RESUMEN
Common rust (CR) caused by Puccinia sorghi Schwein is one of the major foliar diseases of maize (Zea mays L.) in Eastern and Southern Africa. This study was conducted to (i) evaluate the response of elite tropical adapted maize inbred lines to Puccinia sorghi and identify resistant lines (ii) examine associations between CR disease parameters and agronomic traits, and (iii) assess the genetic diversity of the inbred lines. Fifty inbred lines were evaluated in field trials for three seasons (2017-2019) in Uganda under artificial inoculation. Disease severity was rated on a 1-9 scale at 21 (Rust 1), 28 (Rust 2), and 35 (Rust 3) days after inoculation. Area under disease progress curve (AUDPC) was calculated. The genetic diversity of the lines was assessed using 44,975 single nucleotide polymorphism markers. Combined ANOVA across seasons showed significant (P < .001) line mean squares for the three rust scores and AUDPC. Heritability was high for Rust 2 (0.90), Rust 3 (0.83), and AUDPC (0.93). Of the 50 lines, 12 were highly resistant to CR. Inbred lines CKL1522, CKL05010, and CKL05017 had significantly lower Rust 3 scores and AUDPC compared to the resistant check CML444 and are potential donors of CR resistance alleles. The genetic correlations between CR disease resistance parameters were positive and strong. A neighbor-joining (NJ) tree and STRUCTURE suggested the presence of three major groups among the lines, with lines highly resistant to CR spread across the three groups. The genetic diversity among the highly resistant lines can be exploited by recycling genetically distant lines to develop new multiple disease resistant inbred lines for hybrid development and deployment.
RESUMEN
Maize (Zea mays L.) yield in sub-Saharan Africa (SSA) is low because of both abiotic and biotic constraints, and limited availability or use of improved seed in some areas. This study was conducted (i) to estimate combining ability and heterosis among seven stress-tolerant populations, and (ii) to assess diversity among the populations and the relationship between diversity and heterosis. Twenty-one hybrids developed from diallel crosses of seven populations, parents, and two checks were evaluated in 10 optimal and 11 stressed environments (drought, low N, and random stress) in Kenya, Ethiopia, Uganda, and Zimbabwe for 2 yr. Analysis II of Gardner and Eberhart showed that variety and heterosis were significant for grain yield (GY) under optimal and managed stress, and across environments. Heterosis accounted for most of the variation for GY among populations under optimal conditions (67%) and drought stress (53%), which suggested the importance of dominance in inheritance of GY under these conditions. Genetic distance (GD) among populations ranged from 0.328 to 0.477 (mean = 0.404). The correlation between GD and heterosis was low (r = 0.14-0.40) in all environments. The simple sequence repeat (SSR) marker-based and GY-based clustering of parental populations showed similar patterns, with three populations distinct from the rest, suggesting significant differentiation of allelic variation in these three populations. The SSR-based diversity and phenotypic analysis results should be useful in defining breeding strategies and maintaining heterotic patterns among these populations.
RESUMEN
Stable quantitative trait loci (QTL) are important for deployment in marker assisted selection in wheat (Triticum aestivum L.) and other crops. We reported QTL discovery in wheat using a population of 217 recombinant inbred lines and multiple statistical approach including multi-environment, multi-trait and epistatic interactions analysis. We detected nine consistent QTL linked to different traits on chromosomes 1A, 2A, 2B, 5A, 5B, 6A, 6B and 7A. Grain yield QTL were detected on chromosomes 2B.1 and 5B across three or four models of GenStat, MapQTL, and QTLNetwork while the QTL on chromosomes 5A.1, 6A.2, and 7A.1 were only significant with yield from one or two models. The phenotypic variation explained (PVE) by the QTL on 2B.1 ranged from 3.3-25.1% based on single and multi-environment models in GenStat and was pleiotropic or co-located with maturity (days to heading) and yield related traits (test weight, thousand kernel weight, harvest index). The QTL on 5B at 211 cM had PVE range of 1.8-9.3% and had no significant pleiotropic effects. Other consistent QTL detected in this study were linked to yield related traits and agronomic traits. The QTL on 1A was consistent for the number of spikes m-2 across environments and all the four analysis models with a PVE range of 5.8-8.6%. QTL for kernels spike-1 were found in chromosomes 1A, 2A.1, 2B.1, 6A.2, and 7A.1 with PVE ranged from 5.6-12.8% while QTL for thousand kernel weight were located on chromosomes 1A, 2B.1, 5A.1, 6A.2, 6B.1 and 7A.1 with PVEranged from 2.7-19.5%. Among the consistent QTL, five QTL had significant epistatic interactions (additive × additive) at least for one trait and none revealed significant additive × additive × environment interactions. Comparative analysis revealed that the region within the confidence interval of the QTL on 5B from 211.4-244.2 cM is also linked to genes for aspartate-semialdehyde dehydrogenase, splicing regulatory glutamine/lysine-rich protein 1 isoform X1, and UDP-glucose 6-dehydrogenase 1-like isoform X1. The stable QTL could be important for further validation, high throughput SNP development, and marker-assisted selection (MAS) in wheat.