Microsoft Word - 16-Bio_7595.doc 473 Original Article Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 GENETIC STRUCTURE OF MANGABA (Hancornia speciosa Gomes) POPULATIONS IN THE CERRADO REGION OF CENTRAL BRAZIL ESTRUTURA GENÉTICA DE SUBPOPULAÇÕES DE MANGABEIRA (Hancornia speciosa Gomes) NO CERRADO DO BRASIL CENTRAL Nara Fernandes MOURA 1 ; Lázaro José CHAVES 2 ; Roland VENKOVSKY 3 ; Ronaldo Veloso NAVES 4 ; Ananda Virgínia AGUIAR 5 ; Mara Fernandes MOURA 6 1. Doutoranda em Agronomia EA/UFG. Goiânia-GO. Brasil. naramf2001@yahoo.com.br. 2. Prof. Dr. EA/UFG, Goiânia-GO. Brasil. 3. Prof. Dr. ESALQ/USP, Escola Superior de Agricultura “Luiz de Queiroz”. Piracicaba, SP. Brasil. 4. Prof. Dr. EA/UFG, Goiânia-GO. 5. Pesquisadora Dra. Embrapa Florestas, Colombo – PR. Brasil. 6. Pesquisadora Dra. Instituto Agronômico de Campinas, Centro APTA Frutas. Jundiaí, SP. Brasil. ABSTRACT: Mangaba tree (Hancornia speciosa Gomes) is a fruit species with great potential for commercial exploitation, found in most regions of Brazil and highly frequent in the Cerrado biome. Information on the genetic variability in this biome is scarce. This study aimed at characterizing the genetic structure of eight populations found in the states of Goiás, Bahia and Minas Gerais by using RAPD markers. Genetic variability was assessed by the analysis of variance on molecular data. A significant proportion of genetic variance was found among populations, corresponding to 19.6 % of the total variation. The correlation between the matrices of geographic and genetic distances was not significant, indicating that there is no spatial structure of the genetic variability among the populations, for the RAPD marker loci used here. The high variability suggests that a strategy for in situ, as well as ex situ conservation, must be based on sampling a large number of local populations. KEYWORDS: Mangaba. Molecular marker. Population genetics. Native species. INTRODUCTION Mangaba (Hancornia speciosa Gomes) is a tropical tree of promising prospect for commercial exploitation. The species, found in most regions of Brazil, is frequent in the Cerrado biome. Mangaba fruits are consumed in natura, dried and used in preparations such as jelly or ice cream; the exquisite taste, makes mangaba highly appreciated by the population of the Center-West and Northeast regions. Due to significant reduction of the original area of ecosystems where it occurs, mangaba is one of the fruit species most threatened by genetic erosion in Brazil. Knowledge on the genetic variability of the remaining populations in the Cerrado biome is scarce. The characterization of genetic variability is essential in support for domestication and breeding. In germoplasm sampling, it is important to take into account the population characteristics as well as the knowledge on the reproductive biology, the genetic structure and the effective size, since these can directly affect the viability of the populations in the long term (MOURA, 2003). Study on the reproductive behavior of H. speciosa var speciosa characterized this botanical variety as cross pollinating and self incompatible (DARRAULT; SCHLINDWEIN, 2006; PINTO et al., 2008). To our knowledge, there is no study on the reproductive biology of the botanical varieties occurring in the core area of Brazilian Cerrado, but the specialized floral morphology suggests that cross pollination can be assumed as the rule for the whole species. Some studies have been carried out with native Cerrado fruit species, focusing on cultivation techniques (SOUZA et al., 2007; BERNARDES, et al. 2008; CAVALCANTE et al., 2008), nutritional aspects (VERA et al., 2007), and the biology and ecology (VERA et al., 2005). Little is known, however, about the genetic structure of natural populations. Some studies have shown the existence of high genetic variability, indicating the perspective for selection and breeding of these species (COLLEVATTI et al., 2001; COLLEVATTI, 2003; ZUCCHI et al., 