RESUMO
Polyploid induction has been used for plant breeding to produce bigger and more robust plants than diploid types. The present study aimed to develop a methodology for in vitro induction of polyploidy in cassava. Apical and lateral microcuttings from the BRS Formosa variety were treated with six oryzalin concentrations for 24 and 48 h. The same methodology was used for colchicine with different concentrations. After 45 days of cultivation and an additional 45 days of subculture, the viability of the explants was assessed and plant acclimatization was performed. Ploidy was determined using flow cytometry. Oryzalin dose and exposure negatively affected cassava explant growth and development compared to untreated explants. Furthermore, apical and lateral explants responded differently to the treatments, showing a diversity in antimitotic sensitivity and effect that is tissue-type specific. In contrast, the doses of 1.25 to 6.25 mM colchicine resulted in high mortality of cassava explants. Therefore, the type of antimitotic affects the morphophysiological behavior of cassava plants in vitro, although apical explants have higher viability and regenerative capacity compared to lateral explants. In addition, the lateral explants have lower mixoploid rates compared to apical explants. Of the 310 plants generated by oryzalin treatments, 277 were diploid, 31 were mixoploid, and 2 were tetraploid. Exposure to oryzalin led to low rate of tetraploids and colchicine caused phytotoxic reactions and death of the explants. The tetraploids were multiplied in vitro to evaluate their yield in the field as well as their behavior against abiotic and biotic stress.
Assuntos
Colchicina/farmacologia , Dinitrobenzenos/farmacologia , Manihot/genética , Mutagênicos/farmacologia , Sulfanilamidas/farmacologia , Tetraploidia , Moduladores de Tubulina/farmacologia , Manihot/efeitos dos fármacos , Melhoramento Vegetal/métodosRESUMO
Our objective was to establish a minimum number of morphological descriptors for the characterization of banana germplasm and evaluate the efficiency of removal of redundant characters, based on univariate and multivariate statistical analyses. Phenotypic characterization was made of 77 accessions from Bahia, Brazil, using 92 descriptors. The selection of the descriptors was carried out by principal components analysis (quantitative) and by entropy (multi-category). Efficiency of elimination was analyzed by a comparative study between the clusters formed, taking into consideration all 92 descriptors and smaller groups. The selected descriptors were analyzed with the Ward-MLM procedure and a combined matrix formed by the Gower algorithm. We were able to reduce the number of descriptors used for characterizing the banana germplasm (42%). The correlation between the matrices considering the 92 descriptors and the selected ones was 0.82, showing that the reduction in the number of descriptors did not influence estimation of genetic variability between the banana accessions. We conclude that removing these descriptors caused no loss of information, considering the groups formed from pre-established criteria, including subgroup/subspecies.
Assuntos
Ecótipo , Musa/genética , Agricultura , Algoritmos , Entropia , Variação Genética , Análise Multivariada , Musa/anatomia & histologia , Fenótipo , Análise de Componente Principal , Sementes/crescimento & desenvolvimentoRESUMO
Several chromosome types have been recognized in Citrus and related genera by chromomycin A(3 )(CMA) banding patterns and fluorescent in situ hybridization (FISH). They can be used to characterize cultivars and species or as markers in hybridization and backcrossing experiments. In the present work, characterization of six cultivars of P. trifoliata ("Barnes", "Fawcett", "Flying Dragon", "Pomeroy", "Rubidoux", "USDA") and one P. trifoliata x C. limonia hybrid was performed by sequential analyses of CMA banding and FISH using 5S and 45S rDNA as probes. All six cultivars showed a similar CMA(+) banding pattern with the karyotype formula 4B + 8D + 6F. The capital letters indicate chromosomal types: B, a chromosome with one telomeric and one proximal band; D, with only one telomeric band; F, without bands. In situ hybridization labeling was also similar among cultivars. Three chromosome pairs displayed a closely linked set of 5S and 45S rDNA sites, two of them co-located with the proximal band of the B type chromosomes (B/5S-45S) and the third one co-located with the terminal band of a D pair (D/5S-45S). The B/5S-45S chromosome has never been found in any citrus accessions investigated so far. Therefore, this B chromosome can be used as a marker to recognize the intergeneric Poncirus x Citrus hybrids. The intergeneric hybrid analyzed here displayed the karyotype formula 4B + 8D + 6F, with two chromosome types B/5S-45S and two D/5S-45S. The karyotype formula and the presence of two B/5S-45S chromosomes clearly indicate that the plant investigated is a symmetric hybrid. It also demonstrates the suitability of karyotype analyses to differentiate zygotic embryos or somatic cell fusions involving trifoliate orange germplasm.