Introduction

Top Abstract Introduction Materials and methods Results Discussion References

The tree Pycnanthus angolensis of the Myristicaceae family, also known as "African nutmeg," is widely used for ethnomedical purposes (Akendengue and Louis, 1994). Of particular relevance to our drug discovery program (Oubré et al., 1997) was information gathered during a field trip to Nigeria on its use to treat chronic fungal infections, a clinical problem commonly seen in patients with uncontrolled hyperglycemia (Gill, 1991). Based on this information, extracts of the leaves of P. angolensis were used to initiate an in vivo guided fractionation program (Oubré et al., 1997; Luo et al., 1998a) in an effort to discover new drugs for the treatment of type 2 diabetes. These efforts resulted in the isolation of two compounds that lowered plasma glucose after the oral administration in mouse models of type 2 diabetes. These compounds, shown in Fig. 1, represent a new class of terpenoid-type quinones. Furthermore, they are structurally distinct from the currently available oral compounds used to treat type 2 diabetes: sulfonylureas, biguanides, disaccharidase inhibitors, and thiazolidinediones. In this presentation, we evaluated the pharmacological effects of these two compounds on various aspects of glucose and insulin metabolism in mouse models of diabetes.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Chemical structures of SP-18904 and SP-18905

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Compounds SP-18904 and SP-18905 were isolated from an ethanolic extract of the leaves of P. angolensis with a series of in vivo guided fractionation steps (Oubré et al., 1997; Luo et al., 1998a) involving liquid/liquid partition, LH-20 column chromatography, and high pressure liquid chromatography.

Male C57BL/6J-ob/ob mice (ob/ob) and C57BL/ks-db/db mice (db/db) (7-8 weeks old) were used to both guide the fractionation process and evaluate the purified compound. Mice were purchased from the Jackson Laboratory (Bar Harbor, ME), housed (4 mice/cage) in a temperature- (22 ± 3°C) and humidity- (50 ± 20%) controlled room with a 12-h light (6 AM-6 PM)/dark cycle, and maintained on a diet of Purina rodent chow and water ad libitum. In addition, experiments were also performed on C57BL/ks mice in whom insulin deficiency was induced by the i.p. injection of streptozotocin (150 mg/kg) (Sigma Chemical, St. Louis, MO) according to Rossini et al., (1977).

Mice were bled and prescreened for plasma glucose. Mice selected for study had glucose concentrations of 300 to 600 mg/dl. Each treatment group consisted of five to eight mice, distributed so that the mean glucose levels were equivalent in each group at the start of each study. Mice were dosed orally once a day by gavage with either vehicle, SP-18904, or SP-18905. Testing materials were delivered in a liquid vehicle containing 0.25% (w/v) carboxymethylcellulose, 1% Tween 60, and 10% dimethylsulfoxide (DMSO) (all from Sigma). The amount of material given and the duration of treatment varied among experiments, and details are given with the experimental results. Blood samples were taken from the tail vein 3 h after the dosing on the corresponding sampling day in nonfasted conditions unless indicated otherwise. Individual body weight and mean food consumption (each cage) were measured daily.

Estimates of insulin-mediated glucose uptake were obtained after 4 days of treatment with SP-18905 following the simplified insulin suppression test method described by Luo et al., (1998b). Food was removed on the day of the experiment, with the last oral dose of test compounds was given 1 h later. Four hours after the withdrawal of the food, mice were anesthetized with i.p. sodium pentobarbital (Sigma) at 100 mg/kg. The abdominal cavity was opened, and the main abdominal vein exposed and catheterized with a 24-gauge i.v. catheter (Johnson-Johnson Medical Inc., Arlington, TX). The catheter was secured to muscle tissue adjacent to the abdominal vein, cut on the bottom of the syringe connection, and hooked to a prefilled PE50 plastic tube, which in turn was connected to a syringe with infusion solution. The abdominal cavity then was sutured closed. With this approach, there is no blockage of the back flow of the blood from the lower part of the body. Mice were infused continuously with glucose (20 mg/kg/min) and insulin (20 mU/kg/min) (both from Sigma) at a rate 10 µl/min. Retro-orbital blood samples (70 µl each) were taken 105, 120, and 135 min after the start of infusion for the measurement of plasma glucose and insulin concentration. The mean of these three samples was used to estimate the steady-state plasma glucose (SSPG) and insulin (SSPI) concentrations for each animal.

