Genetic improvement of pigeonpea [Cajanus cajan (L.) millspaugh] for yield, earliness and resistance to Fusarium wilt (Fusarium udum Butler) in Malawi.

Abstract

Pigeonpea [Cajanus cajan (L.) Millspaugh, 2n=2x=22] is a one of the important food legumes in Sub-Saharan Africa and Asia. Malawi is a major pigeonpea grower in Africa with production of 403,519 tonnes produced in 248,400 hactares. Pigeonpea is good source of protein and cash income to millions of farmers. Pigeonpea crop residues form excellent animal feed. It serves in atmospheric nitrogen fixation and biomass allocation in the soil. Despite Malawi being the highest pigeonpea producer, grain yield of pigeonpea is low (< 700 kg ha-1) compared with the potential yield of the crop (2000 kg ha-1). The yield gap is due to various production constraints, including Fusarium wilt disease, insect pests, and lack of early maturing and high yielding varieties that are photoperiod insensitive. Breeding and deployment of high yielding, early maturing, and Fusarium wilt resistant cultivars have the potential to enhance pigeonpea production and productivity. The overall objective of this study was to contribute to food security in Malawi through breeding high yielding and farmer-preferred pigeonpea varieties. The specific objectives were: (1) to determine the production constraints affecting pigeonpea, and to identify farmer-preferred traits in Malawi to guide future breeding of pigeonpea; (2) to determine the diversity among pigeonpea germplasm collections using agro-morphological traits to enable selection of genetically distinct lines for breeding; (3) to determine the genetic diversity among the tested pigeonpea germplasm, using single nucleotide polymorphism (SNP) markers to select genetically distinct lines for breeding; (4) to determine the combining ability effects and gene action controlling agro-morphological traits and resistance to Fusarium wilt; and to select the best parents and families from the test population for further breeding. In the first study, a participatory rural appraisal study was conducted in four major pigeonpea-growing districts in southern Malawi (Chiradzulu, Mulanje, Thyolo and Zomba), using a semi-structured questionnaire, transect walks and focus group discussions (FGDs). The results revealed that a landrace pigeonpea variety, ‘Mthawajuni’, was preferred by famers due to its positive attributes such as good taste, early to medium maturity, short cooking time and tolerant to pod borer (Helicoverpa armigera Hubner). Pigeonpea trait preference was dependent on gender, with female respondents preferring rapid cooking, early maturity, long storage and good pest resistance, whereas men focussed on high yields, large seed size, cream seed colour and disease resistance. The study identified the pod borer (H. armigera), Fusarium wilt disease (Fusarium udum Butler), low yields of the existing varieties, drought, and unreliable market prices as the leading challenges affecting pigeonpea production in southern Malawi. A second part of the study focused on phenotypic and genetic diversity and yield stability analyses among pigeonpea accessions in selected target production environments, as a basis to select complementary and unique genotypes for breeding. Eighty-one pigeonpea genotypes were evaluated in six environments in Malawi using a 9 × 9 alpha-lattice design with two replications. Significant genotype variation were recorded for qualitative traits including flower colour, flower streak pattern, pod colour, seed coat colour pattern, seed coat main colour, seed shape and seed eye colour. All evaluated quantitative traits initially were significantly affected by genotype × environment interaction effects except the number of seeds per pod. Genotypes MWPLR 14, ICEAP 01170, ICEAP 871091 and ICEAP 01285 were identified as early maturing varieties, maturing in 125 to 137 days. The genotypes Kachangu, MWPLR 16, TZA 5582, No. 40 and MWPLR 14 had the highest number of pods per plant (NPP) and highest grain yields (GYD). Grain yield was positively and significantly correlated with days to flowering (DTF) (r=0.23, p<0.01), NPP (r=0.35, p<0.01) and hundred seed weight (HSWT) (r=0.50, p<0.01), suggesting the usefulness of these traits for selection to enhance grain yield improvement when assessing pigeonpea populations. Using principal component analysis, three principal components (PCs) accounted for 57.7% of the total variation. The most important traits that reliably discriminated between the test genotypes were DTF, days to maturity (DTM), number of primary (NPB) and secondary branches (NSB), HSWT and GYD. Genotype, environment and genotype × environment interaction accounted