The Journal of Nutrition Vol. 128 No. 12 December 1998, pp. 2319-2323

Antihyperglycemic Actions of Eucalyptus globulus (Eucalyptus) are Associated with Pancreatic and Extra-Pancreatic Effects in Mice^1,2

Alison M. Gray³ and Peter R. Flatt

School of Biomedical Sciences, University of Ulster, Coleraine, Northern Ireland, BT52 1SA, UK

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Eucalyptus globulus (eucalyptus) is used as a traditional treatment for diabetes. In this study, incorporation of eucalyptus in the diet (62.5 g/kg) and drinking water (2.5 g/L) reduced the hyperglycemia and associated weight loss of streptozotocin-treated mice. An aqueous extract of eucalyptus (AEE) (0.5 g/L) enhanced 2-deoxy-glucose transport by 50%, glucose oxidation by 60% and incorporation of glucose into glycogen by 90% in mouse abdominal muscle. In acute, 20 min incubations, 0.25-0.5 g AEE/L evoked a stepwise 70-160% enhancement of insulin secretion from the clonal pancreatic -cell line (BRIN-BD11). The stimulatory effect of 0.5 g/L AEE was unaltered by the presence of 400 µmol diazoxide/L and prior exposure to AEE did not alter subsequent insulin secretory response to L-alanine, thereby negating adetrimental effect on cell viability. The effect of AEE was not potentiated by glucose or demonstrable in cells exposed to a depolarizing concentration of KCl. Further study of the insulin-releasing effects of AEE revealed the activity to be heat stable, acetone insoluble, stable to acid, but abolished by exposure to alkali. Sequential extraction with solvents revealed activity in both methanol and water fractions, indicating the presence of more than one biologically active extract constituent. These data indicate that Eucalyptus globulus represents an effective antihyperglycemic dietary adjunct for the treatment of diabetes and a potential source for discovery of new orally active agent(s) for future therapy.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

Plants have long been considered as sources of medicinal agents for the treatment of disease. Although present therapy for diabetes mellitus relies heavily on an arsenal of drugs developed since the introduction of insulin (Bailey and Flatt 1990), prior to 1922 diabetes therapy revolved around dietary measures including the use of traditional antihyperglycemic plants (Bailey and Day 1989, Gray and Flatt 1997a). Additionally, whereas many traditional plant treatments for diabetes were described (Bailey and Day 1989, Gray and Flatt 1997a), few received scientific or medical scrutiny. The World Health Organization has recommended, accordingly, that traditional plant treatments for diabetes warrant further evaluation (World Health Organization Expert Committee on Diabetes Mellitus 1980). Such studies might reveal effective dietary adjuncts for the treatment of the disease or the discovery of natural products for developing new antidiabetic drugs.

View this table: [in this window] [in a new window]   Table 2. Effect of aqueous extract of eucalyptus (AEE) on glucose uptake and metabolism by isolated mouse abdomen muscle during incubations in the absence and presence of 108 mol/L insulin.1

View larger version (11K): [in this window] [in a new window]   Fig 1. Effects of aqueous extract of eucalyptus (AEE) on clonal rat insulin-secreting cells. Values are means ± SE, n = 6. *P < 0.05 compared with control incubations without extract.

Eucalyptus globulus (eucalyptus; blue gum tree), although indigenous to Tasmania, is traditionally used to treat diabetes in South America and Africa (Duke 1985, Lewis 1949). The medicinal part of the plant are the leaves from which tea is made (Chiej 1988). The use of Eucalyptus globulus leaves in the treatment of diabetes mellitus was first advocated by Faulds (1902). Recent studies in streptozotocin-diabetic mice confirmed the antihyperglycemic efficacy of Eucalyptus globulus (Swanston-Flatt et al. 1990).

The present study was undertaken to confirm the antidiabetic properties of Eucalyptus globulus and to investigate the mechanism(s) responsible for the reported anti-hyperglycemic effects. For the latter purpose, glucose transport and metabolism in mouse (abdominal) skeletal muscle and insulin secretion by clonal BRIN-BD11 cells were determined during in vitro incubations with an aqueous extract of Eucalyptus globulus (AEE).