Murine 3T3-L1 preadipocytes (American Type Culture Collection CL 173) were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 10% (v/v) supplemented calf serum, antibiotics, and 25 mM glucose. Cells were seeded onto 24-well cluster plates (10,000 cells/well), grown to confluence (typically 5 days), and induced to differentiate 2 days after confluence (day 0) according to the standard protocol of Frost and Lane (1985). After differentiation, adipocytes were maintained in DMEM containing 10% fetal bovine serum and provided with fresh medium every 2 to 3 days. Adipocytes used in this study were used on days 7 to 10 after differentiation. On the day of the experiment, adipocytes were washed with phosphate-buffered saline and switched to serum-free DMEM. Adipocytes were treated (in triplicate) for 18 h with 10 µM SP-18904 or SP-18905. Concentrated stock solution of SP-18904 or SP-18905 was freshly prepared in DMSO and diluted into culture medium. The final concentration of DMSO was 0.2% (v/v), which was also included in basal conditions. After overnight (18 h) treatment, the culture medium was aspirated, and the monolayers were washed with Krebs-Ringer-HEPES buffer. To assess the effects of the compounds on basal glucose transport, 2-deoxy-D-glucose uptake (an indicator of glucose transport) was measured in the absence of insulin stimulation. To determine whether 18-h exposure to compounds potentiated the stimulatory effect of insulin, adipocytes were further treated with 0.5 nM insulin (a submaximal concentration) for 30 min at 37°C. Glucose transport assays were initiated by the addition of 2-deoxy-D-[³H]glucose (0.5 mCi/ml; 100 µM final concentrations) to each well followed by incubation for 10 min at 22°C. Assays were terminated by aspirating the media and rapidly washing the monolayer 2 times with ice-cold phosphate-buffered saline solution. Cell monolayers were solubilized in 0.1 N NaOH and transferred to scintillation vials, and radioactivity was determined by liquid scintillation counting. All data were corrected for nonspecific hexose uptake determined in parallel samples treated for 5 min with 200 µM cytochalasin B. Chemicals and medium used in the in vitro assay were from Sigma Chemical (St. Louis, MO).

Plasma glucose concentrations were determined using the Glucose Diagnostic Kit (Sigma 315), an enzyme colorimetric assay. Plasma insulin levels were determined by using the Rat Insulin RIA Kit from Linco Research Inc. (Cat. No. RI-13K; St. Charles, MO). Data are expressed as mean ± S.E.M., and one-way analysis of variance (ANOVA) with Fisher's PLSD post hoc test was used for assessing statistical significance of differences, with a p value of <.05 used as measure of significance.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

Plasma glucose concentrations measured at baseline and 3 h after daily treatment with various doses of SP-18904 for 4 days in ob/ob mice are shown in Fig. 2 (top). These data demonstrate that glucose concentrations were significantly reduced after 1 day of the oral administration of SP-18904 and remained lower on every day of dosing. Furthermore, a dose-response effect is evident. In contrast, plasma glucose concentration did not fall in vehicle-treated mice.

View larger version (43K): [in this window] [in a new window]   Fig. 2.   Dose-dependent effects of SP-18904 (top) and SP-18905 (bottom) on plasma glucose concentrations in ob/ob mice. There were eight mice in each group. *p < .05 (ANOVA, Fisher's PLSD post hoc test).

Somewhat similar data were seen in ob/ob mice treated with SP-18905 (Fig. 2, bottom), only in this instance data are only available on days 2 and 3 after initiation of treatment. Although the decline in plasma glucose concentration was statistically significant on days 2 and 3 in response to 50 and 100 mg/kg SP-18905, this compound seems to be somewhat less potent than SP-18904. As before, there was no change in plasma glucose concentration in the vehicle-treated mice.

Food intake of all the groups of mice shown in Fig. 2 ranged from a mean of 4.9 to 5.7 g/mouse/day, with no difference between the groups. Similarly, weight gain over the 4 days averaged ~1.0 g and was comparable in each of the groups of mice.

Plasma insulin concentrations 3 h after the last dose of SP-18904 (4 days of treatment) and SP-18905 (3 days of treatment) to ob/ob mice are seen in Fig. 3. It is apparent that the fall in plasma glucose concentration shown in Fig. 2 was associated with a decrease in plasma insulin after treatment with either SP-18904 (top) and SP-18905 (bottom). Furthermore, a dose-response effect is also obvious from insulin concentrations.

View larger version (43K): [in this window] [in a new window]   Fig. 3.   Dose-dependent effects of SP-18904 (top) and SP-18905 (bottom) on plasma insulin concentrations in ob/ob mice. There were eight mice in each group. *p < .05 (ANOVA, Fisher's PLSD post hoc test).

The antihyperglycemic effect of these compounds were evaluated in db/db mice, a second model that is hyperglycemic and hyperinsulinemic (~400 µU/ml), as well as in mice made insulin deficient (~10 µU/ml) with streptozotocin. These studies were performed after 2 days of treatment and indicated that the plasma glucose concentration of db/db mice fell significantly (p < .05) after oral administration of either 100 mg/kg SP-18904 (411 ± 22 mg/dl) SP-18905 (368 ± 18 mg/dl) compared with vehicle (523 ± 24 mg/dl)-treated mice. In contrast, plasma glucose concentrations in insulin-deficient mice did not decrease in response to treatment with either SP-18904 (683 ± 32 versus 657 ± 25 mg/dl) or SP-18905 (507 ± 17 versus 523 ± 14 mg/dl).