2005; TRINDADE; CHAVES, 2005; MARTINS et al., 2006; SOARES et al., 2008a; SOARES et al. 2008b). Study with morphological traits of fruits and trees from natural populations of H. speciosa from Cerrado region showed the existence of high variation among botanical varieties, among populations within botanical varieties and within populations (GANGA et al., 2010). High levels of genetic variation were also observed in a progeny study focusing on initial development traits both in nursery and field environment (GANGA et al., 2009). A few studies were performed with H. speciosa populations from the Brazilian Cerrado based on molecular markers (SILVA ,2006; Received: 23/06/10 Accepted: 15/11/10 474 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 RODRIGUES, 2009). The results from these studies have shown high levels of genetic diversity among populations, suggesting that a great number of populations must be taken for genetic conservation of the species. The purpose of this study was to characterize the genetic structure of eight mangaba populations from four regions of Brazilian Cerrado based on RAPD markers, in search of an applicable contribution for in situ and ex situ conservation of the genetic variability of the species. MATERIAL AND METHODS Eight local populations (or subpopulations) were sampled from August to December 2001 in four regions of the Brazilian Cerrado (Figure 1). Areas with little human influence and with 30 or more adult plants within a radius of 5 km were chosen (minimum distance of 20 km between populations). The altitude of each area was determined by an altimeter and the geographic coordinates determined using a GPS receptor (Global Positioning System, model Geo Explorer). A matrix of geographic distances among subpopulations was established based on the geographic coordinates. Each population was represented by 30 randomly sampled plants. Young leaves were collected from adult plants, placed in paper bags and stored on ice, until arrival at the laboratory, where they were deep-frozen, and prior to extraction. The total genomic DNA was extracted by the CTAB method, described by Murray and Thompsom (1980) and Rogers and Bendich (1985), with modifications. DNA was extracted from a total of 217 plants, varying from 25 to 30 plants per subpopulation. The DNA concentration in the samples was estimated by comparison with a sample of known DNA concentration of the whole phage λ, on 0.8 % agarose gel. The DNA extracted from each plant was diluted at a concentration of 1 ng/µL and the stock solution maintained in the refrigerator until use. Figure 1. Spatial distribution of 8 populations of Hancornia speciosa Gomes from Brazilian Cerrado. Each amplification reaction consisted of a volume of 25 µL containing: 2 ng DNA template; 50 mM KCl; 10 mM Tris-HCl pH 8.3; 2.0 mM MgCl; 250 µM dNTP; 250 nM primer and 2 units of Taq DNA polymerase. The reactions were subjected to 48 amplification cycles after initial denaturation at 94o C for five min. Each cycle consisted of 30 s at 92o C, 1 min and 30 s at 37o C and 1 min and 30 s at 72o C. After 48 cycles, a final extension step of 6 min at 72o C was performed. The amplified fragments were electrophoresed in horizontal 1.4 % agarose gels at 1 % TBE (Tris- borate 90 mM and EDTA 2 mM) buffer to detect polymorphisms. The gels were stained with 475 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 ethidium bromide and pictures taken under UV light (Kodak EDA camera). The gel images were interpreted and each plant was genotyped according to the presence (1) or absence (0) of each band for all primers analyzed; only the clearest and well- separated bands were selected. A binary data matrix was established based on this genotyping and the polymorphism level calculated for each locus. The total variation was partitioned into components using the analysis of variance of molecular data (AMOVA) (EXCOFFIER et al., 1992) based on the binary data matrix, using software Arlequim (SCHNEIDER et al., 2000). AMOVA aims mainly at the estimation of parameter STφ , which measures the fraction of molecular genetic diversity (of RAPD) among subpopulations. This parameter was partitioned according to the structure of the subpopulations, into the following components of variability: STφ - fraction of the total variation that is due to the subpopulations altogether, RTφ - fraction of the total variation that is due to the differences between regions and SRφ - variation due to the subpopulations within regions. The components of variance of each hierarchical level were calculated based on the mean square expectations, using the method of moments. Using the binary data, pair wise similarity measures (Sij) were obtained from all plants using the Jaccard method. The respective dissimilarities or genetic distances were computed by the complement of each similarity measure relative to 1.0 (Dij = 1 - Sij). Based on the Dij values the mean dissimilarity values within populations and among populations were calculated pairwise, resulting in a matrix of mean dissimilarity among subpopulations. To investigate the spatial variation patterns the Pearson correlation coefficient (r) was estimated between the matrices of genetic dissimilarities and geographic distances, using software NTSYS 1.7 (Numerical Taxonomy and Multivariate Analysis System) (Rohlf, 1989). The significance of this correlation was verified by Mantel's Z statistic (1967), by using 9999 random permutations. To evaluate the variation pattern among populations, a dendrogram was constructed based on the mean dissimilarities among populations, using the UPGMA grouping criterion (Unweighted Pair-Group Method Using Arithmetic Averages) (Sneath and Sokal, 1973). The cluster stability was also tested by the bootstrap resampling procedure, using 10,000 bootstraps. RESULTS AND DISCUSSION Of the 46 primers analyzed by Moura et al. (2005), 11 primers were pre-selected, totalizing 60 polymorphic loci. Subsequently the primers with a homogeneous profile and constant genomic DNA amplification for the populations were selected: OPB-12, OPC-04, OPH-18 and OPH-19. The RAPD profiles detected in the primer selection trial (Figure 1) determined the requirements for the maintenance of the PCR conditions invariably. This procedure made possible the control of profile reproducibility in the amplification reactions. Figure 1. Profile of a RAPD gel using primer OPC04 of 30 mangaba trees, on the sides the DNA ladder (250 pb). The results of AMOVA indicated a total variance of 4.45 (Table 1). The genetic variance among populations was significant (P<0,01) and corresponded to 19.67 % of the total variance. Of this, 18.73 % corresponded to the variation among populations within regions, a significant value at 1 % probability. The variation among regions represented only 0.94 %, which resulted non- significant. As expected, the major part of the variation (80.33 %) was concentrated within populations, a significant variation at 1 % probability. 476 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 Table 1. Analysis of variance of molecular data (AMOVA) for eight subpopulations of H. speciosa Gomes in the Cerrado of Central Brazil, based on thirty one RAPD bands. Source of variation D. F S. S. Variance components Variation (%) p-value Regions 3 85,360 0,0419 0,94 0.400 Population/Regions 4 104,540 0,8337 18,73 0,001 Within populations 209 747,428 3,5762 80,33 0,001 Population 1 26 100,296 3,9761 13,81 Population 2 29 100,833 3,5400 12,30 Population 3 27 95,964 3,6199 12,58 Population 4 26 101,370 3,8858 13,50 Population 5 28 105,793 3,8781 13,47 Population 6 29 112,033 3,8752 13,46 Population 7 20 57,857 2,9220 10,15 Population 8 24 73,280 3,0876 10,73 Total 216 937,327 4,4518 If the over all population is panmitic, supposing any restriction to gene flow, it is expected that the variation among randomly sampled populations would be zero and, therefore, all genetic variation would