Vol. 1X1, No. 4 Punted in U.S.A.
Glyburide and Tolbutamide Induce Desensitization Insulin Release in Rat Pancreatic Islets by Different Mechanisms AGATA M. RABUAZZO, MASSIMO BUSCEMA, MARIO VETRI, FIORELLA FORTE, RICCARDO Institute Catania,
of Internal Italy
Medicine,
Metabolism,
CARMELA VIGNERI,
and Endocrinology,
VINCI, VENERA AND FRANCE%0
University
of
of
CALTABIANO, PURRELLO
Catania Medical
School,
ABSTRACT Insulin secretion was studied in rat pancreatic islets after 24-h exposure to various glyburide or tolbutamide concentrations. Glucoseinduced insulin release was significantly (P < 0.05) reduced in islets cultured with 0.1 pM glyburide or 100 pM tolbutamide (2098 + 187,832 + 93, and 989 + 88 pg/islet. h in control, glyburide-exposed, and tolbutamide-exposed islets, respectively). When glyburide-treated islets were stimulated with glyburide or tolbutamide, insulin release was also impaired compared to that in control islets (P < 0.05). In contrast, tolbutamide-exposed islets showed an impaired response to tolbutamide, but a normal response to glyburide. To investigate the mechanism of the sulfonylurea-induced impairment of insulin secretion, we measured insulin release and Rb’ efflux (a marker of the K’ channel activity) under static conditions. in a perifusion system and islet Ca’+ uptake Insulin release in response to 16.7 mM glucose increased in control islets from 9.4 f 1.1 to 131 f 19 pg/islet.min (first phase secretion peak). Simultaneously, the fractional *‘Rb+ efflux declined from 0.015 + 0.002% to 0.006 + 0.001% (change in decrement, -63.5%). Glucoseinduced insulin release in glyburideand tolbutamide-treated islets was significantly reduced (first phase peak, 22.1 + 5 and 39.7 f 8 pg/islet. min, respectively; P < 0.05), and the fractional RfiRb+ efflux decrement was -21 + 6% for glyburide (P < 0.005 US. control islets) and -65 +
4% (not different from control) for tolbutamide. When glyburideor tolbutamide-exposed islets were stimulated with the corresponding sulfonylurea, insulin release was impaired compared to that in control islets (P < 0.05), but, again, 86Rb+ efflux was impaired (P < 0.05) only in glyburide-exposed islets. When 45CaZ+ uptake was studied, the increase in glucose concentration from 2.8 to 16.7 mM increased calcium uptake in control islets from 1.76 f 0.58 to 7.27 + 1.36 pmol/islet. 2 min (n = 4). Preexposure to 0.1 pM glyburide did not change calcium uptake at a glucose concentration of 2.8 mM (1.44 + 0.45 pmol/islet.2 min) but significantly reduced calcium uptake stimulated by 16.7 mM glucose (3.21 f 0.35 pmol/islet. 2 min; n = 4; P < 0.005 compared to control islets). In contrast, preexposure to 100 pM tolbutamide did not change either basal or glucose-stimulated calcium uptake (1.44 + 0.45 and 6.90 + 0.81 pmol/islet.2 min, respectively; n = 4). These data show that in uitro chronic exposure of pancreatic islets to the sulfonylureas glyburide and tolbutamide impairs their ability to respond to a subsequent glucose or sulfonylurea stimulation. The impaired insulin secretion is accompanied by alterations of ionic fluxes (reduced K’ channel closure and reduced calcium uptake) only in islets preexposed to glyburide, but not to tolbutamide, suggesting a different mechanism for the two sulfonylureas in inducing islet desensitization. (Endocrinology 131: 18151820,1992)
S
underlying the P-cell hyporesponsivity after prolonged exposure to sulfonylureas is not known. Sulfonylureas stimulate insulin release by binding to a plasma membrane receptor on the P-cell surface (8). An ATP-sensitive K+ channel or a closely associatedprotein has been suggestedto be the sulfonylurea receptor (9, 10). After sulfonylurea stimulation, K+ efflux through this ATP-sensitive K+ channel decreases,then causing cell depolarization and the subsequent influx of Ca*+ through the voltagesensitive Ca*+ channels. The increasein intracellular calcium then triggers the exocytosis of the insulin granules (8, 11, 12). The aim of this study was to investigate the mechanism of the impaired insulin secretion in rat pancreatic islets chronically exposed to glyburide or tolbutamide. In the light of the critical role of ion channels in the mechanism of action of sulfonylureas, we measuredRb+ efflux (an index of the ATPsensitive K’ channel activity) and Ca*+ uptake in control islets and islets preexposed to either glyburide or tolbutamide. Whereas insulin secretion and ionic fluxes were both altered in islets preexposed to glyburide, this was not so in islets preexposed to tolbutamide, where Rb’ efflux and Cal+
ULFONYLUREAS are widely used in the treatment of noninsulin-dependent (type II) diabetic patients. After acute or short term administration, their hypoglycemic action is mainly due to a direct stimulation of insulin secretion. After chronic therapy, their mechanism of action is not clear, since many studies have shown that in treated patients plasma insulin concentrations return toward pretreatment levels after the initial elevation. This does not necessarily mean that in this condition sulfonylureas do not stimulate insulin secretion, since the reduced insulin levels may be a consequence of the amelioration of blood glucose levels. However, chronic sulfonylurea therapy may lead to a selective desensitization of pancreatic p-cells to sulfonylureas (l3) and, in some patients, may contribute to the secondary failure of these drugs (4). Moreover, in vitro chronic exposure of rat pancreatic islets to tolbutamide or glyburide causesa reversible impairment of glucose-induced insulin release(57), an effect that may be clinically relevant. The mechanism Received May 4, 1992. Address requests for reprints to: Francesco di Endocrinologia, Ospedale Garibaldi, Piazza Catania, Italy.
Purrello, S. Maria
M.D., Cattedra di Gesh, 95123
1815
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
1816
OF INSULIN
uptake were normal in spite of the impaired insulin secretion. These findings suggest a different mechanism of action for the two sulfonylureas in inducing islets desensitization.
SECRETION
and Methods
Materials Crude collagenase was obtained from Boehringer (Mannheim, Germany). Culture medium CMRL-1066, heat-inactivated fetal calf serum, glutamine, and antibiotics were obtained from Gibco (Glasgow, Scotland). Glyburide and tolbutamide were purchased from Sigma Chemical Co. (London, United Kingdom). 86RbCl (0.33 Ci/mmol) and 45CaC12 (27 mCi/mg) were purchased from Amersham (Aylesbury, United Kingdom). Silicone oil (density, 1.040) was purchased from Merck (Rahway, NJ). All other chemicals were of analytical grade.
Islet isolation
and culture conditions
Pancreatic islets were isolated from 200- to 250-g fed male Wistar rats by the collagenase method, as previously described (13). Islets were hand-picked under stereomicroscope observation. Using this technique, 300-400 islets were isolated from each pancreas. The procedure was completed within 120 min. Batches of 100 islets were maintained at 37 C in a 95% 02-5% CO, atmosphere for 24 h in a plastic petri dish containing 3 ml CMRL-1066 (glucose 5.5 mM) supplemented with 10% heat inactivated calf serum and 2 mmol/liter glutamine in the presence or absence of various glyburide and tolbutamide concentrations. After 24-h incubation in the presence or absence of the sulfonylureas, islet secretory activity was measured either under static incubation or in perifusion.
