Stevioside induces antihyperglycaemic, insulinotropic and glucagonostatic effects in vivo: studies in the diabetic Goto-Kakizaki (GK) rats.
Extracts of leaves from the plant Stevia rebaudiana Bertoni have been used in the traditional treatment of diabetes in Paraguay and Brazil. Recently, we demonstrated a direct insulinotropic effect in isolated mouse islets and the clonal beta cell line INS-1 of the glycoside stevioside that is present in large quantity in these leaves. Type 2 diabetes is a chronic metabolic disorder that results from defects in both insulin and glucagon secretion as well as insulin action. In the present study we wanted to unravel if stevioside in vivo exerts an antihyperglycaemic effect in a nonobese animal model of type 2 diabetes. An i.v. glucose tolerance test (IVGT) was carried out with and without stevioside in the type 2 diabetic Goto-Kakizaki (GK) rat, as well as in the normal Wistar rat. Stevioside (0.2 g/kg BW) and D-glucose (2.0 g/kg BW) were administered as i.v. bolus injections in anaesthetized rats. Stevioside significantly suppressed the glucose response to the IVGT in GK rats (incremental area under the curve ( IAUC): 648 [+ or -] 50 (stevioside) vs 958 [+ or -] 85 mM x120 mm (control); P < 0.05) and concomitantly increased the insulin response (IAUC: 51116 [+ or -] 10967 (stevioside) vs 21548 [+ or -] 3101 [mu]U x 120 min (control); P<0.05). Interestingly, the glucagon level was suppressed by stevioside during the IVGT, (total area under the curve (TAUC): 5720 [+ or -] 922 (stevioside) vs 8713 [+ or -] 901 pg/ml x 120 min (control); P<0.05). In the normal Wistar rat stevioside enhanced insulin levels above basal during the IVGT (IAUC: 79913 [+ or -] 3107 (stevioside) vs 17347 [+ or -] 2882 [mg]U x 120 mm (control); P < 0.001), however, without altering the blood glucose response (IAUC: 416 [+ or -] 43 (stevioside) vs 417 [+ or -] 47 mM x 120 mm (control)) or the glucagon levels (TAUC: 5493 [+ or -] 527 (stevioside) vs 5033 [+ or -] 264 pg/ml x 120 mm (control)). In conclusion, stevioside exerts antihyperglycaemic, insulinotropic, and glucagonostatic actions in the type 2 diabetic GK rat, and may have the pote ntial of becoming a new antidiabetic drug for use in type 2 diabetes.
Key words: Stevioside, insulin, glucose, glucagon, type 2 diabetes, Goto-Kakizaki rat, Wistar rat
Extracts of leaves from the plant Stevia rebaudiana Bertoni (SrB), have been used for many years in traditional medicine in South America in the treatment of diabetes (Soejarto et al., 1983; Mosettig et al., 1955; Kohda et al., 1976). Interestingly, Gun et al. (1986) showed that oral intake of extracts of SrB for 3 days slightly suppressed plasma glucose during an oral glucose tolerance test in healthy subjects. A 35% reduction in blood glucose has also been observed in diabetic subjects after an oral intake of extracts from SrB (Oveido et al. 1979). Various substances are present in the leaves of SrB e.g. the glycoside Stevioside (Bridel and Lavielle, 1931; Wood et al., 1955), which we recently have demonstrated exerts a direct insulinotropic action in isolated mouse islets and in the clonal rat beta cell line, INS-l (Jeppesen et al., 1996; 2000). Type 2 diabetes mellitus is a chronic metabolic disorder that results from defects in both insulin secretion and insulin action (DeFronzo, 1988). In addition to th e insulin abnormalities, a relative glucagon excess and a pancreatic alpha-cell dysfunction have been demonstrated (Unger 1978). An elevated rate of basal hepatic glucose production is the primary cause of fasting hyperglycaemia while impaired suppression of hepatic glucose production and decreased glucose uptake by muscle contribute almost equally to the postprandial hyperglycaernia in type 2 diabetes. The present study was carried out to elucidate if stevioside possesses a potential as an antidiabetic treatment. Since the appropriate treatment of any disease is based on an understanding of its pathophysiology we wanted to reveal if stevioside possesses antihyperglycaemic, insulinotropic and glucagonostatic properties. For this purpose we have applied the Goto-Kakizaki-rat (GK), which is a nonobese animal model of type 2 diabetes charactensed by deficient insulin response to glucose in vivo and in vitro and insulin resistance (Suzuki and Toyota, 1992, Ostenson et al., 1993). The influence of stevioside on ci rculating glucose, insulin and glucagon was studied during an i.v. glucose tolerance test (IVGT) both in the anaestethized diabetic GK rat and the anaestethized normal Wistar rat.