be found within them. Actual significant difference among populations however, demonstrates that these have been influenced by microevolution factors, differentiating them. The estimated value of the φST statistic was 0.197, significant at 1% probability. This parameter derived from AMOVA is analogous to Wright’s FST statistic and is a measure of the differentiation among populations caused by genetic drift. In the case of gene flow among geographically close populations, these will tend to become similar than distant ones, according to a microevolution isolation-by-distance model or stepping-stones model (FUTUYMA, 1992). In this case the genetic variability found among regions was not significant, indicating that the dissimilarities among populations within regions tend to be similar to the ones among different regions. This can be explained by a possible restriction of gene flow even at smaller distances, among populations, by possible genetic drift, generating random variation pattern, which would correspond to the Wright’s island model (FUTUYMA, 1992). The genetic variability in a population is the result of the balanced joint action in microevolution processes (selection, mutation, migration and genetic drift) responsible for the genetic and genotypic changes. Since the RAPD markers are considered selectively neutral, the differentiation among populations found here must be result of genetic drift, not counterbalanced by gene flow. Although being a cross pollinated species, H. speciosa presents mass-flowering and pollination is mainly by geitonogamy (PINTO et al., 2008). So the fertilization can occurs mainly between neighbor plants. Seed dispersion is another factor responsible for gene flow in plant populations. The strong structure found in the mangaba populations suggests that seed dispersion is insufficient to maintain the genetic connection among the sampled populations. The fragmentation of the Cerrado vegetation can be contributing to this, restricting the movement of pollen and seed dispersants among remaining populations. The apparent gene flow (Nm) for a set of subpopulations can be estimated indirectly via the FST statistic. In this method Wright’s island model (1951) is admitted and the equilibrium of migration and genetic drift is assumed. Thus, when the Nm values are higher or equal to 1, i.e., when the apparent gene flow is equivalent to one or more individuals migrated per generation, the gene flow effects are sufficient to counterbalance genetic drift effects and impairs the divergence of populations. For Nm ≥ 1, FST must therefore be ≤ 0.20 (WRIGHT, 1951). The φST parameter, based on dominant markers, is analog to FST, since it measures the differentiation among subpopulations caused by genetic drift. However, it may not be used indiscriminately to determine the number of migrants per generation, since φST and FST estimates are frequently discrepant for the same set of subpopulations. With these restrictions, a φST value of 0.1956, as found in the present case, could suggest that the subpopulations under study are in a condition close to equilibrium between gene drift and gene flow. Some studies with fruit species of the Brazilian Cerrado have shown considerable genetic variability among populations which is, in general, spatially structured (TELLES et al., 2003; ZUCCHI 477 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 et al., 2003; TRINDADE; CHAVES, 2005; SOARES et al., 2008a; SOARES et al., 2008b), this fact requires broad sampling for conservation purposes. Studies with H. speciosa based on molecular markers are scarce. In a study based on polymorphism of two non-coding regions of chloroplast DNA, significant variation was found among populations, corresponding to 7 to 9% of the total genetic variation (SILVA, 2006). Using SSR markers to analyze populations of four botanical varieties of the Cerrado region, Rodrigues (2009) found a FST estimate of 0.19, indicating high level of genetic