There is strong epidemiological evidence that more physical activity is associated with reduced risk of colon cancer, but the amount or type of activity necessary to invoke this protection is disputed, and the mechanism that is responsible has not been elucidated. The present study compared the effects of two contrasting exercise regimens on intestinal tumourigenesis in Min mice, and investigated two novel mechanistic factors: colonic butyrate and sleep. From 5 weeks of age, Min mice were exercised by running on a treadmill (TR; ≤ 21 m/min, 30–60 min/d, 5 d/week, ≤ 12 weeks). Additional groups of mice were provided with an exercise wheel (WH) or no exercise (CON). Mice had free access to a Western-style, high-fat diet. WH mice ran 3·97 km (females) and 1·92 km (males) daily (P = 0·002). There were no differences in body weight gain or body composition between treatment groups. Treadmill running reduced the numbers of larger ( ≥ 2 mm diameter) tumours (P = 0·042), and tended to reduce tumour multiplicity in the colon (P = 0·049). TR mice had a higher molar proportion of butyrate in colonic digesta than CON mice (P = 0·030), and when treatment groups were combined, there was a weak negative correlation (r − 0·174, P = 0·061) between butyrate molar proportion and total tumour number. In a subset of animals in which non-exercise physical activity was monitored, there were strong positive correlations between sleep duration and both tumour multiplicity (P < 0·001) and tumour burden (P = 0·001). More studies of the effects of sleep and of colonic butyrate in mediating the effects of physical activity on intestinal tumourigenesis are warranted.

The study was designed to evaluate whether two types of rye-bran fractions result in distinct bifidogenic effect or enterolactone production in multiple intestinal neoplasia (Min) mice and whether these parameters are associated with intestinal tumorigenesis in this animal model. The experimental diets were a non-fibre diet (control), a rye-bran diet, and diets containing either the soluble extract or the insoluble fraction prepared from rye bran. The main result on adenoma formation in these experiments was the observation that the soluble extract increased number (P=0·012) and size (P=0·008) of adenomas in the distal small intestine when compared with the non-fibre group. All rye-supplemented diets supported similarly the in vivo growth of Bifidobacterium (108–109 colony forming units/g) in Min mice, whereas the non-fibre diet lowered intestinal Bifidobacterium below the level of detection. The results show that water solubility does not affect the bifidogenicity of rye bran. Mean plasma enterolactone concentration was highest in the rye-bran group (30·0 nmol/l; P=0·002), which along with the soluble-extract group (16·2 nmol/l; P=0·024) differed significantly from the non-fibre diet group (7·5nmol/l). Thus, the mice fed with the rye bran were the best enterolactone producers. In conclusion, rye bran and rye fractions influence adenoma formation in Min mice to a varying degree but plasma enterolactone levels or the production of bifidogenic bacteria do not mediate the effect.