Breeding sweetpotato for improved yield and related traits and resistance to sweetpotato virus disease in Eastern Tanzania.
Sweetpotato production contributes significantly to food security and incomes of subsistence farmers in Tanzania. However, productivity of the crop is constrained by several biotic, abiotic and socio-economic factors. Amongst the biotic constraints, the sweetpotato virus disease (SPVD) causes significant yield losses in the country. Improved cultivars and landraces that are grown succumb to SPVD. Both chemical and biological control methods are not fully effective against SPVD. The use of resistant varieties remains the most effective and cheapest method for subsistence farmers. Therefore, breeding for SPVD resistance and high yields is an important consideration to develop and release improved sweetpotato varieties with end users preferences. Therefore, the objectives of the study were to: 1) assess the present sweetpotato farming systems, farmers’ preferences, production constraints and breeding priorities in eastern Tanzania, 2) determine genetic variation among diverse sweetpotato germplasm with regards to yield, dry matter content and SPVD resistance and to identify suitable clones for breeding, 3) investigate the genetic diversity of 48 Tanzanian sweetpotato genotypes using nine selected polymorphic simple sequence repeat (SSR) markers and to determine genetic relationship and select unique parents for breeding, 4) determine the general combining ability (GCA) and specific combining ability (SCA) effects of selected sweetpotato clones for the number of storage roots, fresh storage root yield, dry matter content (DMC) and resistance to sweetpotato virus disease (SPVD) for further selection and breeding, and 5) determine the magnitude of genotype-by-environment and stability for yield and yield related traits and SPVD resistance among newly developed sweetpotato clones in eastern Tanzania. Participatory rural appraisal study was conducted involving 138 and 149 farmers sampled for household interviews and focus group discussion, respectively at Gairo, Kilosa and Kilombero districts of Morogoro Region and Mkuranga district of the Coast Region of Tanzania. More than 94% of the respondents depended on crop farming for their livelihoods. Farmers preferred sweetpotato varieties with high yields, high dry matter content, tolerance to diseases and early maturing. Sweetpotato virus disease and pests, drought, unavailability of markets, lack of transport, low prices, inadequate extension services and post-harvest losses were identified by farmers being the most important production constraints. Improved extension service delivery, SPVD tolerant cultivars and reliable and coordinated market systems of sweetpotato were the most immediate needs for improved sweetpotato production and productivity. Field experiments consisting 144 sweetpotato genotypes were conducted at two sites in Tanzania using a 12x12 simple lattice design in 2013 to screen genotypes for yield, dry matter content and sweetpotato virus disease resistance and to identify suitable clones for breeding. The genotypes differed in time to 50% flowering, number of roots per plant, root yield, dry matter content and resistance to SPVD. Seven clones including Simama, Ukerewe, Mataya, Resisto, 03-03, Ex-Msimbi-1 and Gairo were selected as potential parents for sweetpotato breeding for high storage root yield and related traits or SPVD resistance. Nine polymorphic simple sequence repeat markers (SSR) were used to determine genetic relationship among 48 Tanzanian sweetpotato genotypes to identify unique parents useful for future breeding. The SSR markers were highly polymorphic and allocated the genotypes into three major genetic clusters. Ex-Ramadhani, Kibakuli, Mkombozi, Mjomba, Ex-Halima-3 and Kabuchenji were identified as genetically unrelated and complementary genotypes and recommended for future breeding programmes. Eight genotypes contrasting for their yield, dry matter content or SPVD resistance were selected and crossed using an 8x8 half diallel mating design. The families were evaluated in the field using a 6x6 triple lattice design at Sugarcane Research Institute (SRI) at Kibaha, Kilombero Agricultural Training and Research Institute (KATRIN) and Sokoine University of Agriculture (SUA) in Tanzania. There were highly significant differences among families (P< 0.001) for all studied traits across sites. Clonal parents with highest general combining ability (GCA) were 03-03 and Resisto for storage root yield, Ukerewe for dry matter content (DMC) and Ex-Msimbu-1 which displayed negative and significant GCA effect for SPVD resistance. Therefore, the parents Resisto, Ukerewe and Ex-Msimbu-1 could be used for future sweetpotato breeding programmes to improve yield, DMC or resistance to SPVD. The following crosses were best combiners displaying positive and significant SCA effects: Mataya x Gairo and Simama x Gairo for number of roots per plant, Mataya x Ex-Msimbu-1 and 03-03 x Ex-Msimbu-1 for root yield and, Mataya x 03-03, 03-03 x Ukerewe and Resisto x SPKBH008 for DMC, and Mataya x SPKBH008 and Mataya x Gairo had negative and significant SCA effect for resistance to SPVD. The selected parents and crosses were the best candidates to develop improved sweetpotato varieties with high root yield, DMC or SPVD resistance. The magnitude of genotype-by-environment interaction for yield and related traits and SPVD resistance of 26 selected sweetpotato clones was investigated across six diverse environments; namely Gairo, Kilombero Agricultural Training Research Institute (KATRIN), Sokoine University of Agriculture (SUA), Sugarcane Research Institute (SRI), Chambezi and Mkuranga. The Additive Main Effect and Multiplicative Interaction (AMMI) and genotype and genotype-by-environment interaction (GGE) biplot analyses were used to determine the GxE interaction and stability of the genotypes. The genotypes were ranked differently for yield and related traits and SPVD resistance. AMMI and GGE biplot analyses identified the following genotypes: G5, G11, G23, G9, G7, G18 and G17 being high yielding and resistant to SPVD which could be further evaluated in multi-environment yield trials (MEYTs) in eastern Tanzania. Also, the genotypes G22 and G3 were isolated as high yielding and resistant to SPVD but specifically suited to Chambezi and Gairo. Test environments sufficiently discriminated the candidate genotypes for the traits studied. MEYTs are required for selection and recommendation of high yielding, SPVD resistant and stable sweetpotato clones for eastern Tanzanian or similar environments. Overall, the study identified valuable sweetpotato parents and families with high combining ability for number and yield of storage roots, dry matter content and SPVD resistance from which new clones can be selected for future evaluation and release as new cultivars.