differentiation among populations. This result is similar to the ΦST estimates found here. Variation among subpopulations is also possible at a larger geographic scale, in other words, the more distant they are from each other, the more they differ in allele frequencies and in the value of phenotypic traits, although the correlation is often not high. The study of spatial distribution of genetic variability is a strong support for evolution studies of the species. The results of AMOVA did not detect significant variation among regions, indicating that the spatial structure could not be detected with the molecular marker used here. This indicates that the effect of genetic drift superposes that of gene flow, since genetic drift causes a random structure of variability among populations. Results suggest that the gene flow did not counterbalance this differentiation, even at relatively small distances. A possible explanation for this variation pattern could be a restricted seed and pollen dispersion. In fact, mangaba subpopulations tend to form plant aggregates in isolated demes. Mangaba is predominantly pollinated by night-active Lepidoptera (DARRAULT; SCHLINDWEIN, 2006). The spatial pattern of plant distribution, together with the short reach of the pollinating insects favors preferential pollination of nearby plants, which in turn favors the differentiation of populations. The actual fragmentation of the Cerrado areas may lead to a greater isolation among subpopulations which could affect gene flow even more between them. The variance found within subpopulations corresponded to 80.33% of the total variance and, when this source of variation was partitioned, the variability within each subpopulation varied only slightly, in other words, the contribution of each population to the total variation was similar. The distribution of intrapopulation genetic variation reflects, at least partly, the action of genetic processes from the population point of view, i.e., of selection and genetic drift (LOVELESS; HAMRICK, 1984). When a large discrepancy between genetic variability within populations is observed, the action of these genetic processes differs for the subpopulations. Our results therefore suggest that the genetic erosion is not threatening the variability among plants within populations. In the sampling process the variability in any one of the subpopulations studied is practically equal to the others. This is important for the maintenance of the genetic variability, since the conservation strategy is facilitated. The mean dissimilarity among plants within populations varied from 0.26 to 0.39 (Table 2). The AMOVA-based variances and the dissimilarities therefore indicate that the maintenance of internal variability in the populations is relatively similar. Table 2. Mean similarity (Jaccard) and the respective mean dissimilarity of plants within subpopulations of H. speciosa, based on thirty one RAPD bands. Populations Similarities Dissimilarities 01 0,6749 0,3251 02 0,7207 0,2793 03 0,6525 0,3475 04 0,6066 0,3934 05 0,6880 0,3120 06 0,6239 0,3761 07 0,7412 0,2588 08 0,6631 0,3369 The mean genetic similarity among subpopulations varied from 0.54 to 0.68. No grouping according to regions was observed in the dendrogram (Figure 2). The association of dendrogram nodes was generally weak (highest association 53.56 %) among populations and their grouping pattern was not clear for the analyzed amplified fragments. This result had been expected, 478 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 since the non-significance of the variability found among regions indicates random distribution for the amplified fragments analyzed. The matrix correlation between the genetic dissimilarities and geographic distance matrices was not significant. The Pearson correlation coefficient (r) was 0.005, non-significant by the Mantel test. This result reinforces the assumption that the genetic variability among subpopulations, detected with the