Estrutura populacional de isolados de Magnaporthe oryzae Couch & Kohn de arroz (Oryza sativa) no Brasil

• Autor(a) principal: Gonçalves, Fábio José
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tede/11506
• topic: Brusone; Pyricularia oryzae; Patótipos; Marcadores moleculares; Blast; Oryza sativa; Diversity; Resistance; CIENCIAS AGRARIAS::AGRONOMIA::CIENCIA DO SOLO

Rice blast is caused by Magnaporthe oryzae, an ascomycete which presents an accentuate variability, probably caused by mutation, parassexual recombination, Avr gene deletions and the movement of transposition elements (transposons and retrotransposons). This pathogen is rganized in populations composed of pathotypes with distinct virulence pattern, whose identification is based on the blast leaves reactions of national and international rice differential cultivars. The objectives of this work were to determine national and international pathotypes frequencies of M. oryzae isolates, collected from different brazilian rice producing regions; to identify the diversity index that better fits with M. oryzae population diversity; to detect the presence of some avirulence genes; and investigate the genetic variability of collected isolates using 18 microsatellite markers. National and international differential series and commercial rice cultivars were sown in trays and, 21 days after planting it were spray inoculated with a conidia suspension of M. oryzae, in greenhouse conditions. The plants were evaluated nine days after spray inoculation and notes 0 to 3 were considerate resistant (R) and 5 to 9 susceptible (S). The pathotypes were identified and the frequencies calculated. The indexes of Gleason, Shannon and Simpson were calculated over three successive years and analyzed by multiple linear regression with Factor Analysis throw means of principal components method and stepwise. For the genetic studies 494 isolates were grown in liquid culture medium, followed by mycelium filtration, frozen and lyophilized for DNA extraction, which was quantified, amplified and separated by electrophoresis. The polymorphism data, the Nei's genetic distance index and the AMOVA were calculated with Structure, Genetix and Arlequin softwares. AS results we had 523 pathotyped and the frequencies were IF 1 and BA 1 (18% and 12%) in Goiás; IA 45 and BI 1 (32% and 13%) in Mato Grosso do Sul; BD 16 (33%) in Mato Grosso; IA 45 1 and BF (20% and 11%) in Santa Catarina; IA 1 and BA 105 (33% and 22%) in Rio Grande do Sul; IA 1 and BA 126 (14% and 11%) in Tocantins; IE 1 (45%) and BC8, BD16 and BI 1( 25% each) in Para; IF 1 and 8 (52% and BC 40%) in Rondônia; IC 1 and IG 1(18% each) and BI 1 in Minas Gerais; According to Nei’s index the closest international group were C and D (0.080) and the most distant were groups A and F (0.477), among the national groups, the closest were (C) and (D) (0.014) and the most distance F and H (0.429). Based on the regressions results, Gleason index should be represented by R and E factors, and Shannon by factors N and R. Among the isolates which were compatible to cv. CO-39 (S) 86.7% were incompatible (R) to isolineas C 101 LAC (Pi-1/Avr-1), 81.5% to C 101 PKT (Pi-4a/Avr-4a) and (66.7%) C 101 A51 (Pi-2-Avr-2), for samples from both irrigated and upland conditions. The best grouping value for 494 isolates was K = 19. According to AMOVA analyses the variability among subpopulations was 62.87% and 37.13% within subpopulations. Among the 18 locos studied 16 were polymorphic, the markers Pyrms 077b-078 e Pyrms 319-320 were the best ones for detecting differences among the subpopulations with 80.50% and 80.34% of variation, respectively. The markers Pyrms099b-100b, Pyrms037-038 and Pyrms 409-410 were the best one for detecting differences within subpopulations with 70.16%, 67.06% and 63.98% of variation, respectively. The value of genetic differentiation of populations (FST) was 0, 62868 (p < 0.001) calculated by 10,000 random permutations. According to the Nei's Index, subpopulations 8 and 4 are the ones most distance (1,822) and subpopulations 15 and 19 are the closest (0.119). We concluded that, in Brazil the highest frequency were 45-IA and BA-1 pathotypes and, that the molecular markers used in this work were efficient to detect population structure organizing the 494 representative isolates of rice producing regions in 19 subpopulations, according to the cultivar’s origin of the isolate with a strong environment interaction