Animal studies consistently show that diets containing high levels of RS (16%) reduce body weight gain and/or adiposity in rodents[6, 17, 18]. Moderate levels of RS (4-6%) also reduced weight gain in healthy rats and adiposity during weight regain in obese rats. However, the effects of moderate to high RS levels in the diet on adiposity and weight gain have not been examined in OP or OR rodents without prior exposure to a weight loss diet. The present study demonstrated clearly that dietary RS reduced body weight gain, although not final body weight, and that the effect was dependent on the level of RS in the diet and on the phenotype of the rats. When compared to 0% RS, weight gain was significantly lower in OP rats when fed RS at 4, 12 and 16% whereas in OR rats, weight gain was significantly lower only when fed 16% RS. These differences between obesity phenotype could be due to differences in colonic fermentation. In support of this Zhou et al. demonstrated that RS did not reduce adiposity in genetically obese mice that were unable to ferment RS. Additionally studies by Gordon and colleagues show that differences in microbial populations between obese and lean individuals may explain the phenotypic differences however differences in energy intake may also account for this effect. The current study demonstrated that the level of RS in the diet necessary to reduce adiposity, was independent of the obesity phenotype. Although OP rats fed 4% RS gained less weight than rats on the 0% RS diet, visceral and subcutaneous adipose tissue weights, total adiposity and energy intake did not differ. In humans replacement of 5.4% of total dietary carbohydrate with RS (approximately 5 g RS in the meal) increased postprandial lipid oxidation, however consumption of RS (40 g/d) for 12 wk did not affect body weight or fat storage in muscle, liver and visceral depots in overweight/obese subjects. The higher dietary levels of RS (8%) used in the present study lowered total body and visceral adipose tissue weight, as well as plasma triglycerides, total cholesterol and leptin. This level of RS intake, which equates to about 88 g/d for a 75 kg human, would be difficult to achieve for most adults. However it is possible that lower amounts (between 44 and 88 g/d for an adult) that could be readily accommodated in the diet via dietary supplements and consumption of foods high in RS, may elicit favourable changes in adiposity. Furthermore, at these lower levels of RS a greater reduction in adiposity may also be achieved in a trial of longer duration.
The mechanisms by which RS reduces adiposity are likely to involve a decrease in metabolisable energy intake and an increase in fatty acid catabolism. In the current study gross energy intake declined as the amount of RS increased in the diet (16% RS reduced energy intake by 11%). An even larger reduction in energy intake (16.4%) was reported for rats consuming a high RS diet (equivalent to 16% RS) when total accumulative energy intake was measured daily throughout the 4 wk trial rather than a 2-day measurement period used in the current study. In addition, studies that matched the energy content of the control (0% RS) and RS intervention diets, still resulted in a more effective lowering of body fat than the control[17, 18] which suggests that mechanisms other than reducing energy intake may also play a role in reducing adiposity. Fermentation of RS in the large bowel has been proposed as a mechanism to explain the increase in lipid oxidation[18, 23]. SCFA, the major products of RS fermentation, are absorbed by colonocytes and enter the hepatic portal circulation where they can directly regulate a variety of pathways involved in fatty acid and cholesterol metabolism[24–26]. In particular, propionate can reduce the incorporation of acetate into cholesterol and inhibits fatty acid synthesis[27, 28]. Additionally, a lower ratio of serum acetate to propionate is associated with lower serum cholesterol[29, 30] but was not seen in the current study. Although most SCFA produced in the colon are metabolised by the liver small amounts enter the systemic circulation[31, 32] and there is growing evidence that they may directly regulate adipogenesis and adipokine release in adipose tissue mediated via G-protein coupled receptors. Fermentation of a single meal high in RS (measured using breath hydrogen) was associated with a reduction in plasma NEFA levels in healthy subjects. This reduction in NEFA may be particularly effective in improving insulin sensitivity in obese individuals and those with T2DM as it would reduce fatty acid oxidation and storage, increase muscle glucose uptake and oxidation, and improve β-cell insulin secretory response to glucose. However, in the current study it is not clear why RS did not reduce plasma NEFA, but it may explain in part why an improvement in insulin sensitivity was not seen. Additionally, SCFA may modulate fat metabolism indirectly by stimulating large bowel enteroendocrine cell production of GLP-1 and PYY. In the current study total SCFA pools in caecal digesta were positively associated with plasma concentrations of both hormones. GLP-1 and PYY have been shown to be released into the blood in a sustained day-long manner and to act systemically on white adipose tissue to regulate lipogenesis, lipolysis, fatty acid release and adipocyte differentiation. Additionally, PYY can increase thermogenesis and energy expenditure, thereby reducing adiposity. In support of this, plasma GLP-1 and PYY levels explained 16 – 19% of the change in total body fat mass and visceral fat mass. Furthermore, diets containing added RS reduced plasma concentrations of GIP, a gut hormone known to promote lipogenesis by stimulating adipose tissue blood flow, glucose uptake and fatty acid re-esterification, and, as a consequence, increased triglyceride deposition in abdominal subcutaneous adipose tissue[33, 34].
Foods containing RS elicit lower postprandial insulin responses and short-term consumption of RS (15–40 g/d) by healthy as well as overweight and obese subjects improves insulin sensitivity[21, 35, 36]. Studies in rats also show consistent positive effects for RS on postprandial glycemic and insulinemic responses and whole body insulin sensitivity. Therefore, it was surprising in the current study that resistant starch even at the highest level of dietary inclusion did not improve insulin sensitivity of the rats. It is possible that the duration of the moderate fat diet was too short or that fasting levels of glucose and insulin were not sensitive enough to discern changes in insulin sensitivity. Higgins et al. showed that rats fed an amylopectin-based diet (low in RS) developed insulin resistance (as measured by an intravenous glucose tolerance test) after 3 mo of feeding whereas those on an amylose diet (high in RS) developed insulin resistance after 6 mo. A feeding trial considerably longer than the 4 wk used in the current study may be required to investigate the dose response effects of resistant starch on improving insulin sensitivity.
The low fat background diet used in the present study is not representative of the typical Western diet and may have limited the impact of RS in improving insulin sensitivity. Andersson et al. showed that a low dietary fat content ameliorated the negative effect of a high glycaemic diet on insulin resistance. In our study the low level of dietary fat was more effective in lowering hepatic fat levels than feeding RS. However the length of the intervention was only 4 wk and may have been too short to observe an effect of RS on reducing heptic lipid content.