Gluten is a crucial functional component of bread, but the effect of increasing gluten content on gastrointestinal (GI) function remains uncertain. Our aim was to investigate the effect of increasing gluten content on GI function and symptoms in healthy participants using the unique capabilities of MRI. A total of twelve healthy participants completed this randomised, mechanistic, open-label, three-way crossover study. On days 1 and 2 they consumed either gluten-free bread (GFB), or normal gluten content bread (NGCB) or added gluten content bread (AGCB). The same bread was consumed on day 3, and MRI scans were performed every 60 min from fasting baseline up to 360 min after eating. The appearance of the gastric chime in the images was assessed using a visual heterogeneity score. Gastric volumes, the small bowel water content (SBWC), colonic volumes and colonic gas content and GI symptoms were measured. Fasting transverse colonic volume after the 2-d preload was significantly higher after GFB compared with NGCB and AGCB with a dose-dependent response (289 (sem 96) v. 212 (sem 74) v. 179 (sem 87) ml, respectively; P=0·02). The intragastric chyme heterogeneity score was higher for the bread with increased gluten (AGCB 6 (interquartile range (IQR) 0·5) compared with GFB 3 (IQR 0·5); P=0·003). However, gastric half-emptying time was not different between breads nor were study day GI symptoms, postprandial SBWC, colonic volume and gas content. This MRI study showed novel mechanistic insights in the GI responses to different breads, which are poorly understood notwithstanding the importance of this staple food.

The digestion of non-starch polysaccharides (NSP) and of resistant starch (RS) by rats fed on wholemeal-bread-based diets containing 0-450 g cooked, freeze-dried haricot beans (Phaseolus vulgavis)/kg diet was measured over the final 14 d of a 21 d feeding experiment. The bread and beans provided all the dietary polysaccharide. RS could not be detected consistently in faeces and it was assumed that this fraction was entirely fermented in the large bowel (LB). NSP digestibilities were 0.56 and 0.86 for wholemeal bread and beans respectively with no evidence that the dietary presence of beans affected digestibility of bread NSP. Bean non-cellulosic polysaccharides were highly digestible with values of 0.98, 0.88 and 0.99 for arabinose, xylose and uronic acids components respectively. There were large increases in organic matter flow to the LB when beans were fed which was associated with marked caecal hypertrophy and alterations in caecal volatile fatty acids (VFA) pattern. Calculated VFA absorption from the LB was 5-fold higher with the highest level of beans and this was reflected in higher concentrations of VFA in portal and heart blood.