Effect of cellobiose supplementation and dietary soluble fibre content on in vitro caecal fermentation of carbohydrate-rich substrates in rabbits

The in vitro caecal fermentation of five substrates low in starch and protein content d-(+)-glucose (GLU), d-cellobiose (CEL), sugar beet pectin (PEC), sugar beet pulp (SBP) and wheat straw (WS)] was investigated using soft faeces from rabbits receiving different levels of cellobiose and soluble fibre as inoculum. A total of 24 rabbits were supplemented 3 levels of cellobiose in the drinking water (0.0, 7.5, 15.0 g/l) and fed two experimental diets containing either low soluble fibre (LSF) or high soluble fibre (HSF) levels (84.0 and 130 g/kg dry matter). All substrates were subjected to a two-step pepsin/pancreatin in vitro pre-digestion, and the whole residue was used as substrate for the in vitro incubations. Gas production was measured until 144 h, and volatile fatty acid (VFA) production was determined at 24 h incubation. Experimental treatments did not affect SBP fermentation and had only a subtle influence on fermentation of WS and GLU. In contrast, cellobiose supplementation × donors’ diet interactions were detected for most gas production parameters for CEL. Both the fractional gas production (k) and maximal gas production rates were linearly increased (p ≤ 0.042) and the initial delay in the onset of gas production (Lag) linearly decreased (p < 0.001) by cellobiose supplementation with the HSF inoculum, with no differences between the 7.5 and 15.0 doses. In contrast, with the LSF inoculum cellobiose supplementation only affected k values, which were quadratically increased (p = 0.043) and had maximal values for the 7.5 dose. A quadratic effect (p ≤ 0.018) of cellobiose supplementation was observed for total VFA production at 24 h when CEL and PEC were fermented, obtaining the maximal VFA production for the 7.5 dose of cellobiose. Total VFA production for CEL was greater with LSF than with HSF inoculum (20.7 vs. 12.9 mmol/l; p = 0.014), but the opposite was found for WS (3.97 vs. 6.21 mmol/l; p = 0.005). The use of LSF inoculum for CEL fermentation sharply reduced acetate (p = 0.001) and increased butyrate proportions (p ≤ 0.001) compared with the HSF inoculum. A positive relationship between total VFA caecal concentrations in rabbits receiving the same experimental treatments and in vitro values was only observed when WS was used as substrate (r = 0.90; p = 0.015; n = 6). The results suggest that experimental factors influenced the fermentative activity of caecal digesta, but the observed response differed with the incubated substrate, being the CEL the most affected.