PHYTOTHERAPY RESEARCH, VOL. 12, 98–102 (1998)

Insulinotropic Activity of Tinospora crispa Extract: Effect on ß-cell Ca2‡ Handling Hamdan Noor* and Stephen J. H. Ashcroft Nuffield Department of Clinical Biochemistry, John Radcliffe Hospital, Oxford OX3 9DU, UK

The mechanism of insulinotropic action of Tinospora crispa was investigated in vitro using an insulinsecreting clonal ß-cell line, HIT-T15. The aqueous extract sensitizes the ß-cell to extracellular Ca2‡ and promotes intracellular Ca2‡ accumulation which in turn causes increased insulin release. The increase in cytosolic Ca2‡ concentration is due to stimulation of Ca2‡ uptake from the extracellular medium and inhibition of Ca2‡ efflux from the cytosol. That the mechanism of insulinotropic action of T. crispa is physiological suggests that the insulin secretagogue/s present in the extract could indeed be a potential source of a specific oral hypoglycaemic agent for the treatment of non-insulin-dependent diabetes mellitus. # 1998 John Wiley & Sons, Ltd. Phytother. Res. 12, 98–102, (1998) Keywords: Tinospora crispa; insulinotropic; Ca2‡; insulin secretion; ß-cell; diabetes mellitus

INTRODUCTION Treatment of diabetes mellitus with herbal remedies has been practised since ancient times (Ajgaonkar, 1979). Even after the discovery and use of insulin and oral hypoglycaemics (sulphonylureas and biguanides) as modern antidiabetic agents, a search for safer and more effective drugs of plant origin in the treatment of diabetes continues (Lin, 1992; Famuyiwa, 1993; Baker et al., 1995). In Malaysia, an aqueous extract derived from boiling the stem of a climbing plant, Tinospora crispa (family Menispermaceae) (Burkill, 1966) is taken orally to treat type 2 (non-insulin-dependent) diabetes (Gimlette et al., 1930). We have already verified in in vivo animal models (Noor and Ashcroft, 1989) as well as in in vitro animal and human models (Noor et al., 1989) that the aqueous extract indeed exhibited hypoglycaemic activity. We have presented evidence that the hypoglycaemic activity is probably due to the ability of the extract to stimulate ßcell insulin release. Although the insulinotropic effect of T. crispa observed in in vivo and in vitro models supports the anecdotal claims for its antidiabetic activity, further characterization of the extract with respect to its mechanism of action is necessary before a pharmacological role can be assigned. In the present study, we have evaluated T. crispa as a potential source of a specific oral hypoglycaemic agent by characterizing its insulinotropic activity using HIT-T15 cells as an insulin-secreting cell model (Ashcroft et al., 1986). Since Ca2‡ plays an essential role in the regulation of insulin release * Correspondence to: H. Noor, Biology Department, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor, Malaysia. Contract/grant sponsor: British Diabetic Association. Contract/grant sponsor: Government of Malaysia.

CCC 0951–418X/98/020098–05 $17.50 # 1998 John Wiley & Sons, Ltd.

(Wollheim et al., 1996), we performed experiments to determine the effect of T. crispa on Ca2‡ handling by the ß-cells.

MATERIALS AND METHODS Preparation of T. crispa Extract. The extract was prepared by the method previously described (Noor et al., 1989). Mature T. crispa stems obtained from the herb garden of the Universiti Putra Malaysia, Serdang, Malaysia were cut and air-dried, and ground into powder. It was then boiled and refluxed for 4 h, and centrifuged (40 min, 15 000  g, 4°C) to separate out the solid mass. The supernatant was filtered and freeze-dried for use in the experiments. Culture of Hit-T15 Cells. HIT-T15 cells which were routinely cultured in RPMI 1640 supplemented with 11mM glucose, antibiotics and fetal calf serum (10%, v/v) (Ashcroft et al., 1986) were passaged 2–4 days before each experiment and plated in 24-well Nunclon multiwell plates or tissue culture inserts at a density of 5  105 cells/well (or insert). Measurement of Insulin Secretion. Insulin secretion was measured as previously described (Ashcroft et al., 1986). Multiwells were seeded with 5  105 cells and insulin release measured after 4–5 days as follows. The culture medium was replaced with modified Krebs bicarbonate medium containing 5 mg/mL albumin. After 1 h this incubation medium was replaced with buffer containing zero glucose, 0.10 mg/mL T. crispa, 10 mM glucose or 0.10 mg/mL T. crispa. After a further hour incubation, an aliquot was removed and diluted in phosphate buffer and stored at ÿ20°C. Insulin release Accepted 9 October 1997

