The role of condensed tannins in the nutritional value of Lotus pedunculatus for sheep

1. Vegetative secondary growth Lotus pedunculatus was cut daily, and fed fresh at hourly intervals (600 g dry matter (DM)/d) to three groups each of three sheep fitted with permanent cannulas into the rumen and duodenum. Lotus fed to two of the groups was sprayed with low and high rates of polyethylene glycol (PEG; molecular weight 3350), which specifically binds the condensed tannins (CT). Nutrient intake and faecal excretion were measured directly, duodenal flows estimated from continuous intraruminal infusion of inert ruthenium phenanthroline (Ru-P) and CrEDTA markers, and rumen pool sizes measured at slaughter. 2. Dietary concentrations of total reactive CT (i.e. that not bound to PEG) were 95, 45 and 14 g/kg DM, whilst the corresponding values for free CT were 15, 5 and 2 g/kg DM. 3. Increasing dietary reactive CT concentration linearly increased duodenal flows of non-ammonia nitrogen, but linearly decreased the apparent digestibility of energy and organic matter, and rumen digestion of hemicellulose but not of cellulose. Rumen digestion as a proportion of total digestion was increased by the higher PEG rate for organic matter, energy, pectin and lignin. 4. High dietary CT concentration was associated with increased N retention. Rumen ammonia concentration and pool size showed only a slight decline on this diet, indicating that there must have been increased recycling of N into the rumen. 5. Increasing dietary reactive CT concentration had no effect on the rate at which carbohydrate constituents were degraded in the rumen per unit time (FDR), but increased the rate at which their undegraded residues (FOR) left the rumen per unit time. The latter appeared to be the principal mechanism by which rumen digestion as a proportion of total digestion was reduced at high dietary CT concentrations. From a comparison of FDR and FOR of carbohydrate components in lotus and Brassica oleracea diets, it was concluded that hemicellulose digestion was rate-limiting for rumen cell-wall digestion, probably due to bonding with lignin. However, the considerable post-rumen digestion of hemicellulose was not associated with post-rumen lignin digestion. 6. It was concluded that a desired concentration of CT in Lotus sp. should represent a balance between the positive effect of CT in improving the efficiency of N digestion and their negative effect in depressing rumen carbohydrate digestion. A recommended concentration is 3WOg/kg DM.