Insulin
release and calcium
uptake in static incubation
After 24-h preincubation with or without sulfonylureas, the islets were washed twice and divided into groups. This procedure required 15-20 min. Groups of five islets were incubated for 1 h at 37 C in KrebsRinger buffer containing 2.8 or 16.7 mmol/liter glucose. Insulin was measured in the medium or in the acid-alcohol cell extracts by RIA, ‘5Ca uptake was measured according to using rat insulin as a standard. the method described by Henquin and Lambert (14). Batches of 10 islets were transferred into 50 ~1 modified Krebs-bicarbonate buffer containing 25 rnM HEPES and with phosphate and sulfate replaced by equimolar amounts of chloride (15). Islets were layered on silicone oil, and the uptake period was started by adding 50 ~1 medium containing glucose (final concentration, 2.8 or 16.7 mM) and 45Ca (2.5 mM). The reaction was stopped after 2 min by centrifuging the islets through the silicone oil layer. The tube bottoms were then cut, and the radioactivity was counted. Tubes without islets were run as blanks. The uptake of [U-“C] sucrose was measured to correct for label in the extracellular space.
Insulin
release and ‘“Rb efflux in perifusion
The 86Rb efflux (a marker of K+ permeability) and insulin release kinetics were studied by a perifusion system (16). After the 24-h incubation in the presence or absence of glyburide or tolbutamide, groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 rnM 86Rb and the same glyburide or tolbutamide concentrations used in the 24-h incubation. Islets were then washed three times with fresh Rb’-free medium, placed in an Endotronics chamber in a Bio-GelPz matrix (Bio-Rad, Richmond, CA), and perifused at a flow rate of 1 ml/min at 37 C. The washing procedure required lo-15 min. The perifusion buffer consisted of a modified Krebs-Ringer buffer (115 mmol/liter NaCI, 5.4 mmol/liter KCI, 2.38 mmol/liter CaCl*, 0.8 mmol/ liter MgSO,, 1 mmol/liter Na,HP04, 10 mmol/liter HEI’ES, and 0.5 g/ dl BSA, pH 7.35). After 10.min perifusion with the buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were exposed to different secretagogues (glucose or sulfonylureas). Perifusate samples were collected at I-min intervals, and 400-~1 aliquots were analyzed for
Endo. 1992 Vol 131. No 4
86Rb radioactivity and insulin content. Results are expressed as the mean f SE fractional efflux released per min divided by 86Rb remaining in the islets).
Statistical Materials
BY SULFONYLUREAS
of Rb’ (*6Rb
analysis
The statistical analysis of results and the significance were assessed by Student’s unpaired t test.
of the differences
Results Static experiments release. Experiments were first carried out to determine the dose of glyburide or tolbutamide to be used in subsequent studies of 45Cauptake and 86Rbefflux. In control pancreatic islets cultured for 24 h without sulfonylureas, insulin releasein responseto a subsequentstimulation was 169 f 10 pg/islet +h in the presence of 2.8 mM glucose and 2098 f 187 pg/islet . h in the presence of 16.7 mM glucose (mean k SE; n = 8). Islet exposure for 24 h to increasing glyburide concentrations in the culture medium (containing 5.5 mM glucose) progressively reduced the secretory response to glucose. A significant and near-maximal inhibition of glucose-induced insulin release was observed at 0.1 PM glyburide (832 + 93 pg/islet.h; P < 0.05 us. control islets), and the maximal effect was present at 10 PM glyburide (603 + 47 pg/islet.h; n = 8; P < 0.01 VS.control islets). Exposure of pancreatic islets to increasing tolbutamide concentrations (ranging from l-1000 PM) caused a progressive reduction of glucose-induced insulin release. Insulin release in response to 16.7 mM glucose was 989 + 88 pg/ islet. h in islets preexposed to 100 PM and 801 f 95 pg/islet . h in islets preexposed to 1000 PM tolbutamide (n = 8; P < 0.05 VS.control islets). Basalinsulin secretion in the presence of 2.8 mM glucose was similar in control and sulfonylureatreated islets. Basedon the previous data, the studies of 86Rb efflux and 45Cauptake were performed in islets preexposed to 0.1 FM glyburide or 100 PM tolbutamide. Islet insulin contents were 51 + 8, 41 + 10, and 39 f 9 rig/islet (n = 7) in control islets and isletsexposed to 0.1 I.~M glyburide or 100 PM tolbutamide, respectively. When glyburide-exposed islets were stimulated with the maximal effective glyburide (10 PM) or tolbutamide (1 mM) concentration, insulin releasewas also impaired compared to that in control islets (P < 0.05; Fig. 1). In contrast, tolbutamide-exposed islets showed an impaired response to 1 mM tolbutamide (P < 0.05), but a normal response to 10 PM glyburide (Fig. 1). Inhibition of glucose-induced insulin release was fully reversible; insulin releasein responseto 16.7 mM glucosewas 2230 + 129 and 1900 + 188 (n = 3) in islets preincubated for 24 h with glyburide or tolbutamide, respectively, and then cultured in medium without sulfonylureas for an additional 24 h. Glucose-induced insulin releasewas 2082 f 202 (n = 3) in control islets cultured without sulfonylureas for 48 h.
Insulin
uptake. 45Cauptake was studied in pancreatic islets exposed for 24 h to either 0.1 PM glyburide or 100 PM tolbutamide. The 2 min uptake was measured because it Calcium
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
Control
Glyb-exposed
OF INSULIN
Tolb-exposed
FIG. 1. Effect of preexposure to glyburide (Glyb) or tolbutamide (Tolb) on sulfonylurea-induced insulin release. Pancreatic islets were cultured for 24 h in the absence (control) or presence of 0.1 pM glyburide or 100 FM tolbutamide. Insulin secretion was then measured for 1 h at 37 C in Krebs-Ringer buffer containing 2.8 mM glucose (m), 10 FM glyburide (H), or 1 mM tolbutamide (0). Results are the mean f SE of five experiments. *, P < 0.05 us. control islets. TABLE
1. Effect Treatment
of glucose
during preincubation
None Glyburide (0.1 pM) Tolbutamide (100
pM)
on ‘%a’+
uptake
r5Caz+ uptake at 2.6 mM
‘YY+ uptake at 16.7 mM
1.76 + 0.58 1.44 + 0.45 1.44 + 0.45
7.27 f 1.36 3.21 f 0.35” 6.90 f 0.81
%a’+ uptake was measured for 2 min in buffer containing either 2.8 or 16.7 mM glucose. Ca*+ uptake is expressed as picomoles per islet/ 2 min. a P < 0.05 US. islets preincubated without glyburide.
mainly shows the inward movement of Ca2+ (17). Calcium uptake in control islets was 1.76 f 0.58 pmol/islet.2 min (mean + SE; n = 4) under basal conditions (i.e. in the presence of 2.8 mM glucose) and increased to 7.27 + 1.36 in the presence of 16.7 mM glucose (Table 1). In pancreatic islets preexposed to 0.1 PM glyburide, calcium uptake under basal conditions was similar to that in control islets, but glucosestimulated uptake was significantly reduced (n = 4; P < 0.005 VS. control islets). In contrast, in islets preexposed to 100 PM tolbutamide, both basal and glucose-stimulated calcium uptakes were similar to that in control islets (Table 1).
