Endah Sri Redjeki, Faraz Khan, O. Molosiwa, S. Zehra, Jeremy A. Roberts, Katie Mayes, Nariman Salih Ahmad, Festo Massawe, Sean Mayes, Andrzej Kilian, S.M. Basu, S. Noah, S. Azam-Ali, and Florian Stadler
Bambara groundnut (Vigna subterranea L. Verdc.) is an underutilised, drought tolerant legume that has the potential to form an important part of Food Security for the coming decades. The challenges facing farmers to produce enough food for the growing world population – particularly that of climatic instability – are well documented and together represent probably one of the biggest challenges humanity has faced. Our extreme reliance on a limited number of staple (often nonindigenous and sometimes also poorly adapted) crops represents a clear vulnerability. This can be partly reduced by the development of alternative crops. These currently underutilised crops often have beneficial characteristics not found in conventional main crops and if these traits address either biotic or abiotic stresses in a sustainable way, then there is the potential for agricultural diversification. There are a number of stumbling blocks to developing such crops, including; poor yields, unadapted crop features, limited processing knowledge, few value-added products, poorly developed transport chains and markets, negative cultural perceptions and little perceived profit margin for commercial breeders. An integrated approach is needed to begin to address these problems. As part of this, we have focused on the application of molecular genetics to Bambara groundnut and the opportunities to exploit knowledge from other species, new technologies and new approaches, to establish a framework for genetic improvement through breeding. We also try to draw out lessons from our work in Bambara groundnut which may be relevant in other underutilised species, to try to contribute to the development of a generic approach (and hopefully, a faster and cheaper approach) to tackling these same questions in other underutilised species. In this paper we ask what is the fundamental information we need about the breeding system of an underutilised species and how could this alter our genetic improvement, using Bambara groundnut as an example. INTRODUCTION Developing new crops and expanding the range and level of production and use of existing underutilised crops is a major challenge which could help to generate more resilient agriculture. While this does not underestimate the critical importance of improvements to staple crops, such developments could provide alternatives where climate, demographic or soil nutritional changes prevent current staples from being economically cultivated in future. Bambara groundnut is an indigenous African legume with good drought tolerance which is still grown widely in sub-Saharan Africa, albeit at Proc. 2 Int. Symp. on Underutilized Plants Species “Crops for the Future – Beyond Food Security” Eds.: F. Massawe et al. Acta Hort. 979, ISHS 2013 452 low levels. Many such crops exists which have potential to be more utilised than they currently are, but establishing extensive research programmes on a large number of underutilised crops is not feasible, either in terms of capacity or in terms of finance. BAMLINK was an EU FP6 INCO-DEV programme which aimed to evaluate the use of Bambara groundnut at the molecular, eco-physiological, nutritional and end-user levels to try to identify the constraints to enhanced uptake of this crop for a range of uses. We are also attempting to use this crop as an exemplar for other underutilised crops, to understand what we need to know about a crop to be able to make progress with it. Here, we specifically look at our results for the molecular analysis of this crop to identify what information we need to understand to be able to make focused progress in a generic crop. In Bambara groundnut we have generated within species data, where specific tools were needed for subsequent quality control and breeding and have also applied data across species to try to link this underutilised species with available data from major species. This article focuses on the generation of data within species and how this knowledge has allowed us to map out breeding and genetic improvement options within Bambara groundnut, which could allow a strategic approach to be developed for other underutilised species. MATERIALS AND METHODS Plant Materials All Bambara groundnut genotypes analysed were drawn from either the collection at the International Institute for Tropical Agriculture (IITA, Ibadan, Nigeria) or from the Nottingham University stock collections (Sutton Bonington Campus, Nottingham University, UK). The 24 standard landrace accessions (single genotypes) used for characterising the SSR markers are given in Table 1. Fifty individual seed of UniSwaRed and S19-3 landraces, grown in the Tropical Crops Research Unit (TCRU) glasshouses in 2007, were leaf sampled. DNA was extracted from individual genotypes using standard approaches and all 100 samples were screened against three SSR markers from Basu et al. (2007). Microsatellites were labelled, amplified and analysed as described below. The Correspondence Analysis was generated using the defaults in MVSP v.3 (Kovachs Computing Services) and the first two axes plotted. Development of Microsatellites for Bambara Groundnut Three approaches were used to develop within species microsatellite markers: 1. Development of a genomic fragment microsatellite-enriched library, essentially according to Basu et al. (2007). Sanger sequencing of clones containing repeat sequences was used to allow the construction of primers flanking the microsatellite repeat sequences. 2. The amplified PCR products from the adaptor bearing insert of the library from 1) was sequenced using part of a 1/16 Roche 454 pyrosequencing run (9 species libraries with single base changes in the adaptors were mixed before sequencing; Eurofins MWG). 3. A plate of 454 pyrosequencing (Titanium reagents) from the leaf transcriptome of a single plant of the Namibian accession S19-3 (50 days after sowing; 28°C/23°C cycle with 12 hours photoperiod, planted directly into a sandy loam) was produced. The data were assembled (Newbler, by Deep Seq, University of Nottingham) and microsatellite repeats within the transcriptome had primer pairs designed to them, were possible. In each of the three cases, the MISA.pl script (IPK, Gartersleben, Germany; pgrc.pik-garterleben.de/misa/misa.html) was used to characterise the microsatellites present. Primers were designed using the Primer3 software (frodo.wi.mit.edu) including a minimum primer length of 24 nucleotides and optimal length of 27. For all primers an M13 extension was added to the forward primer of the pair to allow third primer labelling (Schuelke, 2000), with a dye-labelled (Well-RED blue; Sigma-Aldrich) to be incorporated allowing fragment size assessment on a Beckmann CEQ 8000. PCR