* Material and Methods
Adult male GK (originally obtained from Takeda Chemical lad., Tokyo, Japan and breeded locally; age 20 weeks; 250-320 g) and male Wistar rats (Bomholtgaard Breeding and Research Centre Ltd., Ry, Denmark; age 15 weeks; weighing 250-350 g), were used. The animals were kept on a standard pellet diet and had tap water ad libitum and a light cycle of 12h/12h. Prior to the experiments food was withhold from the rats for 12 h.
Glucose tolerance test in anaesthetized animals
The animals were anaestbetized with pentobarbital i.p. (30 mg/kg) and placed on a temperature-regulated table to maintain body temperature (rectal thermometer) during the glucose tolerance testing.
In the GK and Wistar rats venflons (Insyte-N[TM] 0.7 mm x 1.4 cm, Becton Dickinson Vascular Access, Utah, USA) were placed bilaterally in the femoral veins for blood sampling and infusion. After 15 min glucose (2.0 g/kg BW) and stevioside (0.2 g/kg BW), were injected intravenously as a bolus (dissolved in 0.3 ml 0.9% saline). Stevioside (19-0-B-glucopyranosyl-13-0-[B-glucopyranosyl(1-2)]-B-glucopyranosyls teviol) (96% purity) was obtained from Sigma Chemical Co. Mo. USA. In control experiments only glucose (2.0 g/kg BW) was infused intravenously. Blood samples were drawn at time-points -15, 0, 15, 30, 60, 90 and 120 min, respectively . Blood (300 [mu]l) was collected in microtubes containing 3 [mu]l of a aprotinine/hepanne mixture (7.7 mg/ml aprotinine and 2300 IE/ml heparine, respectively). Blood was replaced by isotonic saline to avoid volume depletion (Rao et al., 1992).
Glucose was determined using the glucose oxidase method on an YSI 2300 Stat Plus (Yellow Springs Instrument Co., Ohio, USA). Blood samples were centrifuged (10 mm, 4 [degrees]C, 4000 rpm). Plasma was removed and frozen for subsequent insulin and glucagon determinations. Plasma insulin and glucagon were analyzed by RIA kits (Linco Research INC. St. Charles, Mo. USA). Stevioside did not crossreact in the insulin or glucagon assays.
Statistical analysis was performed by Students unpaired t-test. Significant differences were considered for P < 0.05. Data are expressed as mean [+ or -] SEM. The incremental area was calculated as the area under the curve above basal (IAUC ). Total area under the curve (TAUC) was calculated as the area above zero.
Effects of stevioside in anaesthetized GK rats during IVGT
As seen at Fig. 1 A we found similar fasting blood glucose (5.8 [+ or -] 0.4 mM vs 6.6 [+ or -] 0.5 mM, respectively; NS) and plasma insulin levels (171 [mu]U/ml [+ 0r -] 32 vs 198 [+ or -] 29 [mu]U/ml, respectively; NS) in the two groups of anaestethized GK rats before infusion of stevioside (n = 6) and saline (n = 6), respectively. Stevioside (0.2 g/kg BW) reduced the glucose response to the IVGT (2.0 glucose g/kg BW) (Fig. 1A). Thus, the IAUC of glucose was reduced in GK rats receiving stevioside (648 [+ or -] 50 vs 958 [+ or -] 85 mM x 120 min; p < 0.05). The lower blood glucose levels were detected from min 30 and onward (P < 0.05). As seen in Fig. 1B stevioside caused a pronounced increase in insulin (IAUC: 51116 [+ or -] 10967 vs 21548 [+ or -] 3101 [mu]U/ml x 120 ruin; P < 0.05). The stevioside-induced insulin response increased throughout the entire test period of 120 min. Interestingly, the glucagon level was concomitanly suppressed as given in Fig. 1C. Thus, a decrease in glucagon occurred 15 min after the injection of stevioside and paralleled the control curve until ruin 90 and attained control level at min 120. The TAUC of glucagon was significantly lowered by stevioside in the GK rats (5720 [+ or -] 922 vs 8713 [+ or -] 901 pg/ml x 120 min., P< 0.05).