Dynamic experiments insulin releaseand 86Rbeffllux in responseto glucose. In control islets, the average fractional 86Rb efflux declined slowly during the initial period of perifusion with 2.8 mM glucose and then markedly decreased when the glucose concentration was raised to 16.7 mM (from 0.015 f 0.002% immediately before glucose concentration increase to 0.006 + 0.001% at the nadir value; n = 6; change in (A) decrement, -63 + 5%). In the same perifusion fractions, insulin release increased from 9.4 ~fr 1 .l pg/islet +min under basal conditions to 131 + 19 (peak of the first phase glucose-stimulated insulin release) and then to 96 f 10 (second phase of insulin release maximum value, mean + SE; n = 6; Fig. 2, A and B). In islets preincubated with 0.1 PM glyburide only, a small
SECRETION
BY SULFONYLUREAS
1817
decrease in s6Rb efflux was observed after 16.7 mM glucose stimulation. The average fractional 86Rb efflux declined from 0.011 + 0.002% to 0.0085 + 0.001% at the nadir value (n = 6; A of decrement, -21 f 6%; P < 0.005 uscontrol islets). In these islets both first and second phase glucose-stimulated insulin release were diminished (22.1 + 5 and 33.4 + 9 pg/ islet.min, respectively; n = 6; P < 0.05 VS. control islets; Fig. 2, C and D). In islets preexposed to 100 PM tolbutamide, 86Rb efflux markedly declined in response to 16.7 mM glucose from 0.013 -C 0.001% to 0.005 + 0.0005 at the nadir value (n = 6; A of decrement, -65 + 4%), although both first phase insulin release (39.7 + 8 pg/islet.min) and second phase peak (29.8 + 6 pg/islet.min) were significantly reduced compared to control islets (n = 6; P < 0.05; Fig. 2, E and F).
insulin release and “‘Rb efflux in responseto glyburide. In control islets the addition of 10 PM glyburide to the perifusion medium resulted in a rapid decrease of 86Rb efflux from 0 .O 12 f. 0.001 to 0.007 + 0.001 at the nadir value (n = 5; A of decrement, -44 + 5%). In the same perifusion fractions insulin release was 7.9 f 5 pg/islet . min under basal conditions and 106 + 18 pg/islet/min . at the peak of glyburide stimulation (n = 5; Fig. 3, A and B). In contrast, in glyburidepreincubated islets, 10 PM glyburide induced only a small decrement in 86Rb efflux (from 0.012 + 0.002 to 0.010 + 0.001 at the nadir value; n = 5; A of decrement, -16 + 4%; P < 0.05 IIS. control islets). Under the same experimental conditions, insulin release in response to glyburide was significantly reduced compared to control values (maximum value, 52 + 9 pg/islet.min; n = 5; P < 0.05; Fig. 3, C and D). insulin releaseand 86Rbefflux in responseto tolbutamide. In control islets the addition of 1 mM tolbutamide to the perifusion buffer caused a reduction of the average fractional 86Rb efflux from 0.014 + 0.001 to 0.007 + 0.001 at the nadir value (n = 5; A of decrement, -51 + 5%). In the same perifusion fractions insulin release was 9.7 f 4 pg/islet . min under basal conditions and 78 + 13 at the peak of tolbutamide stimulation (Fig. 4, A and B). In tolbutamide-preexposed islets, 86Rb efflux declined from 0.013 + 0.001 to 0.005 + 0.0005 at the nadir value after tolbutamide stimulation (n = 5; A of decrement, -60 + 9%). In these islets, despite the normal 86Rb efflux, tolbutamide elicited a blunted insulin release (maximum value, 34.5 f 9 pg/islet.min; n = 5; P < 0.05; Fig. 4, C and D).
Discussion Our study demonstrates that the continuous in vitro exposure of pancreatic islets to the sulfonylureas glyburide and tolbutamide impairs their ability to secrete insulin in response to a subsequent glucose or sulfonylurea stimulation (islet desensitization). Similar results have been previously reported for both drugs (5-7, 17). Since ion channels play a critical role in the mechanism of sulfonylurea action, one possible mechanism of sulfonylurea islet desensitization is that the chronic membrane depolarization caused by contin-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
1818
Control
OF INSULIN
islets
SECRETION
BY
SULFONYLUREAS
Glyburide-exposed islets
Endo. 1992 Vol131.No4
Tolbutamide-exposed islets
I 5, ZeRv .C tx
II 0 *E
0,Ol
%' B 0,oo
I
.
I
1
1
I
0
10
20
30
‘
0
10
20
30
4
0
10
20
30
40
0
10
20
30
40
Time
(min)
Time
0
20
10
(mid
Time
30
40
(mid
FIG. 2. Effect of preexposure to glyburide or tolbutamide on glucose-induced insulin release and *Rb efflux. Pancreatic islets were h in the absence (control islets) or presence of 0.1 yM glyburide or 100 KM tolbutamide. After this period, groups of 100 islets were h at 37 C in CMRL-1066 medium with 0.2 mM %Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After 10.min buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed vertical line) with glucose 16.7 samples were collected at 1-min intervals, and 400-~1 aliquots were analyzed for “fiRb radioactivity (top panels) and insulin release Results are expressed as the mean + SE of six experiments.
cultured for 24 incubated for 2 perifusion with mM. Perifusate
(bottom panels).
Gl yburide-exposed islets
Control islets
FIG. 3. Effect of preexposure to glyburide on glyburide-induced insulin release and “Rb efflux. Pancreatic islets were cultured for 24 h in the absence (control islets) or presence of 0.1 pM glyburide. After this period groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 mM =Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After lo-min perifusion with buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed vertical line) with glyburide (10 PM). Perifusate samples were collected at 1-min intervals, and 400-~1 aliquots were analyzed for =Rb radioactivity (top panels) and insulin release (bottom panels). Results are expressed as the mean f SE of six experiments. 0
10
20 0 Time
uous exposure to the sulfonylurea can lead the P-cell to a refractory state in which it responds less effectively to a further stimulation. An altered function of K’ channels has been previously demonstrated after prolonged stimulation of
30 (mid
40 0
10
20 Time
30
40
(mid
P-cells with glucose (18). We investigated this possibility by measuring, in parallel with insulin secretion, Rb’ efflux (an index of the ATP-sensitive K’ channel activity) and Ca’+ uptake in control islets and in islets preexposed to either
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
OF INSULIN
SECRETION
BY SULFONYLUREAS To1butamide-exposed islets
Control islets I
1819
I
I
I
FIG. 4. Effect
of pr-exposure to tolbutamide on tolbutamide-induced insulin release and MRb efflux. Pancreatic islets were cultured for 24 h in the absence (control islets) or presence of 100 FM tolbutamide. After this period, groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 mM ‘“Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After lo-min perifusion with buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed uertical line) with tolbutamide (1 mM). Perifusate samples were collected at lmin intervals, and 400-~1 aliquots were analyzed for “Rb radioactivity (top panels) and insulin release (bottom panels). Results are expressed as mean f SE of six experiments.
