Genetic diversity and combining ability of selected quality protein maize (QPM) inbred lines adapted to the highland agro-ecology of Ethiopia.

Abstract

The high altitude sub-humid agro-ecology in Ethiopia is estimated to cover 20% of the land devoted to annual maize cultivation. More than 30% of small-scale farmers in this agro-ecology depend on maize production for their livelihood. This area is also characterized by high human population density necessitating maize production for food security and alleviating malnutrition. Research on highland maize improvement in Ethiopia has generally lagged behind compared to research undertaken in the other agro-ecologies until it was accelerated in 1998 as part of the Highland Maize Gene Pool Project for east and central African countries (ECA). A strategic research is required to develop nutritionally enhanced maize cultivars such as quality protein maize (QPM) to alleviate food insecurity and nutritional problem of millions of people who depend on this staple food crop. The Highland Maize Breeding Program of Ethiopia in collaboration with CIMMYT-Ethiopia has been systematically developing a pool of highland QPM source germplasm by converting non-QPM highland maize germplasm into QPM counterparts. The converted genotypes should be well studied for strategic breeding or production purposes in the highland sub-humid agroecologies of Ethiopia and other ECA countries with similar environments. The objectives of this study were, therefore, to: (1) establish farmers’ preferences of normal and QPM cultivars and traits of interest, and identify maize production constraints in the highlands of Ethiopian; (2) examine phenotypic variability among Tropical-highland adapted QPM inbred lines; (3) examine genotypic variability among Tropical-highland adapted QPM inbred lines using simple sequence repeat (SSR) markers; (4) estimate heterosis and combining ability for grain yield and yield related agronomic traits; and (5) determine genotype-environment interaction (G x E) and yield stability of experimentalQPM hybrids. A participatory rural appraisal (PRA) research was conducted involving 160 experienced maize farmers selected from four districts of two zones in the central highlands of Ethiopia during 2012. Results showed that few highland maize cultivars are available and adopted by farmers. A two-decade old cultivar, ‘BH660’, originally released for the mid-altitude agro-ecology, has been widely adopted in most highland areas. Grain yield was considered the most important trait for maize cultivar selection. Limited access to inputs (improved maize seeds and inorganic fertilizers) and late on-set and inadequate rainfall were the primary constraints identified by farmers across the study areas. Thirty-six maize inbred lines (30 QPM and six normal) adapted to tropical highlands were phenotyped and genotyped using 18 morpho-agronomic traits and 25 polymorphic SSR markers, respectively, in 2012. Significant phenotypic variation observed among inbred lines for all measured traits. Principal component and unweighted paired group method using arithmetic averages (UPGMA) cluster analyses of the phenotypic data revealed the presence of three distinct clusters. Seven inbred lines [KIT32Q, 142-1eQ, SRSYN20Q, FS67(BC2), FS170Q, FS60, and F7215] with complementary phenotypic traits and relatively better yield performance were selected for further genetic analysis and breeding. Similarly, the SSR data showed that nearly 98% of the pairwise comparisons had genetic distance between 0.30 and 0.78, indicating large genetic differences among tested inbred lines. The model-based population structure, principal coordinate and neighbor-joining cluster analyses also revealed the presence of three genetic groups, which is generally consistent with pedigree information and partly with heterotic grouping. Analysis of molecular variance indicated a considerable genetic difference among heterotic groups explained by 8.6 to 15.4% of the total SSR variance. Sixty-six experimental QPM hybrids obtained from a 12-parent diallel cross was tested for grain yield and related traits under rain-fed conditions to determine combining ability and heterosis. The ratio of dominance to additive genetic variance was relatively larger for grain yield compared to other traits, suggesting the genetic worthiness of QPM hybrid development. Inbred lines KIT32, FS60 and 142-1-EQ were good general combiners (GCA) for grain yield; while FS60 exhibited good GCA for days to anthesis, plant and ear height. The best crosses identified were KIT32 x 142-1-EQ and SRSYN20 x FS60 that yielded 9.6 t ha-1 and 8.8 t ha-1, respectively. The hybrids can be used as potential single cross testers for the development of three-way cross QPM hybrids for the highland agro-ecologies of Ethiopia and ECA. Genotype-environment interaction (GEI) and grain yield stability analyses were conducted using AMMI and GGE biplot models for the newly developed QPM single crosses and two commercial hybrid checks evaluated across seven environments. Hybrid 10 (KIT32 x 142-1-eQ) followed by hybrid 66 (142-1-eQ x CML144) and hybrid 59 (FS60 x 142-1-eQ) with yield levels of 10.3, 9.6 and 9.4 t ha-1, respectively, were selected as the best performers and hence desirable hybrids while KUL13 (Kulumsa) was the most suitable environment identified by both analyses. The GGE analysis divided the highland test environments into two mega-environments. Overall, the study attempted to understand the importance of highland maize and farmers’ production constraints, identified useful highland adapted QPM inbred lines with good combining ability for breeding and good performing QPM single crosses to be used as testers for hybrid development in the highland sub-humid agro-ecologies of Ethiopia, in particular, and the ECA countries, in general.