GENOTYPIC VARIATION IN FLATULENCE CAUSING OLIGOSACCHARIDES IN BIOFORTIFIED AND COMMERCIAL DRY BEAN VARIETIES

INTRODUCTION Elimination of flatulence is a challenging practical problem associated with consumption of legumes. The problem is compounded by the variability in susceptibility among individuals. Rackis (1981) established that the oligosaccharides-verbascose, stachyose and raffinose are major causes of flatulence. They escape digestion and are fermented by intestinal micro floral to form excessive amounts of carbon dioxide and hydrogen. Little has been done to develop bean varieties that combine agronomic superiority with nutritional quality, fast cooking and low flatulence levels in eastern Africa. The objectives of this study were to: (i) determine if there is genotypic variation in concentration of oligosaccharides associated with flatulence in commercial bean varieties, recently released biofortified bean and advanced breeding lines, and (ii) the effect of cooking on oligosaccharide concentration. MATERIALS AND METHODS Study materials were 10 commercial dry bean varieties and seven recently released biofortified cultivars representing the Andean and Mesoamerican gene pools and the major market classes grown in east, central and southern Africa. Verbascose, stachyose and raffinose were extracted twice from ground raw and cooked bean milks using a 3:7 v/v methanol-water mixture and quantified on a high performance chromatography system using analytical grade standard reagents. The oligosaccharides sugars were quantified using a high performance liquid chromatography (Chromatography Systems model 750, Shimadzu, USA) with a differential refractometer. The precipitated material was removed by centrifugation and the supernatant filtered prior to analysis. The column used (250mm*4.6mm i.d) was packed with spherisorb-5- amino, as a slurry in propan-2-ol. Sample injection valve, model 7120 was used. The eluting solvent was acetonitrile/water (67:33, v/v) with a flow rate of 2.0 ml min-1 at a temperature of 40°C. Quantification was carried out by peak area comparisons of sample and standards of known concentration (Pinthong et al, 1980). Standards were obtained from SIGMA-Aldrich, USA. Data was analyzed using Genstat statistical software (v15)