C
-1 0
10
20
30
40
0
10
20
30
0
10
20
30
40
0
10
20
30
40
200
X2 h9 ‘5 7. :; g 1c 3’
100
.g J 0
Time
glyburide or tolbutamide, and we found that both ATPregulated K’ channel activity and Ca2+influx were impaired in pancreatic islets preexposed for 24 h to 0.1 PM glyburide. In contrast, both K’ channel activity and Ca2+ influx were normal in islets preexposed to 100 PM tolbutamide. These results suggest a different mechanism for glyburide and tolbutamide in inducing islet desensitization by prolonged action. The glyburide-induced impairment of insulin secretion, but not the tolbutamide effect, is associatedwith alterations in ionic fluxes. Therefore, the desensitization of tolbutamide-exposed islets seems to be caused by different mechanisms. There are known differences in glyburide and tolbutamide interaction with pancreatic P-cells. First, glyburide, but not tolbutamide, is internalized (19, 20) and produces a prolonged stimulation of insulin release even after the drug is no longer present in the extracellular medium (19). At equimolar concentrations, glyburide is about loo-fold more powerful than tolbutamide in inducing insulin secretion, a possible consequenceof its intracellular accumulation. Second, glyburide has been reported to affect ATP-regulated K+ channel activity with a slower onset, poorer reversibility, and a more potent action than tolbutamide (21). Diazoxide, a benzothiazedine that induces the ATP-sensitive K+ channel opening, is able to reverse the inhibition of s6Rb’ efflux causedby tolbutamide, but not that causedby glyburide (22). These observations seem consistent with our finding of K’ channel blockade after glyburide and not after tolbutamide. The possibility that the differential effects of chronic exposure to the two drugs on K+ channels may arise at least in part from artifacts of the experimental design should also be considered. However, we believe that it is unlikely that the
(min)
Time
40
(min)
lo- to 15-min the washing procedure may be sufficient for tolbutamide to completely dissociatefrom its binding sitesat the K+ channel and then normalize the subsequent s6Rb’ efflux. Third, tolbutamide, but not glyburide, has been shown to increase phosphoinositide hydrolysis in rat islets (5, 23), an effect that appears to play an important role in the P-cell secretory response through plasma membrane diacylglycerol increase and its interaction with protein kinase-C. This mechanism has been specially related to the second phase of insulin secretion (24) and has been found to be impaired after a 2-h continous tolbuutamide exposure (5). Interestingly, in our experiments, tolbutamide-exposed islets exhibited a preferential loss of the second phase of insulin release,whereas glyburide-exposed islets, whose primary defect is at the ionic flux level, showed a preferential loss of the first phase of insulin release. Finally, a further explanation for the difference between glyburide and tolbutamide is the possibility that the two sulfonylureas bind to a different binding site on the P-cell membrane, since the presenceof more than one binding site for sulfonylureas has been recently suggested(25). Tolbutamide-exposed islets were able to respond to the glyburide stimulation. This finding supports the view that glyburide and tolbutamide may stimulate insulin release through at least partly different mechanisms and suggests that the glyburide ability to stimulate insulin release in tolbutamide-exposed islets is allowed by the normal K+ and Ca*’ channel activity that we found in these islets. The findings of the present study contribute to a better understanding of the mechanism of chronic glyburide and tolbutamide interaction with pancreatic P-cells. They also have important implications for the therapeutic use of these
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
1820
DESENSITIZATION
OF INSULIN
drugs in patients. In fact, they support the view that the discontinuous exposure to sulfonylureas may be the best approach to maintain their effectiveness in stimulating insulin secretion and avoid pancreatic desensitization, a possible cause of the secondary failure of these agents.
SECRETION
BY
Endo. Voll31.
SULFONYLUREAS
1992 No 4
Misler S 1989 Effects of sulfonamides on a metabolite-regulated ATP-sensitive K’ channel in rat pancreatic B-cells. Am J Physiol 257:C1119-Cl127 13. Purrello F, Vetri M, Gatta C, Gullo D, Vigneri R 1989 Effects of high glucose on insulin secretion by isolated rat islets and purified B-cells and possible role of glycosylation. Diabetes 38:1417-1422 14. Henquin JC, Lambert AE 1975 Cobalt inhibition of insulin secretion and calcium uutake bv, isolated rat islets. Am J Physiol 228:16691677
References 1. Karam JH, Sanz E, Salomon E, Nolte MS 1986 Selective unresponsiveness of pancreas B-cells to acute sulfonylurea stimulation during sulfonylurea therapy in NIDDM. Diabetes 35:1314-1320 2. Filipponi P, Marcelli M, Nicoletti I, Pacifici R, Santeusanio F, Brunetti P 1983 Suppresive effect of long-term sulfonylurea treatment on A, B, and D cells of normal rat pancreas. Endocrinology 113:1972-1979 3. Dunbar JC, Foi PP 1974 An inhibitory effect of tolbutamide and glibenclamide on the pancreatic islets of normal animals. Diabetologia 10:27-32 4. Groop LC, Pelkonen R, Koskimies S, Bottazzo GF, Doniach D 1986 Secondary failure to treatment with oral antidiabetic agents in non-insulin-dependent diabetes, Diabetes Care 9:129-133 5. Zawalich WS 1989 Phosphoinoside hydrolysis and insulin secretion in response to glucose are impaired in isolated rat islets by prolonged exposure to the sulfonylurea tolbutamide. Endocrinology 125:281-
15.
16. 17.
18.
I
7.
8. 9.
10.
11.
12.
Bolaffi JL, Heldt A, Lewis LD, Grodsky GM 1986 The third phase of in vitro insulin secretion: evidence for glucose insensitivity. Diabetes 35:370-373 Gullo D, Rabuazzo AM, Vetri M, Gatta C, Vinci C, Buscema M, Vigneri R, Purrello F 1991 Chronic exposure to glibenclamide impairs insulin secretion in isolated rat pancreatic islets. J Endocrinol Invest 14:287-291 Boyd AE 1988 Sulfonylurea receptors, ion channels and fruit flies. Diabetes 37:847-850 Niki I, Nicks JL, Aschcroft SJ 1990 The beta cell glibenclamide receptor is an ADP-binding protein. Biochem J 268:713-718 Siconolfi-Baez L, Banerj MA, Lebovitz HE 1990 Characterization and significance of sulfonylurea receptors. Diabetes Care [Suppl 31 13:2-8 Henquin JC, Meissner HP 1982 Opposite effects of tolbutamide and diazoxide on 86 Rb fluxes and membrane potential in pancreatic B-cells. Biochem Pharmacol 31:1407-1415 Gillis KD, Gee W M, Hammoud A, McDaniel ML, Falke LC,
I
399 19.
20.
286 6.
Flatt PR, Berggren PO, Gylfe E, Hellman BO 1980 Calcium and pancreatic B-cell function. IX. Demonstration of lanthanid-induced inhibition of insulin secretion independent of modifications in transmembrane Ca+ fluxes. Endocrinoloev 107:1007-1013 Henquin JC 1978 D-Glucose inhit% potassium efflux from pancreatic islet cells. Nature (Lond) 271:271-273 Henquin JC 1980 Tolbutamide stimulation and inhibition of insulin release: studies of the underlying ionic mechanisms in isolated rat islets. Diabetologia 18:151-160 Purrello F, Vetri M, Vinci C, Gatta C, Buscema M, Vigneri R 1990 Chronic exposure to high glucose and impairment of K’channel function in uerifused rat uancreatic islets. Diabetes 39:397-
21.
22.
23.
24.
25.