Because the chemical structure and the effects on plasma glucose and insulin concentrations of both compounds were similar, only SP-18905 was used to determine whether the compound could counteract insulin resistance in ob/ob mice. As shown in Fig. 4, SSPG concentrations (right) after the administration of 50 mg/kg SP-18905 for 4 days were significantly lower compared with vehicle-treated mice. Because the SSPI concentrations were similar in the two groups, these data indicate that insulin-mediated glucose disposal had increased in association with SP-18905 administration.

View larger version (41K): [in this window] [in a new window]   Fig. 4.   Effects of SP-18905 on SSPI and SSPG concentrations during the last 30 min of a 135-min infusion of insulin (20 mU/kg/min) and glucose (20 mg/kg/min) in ob/ob mice. There were eight and seven mice in the vehicle and treatment groups, respectively. *p < .05 (ANOVA, Fisher's PLSD post hoc test).

Both compounds were evaluated for their effects on 2-deoxy glucose transport in 3T3-L1 adipocytes. In the absence of insulin, neither SP-18904 nor SP-18905 (10 µM) increased glucose transport into adipocytes (Fig. 5). However, at the same concentrations, in the presence of a submaximal insulin concentration (5 nM), both SP-18904 and SP-18905 potentiated insulin-stimulated glucose uptake.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Effects of SP-18904 and SP-18905 on 2-deoxyglucose uptake in 3T3-L1 adipocytes with or without the presence of insulin. Results are the mean of three experiments, each experiment done in triplicate. *p < .01 compared with insulin-stimulated glucose uptake (ANOVA, Fisher's PLSD post hoc test).

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

To the best of our knowledge, there is no previous ethnomedical evidence that extracts of P. angolensis would be useful in the treatment of hyperglycemia. Our decision to explore this possibility was based on the observation of its use as an antifungal agent and the knowledge that uncontrolled hyperglycemia increases the risk of fungal infections. Based on this association, we evaluated the ability of crude extracts of P. angolensis to lower plasma glucose concentrations in animal models of type 2 diabetes and, if so, to initiate an in vivo based fractionation effort (Oubré et al., 1997; Luo et al., 1998a) to identify the compound or compounds responsible for this effect. We used ob/ob and db/db mice to identify and evaluate the antihyperglycemic compounds present in P. angolensis. They are well recognized animal models of type 2 diabetes (Shafrir, 1992) that are used extensively in drug discovery and evaluation of potential pharmacological approaches to treat the clinical syndrome (Chang et al., 1983; Shafrir, 1992). This effort was successful, as attested to by the results shown in Fig. 2 documenting the fact that two compounds isolated from P. angolensis, SP-18904 and SP-18905, were capable of significantly lowering plasma glucose concentrations when given orally to a mouse model of type 2 diabetes.

The two antihyperglycemic compounds isolated from P. angolensis are terpenoid-like quinones not previously identified. As such, they are both novel compounds, as well as newly recognized antihyperglycemic agents. It is apparent from the structures shown in Fig. 1 that both SP-18904 and SP-18905 are chemically distinct from the four classes of compounds currently approved to treat type 2 diabetes: sulfonylureas, biguanides, thiazolidinediones, and disaccharidase inhibitors. Consequently, the identification of these terpenoid-like quinones offers a new approach to the development of drugs that may be useful in the treatment of type 2 diabetes.

Although not the primary goal of this study, the results presented also provide some insight into the means by which SP-18904 and 18905 lower plasma glucose concentration. Specifically, the observation that the decline in plasma glucose concentration was associated with lower plasma insulin concentrations indicates that the compounds are not insulin secretagogues but seem to be enhancing the ability of insulin to stimulate glucose disposal. This possibility is further supported by the infusion studies in which treated mice had similar SSPI concentrations during the infusion but lower SSPG concentrations (Fig. 4). The fact that these compounds had little, if any, effect on plasma glucose concentrations in insulin-deficient mice provides further evidence that it is acting to enhance insulin-mediated glucose disposal. Furthermore, consistent with the improvement of insulin action in in vivo, both SP-18904 and SP-18905 were found to enhance insulin-stimulated glucose uptake by 3T3-L1 adipocytes (Fig. 5). These data suggest that SP-18904 and SP-18905 improve insulin action at the cellular level, the underlying mechanism of which awaits further study. Given the extensive evidence that insulin resistance is the basic defect in patients with type 2 diabetes (Reaven, 1995), it is not surprising that recent drug development for the treatment of type 2 diabetes has focused on agents that reduce insulin resistance (Saltrel and Olefsky, 1996; Imura, 1998). In this context, the fact that SP-18904 and SP-18905 appear to enhance insulin-mediated glucose disposal suggests that they represent a new and physiologically relevant approach to the treatment of type 2 diabetes.

In summary, SP-18904 and SP-18905 represent a new class of terpenoid-type quinones that have marked antidiabetic effects in mouse models of type 2 diabetes. The antidiabetic effects of this class of compound may rely on their ability to improve insulin-mediated glucose disposal. These characteristics define compounds of potential great use in the treatment of type 2 diabetes.