Horm. Metab. Res. 8 (\ 976) 413-419

© Georg Thieme Verlag Stuttgart

Studies on the Dynamics and Mechanism of Glibenclamide-Induced Insulin Secretion J.C. Basabe*, J.M.S. Farina and R.A. Chieri** Instituto de Fisiologla, Facultad de Medicina, Universidad de Buenos Aires and Fundacion Laboratorios de I nvestlgaciones Pediatricas, Hospital de Pediatria "Pedro de E lizalde", Buenos Aires, Argentina

Sustaincd, 60-minutc perfusion of gJibenclamide (0.5, \.5 and 10 J..Ig/ml) elicits a one-phase insulin release profile, formed by a rapid secretion peak followed by a second peak with lower insulin levels than the former. Basal insulin secretion valucs are observed during the period comprised between 13 and 60 minutes of perfusion, Concurrent stimulation with glucose (100, 150, 200 and 300 mg'f,,) plus glibenclamide (1 J..Ig/ml) causes a marked rise in both phases of insulin secretion. The addition of glibenclamide does not modify the biphasic secrction pattern caused by maximal glucose concentration (400 mg'}t,,). Thc maximal values of both phases of secretion in the dose-response curve clicited by different glucose concentrations shift to thc left when glibenclamide is added to the perfusate. Thc increase in insulin secretion caused by glibenclamide is not inhibited by puromycin.

to demonstrate such an effect (Loubatieres et al. 1970, Laube et al. 1971, Fussgänger et al. 1969, Malaisse 1969, Grodsky et al. 1967, Windsträm and Cerasi 1973b, Basabe et al. 1971 a)_ The first part of this paper deals with the possible glibenc\amide-glucose inter-relation regarding the rapid-release phase (1 st phase) and the provisionary phase (2nd phase) of insulin secretion.

In previous studies we found g1ibenc\amide to modulate glucose metabolism within the beta cell by producing the ATP levels required for nonnal insulin secretion (Basabe et al. 1975). Furthennore, glibenc1amide was shown to become bound to the pancreatic beta cell without entering the cell itself Both theophylline and phentolaminc modify and increase the glibenclamide-induced insulin release pattern. Propranolol (Hel/man et al. 1973). The relation of glibenc\amide and imidazole inhibit glibendamide-induced insulin release. with the adrenergic receptors and compounds affectOur results suggcst that: 1. Glibcnclamide increases beta eell ing the concentration of 3'5' adenosine monophossensitivity to glucose stimulation. phate (cAMP) was studied in the second part of this 2. Glibenclamide and glucose inducc secretion of insulin paper in an attempt to complement the studies quoted originating in the same compartment. above. 3. Modification of alpha and beta adrenergic receptors may modify glibendamide-induced insulin secretion. 4. 3'5' adenosine monophosphate levels modulate the beta cell response to glibenclamidc.

Material and Methods

Key-Words: Insulin Secretion- Glibenclamide -- Adrenergic Receptors -- Sulfonylurea-Glucose Interaction

Male rats from the strain of the Instituto de Fisiologla, Universidad de Buenos Aues, weighing 360 to 400 gm, were used. All the animals received a standard tube diet ad /ibitum, and in all cases were fasted overnight before each perfusion.

lntroduction Since the initial reports by Loubatieres (1944, 1946), the blood glucose lowering sulfonylureas have been widely used for nearly two decades. Despite a great number of papers and reviews dealing with their effect of insulin secretion, several points are still obscure (Levine and Pfeiffer 1969, Butterjield and Van Westering 1967)_ Studies on the sulfonylurea-g1ucose inter-relation have been the subject of controversied_ While some investigators have shown that sulfonylureas potentiate glucose-induced insulin secretion, other failed "Present address: Fundacion Laboratorios de Investigaciones Pediatricas, Hospital de Pediatrla "Pedro de Elizalde", Buenos Aires, Argentina. .... Member of the Carrera dei Investigador Cientlfico, Consejo Nacional de Investigaciones Cientlficas y Tecnicas, Buenos Aires, Argentina. Reeeived: 12 Sept. 1975

Accepted: 30 Apr. 1976

Pancreatic perfusion was performed according to the technique described by Penhos et al. (\ 969). Krebs-Ringer-bicarbonate buffer containing 3% dextran 70, 1% bovine serum albumin, and 50 mg% glucose was used (in only one group, glibenclamide, I ug/ml, was perfused without adding glucose to the buffer). The buffer was equilibrated with 95 percent oxygen and 5 percent CO 2 , resulting in a pH ranging from 7.35 to 7.40. The perfusion flow rate was kept between 7 and 10 mi/minute, with pressure variations between 45 and 65 mmHg. None of the substances used caused any change in the perfusion flow rate. The buffer was recirculated for twelve to fifteen minutes in order to stabilize temperature and attain eonstant perfusion conditions. F ollowing this stabilization period , all of the buffer flowing out of the portal vein was collected in graduated tubes at 60-second intervals during the first 15 minutes of perfusion, then at the intervals indicated in each figures, for a total time of 60 minutes. Sam pies from zero to two minutes were considered basal levels, In all cases, the substances under study were perfused continuously and at the same concentration, from 2 to 60 minutes. The results were evaluated using standard procedures for statistical analysis, the differences being considered si~ni­ ficant only if p values were lower than 0.05. The brackets

Downloaded by: University of Illinois. Copyrighted material.