molecular markers used, is randomly distributed in space. This situation of random genetic spatial distribution, though atypical, was also found in same instances (e.g. SAENZ-ROMERO et al. 2001). Among Cerrado species, a similar situation was verified in 11 natural populations of Annona crassiflora in the state of Goiás (BLANCO et al., 2007). Study with 32 H. speciosa populations based on SSR markers, also showed no significant correlation between genetic and geographical distances (Rodrigues, 2009), corroborating the random pattern of the genetic variation among populations and the occurrence of restricted gene flow just between neighbor populations. Figure 2. Grouping pattern of eight mangaba (H. speciosa) subpopulations defined by the UPGMA method on the mean similarity index of Jaccard (cophenetic correlation = 0.771**) and thirty one RAPS bands. Vencovsky and Crossa (2003) showed that the higher the value of FST, the larger must be the number of subpopulations sampled to ensure an adequate effective size. Although some studies show that φST estimates tend to be higher than those of FST (ZUCCHI et al., 2005), it is admitted that a high φST value corresponds also to significant, FST values. The high variability found in the subpopulations therefore suggests that a strategy for in situ, as well as ex situ conservation for the species must be based on sampling a large number of subpopulations. The fact that no spatial structure of genetic variability was detected indicates that subpopulations may be sampled randomly. It must however be emphasized that aside from the quantification and distribution of genetic variability detected by molecular markers, other variability descriptors, such as phenotypic variation of morphological traits, must also be considered; this includes variation in leaf pilosity and color, color pattern, fruit size, flower morphology and others. Pronounced differences between the subpopulations studied were observed but not quantified, with regard to some of these traits. This indicates that the sampling of subpopulations should cover different regions of occurrence of the species. CONCLUSIONS The natural populations of H. speciosa present highly significant interpopulation genetic variability, suggesting the need of sampling a large number of areas for collecting germplasm or in situ conservation. The natural populations of H. speciosa have similar within population genetic variability, showing no spatial genetic structure among populations, for the RAPD loci used here. ACKNOWLEDGMENT We are grateful to ASG Coelho and JL Nascimento for helping in field collection and to an anonymous reviewer for helpful suggestions and comments on the manuscript. CNPq – Brazil provided financial support and research grants to LJ Chaves and R Vencovsky. NF Moura received master fellowship from Capes – MEC – Brazil. 479 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 RESUMO: A mangabeira (Hancornia speciosa Gomes) é uma espécie frutífera com grande potencial para exploração econômica, encontrada em várias regiões do Brasil, sendo freqüente no bioma Cerrado. Informações sobre a variabilidade genética das populações existentes neste bioma são escassas. Este estudo teve como objetivo caracterizar a estrutura genética de oito populações situadas nos Estados de Goiás, Bahia e Minas Gerais por meio de marcadores RAPD. A estrutura da variabilidade genética foi avaliada utilizando-se a análise de variância de dados moleculares. Uma proporção significativa da variância genética foi encontrada entre populações, correspondendo a 19,6 % da variação total. A correlação entre as matrizes de distâncias geográficas e genéticas não foi significativa, indicando que não existe estruturação espacial da variabilidade genética entre populações, para os locos marcadores RAPD utilizados. A alta variabilidade sugere que estratégias para conservação in situ, bem como ex situ, devem ser baseadas em amostragem de um grande número de populações