Gorus FK, Schuit FC, In’t Veld PA, Gepts W, Pipeleers DG 1988 Interaction of sulfonvlureas with uancreatic B-cells. A study with glyburide. Diabetes 37:1090-1094 1 Carpentier JL, Sawano F, Ravazzola M, Malaisse WJ 1986 Internalization of 3H-glibenclamide in pancreatic islet cells. Diabetologia 29:259-261 Panten U, Burgfeld J, Goerke F, Rennicke M, Schwanstecher M, Wallsch A, Zunkler BJ, Lenzen S 1989 Control of insulin secretion by sulfonylureas, meglitinide and diazoxide in relation to their binding to the sulfonylurea receptor in pancreatic islets. Biochem Pharmacol38:1217-1229 Sturgess NC, Koslowski RZ, Carrington CA, Hales CN, Ashford ML1 1988 Effects of sulfonvlureas and diazoxide on insulin secretion and nucleosidesensitive channels in an insulin-secreting cell line. Br J Pharmacol95:83-94 Best L, Malaisse WJ 1983 Stimulation of phosphoinositide breakdown in rat pancreatic islets by glucose and carbamylcholine. Biochem Biophys Res Commun 116:9-16 Rasmussen H, Zawalich KC, Ganesan S, Calle R, Zawalich WS 1990 Physiology and pathophysiology of insulin secretion. Diabetes Care 13~655-666 Verspohl EJ, Ammon HPT, Mark M 1990 Evidence for more than one binding site for sulfonylureas in insulin-secreting cells. J Pharm Pharmacol42:230-235
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
Glyburide and Tolbutamide Induce Desensitization Insulin Release in Rat Pancreatic Islets by Different Mechanisms AGATA M. RABUAZZO, MASSIMO BUSCEMA, MARIO VETRI, FIORELLA FORTE, RICCARDO Institute Catania,
of Internal Italy
Medicine,
Metabolism,
CARMELA VIGNERI,
and Endocrinology,
VINCI, VENERA AND FRANCE%0
University
of
of
CALTABIANO, PURRELLO
Catania Medical
School,
ABSTRACT Insulin secretion was studied in rat pancreatic islets after 24-h exposure to various glyburide or tolbutamide concentrations. Glucoseinduced insulin release was significantly (P < 0.05) reduced in islets cultured with 0.1 pM glyburide or 100 pM tolbutamide (2098 + 187,832 + 93, and 989 + 88 pg/islet. h in control, glyburide-exposed, and tolbutamide-exposed islets, respectively). When glyburide-treated islets were stimulated with glyburide or tolbutamide, insulin release was also impaired compared to that in control islets (P < 0.05). In contrast, tolbutamide-exposed islets showed an impaired response to tolbutamide, but a normal response to glyburide. To investigate the mechanism of the sulfonylurea-induced impairment of insulin secretion, we measured insulin release and Rb’ efflux (a marker of the K’ channel activity) under static conditions. in a perifusion system and islet Ca’+ uptake Insulin release in response to 16.7 mM glucose increased in control islets from 9.4 f 1.1 to 131 f 19 pg/islet.min (first phase secretion peak). Simultaneously, the fractional *‘Rb+ efflux declined from 0.015 + 0.002% to 0.006 + 0.001% (change in decrement, -63.5%). Glucoseinduced insulin release in glyburideand tolbutamide-treated islets was significantly reduced (first phase peak, 22.1 + 5 and 39.7 f 8 pg/islet. min, respectively; P < 0.05), and the fractional RfiRb+ efflux decrement was -21 + 6% for glyburide (P < 0.005 US. control islets) and -65 +
4% (not different from control) for tolbutamide. When glyburideor tolbutamide-exposed islets were stimulated with the corresponding sulfonylurea, insulin release was impaired compared to that in control islets (P < 0.05), but, again, 86Rb+ efflux was impaired (P < 0.05) only in glyburide-exposed islets. When 45CaZ+ uptake was studied, the increase in glucose concentration from 2.8 to 16.7 mM increased calcium uptake in control islets from 1.76 f 0.58 to 7.27 + 1.36 pmol/islet. 2 min (n = 4). Preexposure to 0.1 pM glyburide did not change calcium uptake at a glucose concentration of 2.8 mM (1.44 + 0.45 pmol/islet.2 min) but significantly reduced calcium uptake stimulated by 16.7 mM glucose (3.21 f 0.35 pmol/islet. 2 min; n = 4; P < 0.005 compared to control islets). In contrast, preexposure to 100 pM tolbutamide did not change either basal or glucose-stimulated calcium uptake (1.44 + 0.45 and 6.90 + 0.81 pmol/islet.2 min, respectively; n = 4). These data show that in uitro chronic exposure of pancreatic islets to the sulfonylureas glyburide and tolbutamide impairs their ability to respond to a subsequent glucose or sulfonylurea stimulation. The impaired insulin secretion is accompanied by alterations of ionic fluxes (reduced K’ channel closure and reduced calcium uptake) only in islets preexposed to glyburide, but not to tolbutamide, suggesting a different mechanism for the two sulfonylureas in inducing islet desensitization. (Endocrinology 131: 18151820,1992)
S
underlying the P-cell hyporesponsivity after prolonged exposure to sulfonylureas is not known. Sulfonylureas stimulate insulin release by binding to a plasma membrane receptor on the P-cell surface (8). An ATP-sensitive K+ channel or a closely associatedprotein has been suggestedto be the sulfonylurea receptor (9, 10). After sulfonylurea stimulation, K+ efflux through this ATP-sensitive K+ channel decreases,then causing cell depolarization and the subsequent influx of Ca*+ through the voltagesensitive Ca*+ channels. The increasein intracellular calcium then triggers the exocytosis of the insulin granules (8, 11, 12). The aim of this study was to investigate the mechanism of the impaired insulin secretion in rat pancreatic islets chronically exposed to glyburide or tolbutamide. In the light of the critical role of ion channels in the mechanism of action of sulfonylureas, we measuredRb+ efflux (an index of the ATPsensitive K’ channel activity) and Ca*+ uptake in control islets and islets preexposed to either glyburide or tolbutamide. Whereas insulin secretion and ionic fluxes were both altered in islets preexposed to glyburide, this was not so in islets preexposed to tolbutamide, where Rb’ efflux and Cal+
ULFONYLUREAS are widely used in the treatment of noninsulin-dependent (type II) diabetic patients. After acute or short term administration, their hypoglycemic action is mainly due to a direct stimulation of insulin secretion. After chronic therapy, their mechanism of action is not clear, since many studies have shown that in treated patients plasma insulin concentrations return toward pretreatment levels after the initial elevation. This does not necessarily mean that in this condition sulfonylureas do not stimulate insulin secretion, since the reduced insulin levels may be a consequence of the amelioration of blood glucose levels. However, chronic sulfonylurea therapy may lead to a selective desensitization of pancreatic p-cells to sulfonylureas (l3) and, in some patients, may contribute to the secondary failure of these drugs (4). Moreover, in vitro chronic exposure of rat pancreatic islets to tolbutamide or glyburide causesa reversible impairment of glucose-induced insulin release(57), an effect that may be clinically relevant. The mechanism Received May 4, 1992. Address requests for reprints to: Francesco di Endocrinologia, Ospedale Garibaldi, Piazza Catania, Italy.
Purrello, S. Maria
M.D., Cattedra di Gesh, 95123
1815
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
1816
OF INSULIN
uptake were normal in spite of the impaired insulin secretion. These findings suggest a different mechanism of action for the two sulfonylureas in inducing islets desensitization.
SECRETION
and Methods
Materials Crude collagenase was obtained from Boehringer (Mannheim, Germany). Culture medium CMRL-1066, heat-inactivated fetal calf serum, glutamine, and antibiotics were obtained from Gibco (Glasgow, Scotland). Glyburide and tolbutamide were purchased from Sigma Chemical Co. (London, United Kingdom). 86RbCl (0.33 Ci/mmol) and 45CaC12 (27 mCi/mg) were purchased from Amersham (Aylesbury, United Kingdom). Silicone oil (density, 1.040) was purchased from Merck (Rahway, NJ). All other chemicals were of analytical grade.
Islet isolation
and culture conditions
Pancreatic islets were isolated from 200- to 250-g fed male Wistar rats by the collagenase method, as previously described (13). Islets were hand-picked under stereomicroscope observation. Using this technique, 300-400 islets were isolated from each pancreas. The procedure was completed within 120 min. Batches of 100 islets were maintained at 37 C in a 95% 02-5% CO, atmosphere for 24 h in a plastic petri dish containing 3 ml CMRL-1066 (glucose 5.5 mM) supplemented with 10% heat inactivated calf serum and 2 mmol/liter glutamine in the presence or absence of various glyburide and tolbutamide concentrations. After 24-h incubation in the presence or absence of the sulfonylureas, islet secretory activity was measured either under static incubation or in perifusion.
