The beneficial actions of diets high in carbohydrate and fiber on insulin sensitivity in noninsulin-dependent diabetes mellitus (NIDDM)⁴ and insulin-dependent diabetes mellitus (IDDM) are well documented (Anderson et al. 1991, Fukagawa et al. 1990, Simpson et al. 1981). However, considerable debate exists as to the effectiveness of such diets given the potentially adverse actions of high carbohydrate diets on plasma lipids (Grundy 1991). The supplementation of high carbohydrate diets with guar gum, a viscous galactomannan extracted from the Indian cluster bean (Cyamopsis tetragonolobus), has been demonstrated to effectively enhance insulin sensitivity in individuals with either NIDDM or IDDM (Ebeling et al. 1988, Lalor et al. 1990, Tagliaferro et al. 1985).

The actions of guar gum to lower blood glucose levels may be the result of the increased viscosity of the stomach and small intestine gastrointestinal contents, impeding carbohydrate digestion and absorption. Increasing the viscosity of the gastrointestinal contents with guar gum (Cameron-Smith et al. 1994a) acts as a physical impediment to carbohydrate digestion and absorption (Johnson 1991, Leclere et al. 1994). This has been supported in numerous studies in which guar gum lowers the postprandial glucose response when mixed into a variety of test meals (Collier et al. 1986, Jenkins et al. 1977, Leclere et al. 1994).

The longer-term reductions in the postprandial rate of carbohydrate absorption with guar gum supplementation is thought to improve insulin sensitivity through the combined actions of reduced diurnal insulin excursions (Jenkins et al. 1989), lower postprandial counterregulatory hormone release (Collier et al. 1987) and reduced gastric inhibitory polypeptide secretion (Morgan et al. 1990). However, despite our knowledge of the whole-body improvements in carbohydrate metabolism, the tissue-specific improvements in carbohydrate metabolism following guar gum supplementation have not been investigated. Therefore, the aims of the present study are to investigate in streptozotocin-induced (STZ) diabetic rats the effects of guar gum, when incorporated into a high carbohydrate diet, on tissue-specific and whole-body insulin sensitivity. Alterations in basal and insulin-stimulated glucose metabolism were determined using the hyperinsulinemic clamp technique, in combination with infusion of labeled 2-deoxy-D-glucose, a nonmetabolizable glucose analog which provides a sensitive in vivo measure of tissue-specific rate of glucose disposal (Kraegen et al. 1985).

MATERIALS AND METHODS

Table 1. Composition of diets [View Table]

Prior to the commencement of tracer infusions, blood samples (300 µL) were collected from rats food-deprived overnight into heparinized tubes for determination of plasma glucose, cholesterol and triglycerides. Basal rates of glucose turnover were determined by the isotope dilution of infused ³(H)-6-glucose (Du Pont, Wilmington, DE) as described previously (Kraegen et al. 1983). Glucose turnover was measured following a priming infusion [5-min priming dose, 88 kBq/(kg·min)] followed by a constant dose [8.8 kBq/(kg·min)] of ³(H)-6-glucose (Du Pont, Boston, MA).

Following the basal infusion, insulin-stimulated rates of glucose turnover were measured following a primed (5-min) continuous exogenous insulin infusion. Insulin was added to the ³(H)-6-glucose infusion to achieve a dose of 18 pmol/(min·kg body weight) (Actrapid, Novo, Denmark). A variable rate of 250 g/L glucose, containing 240 MBq/L ³H-6-glucose, was infused to maintain euglycemia. Triplicate blood samples (300 µL) were taken at 5-min intervals 70 min after initiation of the insulin infusion to determine plasma ³H specific activity.

An estimate of individual tissue glucose metabolic rate, the glucose utilization index (Rg), was measured as the tissue accumulation of phosphorylated 2DG; Rg [µmol/(100 g of tissue·min)] was calculated using the following expression: where Cp is the steady-state plasma glucose concentration (mmol/L), Cm·(T) is the total phosphorylated 2DG accumulated in 45 min and _o⁴⁵ C_p·dt is the integrated value for plasma 2DG specific activity over 45 min, as described in detail by Kraegen et al. (1985).

RESULTS

The effects of STZ-induced diabetes and fiber supplementation on daily food intakes are shown in Figure 2. STZ diabetic rats fed the bran diet were hyperphagic throughout the study, consuming on average 42.9 ± 1.0 g diet/d. In the STZ diabetic guar-fed rats, compared with the STZ diabetic bran-fed rats, food intake was lower at all weeks, reaching significance in the first and third weeks of the study. No significant differences in food consumption were evident between the bran- and guar-supplemented control groups.