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Plant material.  Dried eucalyptus leaves were obtained from a commercial source (The Health Food Center, Birmingham, UK). Leaves were homogenized to a fine powder and stored at room temperature (20 ± 2°C) until use. For animal diets eucalyptus was incorporated into powdered mouse diet, thoroughly mixed, distilled water added and mixed to a stiff paste. The mixture was then pelleted and dried at 45°C until dry. Control diet was prepared by the same method to ensure there were no end differences in vitamin and mineral content as a result of the drying process. Aqueous extract of eucalyptus (AEE) was prepared by 15 min decoction of the powdered material at 250 g/L. In brief, 1g of powdered material was placed in 40 mL of cold (distilled) water, brought to the boil, then removed from the heat source and allowed to infuse for 15 min. This suspension was filtered (Whatman no. 1) and the volume readjusted with distilled water to 40 mL. For in vivo studies, 10 mL aliquots of AEE were stored at -20°C until use when they were diluted 10-fold with tap water (25 g/L). For in vitro studies, 1 mL aliquots of extract were brought to dryness under vacuum (Savant speedvac, Savant Instrumentation Incorp., Framingdale, NY) and stored at -20°C until use. AEE doses are expressed as total plant material weight (rather than dry residue) per L of water. To account for possible differences in potency, test and control incubations within a single experiment using isolated muscle or insulin-secreting cells were always conducted using the same batch of AEE. This allowed for variation in potency of different batches of AEE apparent in the same insulin release experiments.

Animal studies.  Male mice derived from a colony maintained at Aston University, Birmingham, UK (Flatt and Bailey 1981) were used at 21-24 wk of age. Groups of mice were housed in an air conditioned room at 22 ± 2°C with a lighting schedule of 12 h light (0800-2000 h) and 12 h dark. Animals had free access to a standard pellet diet (Mouse breeding diet, Pillsbury Ltd., Birmingham, UK) and tap water. The overall nutrient composition of the diet was 36.2% carbohydrate, 20.9% protein, 4.4% fat and 38.5% fiber with added vitamins and minerals and with a metabolizable energy content of 1.18 MJ/100g. The experimental procedure for in vivo studies was similar to that previously described, namely three groups: (1) control mice receiving unsupplemented diet and drinking water, (2) streptozotocin-treated mice receiving unsupplemented diet and drinking water and (3) streptozotocin-treated mice receiving eucalyptus supplemented diet and drinking water (Gray and Flatt 1997b). Eucalyptus (62.5 g/kg) was incorporated into the diet and drinking water (2.5 g/L) of mice 5 d before and after intraperitoneal administration of streptozotocin (STZ; Sigma Chemical Co., Poole, Dorset, UK) at 200 mg/kg body weight in 0.1 mol/L sodium citrate buffer (pH 4.5). Daily measures of body weight, food intake and fluid intake were made. Postprandial blood samples obtained from the cut tail tip of conscious mice were collected at the same time (9000-1000 h) on Days 12 and 20 for plasma glucose analysis (Stevens 1971). Groups of normal mice and STZ-treated mice with free access to unsupplemented diet and drinking water were used as controls. All animal studies were performed in accordance with the Animals (Scientific Procedures) Act, 1996.

Glucose transport and glucose metabolism in vitro.  Recently weaned postprandial male mice (3-5 wk) were killed by cervical dislocation and pieces of abdominal muscle (~10-20 mg) were prepared. To replicate by mouse, strips of muscle from each mouse were designated to each of the four treatment groups. Incubations were performed using Krebs-Ringer bicarbonate buffer supplemented with 2 g insulin-free bovine serum albumin /L (KRB-BSA; 118 mmol NaCl,/L 25 mmol NaHCO₃/L, 5 mmol KCl/L, 1.28 mmol CaCl₂/L, 1.18 mmol MgSO₄/L, 1.17 mmol KH₂PO₄/L).

Glucose uptake was determined as described previously (Gray and Flatt 1997b). In brief, muscle pieces were incubated at 30°C for 30 min in 1 mL KRB-BSA supplemented with 2 mmol sodium pyruvate/L , 3.7 Bq 2-deoxy-D-[1-³H]glucose/L, 0.37 Bq L-[1-¹⁴C]glucose/L in the presence and absence of 0.01 µmol human insulin/L and 0.5 g AEE/L (2 [times] 2 factorial design). After incubation, tissue was hydrolyzed using 1 mol NaOH/L and counted for ³H and ¹⁴C radioactivity. The extracellular fluid volume of the muscle was determined from the amount of the nontransported L-[1-¹⁴C]glucose, and this was taken into account in the calculation of tissue 2-deoxy-D-[1-³H]glucose uptake, expressed as nBq/(mg wet muscle · h).