for 16.4, 33.5 and 49.6% to the total variation for quantitative traits, respectively. The test environments were delineated into three mega-environments, based on site and seasonal variability. MWPLR 14 (G51), MWPLR 24 (G26) and ICEAP 01155 (G27) were the most stable genotypes for yield across environments, while MWPLR 14, TZA 5582 and MWPLR 4 were the highest yielding genotypes across environments. To broaden the genetic base of the pigeonpea for selection, divergent genotypes such as MWPRL 14, TZA 5582, MWPLR 4, MWPLR 16, Sauma and Kachangu are recommended as parents for targeted crosses. The fourth part of the study examined genetic relationships among 81 genotypyes using 4122 single nucleotide polymorphism (SNP) markers. The SNP markers also confirmed the genetic diversity among the genotypes. The mean gene diversity and the polymorphic information content (PIC) were 0.14 and 0.11, suggesting moderate genetic differentiation among the genotypes. The low genetic diversity and PIC could hinder genetic gains in future pigeonpea breeding programs using this population. The genotypes were delineated into three groups based on population structure and the joint analysis of the phenotypic and genotypic data. The analysis of molecular variance (AMOVA) revealed that differences among clusters accounted for only 2.7% of the variation, while within-cluster variation among individuals accounted for 97.3% of the variation. This suggested that unique breeding populations could be created by identifying and selecting divergent individuals as parental lines. There is a need to create new genetic variation or introgress genes from close relatives to increase the genetic base of pigeonpea since the available genetic variability may not meet the demand for improved cultivars. The phenotypic diversity assessment using morphological attributes grouped the genotypes into three distinct clusters. The mean gene diversity and polymorphic information content were 0.14 and 0.11, respectively, suggesting moderate genetic differentiation among the genotypes. The genotypes were delineated into three heterotic groups based on population structure and the joint analysis of the phenotypic and genotypic data, suggesting the possibility of creating unique breeding populations through targeted crosses of parents from divergent heterotic groups. In a third study, the best and most diverse genotypes from the diversity studies with early maturity, Fusarium wilt (FW) resistance from previous studies and farmer-preferred traits were selected for crosses. Finally, the ten selected parental lines were crossed using a factorial mating design and 25 progenies were successfully developed. The parents and progenies were field evaluated at two locations; 1) Chitedze Agricultural Research Station and 2) Makoka Agricultural Research Station in Malawi. The trial design was 7 × 5 alpha lattice design with two replications. The test genotypes were evaluated for FW resistance through a root dip inoculation technique. There was significant genetic variation among parental lines and families for days to 50% flowering (DTF), days to 75% maturity (DTM), plant height (PH), 100 seed weight (HSWT), FW resistance, and grain yield (GYD). Parental lines, ICEAP 87105, and ICEAP 01285 had desirable general combining ability (GCA) (-32.90 and -14.16 respectively) for days to 75% maturity (DTM), parental lines, MWPLR 16, Sauma and Mwaiwathualimi had desirable GCA (319.11, 168.8 and 46.45 respectively) for grain yield (GYD) and parental lines, TZA 5582, ICEAP 00554, Mwayiwathualimi and Sauma had desirable GCA effects (-3.16, -0.54, -0.24 and 0.17 respectively) for FW resistance. Hybrids such as TZA 5582 × MWPLR 22, TZA 5582 × MWPLR 14, and Mwayiwathualimi × MWPLR 22 had desirable specific combining ability (SCA) effects for DTM (-1.22 -1.51 and -0.91 respectively), GYD (80.93, 42.67 and 79.55 respectively) and FW resistance (-1.10, -0.15, and -1.66 respectively). The study further revealed that additive gene effects were important in inheritance of DTF, DTM and PH traits and non-additive gene effects were important in inheritance of GYD, 100 seed weight (HSWT) and FW resistance. This suggest that both pedigree and recurrent selection are important to achieve pigeonpea improvement. Overall, this study determined the present pigeonpea production constraints and farmer-preferred traits in Malawi. Further, significant genetic variations were detected among a diverse set of pigeonpea germplasm for breeding early maturing/short-duration, high yielding and FW resistant varieties. The study developed new breeding populations based on selected complementary parents for variety development and release in Malawi.