INSULINOTROPIC ACTIVITY OF T. CRISPA

was measured by radioimmunoassay (Ashcroft and Crossley, 1975). Measurement of Cytosolic Free Ca‡‡. Cytosolic Ca2‡ was measured using the Ca2‡ binding fluorescent indicator quin 2 as described by Hughes et al. (1989). HIT-T15 cells were detached from the culture flask, loaded with quin 2 (50 mM), centrifuged and resuspended in 2 mL Hepes-buffered Krebs medium. The cells were stored on ice at a density of 25  106 cells/mL prior to use. Fluorescence measurements were carried out in a Perkin Elmer LS5 luminescence spectrometer. HIT-cells (5  106) were preincubated at 37°C for 5 min with continuous stirring in medium containing zero glucose, 0.10 mg/mL T. crispa, 10 mM glucose or 10 mM glucose plus 0.1 mg/mL T. crispa. At the end of this time, the cells were quickly centrifuged, the supernatant aspirated and the pellet resuspended in 2 mL of fresh basal medium. The cell suspension was transferred into a cuvette and placed in a luminescence spectrophotometer which was equilibrated at 37°C. Fluorescence was recorded at 15 intervals starting exactly 30 s after the incubation medium was replaced. Cytosolic free Ca2‡ was calculated according to Rorsman and Abrahamson (1985). Measurement of 45Ca2‡ uptake. Ca2‡ uptake experiments were carried out using a non-wash double-isotope technique as described by Henquin and Lambert (1975). HIT-T15 cells were suspended in Hepes-Krebs buffer to a density of 107 cells per mL. The incubations were carried out in Beckman polyethylene tubes into which had been layered (from the bottom up) 25 mL of 6 M urea, 200 L of a mixture of dibutylphthalate and dinonylphthalate (10:3 v/v) and 50 mL of HIT cell suspension. The tubes were first pre-incubated for 30 min in a non-shaking water bath. The incubation period (2–60 min) was started by addition of 50 mL of warm (37°C) Hepes-Krebs medium containing 45CaCl2 (1.0 mCi) and 3H-sucrose (1.4 mCi) and test agents (zero glucose, 0.10 mg/mL T. crispa, 10 mM glucose or 10 mM glucose plus 0.1 mg/mL T. crispa) to the top layer of each tube. The incubations in all experiments was ended by a 20 s centrifugation of the tubes. This rapidly forced the cells through the phthalate mixture into the urea layer below, while leaving most extracellular fluid behind. The tubes were then dipped in liquid nitrogen and the bottom containing the cells in the urea layer and a small portion of the oil layer was cut off and placed in scintillation vials containing scintillation fluid. 45Ca and 3H were counted simultaneously in a scintillation counter. Calcium uptake was measured as 45 Ca taken up by the cells in excess of the extracellular 3 H-sucrose space for each tube. Measurement of 45Ca2‡ efflux. 45Ca2‡ efflux from HITT15 cells was measured according to the technique described by Malaisse et al. (1973) in a perifusion system (Noor et al., 1989). HIT-T15 cells in tissue culture inserts were labelled with the isotope by incubating for 60 min with 0.2 mL Hepes-Krebs medium containing 45CaCl2 (0.10 mCi, 2.5 mM) and 10 mM glucose. The cells were then perifused in parallel chambers with Ca2‡-free (containing 0.1 mM EGTA) basal medium without glucose for 60 min at a constant flow rate of 1 mL/min. The perifusion medium was then changed to one supplemented with 10 mM glucose, 10 mM glucose plus # 1998 John Wiley & Sons, Ltd.