Effects of stevioside in anaesthetized Wistar rats during IVGT
Similar basal levels of blood glucose (4.3 [+ or -] 0.3 mM vs 4.1 [+ or -] 0.2 mM) and insulin (140 [+ or -] 24 [mu]U/ml vs 147 [+ or -] 26 [mu]U/ml) were observed in the stevioside (n = 14) and the control group (n = 12) of anaesthetized Wistar rats. As seen at Fig. 2A no difference in blood glucose responses was seen during the IVGT (IAUG): 416 [+ or -] 43 vs 417 [+ or -] 47 mM x 120 min) irrespective stevioside was administrered or not. However, stevioside induced a marked, shortlived increase in insulin levels (IAUC: 79913 [+ or -] 3107 vs 17347 [+ or -] 2882 [mu]U x 120 min; P<0.001) that returned to control levels after 90 min (Fig. 2B). Stevioside caused no change in the glucagon level in normal Wistar rats (TAUC: 5493 [+ or -] 527 vs 5033 [+ or -] 264 pg/ml x 120 min) (Fig. 2C).
This study shows that stevioside induced a reduction in the blood glucose response to an IVGT in anaesthetized, diabetic GK rats, an increase in the insulin secretion as well as a suppression of the glucagon level. In contrast, stevioside did not cause any change in blood glucose to an IVGT in the anaesthetized, normal Wistar rat irrespective a high, transient insulin increment.
At all time points stevioside caused a suppression of the glucose level of 3-4 mM in the GK rat and about a 30% reduction of the total glucose response. As expected higher glucose levels were induced by the IVGT in the GK rat compared to the Wistar rat. In the GK rat basal glucose levels were not reached during the 120 min test period after IVGT while control levels were obtained after 60-90 min in the Wistar rat. However, after stevioside infusion basal blood glucose levels were attained in the GK rat after 90 min in the GK rat. It occurs puzzling that stevioside did not cause a blood glucose lowering effect in normal Wistar rats during IVGT irrespective of the prompt and transient insulin release. The reason for this is not known. However, also Suanarunsawat and Chaiyabutr (1997) found no suppression of blood glucose in the normal rat during stevioside infusion at euglycaemica. In contrast they observed a slight increase in circulating blood glucose to stevioside infusion while blood glucose levels were una ffected by long-term stevioside feeding. The lack of any glucagon suppression to stevioside in the Wistar rat does not seem to be a likely explanation for the absent lowering of blood glucose.
The GK rat represents a mild type 2 diabetic model. It is not clear whether the histopathological changes in the endocrine pancreas are related to the pathogenesis of hyperglycaemia or if they occur secondarily to the abnormal metabolic alterations. Movassat et al. (1997) found that the plasma glucose levels in the GK rat were higher and plasma insulin levels were lower than in Wistar rats. Also the pancreatic insulin content, the relative beta cell volume and total mass of beta cells were lower, representing only 23% of a Wistar control group (Movassat et al., 1997). In the present study, insulin levels are also lower in the GK rat compared to the Wistar rat and the GK rat displayed mild hyperglycaemia and higher glucagon levels after the IVGT. Other studies have found defects late in the signal transduction in islets of GK rats, possibly occurring at the site of activation of cytosolic nucleoside diphosphokinase, a G-protein-dependent step in the exocytosis (Metz et al., 1999). Whether stevioside affects G- protein-dependent steps in the insulin exocytosis is not known. The mechanisms underlying the anti-hyperglycaemic action of stevioside in diabetes is thus still poorly understood. Interestingly, we have recently demonstrated that the insulinotropic effect of stevioside unlike the classical sulphonylureas is not associated with closure of the ATP sensitive potassium channels in the beta cells (Jeppesen et al., 2000). In addition to the glucagonostatic effect of stevioside that has the potential of reducing hepatic glycogenolysis other mechanisms may play a role regarding the glucose-lowering action of stevioside. These mechanisms could involve increased tissue glycolysis, increased muscle glycogen storage, less hepatic glycogenolysis or increased urinary glucose loss. Consequently, it has been described that an intravenous infusion in rats of stevioside at a dose of [greater than or equal to]8 mg/kg/h caused an increased urinary glucose clearence (Melis, 1992). In our experiment a much higher stevioside dose ( 0.2 g/kg i.v. as bolus injection) was applied and we cannot exclude that an increased renal glucose clearence in part may contribute to the blood glucose lowering effect. However, in a study in type 2 diabetic humans we have recently demonstrated that the oral intake of stevioside causes a clearcut reduction in the glycaemic response to a test meal, a rise in insulin and suppression of glucagon without causing any significant change in urinary glucose loss (unpublished results). This supports our hypothesis that the antihyperglycaemic effect of stevioside is primarily due to a direct action on the insulin producing beta cells and glucagon producing alpha cells. In addition, it has been suggested that stevioside inhibits phosphorylation in the rat liver mitochondrias that may be attributed to an inhibition of the ADP/ATP exchange. Thus it is known that inhibition of ATP synthesis in mitochondrias can result in increased glycolysis and decreased gluconeogenesis (Vignais et al., 1966; Bracht et al., 1985).
