Dietary fibres reduce diabetes prevalence in NZO mice independently of obesity development

Introduction: Epidemiological studies provide evidence for a correlation between an increased intake of fibre-rich whole-grain cereals and a reduced diabetes risk. The underlying mechanism is unknown. The aim of this study was to identify possible mechanisms of action by feeding diets enriched with different fibre types, e.g. soluble and insoluble fibres. New Zealand Obese (NZO) mice, which develop obesity-dependent diabetes with beta-cell failure, were used as a polygenic type 2 diabetes model.

Methods: Up to 16 weeks of age, male NZO mice were fed with high-fat diets (15% fat w/w) containing different amounts and types of fibre: Control diet (5% wheat-cellulose w/w) and 3 experimental diets containing 10% (w/w) fibre, either as wheat-cellulose, resistant starch, or beta-glucan. Body weight/composition, food intake and blood glucose were measured regularly. The fermentation rate was estimated by measuring the amount of short-chain fatty acids (SCFA) in caecal content and faeces. Glucose excursions following a meal (week 6) and oral glucose tolerance (week 8) were assessed.

Results: Both the control diet and the fibre-enriched diets caused robust weight gain, whereas the beta-glucan group showed a diminished weight in comparison to the control group (46.6±0.7g vs. 51.2±1.0g, p<0.001, week 9). This could be due to the reduced metabolised energy of the beta-glucan diet in comparison to the control diet (3.0 kJ/g lean mass/d vs. 3.6 kJ/g lean mass/d, p<0.01). Increased production of SCFA by caecal fermentation was verified for the diets containing fermentable fibres (resistant starch and beta-glucan) compared to cellulose containing diets. All fibre-enriched diets significantly reduced the prevalence of diabetes (32–39% vs. 70% control diet; week 14). Glucose excursions after a meal of the fibre-enriched diets were not significantly different between control, cellulose and resistant-starch diets, but were lower in the beta-glucan group. However, the corresponding insulin response was significantly lower for the cellulose diet (15min insulin values control=4.5±0.6µg/l vs. cellulose=2.8±0.4µg/l; p<0.05) but unchanged for all other diets. An oral glucose tolerance test did not reveal differences in glucose and corresponding insulin response between the groups.

Conclusion: Increasing fibre content from 5% to 10% reduces diabetes prevalence in NZO mice independent of obesity development, fibre fermentation and postprandial glucose and insulin response. These data are consistent with the hypothesis of an anti-diabetogenic effect of all fibre types possibly resulting from the mechanical impact of fibres on gastrointestinal filling. In addition, fibre-specific effects on insulin sensitivity and insulin secretion might contribute to the anti-diabetogenic action.