Insulin
release and calcium
uptake in static incubation
After 24-h preincubation with or without sulfonylureas, the islets were washed twice and divided into groups. This procedure required 15-20 min. Groups of five islets were incubated for 1 h at 37 C in KrebsRinger buffer containing 2.8 or 16.7 mmol/liter glucose. Insulin was measured in the medium or in the acid-alcohol cell extracts by RIA, ‘5Ca uptake was measured according to using rat insulin as a standard. the method described by Henquin and Lambert (14). Batches of 10 islets were transferred into 50 ~1 modified Krebs-bicarbonate buffer containing 25 rnM HEPES and with phosphate and sulfate replaced by equimolar amounts of chloride (15). Islets were layered on silicone oil, and the uptake period was started by adding 50 ~1 medium containing glucose (final concentration, 2.8 or 16.7 mM) and 45Ca (2.5 mM). The reaction was stopped after 2 min by centrifuging the islets through the silicone oil layer. The tube bottoms were then cut, and the radioactivity was counted. Tubes without islets were run as blanks. The uptake of [U-“C] sucrose was measured to correct for label in the extracellular space.
Insulin
release and ‘“Rb efflux in perifusion
The 86Rb efflux (a marker of K+ permeability) and insulin release kinetics were studied by a perifusion system (16). After the 24-h incubation in the presence or absence of glyburide or tolbutamide, groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 rnM 86Rb and the same glyburide or tolbutamide concentrations used in the 24-h incubation. Islets were then washed three times with fresh Rb’-free medium, placed in an Endotronics chamber in a Bio-GelPz matrix (Bio-Rad, Richmond, CA), and perifused at a flow rate of 1 ml/min at 37 C. The washing procedure required lo-15 min. The perifusion buffer consisted of a modified Krebs-Ringer buffer (115 mmol/liter NaCI, 5.4 mmol/liter KCI, 2.38 mmol/liter CaCl*, 0.8 mmol/ liter MgSO,, 1 mmol/liter Na,HP04, 10 mmol/liter HEI’ES, and 0.5 g/ dl BSA, pH 7.35). After 10.min perifusion with the buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were exposed to different secretagogues (glucose or sulfonylureas). Perifusate samples were collected at I-min intervals, and 400-~1 aliquots were analyzed for
Endo. 1992 Vol 131. No 4
86Rb radioactivity and insulin content. Results are expressed as the mean f SE fractional efflux released per min divided by 86Rb remaining in the islets).
Statistical Materials
BY SULFONYLUREAS
of Rb’ (*6Rb
analysis
The statistical analysis of results and the significance were assessed by Student’s unpaired t test.
of the differences
Results Static experiments release. Experiments were first carried out to determine the dose of glyburide or tolbutamide to be used in subsequent studies of 45Cauptake and 86Rbefflux. In control pancreatic islets cultured for 24 h without sulfonylureas, insulin releasein responseto a subsequentstimulation was 169 f 10 pg/islet +h in the presence of 2.8 mM glucose and 2098 f 187 pg/islet . h in the presence of 16.7 mM glucose (mean k SE; n = 8). Islet exposure for 24 h to increasing glyburide concentrations in the culture medium (containing 5.5 mM glucose) progressively reduced the secretory response to glucose. A significant and near-maximal inhibition of glucose-induced insulin release was observed at 0.1 PM glyburide (832 + 93 pg/islet.h; P < 0.05 us. control islets), and the maximal effect was present at 10 PM glyburide (603 + 47 pg/islet.h; n = 8; P < 0.01 VS.control islets). Exposure of pancreatic islets to increasing tolbutamide concentrations (ranging from l-1000 PM) caused a progressive reduction of glucose-induced insulin release. Insulin release in response to 16.7 mM glucose was 989 + 88 pg/ islet. h in islets preexposed to 100 PM and 801 f 95 pg/islet . h in islets preexposed to 1000 PM tolbutamide (n = 8; P < 0.05 VS.control islets). Basalinsulin secretion in the presence of 2.8 mM glucose was similar in control and sulfonylureatreated islets. Basedon the previous data, the studies of 86Rb efflux and 45Cauptake were performed in islets preexposed to 0.1 FM glyburide or 100 PM tolbutamide. Islet insulin contents were 51 + 8, 41 + 10, and 39 f 9 rig/islet (n = 7) in control islets and isletsexposed to 0.1 I.~M glyburide or 100 PM tolbutamide, respectively. When glyburide-exposed islets were stimulated with the maximal effective glyburide (10 PM) or tolbutamide (1 mM) concentration, insulin releasewas also impaired compared to that in control islets (P < 0.05; Fig. 1). In contrast, tolbutamide-exposed islets showed an impaired response to 1 mM tolbutamide (P < 0.05), but a normal response to 10 PM glyburide (Fig. 1). Inhibition of glucose-induced insulin release was fully reversible; insulin releasein responseto 16.7 mM glucosewas 2230 + 129 and 1900 + 188 (n = 3) in islets preincubated for 24 h with glyburide or tolbutamide, respectively, and then cultured in medium without sulfonylureas for an additional 24 h. Glucose-induced insulin releasewas 2082 f 202 (n = 3) in control islets cultured without sulfonylureas for 48 h.
Insulin
uptake. 45Cauptake was studied in pancreatic islets exposed for 24 h to either 0.1 PM glyburide or 100 PM tolbutamide. The 2 min uptake was measured because it Calcium
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
Control
Glyb-exposed
OF INSULIN
Tolb-exposed
FIG. 1. Effect of preexposure to glyburide (Glyb) or tolbutamide (Tolb) on sulfonylurea-induced insulin release. Pancreatic islets were cultured for 24 h in the absence (control) or presence of 0.1 pM glyburide or 100 FM tolbutamide. Insulin secretion was then measured for 1 h at 37 C in Krebs-Ringer buffer containing 2.8 mM glucose (m), 10 FM glyburide (H), or 1 mM tolbutamide (0). Results are the mean f SE of five experiments. *, P < 0.05 us. control islets. TABLE
1. Effect Treatment
of glucose
during preincubation
None Glyburide (0.1 pM) Tolbutamide (100
pM)
on ‘%a’+
uptake
r5Caz+ uptake at 2.6 mM
‘YY+ uptake at 16.7 mM
1.76 + 0.58 1.44 + 0.45 1.44 + 0.45
7.27 f 1.36 3.21 f 0.35” 6.90 f 0.81
%a’+ uptake was measured for 2 min in buffer containing either 2.8 or 16.7 mM glucose. Ca*+ uptake is expressed as picomoles per islet/ 2 min. a P < 0.05 US. islets preincubated without glyburide.
mainly shows the inward movement of Ca2+ (17). Calcium uptake in control islets was 1.76 f 0.58 pmol/islet.2 min (mean + SE; n = 4) under basal conditions (i.e. in the presence of 2.8 mM glucose) and increased to 7.27 + 1.36 in the presence of 16.7 mM glucose (Table 1). In pancreatic islets preexposed to 0.1 PM glyburide, calcium uptake under basal conditions was similar to that in control islets, but glucosestimulated uptake was significantly reduced (n = 4; P < 0.005 VS. control islets). In contrast, in islets preexposed to 100 PM tolbutamide, both basal and glucose-stimulated calcium uptakes were similar to that in control islets (Table 1).