Table 2. Body weight and plasma metabolite and insulin concentrations in wheat bran- and guar gum-fed nondiabetic and streptozotocin (STZ) diabetic rats1 [View Table]

The effect of the bran or guar diets in normal and STZ diabetic rats on basal and hyperinsulinemic glucose turnover is described in Table 3. Hepatic glucose output (HGO) was elevated in STZ diabetic rats compared with the nondiabetic groups. However, within the STZ diabetic rats, the guar diet resulted in a significant reduction in basal glucose turnover (P < 0.05) and an elevation in the MCR (P < 0.05).

Table 3. Glucose turnover variables in wheat bran- and guar gum-fed nondiabetic and streptozotocin (STZ) diabetic rats [View Table]

The insulin-stimulated rate of glucose disposal and the insulin-suppressed rate of HGO are shown in Table 3. HGO did not differ among groups during hyperinsulinemia. The rate of glucose disposal (MCR) was significantly greater in both the nondiabetic and STZ-induced diabetic rats fed the guar-containing diet compared with the bran-fed rats.

Whole-body insulin sensitivity following bran or guar feeding in nondiabetic or STZ diabetic rats, as measured by the rate of glucose infused (GIR) to maintain euglycemia in the hyperinsulinemic state, is shown in Figure 3. In nondiabetic rats, the rate of glucose infusion to maintain euglycemia was 32% greater in those fed guar than in those fed bran. In STZ diabetic rats fed the guar gum diet, a significantly greater rate of glucose infusion was required to maintain euglycemia during hyperinsulinemia, compared with the STZ diabetic rats fed bran.

Table 4. 2-Deoxyglucose utilization index (Rg) in individual tissues measured during hyperinsulinemic clamp conditions in wheat bran- and guar gum-fed nondiabetic and streptozotocin (STZ) diabetic rats1 [View Table]

DISCUSSION

The development of hyperglycemia and insulin resistance in the STZ model of diabetes is preceded by the partial destruction of the pancreatic -cell mass by STZ (Kruszynska et al. 1986); however, at the dose of STZ used in the current study, sufficient -cell capacity remained to maintain adequate plasma insulin concentrations in rats deprived of food overnight. Therefore, in STZ diabetic rats, insulin resistance develops when both insulin secretion is reduced and plasma glucose levels are elevated. However, it has been suggested that it is the maintenance of chronic hyperglycemia which is the primary factor responsible for the development and maintenance of insulin resistance in both rodent models of diabetes and human NIDDM subjects (Rossetti et al. 1987, Yki-Jarvinen 1992).

Chronic hyperglycemia is detrimental to both insulin sensitivity and -cell function (Yki-Jarvinen 1992). In STZ-induced and partially pancreatomized diabetic rats, phlorizin treatment, which induces persistent glycosuria by inhibiting renal glucose reabsorption, normalizes blood glucose levels and restores peripheral insulin action, independent of the changes in insulin secretion (Lisato et al. 1992, Rossetti et al. 1987). The present results demonstrate that the supplementation of a high carbohydrate diet with guar gum fed to STZ diabetic rats resulted in a pronounced reduction in plasma glucose concentrations in the food-deprived STZ rats without significant alterations to the insulin concentrations. Because the majority of the infused glucose load during hyperinsulinemia is cleared from the plasma by skeletal muscle (Beck-Nielsen et al. 1992), the detrimental actions of hyperglycemia are likely to be directed towards insulin-sensitive glucose transport and metabolism within skeletal muscle (Lang et al. 1991). In the current study, the improvements in insulin action following guar supplementation were shown to be mediated by increased peripheral tissue insulin sensitivity. This finding confirms previous studies in normal and NIDDM human subjects (Landin et al. 1992, Taglioferro et al. 1985). However, in the current study, the inability of similar levels of hyperinsulinemia to suppress the persistent hyperglycemia in the STZ diabetic bran-fed rats following the infusion of exogenous insulin, while allowing for the normalization of blood glucose concentrations in the STZ diabetic guar-fed rats, prevents the accurate use of calculations dependent upon isotope dilution for the comparison of the changes in peripheral glucose disposal. Nevertheless, this persistent hyperglycemia in the STZ diabetic bran-fed rats provides strong evidence for the marked changes in peripheral insulin action with the addition of guar to the diet of the diabetic rats.