99

0.10 mg/mL T. crispa or 0.1 mg/mL T. crispa. Stimulation lasted 30 min, thereafter basal conditions were restored for a further 16 min. For the control in the parallel chamber, the perifusion medium consisted of the basal medium throughout. Perifusate was collected every 2 min starting from min 45, and samples were analysed for radioactivity by liquid scintillation counting. The release of 45Ca2‡ from the prelabelled cells was expressed as percentage instantaneous fractional outflow rate (FOR) based on the radioactive content of the cells at the end of the perifusion period after they had been lysed with borate buffer.

RESULTS Effect of T. crispa on insulin release at different extracellular Ca2‡ concentrations The effect of glucose and T. crispa on HIT-cell insulin release rates was examined at different extracellular concentrations of Ca2‡, ranging from zero to 5.0 mM (Fig. 1). The data presented are relative to basal insulin release (123.2  19.1 mU/mL/h) at 2.5 mM extracellular Ca2‡. Raising the extracellular Ca2‡ concentration from zero to 5.0 mM in the absence of glucose or T. crispa did not significantly increase insulin secretion as analysed by ANOVA. However, on addition of T. crispa to the basal medium at a concentration of 0.10 mg/mL, there was a significant increase in insulin release when the concentration of extracellular Ca2‡ was raised to 0.5 mM. Below 0.5 mM extracellular Ca2‡ T. crispa had no significant effect on basal insulin release. Therefore, the threshold extracellular Ca2‡ concentration at which T. crispa stimulated insulin release under basal conditions was approximately 0.50 mM. Above this concentration, T.

Figure 1. Effect of T. crispa on insulin release under different extracellular Ca2‡ concentrations. Data are presented as means  SEM (n = 12) of insulin release relative to basal secretion at 2.5 mM external Ca2‡. HIT-cells were treated with zero glucose (&), zero glucose with 0.10 mg/mL T. crispa (&), 10 mM glucose without T. crispa (* ) or 10 mM glucose with 0.10 mg/mL T. crispa (*) at various extracellular Ca2‡ concentrations. Phytother. Res. 12, 98–102 (1998)

100

H. NOOR AND S. J. H. ASHCROFT

Figure 2. Effect of T. crispa on intracellular Ca2‡ accumulation. Cytosolic free Ca2‡ concentrations were measured at 15 s intervals after HIT-cells had been treated with zero glucose (&), zero glucose with 0.10 mg/mL T. crispa (&), 10 mM glucose without T. crispa (*) or 10 mM glucose with 0.10 mg/mL T. crispa (*).

Figure 3. Effect of T. crispa on 45Ca2‡-uptake. Calcium in¯ux into HIT-cells was measured as 45Ca2‡-uptake after the cells were treated with zero glucose (&), zero glucose with 0.10 mg/mL T. crispa (&), 10 mM glucose without T. crispa (*) or 10 mM glucose with 0.10 mg/mL T. crispa (*).

crispa significantly stimulated basal insulin secretion by 39–191%. Glucose (10 mM) induced an increase of 77–868% in insulin release as the extracellular Ca2‡ concentration was raised from 0.5 to 5.0 mM. Addition of T. crispa to the glucose containing medium resulted in potentiation of insulin release at every extracellular Ca2‡ concentration (0.25–5.0 mM) tested. The stimulation of insulin release over that induced by glucose alone ranged between 100%–536%. It is clear that T. crispa shifts the Ca2‡dependent insulin secretion curve to the left, suggesting that the ß-cell sensitivity to extracellular Ca2‡ is increased during incubation with the extract.

crispa in the high glucose medium during the initial 5min incubation significantly potentiated the glucose effect. Again, the level decreased with time.

