Gláucia Salles Cortopassi Buso

The garlic cultivars grown in Brazil evolved from somatic mutations and clone selection by breeding programs and by the introduction of germplasm from other countries. Morphological characters have been used to differentiate these cultivars. Two hundred and six random amplified polymorphic DNA markers were utilized for a diversity analysis of the 17 most planted garlic cultivars in Brazil. Bootstrap analysis showed that the number of markers was efficient and sufficient to obtain a coefficient of variation of 10%. Similarity varied between 16 and 98% and cluster analysis showed that, in general, genetic similarities correlate with morphological characters of the cultivars and production cycle variation. High bootstrap values at most of the nodes supported the dendrogram stability. The grouping of most varieties agreed well with previous reports based on morphological characters. As a vegetative-propagated species, viral diseases are a key problem regarding production and quality of the bulbs, causing gradual loss of yield and decrease in storage capacity. To improve the health quality of garlic seed, a virus-free stock of garlic cloves of the Amarante cultivar was obtained. The ability to distinguish garlic cultivars to detect varietal mixing after in vitro multiplication is extremely important, since correct identification is not possible until bulbs are produced. Random amplified polymorphic DNA markers were also used to differentiate cultivars while they are in vitro and not amenable to morphological discrimination. No difference was identified between the fingerprints of the virus-free or of the infected bulks of Amarante, showing that there was no clove mixing in the handling of material in the clonal multiplication phase.

Acessos de melancia do Banco de Germoplasma da Embrapa Semi-Árido, coletados a partir de sementes de polinização livre na Chapada Diamantina, em Irecê e em Vitória da Conquista, na Bahia, foram caracterizados através de descritores morfológicos e moleculares, adicionando-se ao grupo, como testemunha a cultivar Crimson Sweet. O experimento de campo, em blocos ao acaso com três repetições, foi conduzido na estação experimental da Embrapa Semi-Árido, em Petrolina-PE. Os dados obtidos foram submetidos à análise uni- e multivariada, com agrupamento pelo método de Tocher. Amostras de cinco a oito indivíduos de cada um dos 43 acessos avaliados, foram submetidas à análise molecular utilizando marcadores RAPD, obtidos através de seis primers. O polimorfismo revelado por estes marcadores foi utilizado para estimar os coeficientes de similaridade de Jaccard médios para cada acesso e realizar a análise de agrupamentos, também pelo método de Tocher. Foram encontradas diferenças significativas entre os acessos para todos os descritores morfológicos aplicados, exceto número de frutos por planta. Oito descritores morfológicos (comprimento do ramo principal, espessura da casca na região da inflorescência, padrão de listras dos frutos, teor de sólidos solúveis, formato dos frutos, cor externa dos frutos, espessura da casca na região distal e peso dos frutos) explicaram 79% da variação entre acessos. Quando a testemunha, a cultivar Crimson Sweet foi incluída na análise multivariada, foram formados apenas dois grupos: um grupo com ela própria e, o outro, com todos os demais acesssos. Quando a testemunha foi excluída da análise, os acessos se distribuíram em nove grupos, sendo que um deles englobou 69% do total de acessos avaliados. Foram identificados 64 marcadores RAPD, sendo que 31 foram polimórficos. O agrupamento dos acessos com base na informação desses marcadores, utilizando o método de Tocher, resultou em 28 grupos, 24 deles com um único acesso. Nenhum dos grupos com mais de um acesso foi constituído por acessos provenientes da mesma região. Os marcadores RAPD revelaram polimorfismo expressivo em vários acessos, o que indica que estes marcadores podem ser utilizados na caracterização de acessos em Bancos de Germoplasma de melancia.

Despite the great advances in genomic technology observed in several crop species, the availability of molecular tools such as microsatellite markers has been limited in melon (Cucumis melo L.) and cucurbit species. The development of microsatellite markers will have a major impact on genetic analysis and breeding of melon, especially on the generation of marker saturated genetic maps and implementation of marker assisted breeding programs. Genomic microsatellite enriched libraries can be an efficient alternative for marker development in such species. Seven hundred clones containing microsatellite sequences from a Tsp-AG/TC microsatellite enriched library were identified and one-hundred and forty-four primer pairs designed and synthesized. When 67 microsatellite markers were tested on a panel of melon and other cucurbit accessions, 65 revealed DNA polymorphisms among the melon accessions. For some cucurbit species, such as Cucumis sativus, up to 50% of the melon microsatellite markers could be readily used for DNA polymophism assessment, representing a significant reduction of marker development costs. A random sample of 25 microsatellite markers was extracted from the new microsatellite marker set and characterized on 40 accessions of melon, generating an allelic frequency database for the species. The average expected heterozygosity was 0.52, varying from 0.45 to 0.70, indicating that a small set of selected markers should be sufficient to solve questions regarding genotype identity and variety protection. Genetic distances based on microsatellite polymorphism were congruent with data obtained from RAPD marker analysis. Mapping analysis was initiated with 55 newly developed markers and most primers showed segregation according to Mendelian expectations. Linkage analysis detected linkage between 56% of the markers, distributed in nine linkage groups. Genomic library microsatellite enrichment is an efficient procedure for marker development in melon. One-hundred and forty-four new markers were developed from Tsp-AG/TC genomic library. This is the first reported attempt of successfully using enriched library for microsatellite marker development in the species. A sample of the microsatellite markers tested proved efficient for genetic analysis of melon, including genetic distance estimates and identity tests. Linkage analysis indicated that the markers developed are dispersed throughout the genome and should be very useful for genetic analysis of melon.

The knowledge of population structure and genetic diversity of wild relatives of rice is needed to investigate their evolutionary history and potential use in breeding programs. Very little is known about the wild rice species (Oryza spp.), particularly those that are native to South America. A study using isozyme and RAPD markers was conducted to estimate the level of genetic diversity of four South American wild rice populations (Oryza glumaepatula) recently collected in the Amazon forest and western Brazil rivers. F‐statistics and genetic diversity parameters calculated from isozyme and RAPD markers indicated high values for inbreeding coefficients and differentiation among the four populations. In agreement with this, a pattern of greater variation between than within populations was observed with both types of markers. These findings were corroborated by an AMOVA analysis, which indicated that a large portion of the total genetic variation was attributed to regional divergence. The partition of the AMOVA analysis among populations showed that most of the genetic diversity was due to differences among populations. This distribution pattern of genetic variation of O. glumaepatula populations is in agreement with the expectation for an autogamous species and provides important baseline data for conservation and collection strategies for this species.