To our knowledge this is the first demonstration of a stevioside-induced potentiation of the glucose-stimulated insulin secretion in vivo. The present data corroborates our recent demonstration in vitro of a glucose dependent potentiation of the insulin secretion from isolated mouse islets (Jeppesen et al., 1996; 2000) and from isolated rat islets (Malaisse et al., 1998). A striking difference between the dynamics of the insulin responses to stevioside was observed in the diabetic and the normal rat. Thus, a steady increasing insulin response appeared to stevioside in the anaestethized GK rats in contrast to the high and transient response in the anaestethized Wistar rat. As expected the maximal insulin levels were higher in the Wistar rather than in the diabetic GK rat. The reason for the slow, steady increase in insulin levels to the IVGT in the GK rat in contrast to the prompt, short-lived increase in the Wistar rat is poorly understood. However, it may reflect at least in part the abnormalities in islet g lucose metabolism that have been demonstrated in the GK rat. Thus, glucose utilization and oxidation are markedly increased in GK rat islets (Ostenson et al., 1993; Ling et al., 1998), islet glucose cycling is significantly higher in GK rats compared with control Wistar rats displaying a higher islet glucose-6-phosphatase activity (Ling et al., 2001). Another explanation could be a high affinity of stevioside to the diabetic beta cells, a priming action of stevioside and/or changed clearence of insulin in the GK rat.
Interestingly, plasma glucagon decreased in anaesthetized GK rats in response to an injection of stevioside. In contrast, the glucagon level was not significantly altered in anaesthetized Wistar rats. The suppression in glucagon occurs immediately and is most pronouced during the first 30 min after the injection of stevioside in the GK-rat. The glucagonostatic effect of stevioside may be caused by a direct inhibitory action on the glucagon producing alpha-cells or an indirect action via insulin-induced suppression of the glucagon secretion. The lack of a prominent first phase insulin response to stevioside in the GK rat and the finding that irrespective a noticeable insulin response in the Wistar rat no change in glucagon occurred seems, however, to make direct insulin related suppression of glucagon less likely. How much the glucagonostatic effect contributes to the differential impact of stevioside on blood glucose in the OK and Wistar rat is not known.
The question arises whether stevioside in the fasting state may elicit hypoglycaemia like the sulphonylureas thereby constituting a potential threat for the diabetic subjects. To answer this question we will have to carry out experiments e.g. with bolus injections of stevioside in the fasting, conscious GK and Wistar rats. In our in vitro studies (Jeppesen et al., 2000) we found that the diterpene stevioside stimulated insulin release from isolated mouse islets at glucose concentrations of 8.3, 11.1 and 16.7 mmol/l, respectively. Even at 6.6 mmol/L glucose, a small increment in insulin could be detected. However, at low glucose of 3.3 mmol/l or less, no insulinotropic action was found (Jeppesen et al., 2000).
It should be underlined that anaesthetizing rats may have significant effects on glucose levels, insulin levels, insulin release and other issues related to these hormones. Caution with the interpretation of data obtained in anaestethized compared to conscious animals should consequently be exercised
In conclusion, stevioside possesses antihyperglycaemic, insulinotropic as well as glucagonostatic effects in the diabetic OK rat. Stevioside seems to possess a promising therapeutical role in type 2 diabetes. However, more information is needed on its mechanism of action, efficacy, nonglycemic benefits, safety profile, and long-term effects.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
The authors wish to thank Kirsten Eriksen, Dorthe Rasmussen and Tove Skrumsager for skilled technical assistance.
Bracht, A. K., Alvarez, M., Bracht, A.: Effects of Stevia rebaudiana natural products on rat liver mitochondria. Biochem. Pharmacol. 34: 873-882, 1985.
Bridel, M., Lavielle, R.: Le principe a'saveur sucre du Kaa' -he'-e'. Bull. Soc. Chim. Biol. 13: 636-655, 1931.
Curi, R., Alvarez, M., Bazotte, R. B., Botion, L. M., Godoy, J. L., Bracht, A.: Effect of Stevia rebaudiana on glucose tolerance in normal adult humans. Braz. J. Med. Biol. Res. 19: 771-774, 1986.