Changes in the rate of glucose disposal in specific peripheral tissues were examined in the current study by measuring the accumulation of the nonmetabolizable glucose analog 2-deoxyglucose (2DG) in selected tissues (Kraegen et al. 1985). Insulin-sensitive glucose disposal tended to be greater in rats fed the guar diet in both the nondiabetic and the STZ diabetic groups, compared with the bran-fed rats, in all muscle groups, including several hindlimb muscles, the heart and diaphragm. However, the interpretation of the 2DG accumulation is confounded by the persistent hyperglycemia in the insulin-infused bran-fed diabetic animals, despite correction for plasma glucose concentrations in the calculation of 2DG tissue accumulation. Although it could have been expected that the combined actions of hyperglycemia and elevated plasma insulin concentrations may have acted to increase tissue 2DG accumulation in the diabetic rats fed the bran diet compared with the diabetic rats fed the guar diet. The failure to demonstrate increased 2DG accumulation in the diabetic bran-fed rats further emphasizes the considerable alterations in insulin sensitivity between the two STZ diabetic groups. Therefore, the improvements in insulin-sensitive glucose disposal in the STZ diabetic rats fed guar gum are likely to be the result of small increases in the rate of glucose disposal by all insulin sensitive tissues, rather than a large elevation in the rate of glucose disposal of isolated tissues.

Although the mechanism by which guar gum supplementation increased insulin action was not specifically addressed in this study, it can be speculated that differing mechanisms may account for the improvements in insulin action in the nondiabetic and diabetic rats. The ability of guar gum to suppress postprandial rises in blood glucose has been extensively documented in both human and animal studies (Collier et al. 1986, Jenkins et al. 1977, Track et al. 1982). This reduction in postprandial glucose and insulin excursions may be important in the improvements in insulin action measured using the euglycemic clamp and 2DG infusion in the nondiabetic animals. The importance of postprandial glycemia in determining insulin sensitivity in normoglycemic nondiabetic rats is supported by improvements in glucose tolerance and insulin action of a similar magnitude in normal rats fed starch diets of differing in vitro digestibility (Higgins et al. 1996).

In the STZ diabetic rats, in which the ability of the pancreas to secrete increased insulin in response to absorbed glucose is largely ablated by the streptozotocin treatment (Cameron-Smith et al. 1994b), reductions in postprandial glucose concentrations following the addition of guar to the diet may lead to improvements in insulin action via other mechanisms. As in previous studies, the STZ diabetic rats were hyperphagic (Katovich et al. 1991), with the bran-supplemented STZ diabetic rats consuming on average 57% more food daily than nondiabetic rats fed the bran-containing diet. A possible mechanism by which the guar diet could improve insulin sensitivity in the STZ diabetic rats is the inhibition of urinary glucose loss resulting in a suppression of the hyperphagia. Mild STZ treatment, as used in this study, has been reported to result in significant urinary glucose loss and polyurea (Katovich et al. 1991), although no measurements of urinary glucose were made in this study. The loss of glucose in the urine may result in the hyperphagia, with the diabetic rats consuming food in excess to account for urinary glucose losses. The slowed glucose release from the gastrointestinal tract in the diabetic rats fed guar may have resulted in a more sustained, although smaller postprandial hyperglycemia in the STZ diabetic rats, limiting the period over which the hyperglycemia exceeded the renal threshold. The reduced glucose loss in the urine may have in turn acted to limit the hyperphagia. However, the possibility also exists that this diabetic hyperphagia may have been limited by the nature of the viscous guar diet. Reduced food consumption has been noted in previous studies utilizing nondiabetic rats (Begin et al. 1989, Track et al. 1982), and the satiating actions of guar gum are well described in human studies (Burley et al. 1987, French and Read 1994) in which guar gum is proposed as a potential agent to promote energy restriction in obese individuals (Krotkiewski 1984). The limiting action of the guar diet on hyperphagia may possibly lead to improvements in insulin action. Previously, mild food restriction in STZ diabetic rats and mice has been shown to result in reduced hyperglycemia and increased insulin sensitivity (Cameron-Smith et al. 1994b, Hasegawa et al. 1990). Energy restriction, per se, may improve insulin sensitivity in NIDDM subjects, and the effectiveness of low energy diets in improving many aspects of carbohydrate metabolism, including insulin sensitivity has been well documented (Goldstein 1992, Henry et al. 1985). Therefore, it is possible that reduction in total carbohydrate availability, in guar gum-fed rats, may have contributed to improvements in peripheral insulin sensitivity.

In summary, inclusion of guar gum into a high carbohydrate diet, compared with wheat bran, is effective in lowering plasma glucose concentrations in STZ diabetic rats that were food-deprived overnight. Furthermore, the guar diet improved insulin sensitivity in STZ diabetic rats, with the majority of the improvements resulting from increased insulin sensitivity of glucose disposal in peripheral tissues.