locais. PALAVRAS-CHAVE: Mangaba. Marcador molecular. Genética de população. Espécie nativa. REFERENCES BERNARDES, T. G.; NAVES, R. V.; REZENDE, C. F. A.; BORGES, J. D.; CHAVES, L. J. Propagação sexuada do pequizeiro (Caryocar brasiliense Camb.) estimulada por ácido giberélico. Pesquisa Agropecuária Tropical, Goiânia, v. 38, p. 71-77, 2008. BLANCO, A. J. V.; PEREIRA, M. F.; COELHO, A. S. G.; CHAVES, L. J. Diversidade genética em populações naturais de Araticunzeiro (Annona crassiflora Mart.) por meio da análise de seqüências de cpDNA. Pesquisa Agropecuária Tropical, Goiânia, v. 37, n. 3, p. 169-175, 2007. CAVALCANTE, T. R. M.; NAVES, R. V.; SERAPHIN, J. C.; CARVALHO, G. D. Diferentes ambientes e substratos na formação de mudas de araticum. Revista Brasileira de Fruticultura, Jaboticabal, v. 30, p. 235- 240, 2008. COLLEVATTI, R. G.; GRATTAPAGLIA, D.; HAY, J. D. Population genetic structure of the endangered tropical tree species Caryocar brasiliense, based on variability at microsatellite loci. Molecular Ecology, Oxford, v. 10, p. 349-356, 2001. COLLEVATTI, R. G.; GRATTAPAGLIA, D.; HAY, J. D. Evidences for multiple maternal lneages of Caryocar brasiliense populations in the Brazilian Cerrado based on the analysis of chloroplast DNA sequences and microsatellite haplotype variation. Molecular Ecology, Oxford, vol. 12, p. 105-115, 2003. DARRAULT, R. O.; SCHLINDWEIN, C. Polinização. In: SILVA-JUNIOR, J.F.; LEDO, A. S.(Ed.). A Cultura da Mangaba. Embrapa Tabuleiros Costeiros, Aracaju, 2006. p. 43-56. EXCOFFIER, L.; SMOUSES, P. E.; QUATTRO, J. M. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitocondrial DNA restriction data. Genetics, United States, v. 131, p. 479-491, 1992. FUTUYMA, D. J. Biologia evolutiva. 2a ed. Ribeirão Preto: Sociedade Brasileira de Genética. 631p. 1992. GANGA, R. M. D.; CHAVES, L. J.; NAVES, R. V. Genetic parameters in Hancornia speciosa Gomes progenies from Cerrado. Scientia Forestalis, Piracicaba, vol. 37, p. 395-404, 2009. GANGA, R. M. D.; FERREIRA, G. A.; CHAVES, L. J.; NAVES, R. V.; NASCIMENTO, J. L. Characterization of fruit and trees from natural populations of Hancornia speciosa Gomes of Cerrado. Revista Brasileira de Fruticultura, v. 32, n. 1, p. 101-113, 2010. LOVELESS, M. D.; HAMRICK, J. L. Ecological determinants of genetic structure in plant populations. Ann. Rev. Ecol. Syst., Palo Alto, v. 15, p. 65-95, 1984. 480 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 MANTEL, N. The detection of disease clustering and a generalized regression approach. Cancer Research, Stanford University Libraries, v. 27, p. 209-220, 1967. MARTINS, K.; CHAVES, L. J.; BUSO, G. S. C.; KAGEYAMA, P. Y. Mating system and fine-scale spatial genetic structure of Solanum lycocarpum St. Hil. (Solanaceae) in the Brasilian Cerrado. Conservation Genetics, The Netherlands, vol.7, n. 6, p. 957-969, 2006. MOURA, N.F. Estrutura genética de subpopulações de mangabeira (Hancornia speciosa Gomez) nos cerrados do Brasil Central. 2003. 70p. Dissertação (Mestrado em Agronomia: Genética e Melhoramento de Plantas) - Escola de Agronomia e Engenharia de Alimentos - Universidade Federal de Goiás. Goiânia, 2003. MOURA, N. F.; CHAVES L. J.; VENCOVSKY R.; ZUCCHI M. I.; PINHEIRO J. B.; MORAIS L. K.; MOURA M. F. Seleção de marcadores RAPD para o estudo da estrutura genética de populações de Hancornia speciosa Gomes. Bioscience Journal, Uberlândia, v. 21, n. 3, p. 119-125, 2005. MURRAY, M. G.; THOMPSON, W. F. Rapid isolation of high molecular weight plant DNA. Nucleic Acids Reserch, Oxford, v. 8, p. 1134-1137, 1980. PINTO, C.E.; OLIVEIRA, R.; SCHLINDWEIN, C. Do consecutive flower visits within a crown diminish fruit set in mass-flowering Hancornia speciosa (Apocynaceae)? Plant Biology, v. 10, p. 408-412, 2008. RODRIGUES, A. J. L. Desenvolvimento de marcadores microssatélites e estrutura genética de Hancornia speciosa Gomes (Apocynaceae). 