DeFronzo R. A.: The triumvirate: beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 37: 667-687, 1988.
Jeppesen, R B., Gregersen, S., Hermansen. K.: Stevioside and steviol stimulate insulin secretion from isolated mouse islets. Diabetologia A125: 472, 1996.
Jeppesen, P. B., Gregersen, S., Poulsen, C. R., Hermansen, K.: Stevioside acts directly on pancreatic beta-cells to secrete insulin: Actions independent of cAMP and ATP-sensitive [K.sup.+]-channel activity. Metabolism 49: 208-214, 2000.
Kohda, H., Hasai, R., Yamasaki, K., Murakami, K., Tanaka, O.: New sweet diterpene glycosides from Stevia rebaudiana. Phytochemistry 15: 981-983, 1976.
Ling, Z. C., Efendic, S., Wibom, R., Abdel-Halim S.M., Osteson, G-C., Landau, B.R., Khan, A.: Glucose metabolism in Goto-Kakizaki rat islets. Endocrinology 139: 2670-2675, 1998.
Ling, Z. C., Hong-Lie, C., Ostenson, C-G., Efendic, S., Khan, A.: Hyperglycemia contributes to impaired insulin response in GK rat islets. Diabetes 50 (Suppl 1): S108-S112, 2001.
Malaisse, W. J., Vanonderbergen, A., Louchami, K., Jijakli, H., Malaisse-Lagae, F.: Effects of artificial sweeteners on insulin release and cationic fluxes in rat pancreatic islets. Cell Signal 10: 727-733, 1998.
Melis, M. S.: Renal excretion of stevioside in rats. J. Nat. Prod. 55: 688-690, 1992.
Metz, S. A., Meredith, M., Vadakekalam, J., Rabaglia, M. E., Kowluru, A.: A defect late in stimulus-secretion coupling impairs insulin secretion in Goto-Kakizaki diabetic rats. Diabetes 48: 1754-1562, 1999.
Mosettig, E., Nes, W. R.: Stevioside II: The structure of the aglucone. J. Org. Chem. 20: 884-99, 1955.
Movassat, J., Saulnier, C., Serradas, P., Portha, B.: Impaired development of pancreatic beta-cell mass is a primaty event during the progression to diabetes in the OK rat. Diabetologia 40: 9 16-925, 1997.
Ostenson, C. G., Khan, A., Abdel-Halim, S. M., Guenifi, A., Suzuki K., Goto Y, Efendic S.: Abnormal insulin secretion and glucose metabolism in pancreatic islets from the spontaneously diabetic GK rat. Diabetologia 36: 3-8, 1993.
Oveido, C. A., Franciani, G., Moreno, R., Maas, L. C.: Accion hipoglucemiante de la Stevia rebaudiana Bertoni (Kaa-he-e). Excerpta Medica 209: 92, 1979.
Rao, R. H.: Changes in insulin sensitivity from stress during repetitive sampling in anesthetized rats. Am. J. Physiol. 262: R 1033-1039, 1992.
Soejarto, D. D., Kinghorn, A. D., Farnsworth, N. R.: Potential sweetening agents of plant origin III: Organoleptic evaluation of Stevia leaf herbarium samples for sweetness. J. Nat. Prod. 45: 590-598, 1983.
Suanarunsawat, T., Chaiyabutr, N.: The effect of stevioside on glucose metabolism in rat. Can. J. Physiol. Pharmacol. 75: 976-982, 1997.
Suzuki, K. I., Toyota, T.: Spontaneously diabetic GK (Goto-Kakizaki) rats. Animal diabetes IV. Ed. E. Shafrir 107-115, 1992.
Unger, R. H.: Role of glucagon in the pathogenesis of diabetes: The status of the controversy. Metabolism 27: 1691-1709, 1978.
Vignais, P. V., Duee, E. D., Vignais, P. M., Huet, J.: Effect of atractyligenin and its structural analogues on oxidative phosphorylation and on the translocation of adenine nucleotides in mitochondria. Biochem. Biophys. Acta. 118: 465-483, 1966.
Wood, Jr. H. B., Allerton, R., Diehl, H. W., Fletcher, H. C.: Stevioside I: The structure of glucose moieties. J. Org. Chem. 20: 875-879, 1955.
Per Bendix Jeppesen, Department of Endocrinology and Metabolism, Aarhus University Hospital, Tage-Hansens gade 2, DK - 8000 Aarhus C. Denmark Tel.: ++45-89497735; Fax: ++45-89497649; e-mail: firstname.lastname@example.org