Oxidative glucose metabolism to CO₂ and incorporation of glucose into glycogen were determined as described previously (Gray and Flatt 1997b). In brief, muscles were incubated at 37°C for 60 min in 1 mL KRB-BSA supplemented with 10 mmol glucose/L, 18.5 Bq D-[U-¹⁴C]glucose/L in the presence and absence of 0.01 µmol human insulin/L and 0.5 g AEE /L (2 × 2 factorial design). Following incubation, carbon dioxide was captured onto a NaOH saturated filter paper and muscles were removed for glycogen analysis (Gray and Flatt 1997b). ¹⁴C radioactivity of the filter paper was counted and CO₂ production was expressed as nmol CO₂/(mg wet muscle · h). The incorporation of glucose into glycogen was expressed as nmol glucose equivalents/(mg wet muscle · h).

Insulin secretion in vitro.  BRIN-BD11 cells, produced by electrofusion of immortal RINm5F cell with New England Deaconess Hospital rat pancreatic -cell, were used to evaluate insulin secretion (Gray and Flatt 1997b, McClenaghan et al. 1996). Cells were seeded at a concentration of 0.2 × 10⁶ cells/well in 24-well plates (Falcon, New Jersey) cultured in RPMI-1640 containing 11.1 mmol glucose/L, 10% fetal calf serum and antibiotics (50,000 IU penicillin-streptomycin/L) to allow attachment overnight prior to acute tests. Cells were washed thrice with Krebs-Ringer bicarbonate buffer (KRB; 115 mmol NaCl/L, 4.7 mmol KCl/L, 1.28 mmol CaCl₂/L, 1.2 mmol KH₂PO₄/L, 1.2 mmol MgSO₄/L, 24 mmol NaHCO₃/L, 10 mmol hepes free acid/L, 1 g bovine serum albumin/L, 1.1 mmol glucose/L; pH 7.4) and preincubated for 40 min at 37°C. Unless otherwise stated, cells were then incubated for 20 min with 1 mL KRB at 1.1 mmol glucose/L in the absence and presence of AEE, diazoxide (an established opener of -cell K⁺-ATP channels) and other test agents as indicated. Following incubation, aliquots were removed from each well and stored at 20°C for insulin assay (Flatt and Bailey 1981). Cell viability following incubations was evaluated by modified neutral red assay (Hunt et al. 1987). Following 20 min incubation with test agent, cells were washed thrice (as previously described) and incubated for 2 h with 1 mL neutral red solution (50 µg neutral red dissolved in 50 µL DMSO and made up to 200 mL with KRB). After washing (as described above), 1 mL 49:50:1 distilled water:ethanol:glacial acetic acid was added and plates were gently agitated for 15 min. Absorbance of each well was read at 540 nm and mean ± SEM calculated. Results were expressed as a percentage of control (incubations performed in the absence of test agent) giving percentage cell viability after 20 min exposure to test agent.

To investigate the importance of heat during extract preparation, aqueous extracts of eucalyptus were prepared by the normal method of decoction (AEE) or by cold infusion (cold extract; plant material placed in cold water, allowed to stand for 15 min, then filtered as before). Modified aqueous extract was freshly reconstituted in KRB and the effect on insulin secretion evaluated at a concentration equivalent to 0.5 g/L compared with AEE.

To evaluate the nature of the insulin-releasing component(s) AEE was subjected to heat, overnight dialysis, acid-alkali or acetone treatment. Heat, AEE was boiled for 1 h immediately after preparation. Dialysis, AEE was dialyzed overnight (Spectra/Por molecular weight cut-off 2000 Da; Spectrum, Los Angeles) against Millipore water at 4[deg]C. Acid-alkali treatment, aliquots of AEE were added to 5 mmol HCl /Lor 5 mol NaOH/L to produce 0.1 mol HCl/L or 0.1 mol NaOH/L, respectively, allowed to stand at room temperature overnight, the neutralized. Acetone treatment, 1 mL of AEE (0.5 mg/mL) was added to 10 mL ice-cold acetone, allowed to stand for 30 min and centrifuged (800 x g, 5 min) to obtain acetone soluble and acetone insoluble fractions. Aliquots of untreated AEE and modified aqueous extracts were dried under vacuum. All modified aqueous extracts were freshly reconstituted in KRB and the effects on insulin secretion at a concentration equivalent to 0.5 g/L compared with AEE.