Effect of T. crispa on intracellular Ca2‡ accumulation The experiment was carried out to verify if insulin secretion stimulated by T. crispa is due to increased intracellular Ca2‡ concentration. The concentration of cytosolic free Ca2‡ in the control group 15 s after the cells had been resuspended in fresh basal medium was 117  8.9 nM (Fig. 2). This level remained unaltered during the next 150 s (ANOVA, n = 9) indicating a steady state throughout the experiment. Incubation of the cells with 0.10 mg/mL T. crispa for 5 min resulted in a significantly higher ( p < 0.01) cytosolic free Ca2‡ level 15 s after resuspension in the basal medium compared with the controls. This level significantly decreased with time (ANOVA, p< 0.001, n = 7), suggesting an efflux of Ca2‡ on resuspension of the cells in basal medium subsequent to removal of T. crispa. Incubation with glucose resulted in a significant increase in cytosolic free Ca2‡ measured 15 s after resuspension in basal medium when compared with the non-stimulated cells. Moreover, cytosolic free Ca2‡ decreased with time, indicating an efflux of Ca2‡ into the glucose-free medium. Addition of 0.10 mg/mL T. # 1998 John Wiley & Sons, Ltd.

Effect of T. crispa on 45Ca2‡-uptake The experiment was carried out to determine if increased intracellular Ca2‡ accumulation is due to increased Ca2‡ uptake. The effect of glucose and T. crispa on 45Ca2‡ uptake by HIT cells was measured at different time points from 2 to 60 min. Figure 3 shows that there was no significant increase in 45Ca2‡ uptake with time when HIT cells were incubated for 2–60 min in basal medium in the absence of glucose or T. crispa (ANOVA, n = 12). However, when T. crispa (0.10 mg/mL) was added to the basal medium, there was a significant time-dependent increase in 45Ca2‡ uptake. This increase was significantly higher (57%–166%) than the control values at all of the time points. 45Ca2‡ uptake also increased when HIT cells were incubated in medium containing 10 mM glucose. Moreover, when 0.10 mg/mL was added in this incubation medium, 45Ca2‡ uptake was significantly potentiated by 31%–51% at 2,5 and 10 min. Effect of T. crispa on 45Ca2‡ efflux The experiment was carried out to determine if increased cytosolic free Ca2‡ brought about by T. crispa is also due to decreased rate of efflux. HIT-cells were simultaneously perifused in two chambers. In the first chamber, the cells were perifused with basal medium throughout whereas in the second, the cells were perifused with medium containing glucose and/or T. crispa. Figure 4 shows that when the cells were perifused with basal medium throughout, there was a progressive decrease in the efflux rate. In contrast, stimulation with 10 mM glucose induced a sudden 20.6% inhibition of 45Ca2‡ efflux rate. Addition of 0.10 mg/mL T. crispa to the Phytother. Res. 12, 98–102 (1998)

INSULINOTROPIC ACTIVITY OF T. CRISPA

101

Figure 4. Effect of T. crispa on 45Ca2‡-ef¯ux. Data are expressed as instantaneous fractional out¯ow rate (FOR). The effects of the stimuli (glucose and T. crispa) were analysed by calculating the percentage difference between the average FOR during the time interval 68±90 min and the average FOR during the time interval 46±60 min. The 45Ca-labelled HIT-cells were perifused in parallel chambers with Ca2‡-free basal medium without glucose for 60 min. The perifusion medium was then changed (left arrow in each panel) to one supplemented with 10 mM glucose (panel A, *), 10mM glucose plus 0.10 mg/mL T. crispa (panel B, *) and 0.10 mg/ mL T. crispa (panel C, *). Basal conditions were restored (right arrow in each panel) for a further 16 min. For the control (*) in the parallel chamber, the perifusion medium consisted of the basal medium throughout.

perifusion medium induced a further 38.3% inhibition of the efflux rate. T. crispa alone induced only 6.3% reduction in fractional outflow rate. This rate remained low throughout the duration of the stimulus. Upon withdrawal of the stimulus, the efflux rate increase to a level above the initial basal efflux and then started to decrease again.

