RESUMO
Despite numerous published reports of quantitative trait loci (QTL) for drought-related traits, practical applications of such QTL in maize improvement are scarce. Identifying QTL of sizeable effects that express more or less uniformly in diverse genetic backgrounds across contrasting water regimes could significantly complement conventional breeding efforts to improve drought tolerance. We evaluated three tropical bi-parental populations under water-stress (WS) and well-watered (WW) regimes in Mexico, Kenya and Zimbabwe to identify genomic regions responsible for grain yield (GY) and anthesis-silking interval (ASI) across multiple environments and diverse genetic backgrounds. Across the three populations, on average, drought stress reduced GY by more than 50 % and increased ASI by 3.2 days. We identified a total of 83 and 62 QTL through individual environment analyses for GY and ASI, respectively. In each population, most QTL consistently showed up in each water regime. Across the three populations, the phenotypic variance explained by various individual QTL ranged from 2.6 to 17.8 % for GY and 1.7 to 17.8 % for ASI under WS environments and from 5 to 19.5 % for GY under WW environments. Meta-QTL (mQTL) analysis across the three populations and multiple environments identified seven genomic regions for GY and one for ASI, of which six mQTL on chr.1, 4, 5 and 10 for GY were constitutively expressed across WS and WW environments. One mQTL on chr.7 for GY and one on chr.3 for ASI were found to be 'adaptive' to WS conditions. High throughput assays were developed for SNPs that delimit the physical intervals of these mQTL. At most of the QTL, almost equal number of favorable alleles was donated by either of the parents within each cross, thereby demonstrating the potential of drought tolerant × drought tolerant crosses to identify QTL under contrasting water regimes.
Assuntos
Adaptação Fisiológica/genética , Genoma de Planta , Locos de Características Quantitativas , Zea mays/genética , Cruzamento , Mapeamento Cromossômico , Secas , Meio Ambiente , Marcadores Genéticos , Quênia , México , Fenótipo , Polimorfismo de Nucleotídeo Único , Estresse Fisiológico/genética , Água/análise , ZimbábueRESUMO
Frutos no estádio vermelho maduro de trinta e três acessos de tomate do banco de germoplasma do CCA-UFES, sendo 15 do "tipo Cereja" e 18 do "tipo Salada", foram caracterizados com base nos seguintes descritores morfoagronômicos: peso médio do fruto; comprimento médio do fruto; diâmetro médio do fruto; diâmetro da cicatriz peduncular; espessura do mesocarpo do fruto; teor de sólidos solúveis e pH do fruto. Com base nas variáveis teor de sólidos solúveis (TSS) e pH do fruto, calculou-se a variável sabor, por meio da relação TSS/pH². Foram realizadas análises de variâncias e análises de correlações simples e parciais, determinando a distribuição dos acessos quanto às classes de sabor. Com base nas distribuições de frequências, formaram-se quatro classes para essa variável, sendo que, a maior parcela (33 por cento) dos acessos "tipo Salada" alocaram-se na classe mais baixa (0,05-0,12 ºBrix.pH-2), correspondendo a frutos de sabor menos desejável, enquanto a maior parcela (30 por cento) dos acessos "tipo Cereja", alocaram-se em classe imediatamente superior (0,13-0,20 ºBrix.pH-2). Os dados demonstram a possibilidade de se obter ganhos genéticos para sabor concomitantemente ao peso, comprimento e espessura do mesocarpo dos frutos.
Fruits in the ripe red stage of thirty three tomato (Lycopersicon esculentum Mill.) accessions from CCA-UFES germplasm collection, being 15 "Cherry type" and 18 "Salad type", were characterized based on the following morphooagronomic descriptors: average fruit weight; average fruit length; average fruit diameter; peduncle scar diameter; fruit mesocarp thickness; soluble solids and fruit pH. Based on the data from the soluble solids and fruit pH, the flavor variable was calculated by the ratio TSS/pH². Variance analysis and simple and partial correlations were calculated, and the accessions were clustered in flavor classes. Based on the statistical model of distribution of frequencies, four classes were formed for flavor variable. The largest portion (33 percent) of the accessions of salad type were located in the lowest class (0.05-0.12 ºBrix.pH-2), corresponding to the accessions with not so desirable a flavor, while the largest portion (30 percent) of the accessions of the "cherry" type were clustered in the next class above (0.13-0.20 ºBrix.pH-2). The possibility of concomitantly obtaining genetic gains for flavor, weight, length and mesocarp thickness of the fruits was demonstrated.