In another series of experiments, eucalyptus leaves were subjected to sequential extraction by increasingly polar solvents. Twenty-five hundredths g of plant material was placed in 5 mL of hexane, agitated for 15 min, centrifuged (950 × g, 5 min). The precipitate was dried under vacuum and extracted with a another 5 mL of hexane and centrifuged as before. The extraction supernatants were pooled, filtered (Whatman no. 1) and the volume adjusted with hexane to 10 mL. The extraction precipitate (dried under vacuum) was subsequently extracted (as above) with 2 × 5 mL volumes of ethylacetate, then methanol and finally with water. All extract fractions were freshly reconstituted in KRB and the effects on insulin secretion at a concentration equivalent to 0.5 g/L compared with AEE.

Statistical analyses.  Data were evaluated using Student`s unpaired t test, one-way or two-way ANOVA where appropriate. Groups were considered to be significantly different if P < 0.05. When a significant F value was obtained for ANOVA, the differences between all pairs were tested using Student-Newman-Keul's multiple comparisons test. If the SD was significantly different (Bartlett`s test for homogeneity of variances) data were transformed (log₁₀[x]).

	RESULTS

Abstract Introduction Methods Results Discussion References

Studies in vivo.  Compared with normal mice, mice administered STZ showed hyperglycemia (P < 0.01, Days 12 and 20), a significant (P < 0.05) weight loss by Day 12 and both polydipsia and hyperphagia by the end of the study period (Table 1). Administering eucalyptus reduced the weight loss (P < 0.05, Day 20) and polydipsia (P < 0.001, Day 20) associated with STZ treatment (Table 1). Plasma glucose concentrations in STZ-treated mice receiving eucalyptus were significantly lower (P < 0.05) than unsupplemented STZ-treated mice by Day 12 (Table 1) and were not different from those of normal mice (Table 1).

Glucose transport and glucose metabolism in vitro.  AEE (0.5 g/L) increased glucose uptake by 50%, ¹⁴CO₂ production by 60% and glycogenesis by 90% during incubations without insulin, but did not significantly alter the stimulatory effect of 0.01 µmol insulin/L (Table 2).

Insulin secretion in vitro.  AEE (0.25-0.5 g/L) exerted a dose-dependent stimulatory effect on insulin secretion from BRIN-BD11 cells at 1.1 mmol glucose/L (Fig. 1). At concentrations of 1 g AEE/L and above, cell viability during the test period was significantly diminished (P < 0.01) as evaluated by modified neutral red assay (Hunt et al. 1987; data not shown). Such an effect was not evident at concentrations of 0.5 g/L and below. Consistent with the lack of toxicity at lower concentrations, prior exposure of BRIN-BD11 cells to 0.5 g AEE/L for 20 min did not alter the subsequent 100-150% increase in insulin secretory response to 10 mmol L-alanine/L (data not shown).

To gain a possible insight into the mechanism underlying the action of AEE, the effects of conditions known to modulate -cell function were tested. As shown in Table 3, the insulin-releasing effects of AEE were not glucose dependent, inhibited by diazoxide, or demonstrable in cells depolarized by combination of 16.4 and 25 mmol KCl/L (Table 3).

Prolonged exposure to heat or dialysis did not significantly alter the insulin-releasing activity of AEE (Table 4). Acid-exposed AEE had significantly greater insulin-releasing effects on BRIN-BD11 cells than did alkaline exposed AEE (Table 4). Insulin-releasing activity was retained in the acetone-insoluble fraction of AEE, whereas the acetone-soluble fraction did not significantly affect insulin secretion (Table 4). Only the methanol and water fractions, produced by sequential extraction of eucalyptus leaves, had significant insulin-releasing effects (0.31 ± 0.02 and 0.43 ± 0.05 pmol/(10^-6 cells, 20 min), respectively; n = 6) as compared with basal insulin release (0.21 ± 0.02 pmol/(10^-6 cells, 20 min); n = 6, P < 0.01) recorded in the absence of eucalyptus. Hexane and ethylacetate fractions failed to alter basal insulin secretion by BRIN-BD11 cells.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Eucalyptus was reported in early studies to be antihyperglycemic in alloxan diabetic rabbits and not to affect blood glucose concentrations in normal rats (Bever and Zahnd 1979). Recently, chronic dietary administration of eucalyptus was shown to ameliorate loss of body weight and polydipsia and to reduce hyperglycemia of streptozotocin diabetic mice (Swanston-Flatt et al. 1990). Because glucose concentrations were held down even when insulin concentrations were very low (Swanston-Flatt et al. 1990), eucalyptus is likely to act by modulating insulin secretion and/or insulin action. In the present study, the antihyperglycemic actions of eucalyptus have been confirmed in diabetic mice. Moreover, novel extra-pancreatic and pancreatic effects of eucalyptus extract that may contribute to glucose lowering activitywere demonstrated. The possibility also exists that dietary administration of eucalyptus was associated with the protection or regeneration of pancreatic -cells following exposure to streptozotocin, but this aspect requires further study.