DISCUSSION The antihyperglycaemic and insulinotropic activities of T. crispa observed earlier (Noor and Ashcroft, 1989; Noor et al., 1989) warrant further characterization of the extract with respect to its mechanism of action in an effort to obtain an alternative hypoglycaemic agent. To perform a valid and meaningful characterization of a potential insulinotropic agent using in vitro models, two important factors have to be considered. First, the crude aqueous extract is probably a complex mixture that may contain established insulin secretagogues; an apparent increase in insulin release after incubation with the extract might be attributable to these secretagogues. Second, the extract may contain cytotoxic substances that might damage ß-cell membrane and cause unphysiological responses. These problems have previously been addressed (Noor et al, 1989); T. crispa-evoked insulin release in all in vitro experiments was not due to the presence of established secretagogues, and there is no evidence to suggest that T. crispa is cytotoxic, causing unphysiological responses. Therefore, stimulation of ßcell insulin secretion by T. crispa can safely be assumed to be attributable to novel secretagogue/s present in the extract. Since Ca2‡ plays a central role in the regulation of insulin secretion (Wollheim et al., 1996), the experiments reported here were designed to determine the effect of T. crispa on Ca2‡ handling by the ß-cells. First, T. crispa was found to stimulate insulin secretion and also potentiate glucose-induced insulin release when the extracellular Ca2‡ concentrations were increased above # 1998 John Wiley & Sons, Ltd.

threshold. The extract alone (without glucose) stimulated insulin release by 39%–191% when the extracellular Ca2‡ concentrations were raised above 0.5 mM. Glucosestimulated insulin secretion also increased as extracellular Ca2‡ concentrations were increased. The threshold for glucose-stimulated insulin secretion (0.5 mM) is in agreement with Hughes et al., (1989). Addition of T. crispa into the extracellular medium augmented this effect by 100%–536% provided the extracellular Ca2‡ level was above 0.25 mM. This observation is critical because it implies that T. crispa-stimulated insulin secretion is physiological since there is no stimulation of insulin release without adequate extracellular Ca2‡. Moreover, the Ca2‡ dose-dependent insulin secretion curve shifts to the left in the presence of T. crispa. This implies that the extract either sensitizes the secretory mechanism to Ca2‡ or it increases the efficiency of the plasma membrane Ca2‡-transport system. The increase in ß-cell sensitivity towards extracellular Ca2‡ was confirmed in the next experiment, where T. crispa was found to cause an increase in intracellular Ca2‡ accumulation. This is associated with the stimulatory effect of T. crispa on insulin secretion under basal as well as glucose-stimulated conditions. An increase in cytosolic Ca2‡ concentration may be due to an increase in Ca2‡-uptake into the ß-cell or a decrease in Ca2‡-efflux into the extracellular medium. In order to elucidate the mechanism whereby T. crispa promotes intracellular Ca2‡ accumulation, 45Ca2‡ influx and efflux studies were carried out. The observation that cytosolic free Ca2‡ concentration increased when the ß-cells were stimulated with T. crispa, coupled with the fact that T. crispa-evoked insulin release is dependent on extracellular Ca2‡ concentration, implies that the extract enhances Ca2‡ uptake. This is substantiated by the results of the studies performed to directly measure uptake using 45Ca2‡, where T. crispa indeed stimulates the influx of 45Ca2‡ from the extracellular medium. In addition to increased influx, a decrease in Ca2‡ efflux would also elevate the cytosolic free Ca2‡ concentration, provided its sequestration into the orgaPhytother. Res. 12, 98–102 (1998)

102

H. NOOR AND S. J. H. ASHCROFT

nelles is not altered (Wollheim and Sharp, 1981). That T. crispa inhibits Ca2‡ efflux measured in the absence of extracellular Ca2‡ provides another mechanism for its ability to increase cytosolic Ca2‡ concentration, and consequently stimulates insulin release. In conclusion, the results of this study suggest that the insulinotropic effect of T. crispa is due to insulin secretagogue/s present in the extract which causes an increase in cytosolic Ca2‡ concentration by promoting Ca2‡-uptake from the extracellular medium and inhibiting Ca2‡-efflux from the cytosol. That the mechanism of insulinotropic action of T. crispa is physiological

suggests that the insulin secretagogue/s present in the extract could indeed be a potential source of a specific oral hypoglycaemic agent for the treatment of noninsulin-dependent diabetes mellitus.