2009.106p. Tese de Doutorado. (Doutorado em Agronomia: Genética e Melhoramento de Plantas) – Escola de Agronomia e Engenharia de Alimentos – Universidade Federal de Goiás. Goiânia, 2009. ROGERS, S. O.; BENDICH, A. J. Extraction of DNA from milligram amounts of fresh, herbarium and mummified plants tissues. Plant Molecular Biology, The Netherlands, v. 5, p. 69-76, 1985. ROHLF, F. J. NTSYSpc: numerical taxonomy and multivariate analysis system. New York: Exeter, 1989. SAENZ-ROMERO, C.; GURIES, R. P.; MONK, A. I. Landscape genetic structure of Pinus banksiana: allozyme variation. Can. J. Bot., Canadá, v. 79, p. 871-878, 2001. SCHNEIDER, S.; ROESSLI, D.; EXCOFFIER, L. (2000) Arlequim ver.: a software for population data analysis. (Software). Geneva: University of Geneva, Genetic and Biometry Laboratory. Disponível em: . Acesso em 12/12/2002. SILVA, M.N. Magnitude e distribuição da diversidade genética em populações naturais de mangabeira pela análise de cpdna. 2006. 72f. Dissertação (Mestrado) – Escola de Agronomia e Engenharia de Alimentos, Universidade Federal de Goiás, Goiânia, 2006. SNEATH, P. H.; SOKAL, R. R. Numerical taxonomy: the principles and pratice of numerical classification. San Francisco, W.H. Freeman, 1973. 573p. SOARES, T. N.; CHAVES, L. J.; TELLES, M. P. C.; DINIZ-FILHO, J. A. F.; RESENDE, L. V. Landscape conservation genetics of Dipteryx alata (“Baru” tree: Fabaceae) From Cerrado region of central Brazil. Genetica, The Netherlands, v. 132, p. 9-19, 2008a. SOARES, T. N.; CHAVES, L. J.; TELLES, M. P. C.; DINIZ-FILHO, J. A. F.; RESENDE, L. V. Distribuição espacial da variabilidade genética intrapopulacional de Dipteryx alata. Pesquisa Agropecuária Brasileira, Brasília, v. 43, n. 9, p. 1151-1158, 2008b. SOUZA, O. A. de; NASCIMENTO, J. L.; NAVES, R. V.; BORGES, J. D. Propagação sexuada de pequizeiro (Caryocar brasilienses Camb.): Efeito de procedências de frutos e do ácido giberélico na emergência de plântulas. Pesquisa Agropecuária Tropical, Goiânia, v.37, n. 3, p.131-136, 2007. 481 Genetic structure... MOURA, N. F. et al. Biosci. J., Uberlândia, v. 27, n. 3, p. 473-481, May/June2011 TELLES, M. P. C.; COELHO, A. S. G.; CHAVES, L. J.; DINIZ-FILHO, J. A. F.; VALVA, F. D. Genetic diversity and population structure of Eugenia disenterica DC ('cagaiteira'- Myrtaceae) in Central Brazil: Spatial analysis and implications for conservation and management. Conservation Genetics, The Netherlands, v.4, p. 685-695, 2003. TRINDADE, M. G.; CHAVES, L. J. Genetic structure of natural Eugenia dysenterica DC (Myrtaceae) populations in northeastern Goiás, Brasil, acessed by morphological traits and RAPD markers. Genetics and Molecular Biology, Ribeirão Preto, v.28, n. 3, p. 407-413, 2005. VERA, R.; SOUZA, E. R. B.; FERNANDES, E. P.; NAVES, R. V.; JÚNIOR, M. S. S.; CALIARI, M.; XIMENES, P. A. Caracterização física e química de frutos do pequizeiro (Caryocar brasiliense camb.) oriundos de duas regiões no estado de Goiás, Brasil. Pesquisa Agropecuária Tropical, Goiânia, v.37, n. 2, p. 93-99, 2007. VERA, R.; NAVES, R. V.; NASCIMENTO J. L.; CHAVES, L. J.; LEANDRO, W. M.; SOUZA, E. R. B. Caracterização física de frutos do pequizeiro (caryocar brasiliense camb.) no estado de Goiás. Pesquisa Agropecuária Tropical, Goiânia, v.35, n. 2, p. 73-79, 2005. VENCOVSKY, R.; CROSSA, J. (2003) Measurements of Representativeness Used in Genetic Resources Conservation and Plant Breeding. Crop Science, v. 43, p. 1912-1921, 2003. WRIGHT, S. The genetical structure of populations. Ann. Eugen., New York, v.15, p. 323-354, 1951. ZUCCHI, M. I.; BRONDANI, R. P. V.; PINHEIRO, J. B.; CHAVES, L. J.; COELHO, A. S. G.; VENCOVSKY, R. Genetic structure and gene flow in Eugenia dysenterica DC in Brasilian Cerrado utilizing SSR markers. Genetics and Molecular Biology, Ribeirão Preto, v.26, n. 4, p. 449-457, 2003. ZUCCHI, M. I.; PINHEIRO, J. B.; CHAVES, L. J.; COELHO, A. S. G.; COUTO, M. A.; MORAIS, L. K.; VENCOVSKY, R. Genetic structure and gene flow of Eugenia dysenterica natural populations Pesquisa Agropecuária Brasileira, Brasília, v. 40, n.10, p. 975-980, 2005.