Using an isolated mouse skeletal muscle preparation, aqueous extract of eucalyptus enhanced glucose transport and glucose metabolism in a similar magnitude to 0.01 µmol insulin/L. Although this effect was observed in the absence of added insulin, it does not rule out possible involvement of insulin already bound to its receptor within the muscle preparation. However, the lack of potentiation in the presence of a high concentration of insulin suggests that the active constituent(s) of AEE is likely to act, at least terminally, by mechanisms that are utilized by insulin rather than entirely different pathways. The effect of AEE on glucose uptake differed from that of metformin, which exerts its effects on glucose transport via insulin-mediated enhanced peripheral glucose uptake (Bailey and Puah 1986, Prager et al. 1986).

Studies using clonal pancreatic -cells showed that AEE exerted a dose-dependent stimulatory effect on insulin secretion, with a maximal response at 0.5 g/L. This stimulatory action was not glucose dependent as indicated by coincubation at 1.1 and 16.7 mmol glucose/L. Prior exposure to 0.5 g AEE/L did not diminish the subsequent insulin secretory response to the established amino acid stimulator L-alanine, thereby negating, in these experiments, a possible deleterious effect of AEE on cell viability. The nontoxic nature of this concentration of AEE was also confirmed by toxicity testing using the modified neutral red assay.

Preliminary studies to evaluate the possible mechanisms underlying the insulin-releasing action of AEE argue against an action similar to sulphonylurea drugs currently used for diabetes therapy (Bailey and Flatt 1995). These agents act by binding to sulphonylurea receptors, resulting in closure of plasma membrane K⁺-ATP channels, depolarization of membranes, opening of voltage-dependent calcium channels and elevation of intracellular Ca²⁺ (Nelson et al. 1992). Diazoxide by holding open K⁺-ATP channels inhibited the stimulatory action of sulphonylureas (Dunne et al. 1994), but failed to diminish, in the present study, the insulin-releasing action of eucalyptus. Furthermore, whereas sulphonylureas exhibit additional intracellular stimulatory effects in chemically depolarized cells (Eliasson et al. 1996, Flatt et al. 1994), AEE did not stimulate insulin release in cells exposed to 25 mmol KCl/L . Possible actions of eucalyptus may include enhancement of -cell glucose metabolism or activation of enzyme systems generating cyclic AMP or phospholipid derived messengers, but further studies are required to address such issues.

Leaves from Eucalyptus globulus are reported to contain a high content of eucalyptol (cineol) together with rutin, terpineol, sesquiterpene, alcohols, aliphatic aldehydes, isoamyl alcohol, ethanol, terpenes and tannins (Duke 1985). However, the chemical nature of the antihyperglycemic constituent(s) of eucalyptus remain to be established. In terms of insulin-releasing action, the present study indicates that the active principal(s) is heat stable, acetone insoluble and unaffected by an acidic environment. AEE activity is unlikely to be due to smaller molecules (<2000 Da) or ions as evidenced by retention of more than 85% insulin-releasing activity following dialysis. Sequential solvent extractions point towards a cumulative effect of more than one active constituent, likely to be polar in nature.

In conclusion, the antihyperglycemic action of eucalyptus is associated with the stimulation of insulin secretion and enhancement of muscle glucose uptake and metabolism, possibly reflecting the effect of more than one active constituent. Eucalyptus globulus, therefore, represents an effective antihyperglycemic dietary adjunct for the treatment of diabetes and a potential source for discovery of new orally active agent(s) for future diabetes therapy.

