In vivo assessment of the ability of condensed tannins to interfere with the digestibility of plant protein in sheep

To assess the effect of condensed tannin (CT) astringency on the digestibility of protein post-ruminally, CT were purified from four types (accessions) of Mulga (Acacia aneura), and a Leucaena pallida and dosed into sheep as complexes with 15N-plant protein (tannin-protein complexes, TPC), together with an indigestible marker, chromium-EDTA (Cr-EDTA). Each CT treatment dose comprised 12 mg CT, 10 mg 15N-protein, and 2.77 mg of Cr. A protein-only control (same as the other TPC solutions but without any CT) treatment was also included to make a total of six treatments. Treatments were applied in two 6 × 6 Latin Square designs with 72 h between each infusion for each sheep. In the first, the solutions were infused post-ruminally via an abomasal cannula. In the second they were dosed directly into the mouth following an intra-nasal dose of an analogue of vasopressin in an attempt to stimulate the oesophageal groove reflex to direct the solutions more efficiently toward the small intestine. Results showed no detectable effect of CT type on the in vivo digestibility of the 15N-protein. Protein digestibilities were uniformly high, indicative of complete dissociation of the TPC. There was no correlation between protein digestibility, mouth to faeces, and protein digestibility, abomasum to faeces (P > 0.05). In vivo digestibility was also uncorrelated with CT astringencies defined in vitro (P > 0.05). Astringency in vitro was defined as the mg of CT required to achieve half-maximal precipitation of 0.5 mg of protein (bovine serum albumin). In vitro, the most astringent CT (A. aneura883558), had at least 1.6 times the astringency of the weakest CT, (A. aneura842394). The A. aneura with the weakest CT also contained less than 1/3 the total amount of CT/g leaf dry matter than that with the strongest, highlighting the scope for selection of more nutritionally useful types of A. aneura. Limitations of the in vivo protocol used are discussed and it is concluded that the effect of CT astringency on the availability of protein post-ruminally is minimal.