Dynamic experiments insulin releaseand 86Rbeffllux in responseto glucose. In control islets, the average fractional 86Rb efflux declined slowly during the initial period of perifusion with 2.8 mM glucose and then markedly decreased when the glucose concentration was raised to 16.7 mM (from 0.015 f 0.002% immediately before glucose concentration increase to 0.006 + 0.001% at the nadir value; n = 6; change in (A) decrement, -63 + 5%). In the same perifusion fractions, insulin release increased from 9.4 ~fr 1 .l pg/islet +min under basal conditions to 131 + 19 (peak of the first phase glucose-stimulated insulin release) and then to 96 f 10 (second phase of insulin release maximum value, mean + SE; n = 6; Fig. 2, A and B). In islets preincubated with 0.1 PM glyburide only, a small
SECRETION
BY SULFONYLUREAS
1817
decrease in s6Rb efflux was observed after 16.7 mM glucose stimulation. The average fractional 86Rb efflux declined from 0.011 + 0.002% to 0.0085 + 0.001% at the nadir value (n = 6; A of decrement, -21 f 6%; P < 0.005 uscontrol islets). In these islets both first and second phase glucose-stimulated insulin release were diminished (22.1 + 5 and 33.4 + 9 pg/ islet.min, respectively; n = 6; P < 0.05 VS. control islets; Fig. 2, C and D). In islets preexposed to 100 PM tolbutamide, 86Rb efflux markedly declined in response to 16.7 mM glucose from 0.013 -C 0.001% to 0.005 + 0.0005 at the nadir value (n = 6; A of decrement, -65 + 4%), although both first phase insulin release (39.7 + 8 pg/islet.min) and second phase peak (29.8 + 6 pg/islet.min) were significantly reduced compared to control islets (n = 6; P < 0.05; Fig. 2, E and F).
insulin release and “‘Rb efflux in responseto glyburide. In control islets the addition of 10 PM glyburide to the perifusion medium resulted in a rapid decrease of 86Rb efflux from 0 .O 12 f. 0.001 to 0.007 + 0.001 at the nadir value (n = 5; A of decrement, -44 + 5%). In the same perifusion fractions insulin release was 7.9 f 5 pg/islet . min under basal conditions and 106 + 18 pg/islet/min . at the peak of glyburide stimulation (n = 5; Fig. 3, A and B). In contrast, in glyburidepreincubated islets, 10 PM glyburide induced only a small decrement in 86Rb efflux (from 0.012 + 0.002 to 0.010 + 0.001 at the nadir value; n = 5; A of decrement, -16 + 4%; P < 0.05 IIS. control islets). Under the same experimental conditions, insulin release in response to glyburide was significantly reduced compared to control values (maximum value, 52 + 9 pg/islet.min; n = 5; P < 0.05; Fig. 3, C and D). insulin releaseand 86Rbefflux in responseto tolbutamide. In control islets the addition of 1 mM tolbutamide to the perifusion buffer caused a reduction of the average fractional 86Rb efflux from 0.014 + 0.001 to 0.007 + 0.001 at the nadir value (n = 5; A of decrement, -51 + 5%). In the same perifusion fractions insulin release was 9.7 f 4 pg/islet . min under basal conditions and 78 + 13 at the peak of tolbutamide stimulation (Fig. 4, A and B). In tolbutamide-preexposed islets, 86Rb efflux declined from 0.013 + 0.001 to 0.005 + 0.0005 at the nadir value after tolbutamide stimulation (n = 5; A of decrement, -60 + 9%). In these islets, despite the normal 86Rb efflux, tolbutamide elicited a blunted insulin release (maximum value, 34.5 f 9 pg/islet.min; n = 5; P < 0.05; Fig. 4, C and D).
Discussion Our study demonstrates that the continuous in vitro exposure of pancreatic islets to the sulfonylureas glyburide and tolbutamide impairs their ability to secrete insulin in response to a subsequent glucose or sulfonylurea stimulation (islet desensitization). Similar results have been previously reported for both drugs (5-7, 17). Since ion channels play a critical role in the mechanism of sulfonylurea action, one possible mechanism of sulfonylurea islet desensitization is that the chronic membrane depolarization caused by contin-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
1818
Control
OF INSULIN
islets
SECRETION
BY
SULFONYLUREAS
Glyburide-exposed islets
Endo. 1992 Vol131.No4
Tolbutamide-exposed islets
I 5, ZeRv .C tx
II 0 *E
0,Ol
%' B 0,oo
I
.
I
1
1
I
0
10
20
30
‘
0
10
20
30
4
0
10
20
30
40
0
10
20
30
40
Time
(min)
Time
0
20
10
(mid
Time
30
40
(mid
FIG. 2. Effect of preexposure to glyburide or tolbutamide on glucose-induced insulin release and *Rb efflux. Pancreatic islets were h in the absence (control islets) or presence of 0.1 yM glyburide or 100 KM tolbutamide. After this period, groups of 100 islets were h at 37 C in CMRL-1066 medium with 0.2 mM %Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After 10.min buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed vertical line) with glucose 16.7 samples were collected at 1-min intervals, and 400-~1 aliquots were analyzed for “fiRb radioactivity (top panels) and insulin release Results are expressed as the mean + SE of six experiments.
cultured for 24 incubated for 2 perifusion with mM. Perifusate
(bottom panels).
Gl yburide-exposed islets
Control islets
FIG. 3. Effect of preexposure to glyburide on glyburide-induced insulin release and “Rb efflux. Pancreatic islets were cultured for 24 h in the absence (control islets) or presence of 0.1 pM glyburide. After this period groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 mM =Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After lo-min perifusion with buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed vertical line) with glyburide (10 PM). Perifusate samples were collected at 1-min intervals, and 400-~1 aliquots were analyzed for =Rb radioactivity (top panels) and insulin release (bottom panels). Results are expressed as the mean f SE of six experiments. 0
10
20 0 Time
uous exposure to the sulfonylurea can lead the P-cell to a refractory state in which it responds less effectively to a further stimulation. An altered function of K’ channels has been previously demonstrated after prolonged stimulation of
30 (mid
40 0
10
20 Time
30
40
(mid
P-cells with glucose (18). We investigated this possibility by measuring, in parallel with insulin secretion, Rb’ efflux (an index of the ATP-sensitive K’ channel activity) and Ca’+ uptake in control islets and in islets preexposed to either
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
DESENSITIZATION
OF INSULIN
SECRETION
BY SULFONYLUREAS To1butamide-exposed islets
Control islets I
1819
I
I
I
FIG. 4. Effect
of pr-exposure to tolbutamide on tolbutamide-induced insulin release and MRb efflux. Pancreatic islets were cultured for 24 h in the absence (control islets) or presence of 100 FM tolbutamide. After this period, groups of 100 islets were incubated for 2 h at 37 C in CMRL-1066 medium with 0.2 mM ‘“Rb, then washed and perifused at a flow rate of 1 ml/min at 37 C. After lo-min perifusion with buffer containing 2.8 mmol/liter glucose to equilibrate the system, the islets were stimulated (dashed uertical line) with tolbutamide (1 mM). Perifusate samples were collected at lmin intervals, and 400-~1 aliquots were analyzed for “Rb radioactivity (top panels) and insulin release (bottom panels). Results are expressed as mean f SE of six experiments.