	FOOTNOTES

Manuscript received 8 April 1998. Initial reviews completed 5 June 1998. Revision accepted 11 August 1998.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Bailey C. J., Day C. Traditional treatments for diabetes. Diabetes Care 1989; 12:553-564[Abstract]
• Bailey, C. J. & Flatt, P. R., eds. (1990) New Antidiabetic Drugs. Smith-Gordon, London, UK.
• Bailey, C. J. & Flatt, P. R. (1995) Development of Antidiabetic Drugs. In: Drugs, Diet and Disease. Vol. 2. Mechanistic Approaches to Diabetes. (Ioannides, C. & Flatt, P. R., eds.), pp. 279-326, Ellis Horwood Ltd., Chicester, UK.
• Bailey C. J., Puah J. A. Effect of metformin on glucose metabolism in mouse soleus muscle. Diabetes Metab. 1986; 12:212-218
• Bever B. O., Zahnd G. R. Plants with oral hypoglycemic action. Q. J. Crude Drug Res. 1979; 17:139-196
• Chiej, R. (1988) The MacDonald Encyclopedia of Medicinal Plants. MacDonald & Co. Ltd., London, UK.
• Duke, J. A. (1985) CRC Handbook of Medicinal Herbs. CRC Press, Florida, US.
• Dunne, M. J., Harding, E. A., Jaggar, J. H., Ayton, B. J. & Squires, P. E. (1994) Endogenous and chemical activators of ATP-regulated potassium channels in insulin-secreting cells: possible mechanisms and physiological significance. In: Frontiers of Insulin Secretion and Pancreatic -Cell Research (Flatt, P. R. & Lenzen, S., eds.), pp. 153-159. Smith-Gordon, London, UK.
• Eliasson L., Renström E., Ammälä C., Berggren P. O., Bertorello A. M., Bokvist K., Chibalin A., Deeney , J. T., Flatt P. R., Gabel J., Gromada J., Larsson O., Lindström P., Rhodes C. J., Rorsman P. PKC-Dependent stimulation of exocytosis by sulfonylureas in pancreatic -cells. Science 1996; 271:813-815[Abstract]
• Faulds G. Eucalyptus in the treatment of diabetes. Glasgow Med. J. 1902; 57:342-348
• Flatt P. R., Bailey C. J. Abnormal plasma glucose and insulin responses in heterozygous lean (ob/+) mice. Diabetologia 1981; 20:573-577[Medline]
• Flatt P. R., Shibier O., Szecowka J., Berggren P. O. New perspectives on the actions of sulphonylureas and hyperglycemic sulphonamides on the pancreatic -cell. Diabetes Metab. 1994; 20:157-162
• Gray A. M., Flatt P. R. Nature's own pharmacy: the diabetes perspective. Proceedings of the Nutrition Society 1997a; 56:507-517[Medline]
• Gray A. M., Flatt P. R. Pancreatic and extra-pancreatic effects of the traditional anti-diabetic plant, Medicago sativa (alfalfa). Brit. J. Nutr. 1997b; 78:325-334[Medline]
• Hunt S. M., Chrzanowska C., Barnett C. R., Brand H. N., Fawell J. K. A comparison of in vitro cytotoxicity assays and their application to water samples. Alternatives to Laboratory Animals 1987; 15:20-29
• Lewis J. J. Diabetes and the insulin administration problem. Physiol. Rev. 1949; 29:75-90[Free Full Text]
• McClenaghan N. H., Barnet C. R., Ah-Sing E., Abdelwahab, Y.H.A. O`Harte, F.P.M., Yoon T. W., Swanston-Flatt S. K., Flatt P. R. Characterization of a novel glucose-responsive insulin-secreting cell line, BRIN-BD11, produced by electrofusion. Diabetes 1996; 45:1132-1140[Abstract]
• Nelson, D. A., Aguilar-Bryan, L., Raef, H. & Boyd, A. E. (1992) Molecular mechanisms of sulphonylurea action in the pancreatic -cell. In: Nutrient Regulation of Insulin Secretion (Flatt, P. R., ed.), pp. 319-339. Portland Press, London, UK.
• Prager R., Schernthaser G., Graf H. Effect of metformin on peripheral insulin sensitivity in non-insulin dependent diabetes mellitus. Diabetes Metab. 1986; 12:346-350
• Stevens J. F. Determination of glucose by automatic analyser. Clin. Chem. Acta 1971; 32:199-201[Medline]
• Swanston-Flatt S. K., Day C., Bailey C. J., Flatt P. R. Traditional plant treatments for diabetes. Studies in normal and streptozotocin diabetic mice. Diabetologia 1990; 33:462-464
• World Health Organization Expert Committee on Diabetes Mellitus (1980) Second Report. Technical Report Series 646. World Health Organization, Geneva, p 66.

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences