass of leaves (MSF), root dry weight (MSR) and total dry matter (MST). In Chapter II, twelve genotypes were collected, propagated vegetatively and transplanted. Twelve months after transplantation were assessed 12 quantitative traits, eight morphological and agronomic four. In Chapter III, the total DNA was extracted, then 20 primers were tested, of which nine were selected because they have better electrophoretic profiles agarose gel (2%). The binary matrix was computed in GEOCOMPAR II. It is estimated the diversity of the genetic structure parameters and the data were subjected to Bayesian analysis, and Neighbor-joining dendrogram and principal component analysis (PCA) based on matrix of Nei distances. And in Chapter IV, samples of 100g of leaves each repetition per genotype were used in the essential oil hydrodistillation in Clevenger type apparatus for three hours, quantifying the content. The identification of the compounds and their contents was performed by GC (FID) and GC / MS data 15 and the major compounds were used in diversity analysis. They have been made to cluster analysis and canonical variables, using as dissimilarity measure the Mahalanobis distance (D2). In the first experiment of Chapter I, significant differences were found for CRE, NBE, MSF, MSR and MST, with the best performance for the commercial substrate. In the second, positive effects have been identified both the addition of AIB as the cultivation time on the CRE variables, NBE, MSF and MSR, reaching maximum increment to the estimated concentration of 1.5 g L-1, at 60 days of cultivation. In Chapter II, there was significant variation by F test (p <0.01) for the CF features, LF, CBD, CBE, LP, and MFF MSF. The genotypes formed two groups for almost all variables, by Scott-Knott test (p <005), except for LP, which formed three. The EF002 and EF003 genotypes presented the highest levels for almost all variables. There was the formation of three groups for both UPGMA and for the canonical variables (CV). The characteristics that most contributed to the formation of groups were CBE, MFF and CF. The genotypes EF002, EF003, EF005 and EF012 stood out because they have higher genetic distances. In CHAPTER III, primers produced 131 polymorphic bands. The diversity index of Nei (Ne) ranged between 0.31 and 0.39, while Shannon (I) ranged between 0.33 and 0.48. The percentage coefficient of genetic differentiation (Gst) was 0.29. In AMOVA most of the variation was within populations (69%), while among populations was 27% and 4% among species, indicating a good genetic structure. The average value of Fst was 0.175, demonstrating intermediate differentiation between populations. The structure of the Bayesian analysis method revealed three possibilities for the formation of groups (K = 2; = 6; 8 =;), however, it presented many migrants and high level of mixing individuals. The dendrogram generated by the Neighbor-Joining method confirmed the formation of two groups, with good support for major clades (100%). PCA analysis in the first two axis explained 21.06% of the total variation among populations. Finally, in Chapter IV, the genotypes were classified into four clusters: 1 - EF001 genotypes, EF006, EF007, EF008, EF010, EF011 and EF012 with E-caryophyllene and bicyclogermacrene as major; 2 - EF002 and EF003 genotypes, with the majority same as the previous group, however, percentage with average about 30% higher; 3 - EF004 and EF005 genotypes that showed a greater production of E-caryophyllene; and 4 - with EF009 genotype, forming a single group to present α-pinene as balanced majority and percentage among the rest. This result was confirmed by canonical variables, which explained 76% of the variation. The bicyclogermacrene compounds, 1,8-cineol, α-copaene and spathulenol represented the most important variables for analysis.