C
-1 0
10
20
30
40
0
10
20
30
0
10
20
30
40
0
10
20
30
40
200
X2 h9 ‘5 7. :; g 1c 3’
100
.g J 0
Time
glyburide or tolbutamide, and we found that both ATPregulated K’ channel activity and Ca2+influx were impaired in pancreatic islets preexposed for 24 h to 0.1 PM glyburide. In contrast, both K’ channel activity and Ca2+ influx were normal in islets preexposed to 100 PM tolbutamide. These results suggest a different mechanism for glyburide and tolbutamide in inducing islet desensitization by prolonged action. The glyburide-induced impairment of insulin secretion, but not the tolbutamide effect, is associatedwith alterations in ionic fluxes. Therefore, the desensitization of tolbutamide-exposed islets seems to be caused by different mechanisms. There are known differences in glyburide and tolbutamide interaction with pancreatic P-cells. First, glyburide, but not tolbutamide, is internalized (19, 20) and produces a prolonged stimulation of insulin release even after the drug is no longer present in the extracellular medium (19). At equimolar concentrations, glyburide is about loo-fold more powerful than tolbutamide in inducing insulin secretion, a possible consequenceof its intracellular accumulation. Second, glyburide has been reported to affect ATP-regulated K+ channel activity with a slower onset, poorer reversibility, and a more potent action than tolbutamide (21). Diazoxide, a benzothiazedine that induces the ATP-sensitive K+ channel opening, is able to reverse the inhibition of s6Rb’ efflux causedby tolbutamide, but not that causedby glyburide (22). These observations seem consistent with our finding of K’ channel blockade after glyburide and not after tolbutamide. The possibility that the differential effects of chronic exposure to the two drugs on K+ channels may arise at least in part from artifacts of the experimental design should also be considered. However, we believe that it is unlikely that the
(min)
Time
40
(min)
lo- to 15-min the washing procedure may be sufficient for tolbutamide to completely dissociatefrom its binding sitesat the K+ channel and then normalize the subsequent s6Rb’ efflux. Third, tolbutamide, but not glyburide, has been shown to increase phosphoinositide hydrolysis in rat islets (5, 23), an effect that appears to play an important role in the P-cell secretory response through plasma membrane diacylglycerol increase and its interaction with protein kinase-C. This mechanism has been specially related to the second phase of insulin secretion (24) and has been found to be impaired after a 2-h continous tolbuutamide exposure (5). Interestingly, in our experiments, tolbutamide-exposed islets exhibited a preferential loss of the second phase of insulin release,whereas glyburide-exposed islets, whose primary defect is at the ionic flux level, showed a preferential loss of the first phase of insulin release. Finally, a further explanation for the difference between glyburide and tolbutamide is the possibility that the two sulfonylureas bind to a different binding site on the P-cell membrane, since the presenceof more than one binding site for sulfonylureas has been recently suggested(25). Tolbutamide-exposed islets were able to respond to the glyburide stimulation. This finding supports the view that glyburide and tolbutamide may stimulate insulin release through at least partly different mechanisms and suggests that the glyburide ability to stimulate insulin release in tolbutamide-exposed islets is allowed by the normal K+ and Ca*’ channel activity that we found in these islets. The findings of the present study contribute to a better understanding of the mechanism of chronic glyburide and tolbutamide interaction with pancreatic P-cells. They also have important implications for the therapeutic use of these
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
1820
DESENSITIZATION
OF INSULIN
drugs in patients. In fact, they support the view that the discontinuous exposure to sulfonylureas may be the best approach to maintain their effectiveness in stimulating insulin secretion and avoid pancreatic desensitization, a possible cause of the secondary failure of these agents.
SECRETION
BY
Endo. Voll31.
SULFONYLUREAS
1992 No 4
Misler S 1989 Effects of sulfonamides on a metabolite-regulated ATP-sensitive K’ channel in rat pancreatic B-cells. Am J Physiol 257:C1119-Cl127 13. Purrello F, Vetri M, Gatta C, Gullo D, Vigneri R 1989 Effects of high glucose on insulin secretion by isolated rat islets and purified B-cells and possible role of glycosylation. Diabetes 38:1417-1422 14. Henquin JC, Lambert AE 1975 Cobalt inhibition of insulin secretion and calcium uutake bv, isolated rat islets. Am J Physiol 228:16691677
References 1. Karam JH, Sanz E, Salomon E, Nolte MS 1986 Selective unresponsiveness of pancreas B-cells to acute sulfonylurea stimulation during sulfonylurea therapy in NIDDM. Diabetes 35:1314-1320 2. Filipponi P, Marcelli M, Nicoletti I, Pacifici R, Santeusanio F, Brunetti P 1983 Suppresive effect of long-term sulfonylurea treatment on A, B, and D cells of normal rat pancreas. Endocrinology 113:1972-1979 3. Dunbar JC, Foi PP 1974 An inhibitory effect of tolbutamide and glibenclamide on the pancreatic islets of normal animals. Diabetologia 10:27-32 4. Groop LC, Pelkonen R, Koskimies S, Bottazzo GF, Doniach D 1986 Secondary failure to treatment with oral antidiabetic agents in non-insulin-dependent diabetes, Diabetes Care 9:129-133 5. Zawalich WS 1989 Phosphoinoside hydrolysis and insulin secretion in response to glucose are impaired in isolated rat islets by prolonged exposure to the sulfonylurea tolbutamide. Endocrinology 125:281-
15.
16. 17.
18.
I
7.
8. 9.
10.
11.
12.
Bolaffi JL, Heldt A, Lewis LD, Grodsky GM 1986 The third phase of in vitro insulin secretion: evidence for glucose insensitivity. Diabetes 35:370-373 Gullo D, Rabuazzo AM, Vetri M, Gatta C, Vinci C, Buscema M, Vigneri R, Purrello F 1991 Chronic exposure to glibenclamide impairs insulin secretion in isolated rat pancreatic islets. J Endocrinol Invest 14:287-291 Boyd AE 1988 Sulfonylurea receptors, ion channels and fruit flies. Diabetes 37:847-850 Niki I, Nicks JL, Aschcroft SJ 1990 The beta cell glibenclamide receptor is an ADP-binding protein. Biochem J 268:713-718 Siconolfi-Baez L, Banerj MA, Lebovitz HE 1990 Characterization and significance of sulfonylurea receptors. Diabetes Care [Suppl 31 13:2-8 Henquin JC, Meissner HP 1982 Opposite effects of tolbutamide and diazoxide on 86 Rb fluxes and membrane potential in pancreatic B-cells. Biochem Pharmacol 31:1407-1415 Gillis KD, Gee W M, Hammoud A, McDaniel ML, Falke LC,
I
399 19.
20.
286 6.
Flatt PR, Berggren PO, Gylfe E, Hellman BO 1980 Calcium and pancreatic B-cell function. IX. Demonstration of lanthanid-induced inhibition of insulin secretion independent of modifications in transmembrane Ca+ fluxes. Endocrinoloev 107:1007-1013 Henquin JC 1978 D-Glucose inhit% potassium efflux from pancreatic islet cells. Nature (Lond) 271:271-273 Henquin JC 1980 Tolbutamide stimulation and inhibition of insulin release: studies of the underlying ionic mechanisms in isolated rat islets. Diabetologia 18:151-160 Purrello F, Vetri M, Vinci C, Gatta C, Buscema M, Vigneri R 1990 Chronic exposure to high glucose and impairment of K’channel function in uerifused rat uancreatic islets. Diabetes 39:397-
21.
22.
23.
24.
25.
Gorus FK, Schuit FC, In’t Veld PA, Gepts W, Pipeleers DG 1988 Interaction of sulfonvlureas with uancreatic B-cells. A study with glyburide. Diabetes 37:1090-1094 1 Carpentier JL, Sawano F, Ravazzola M, Malaisse WJ 1986 Internalization of 3H-glibenclamide in pancreatic islet cells. Diabetologia 29:259-261 Panten U, Burgfeld J, Goerke F, Rennicke M, Schwanstecher M, Wallsch A, Zunkler BJ, Lenzen S 1989 Control of insulin secretion by sulfonylureas, meglitinide and diazoxide in relation to their binding to the sulfonylurea receptor in pancreatic islets. Biochem Pharmacol38:1217-1229 Sturgess NC, Koslowski RZ, Carrington CA, Hales CN, Ashford ML1 1988 Effects of sulfonvlureas and diazoxide on insulin secretion and nucleosidesensitive channels in an insulin-secreting cell line. Br J Pharmacol95:83-94 Best L, Malaisse WJ 1983 Stimulation of phosphoinositide breakdown in rat pancreatic islets by glucose and carbamylcholine. Biochem Biophys Res Commun 116:9-16 Rasmussen H, Zawalich KC, Ganesan S, Calle R, Zawalich WS 1990 Physiology and pathophysiology of insulin secretion. Diabetes Care 13~655-666 Verspohl EJ, Ammon HPT, Mark M 1990 Evidence for more than one binding site for sulfonylureas in insulin-secreting cells. J Pharm Pharmacol42:230-235
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 09 October 2016. at 21:40 For personal use only. No other uses without permission. . All rights reserved.
