An extensive review of the literature on the chemistry and biological activity of condensed tannins (Chapter 1 and 2) revealed some significant gaps in knowledge of these processes, and the following experimental program was designed to investigate some specific aspects of the chemical and biological effects of condensed tannins (CT) in animals fed CT-containing diets.
CT were isolated and purified from various tropical legume forage plants, and the CT binding affinities for bovine serum albumin (BSA) in vitro established using the Ninhydrin method (Chapter 3). There were significant (p<0.05) differences in the binding affinities of the CT (e.g. 0.150 mg L. pallida CT/0.375 mgBSA, and 0.300 mg L.leucocephala CT/0.375 mg BSA).
The inclusion of various Leucaena CT types (differing in binding affinity for BSA in vitro) as 1 % of the diet DM fed to chicks
(Chapter 4) significantly (p<0.05) increased the ileal flow of nitrogen (e.g. 8.26 vs. 4.87 g N/kg DM intake, for L. pallida CT and CT-free diets, respectively), and decreased the apparent digestibility of nitrogen (e.g. 78.4 % vs. 86.9 %, for L. pallida CT and CT-free diets, respectively) relative to the CT-free diet. However, there was no effect of CT type (binding affinity) on these parameters suggesting that dietary CT concentration may be more important than CT type in determining CT effects on nitrogen utilization in animals.
The digestion of CT-bound dietary protein in the gastro-intestinal tract of ruminants was investigated by determining the extent of dissociation of insoluble 125I-BSA+CT complexes administered to abomasally- and intestinally-fistulated sheep (Chapter 5). The extent of dissociation was registered as the true digestibility of 125I- BSA. The
true digestibility of 125I-BSA originally bound to L. pallida CT (72.1 %), was significantly (p<0.05) lower than that of 125I-BSA originally bound to L. leucocephala CT (88.0 %) between the abomasum and terminal ileum. These results indicate that differences in the ability of CT to inhibit 125I-BSA digestion in vivo matched the relative abilities of the same CT to bind BSA in vitro, indicating that the in vitro BSA binding assay for ranking CT behaviour was biologically relevant in vivo. Furthermore, compilation of all the results from this study suggested that the true digestibility of CT-bound 125I-BSA between the mouth and in faeces, allows us to predict the quantitative contribution that CT-bound dietary proteins make to improved N supply to the small intestines.
The significance of pH (3, 5, and 8), pepsin, trypsin, and
PEG per se on the rate of release of BSA from insoluble CT-protein complexes (lPC) under conditions simulating those found in the gastro-intestinal tract was investigated in vitro (Chapter 7). The results (apparent BSA % dissociated) indicated that alkaline pH 8 was most important in the dissociation (78.5 %) of TPC suggesting that TPC are mainly dissociated in the small intestines.
The presence of various CT types in the reaction mixture in vitro inhibited the activity of trypsin with Calliandra CT causing more than a five-fold decline in trypsin activity as compared to the CT-free treatment (Chapter 8). This decline in trypsin activity was calculated to possibly adversely affect protein digestion in the small intestines.
Increasing amounts of purified CT differing in binding affinity were added to substrates in buffered rumen fluid in vitro (Chapter 9). Generally, irrespective of CT source (L. pallida or
L. leucocephala), the successive addition of CT decreased fermentation and digestion activity to similar extents, suggesting that the relative differences in binding capacity were less important than the total amount (or concentration) of CT added.
The effects of infusion of L. pallida CT into the rumen on microbial protein production, dietary protein and carbohydrate digestion in sheep fed on Lucerne hay chaff, were investigated by infusing 0.00, 0.62, 1.23, and 2.44 g CT/100 g DMI, as treatments T1, T2, T3, and T4, respectively (Chapter 10). There was a significant (P<0.05) increase infaecal N (0.16 vs. 0.21 g/kg Lwt0.9/d in Tl and T4 respectively), and a significant (P<0.05) decrease in the apparent digestibility of N (82.4 vs. 76.5 % in T1 and T4 respectively). There were no significant treatment effects on the degradation rates of dietary protein in the rumen, on
microbial protein synthesis, or on the efficiency with which microbes synthesised cell protein (34. 7 g microbial-N/kg OMADR). There were non-significant changes in the post-ruminal N digestion. There were Significant (P<0.05) differences in the abomasal flow of CT, with 89 % of the CT infused apparently disappearing in the rumen in T4.
The effect of consumption of absorbable phenolics on nitrogen and mineral metabolism (Chapter 11) was investigated using rats. Quercetin at 1.5 % in the diet had no effect on nitrogen metabolism. This suggested that the results observed in Chapter 10 were mainly due to CT effects at the intestinal level, and that there were likely to be only minor effects, if any, of the breakdown products of CT, and other flavonoids on sheep metabolism.
It was generally concluded from these studies that, the in vitro BSA binding assay for ranking CT behaviour was biologically relevant in vivo, CT concentration
seems to be more important than CT type with regard to the biological effects of CT, the amounts of dietary CT infused had no effect on the efficiency of microbial protein synthesis, and the extent of dissociation of CT-dietary protein complexes at alkaline pH, and protein digestion by enzymes is the main cause of the reduction in the true digestibility of dietary protein, and ultimately the apparent digestibility of protein in vivo.