The true ileal digestibility assay provides the most informative measure of digestibility to assess bioavailability of amino acids in foods for humans. To determine ‘true’ estimates of ileal amino acid digestibility, requires that endogenous amino acids present in digesta at the terminal ileum be quantified. The amounts of endogenous amino acids in ileal digesta can be determined after feeding an animal or human a protein-free diet (traditional approach) or by various methods after giving a protein-containing diet. When the protein-free method has been applied with adult human subjects an overall mean value (three separate studies) for endogenous ileal nitrogen flow of 800 mg N/d has been reported. This value is considerably lower than a comparable value obtained after feeding protein of 1852 mg N/d (mean of four separate studies), and thus endogenous ileal N and amino acids should be measured under conditions of protein alimentation. There is some confusion concerning the terminology used to define digestibility, with the term “true” digestibility having different adopted meanings. Here, true amino acid digestibility is defined as apparent amino acid digestibility corrected for the basal amino acid losses determined after giving either a protein-free or a protein-containing diet. Basal losses should be determined at a defined dry-matter and protein intake. The protein-free diet approach to determining endogenous amino acids is considered unphysiological and basal losses refer to ileal endogenous amino acid flows associated with digesta dry-matter flow, and not including “specific” effects of dietary factors such as non starch polysaccharides and anti nutritional factors. Arguments are advanced that the enzyme hydrolysed protein/ultra filtration method may be suitable for routine application with a cannulated pig model, to obtain physiologically-valid basal estimates of ileal endogenous amino acids to allow calculation of true ileal amino acid digestibility in the pig, and then prediction (via statistical relationships) of true coefficients of amino acid digestibility in humans.

Sheep fitted with a PVC cannula in the dorsal rumen, a silastic infusion tube in the abomasum, and single ‘T’-shape PVC cannulas in the proximal duodenum (D1), distal duodenum (D2), mid-jejunum (J), and terminal ileum (I), were fed on diets varying in crude protein (Nx6·25) and fibre contents. 15N was administered to the sheep as a 15N-labelled grass meal-beer yeast suspension which was continuously infused into the abomasum over a period of 1 week. Subsequently, over a period of 4 d, 15N excess in the D1, D2, J and I digesta and faeces was measured and related to 15N excess in blood plasma TCA-soluble N. Average daily intakes of DM and N were 1·12 kg and 26·3 g respectively. Average daily endogenous N (g) amounted to 3·1 at D1, 10·2 at D2, 6·1 at J, 5·0 at I, and 5·1 in the faeces (F). Corresponding percentage values for total N were: D1 12·4, D2 32·9, J 46-4, I 48.4 and F 50.7. The secretion, particularly of pancreatic juice and/or bile, varied. The apparent re-absorption of endogenous protein varied among experimental diets, on average by 50·4% between D2 and I. Along the small intestine the percentage of 15N in the TCA-precipitable fraction gradually changed, on average by 66·0, 63·0, 46·3 and 55·7 at D1, D2, J and I respectively. Because of differential rates of enrichment and turnover among protein pools which contribute towards endogenous secretion, care should be taken in interpreting these data as absolute values.

The faecal and ileal true protein digestibilities of the raw pea (Pisum sativum) varieties finale and frijaune and the ileal true protein digestibility of steam-processed common beans (Phaseolus vulgaris) were measured in piglets using the 15N-dilution technique. The faecal true protein digestibility of both pea varieties was about 97. The ileal true protein digestibility was between 93 and 95, indicating that the pea protein is almost completely enzymically digested in the small intestine. The faecal apparent protein digestibility was 85 for both varieties while at the ileal level it was 79 and 74 respectively. The lower ileal apparent protein digestibility of peas can be attributed completely to the excretion of endogenous protein. The ileal apparent protein digestibility of toasted common beans was about zero (–4); the ileal true protein digestibility was about 66. This indicates that the protein of the common bean, although toasted, was highly resistant to enzymic digestion. It was calculated that per kg ingested bean protein, 340 g undigested bean protein and 700 g endogenous protein passed the terminal ileum. The results of the present study explain why in previous experiments a strongly reduced weight gain and even weight loss was observed in piglets fed on raw and toasted common beans.

The effect of fibre level in protein-free diets on the recovery and amino acid composition of endogenous protein collected from the ileum in pigs was studied. Although the Heal nitrogen excretion increased slightly with dietary fibre level, no significant difference was observed on the amino acid recovery between diets containing 30, 60,90 or 120 g wood cellulose per kg.

Three steers cannulated in the rumen and abomasum were given a diet of wheat straw, chopped and soaked in sodium hydroxide and then washed to pH less than 8. The straw was sprayed with sucrose, urea and minerals. The diet was free of digestible protein and given at 1·7 to 4·1 kg/day. Microbial protein flowing to the omasum was measured by 15N. By difference, the flow of endogenous protein nitrogen to the omasum was found to be 2·2 g/kg dry-matter intake. The flow of endogenous protein nitrogen to the abomasum was 3·3 to 9·4 g/kg dry-matter intake, and it decreased with increasing food intake.