Acknowledgement Financial support for these studies was provided by the British Diabetic Association. H. Noor gratefully acknowledges support from the Government of Malaysia.

REFERENCES Ajgaonkar, S. S. (1979). Herbal drugs in the treatment of diabetes mellitus. A review. Int. Diabetes Fed. Bull. 24, 1019. Ashcroft, S. J. H., and Crossley, J. R. (1975). Effects of glucose, N-acetylglucosamine, glyceraldehyde and other sugars on insulin release in vivo. Diabetologia 11, 279284. Ashcroft, S. J. H., Hammonds, P., and Harrison, D. E. (1986). Insulin secretory responses of a clonal cell line of simian virus 40-transformed û-cells. Diabetologia 29, 727-733. Baker, J. T., Borris, R. P., and Carte, B. et al. (1995). Natural product drug discovery and development: New perspective in international collaboration. J. Nat. Prod. 58:(9), 1325-1357. Burkill, L. H. (1966). A Dictionary of the Economic Products of the Malay Peninsula: vol II, pp. 2202-2205. Ministry of Agriculture and Cooperations, Kuala Lumpur. Famuyiwa, O. O. (1993). The ef®cacy of traditional medicine in the management of diabetes in Southwestern Nigeria. Afr. J. Med. Sci. 22:(1), 31-37. Gimlette, J. D., Burkill, L. H., and Ismail Munshi, (1930). The medicinal book of Malayan medicine. Gard. Bull. Str. Settle. 6, 324-474. Henquin, J-C., and Lambert, A. C. (1975). Cobalt inhibition of insulin secretion and calcium uptake by isolated rat islets. Am. J. Physiol. 228, 1669-1677. Hughes, S. J., Chalk, J. G., and Ashcroft, S. J. H. (1989). Effect

# 1998 John Wiley & Sons, Ltd.

of secretagogues on cytosolic free Ca2‡ and insulin release at different extracellular Ca2‡ concentrations in the hamster clonal û-cell line HIT-T15. Mol. Cell. Endocrinol. 65, 35-41. Lin, C. C. (1992). Crude drugs used for the treatment of diabetes mellitus in Taiwan. Am. J. Chin. Med. 20:(3-4), 269-279. Malaisse, W. J., Brisson, G. R., and Baird, L. E. (1973). Stimulus-secretion coupling of glucose-stimulated insulin release. X. Effect of glucose on45Ca ef¯ux from perifused islets. Am. J. Physiol, 224, 389-394. Noor, H., and Ashcroft, S. J. (1989). Antidiabetic effect of Tinospora crispa in rats. J. Ethnopharmacol. 27:(1-2), 149161. Noor, H., Hammonds, P., Sutton, R., and Ashcroft, S. J. (1989). The hypoglycaemic and insulinotropic activity of Tinospora crispa: studies with human and rat islets and HIT-T15 cells. Diabetologia 32:(6), 354-359. Rorsman, P., and Abrahamsson, H. (1985). Cyclic AMP potentiates glucose-induced insulin release from mouse pancreatic islets without increasing cytosolic free Ca2‡. Acta Physiol. Scand. 125, 639-647. Wollheim, C. B., Lang, J., and Regazzi, R. (1996). The exocytotic process of insulin secretion and its regulation by Ca2‡ and G-proteins. Diabetes Rev. 4:(3), 276-297. Wollheim, C. B., and Sharp, G. W. G. (1981). Regulation of insulin release by calcium. Physiol. Rev. 61, 914-973.

Phytother. Res. 12, 98–102 (1998)