Summary

414

1.C. Basabc, I.M.S. Farina and R.A. Chieri

in the figures denote the standard error of the mean. Insulin assays were performed with the method of Herbert et al. (965) modificd by Cresto et al. (1972).

± 30; 407 ± 50 at 4, 5 and 6 minutes, respectively, n = 6) and in a buffer containing 0 mgo/r! glucose (Insulin, uVlmin: 311 ± 29; 258 ± 38; 230 ± 21 at

= 5).

Por insulin was used as a standard. None of the substances used interfered with insulin radioimmunoassay.

4,5, and 6 minutes respectively, n

Chemieals utilized: Glibenclamide was a gift from Hoechst Argentina; puromycin was purchased to Nutritional Biochemicals; phentolamine to Ciba; theophylline, imidazole, bovine serum blumin, and dextran to Sigma c.; L-proprano101 was a gift from Gador Argentina.

Insulin secretion patterns induced by concurrent perfusion of glucose and glibenc/amide

Insulin secretion patterns iruluced by different doses of glucose or glibenc/amide Figure I part A shows the patterns and amount of insulin secretion produced by different glucose concentrations. The insulin levels in both phases were significantly higher than control values (50 mg% glucose), and the levels attained were higher as glucose concentration increased. Insulin release patterns elicited by continuous perfusion of glibenclamide alone, unless otherwise indicated, are shown in Figure I, subunit B (the buffer always contains 50 mg% glucose). At all the concentrations used, glibenclamide causes a rapid phase of insulin release lasting approximately 3 minutes, followed by a small peak of 2 to 3 minutes' duration. In a11 cases, basal insulin levels are regained between 14 and 60 minutes of perfusion. Glibenclamide, I ug/ml, was used to study its relation with different glucose concentrations (p < 0.001 at 4, 5, 6, II and 12 minutes, and p < 0.005 at 7 minutes when compared to the control group). There is a significant difference (p < 0.00 I) between the insulin secretion levels produced by the perfusion of I ug/ml gliben- _ clamide in a buffer containing 50 mg% glucose, as in the control group (Insulin, uVlmin: 618 ± 40; 519 A

When glucose and glibenclamide are perfused together, a biphasic secretion pattern again occurs; in all cases (100 to 300 mg%) the first insulin peak is almost equal to the sum of the values caused by the separate perfusion of each substance. The second phase of secretion shows frank potentiation induced by the concurrent perfusion of glibenclamide and glucose. As can be seen in Figure 2, subunit D the patterns and levels of insulin release elicited by 400 mg% glucose and 400 mg% glucose plus I ug/ml glibencla· mide do not differ from each other significantly.

Downloaded by: University of Illinois. Copyrighted material.

Results

Figure 2, subunits A to D show the insulin secretion patterns induced by perfusion of I ug/ml glibenclamid together with different glucose concentrations.

The results obtained upon perfusion of glibenc1amide alone (50 mg% glucose in the buffer) and of glibenc!amide plus different glucose concentrations are expressed in terms of a dose-response relations between glucose concentrations in the perfusate and insulin response levels. The point at wh ich the maximal value was attained was considered representative of the first phase of insulin release (gene rally at 5 minutes of perfusion), and the values occurring at 60 minutes as representative of the second phase. As can be seen in Figure 3, the addition of glibenclamide shifted the dose response curve to the left in both phases of insulin secretion only when glucose was used at a concentration of 100 to 300 mgo/r', but not at 400 mg%.

....-,... "

2~OO

~--

....

...._

10 . . "(_•• )

...._

'10 . . 110(_.6)

B

...._ 200 . . 110( •• 4) ...... 300 . . "(_.6)

.... _'/'1/'"'

Studies on the dynamics and mechanism of glibenclamide-induced insulin secretion.

Horm. Metab. Res. 8 (\ 976) 413-419 © Georg Thieme Verlag Stuttgart Studies on the Dynamics and Mechanism of Glibenclamide-Induced Insulin Secretion...
265KB Sizes 0 Downloads 0 Views