- Email: [email protected]
Effects of prostacyclin and its stable analog, iloprost, upon insulin secretion in isolated pancreatic islets
PROSTAGLANDINS
EFFECTS OF PROSTACYCLIN AND ITS STABLE ANALOG, ILOPROST, UPON INSULIN SECRETION IN ISOLATED PANCREATIC ISLETS 3.
Sieradzki,
Institute of of Medicine,
H.
Wolan,
and A.
Szczeklik
Internal Medicine, Copernicus Kopernika 17, 31-501 Krakow,
Academy Poland
ABSTRACT We investigated the effects of prostacyclin in three concentrations of 2.7 nM, 53.8 nM and 267 nM on insulin release by isolated rat islets. The lowest concentration produced no significant effects. The middle concentration led to stimulation followed by inhibition of the reaction studied, while the highest concentration strongly depressed insulin release. These effects of prostacy.clin appeared to be specific, because they were mimicked by its stable analog, iloprost; but not by its metabolite, suggest that 6-keto-PGFla . The results in islets might exert locally prostacyclin generated an influence upon insulin release. INTRODUCTION There are two reasons to study the effects of prostacyclin on insulin secretion. Firstly, islets endocrine cells synthesize prostacyclin along with other cyclooxygenase products, e.g. PGE2 or PGF2& /l/. In these islets metabolism of arachidonic acid via the cyclooxygenass pathway reduces glucose-stimulated insulin while metabolism via lipoxygenase pathway has release, opposing effects /2/. Secondly, infusion of exogenous prostacyclin to man results in mild hyperglycemia /3,4/. While the hyperglycemic effects do no.t interfere with the usual therapeutic administration.of prostacyclin might constitute a risk in a nevertheless, /5/, they patient with poorly controlled diabetes. MATERIALS
AND METHOD
The experiments were made on male Wistar rats with the mean body we,ight approximating 180-220 g. The rats, were killed by decapitation. Pancrsatecfed murigran, tomy was performed and the pancreas was rinsed with the the temperature of which was O’C. Krebs-Ringer buffer, The islets were isolated according to Lacy and Kostyanowski /6/ using Serva /Heidelberg/ collagenase in the amount of approximately 10 mg per 3 cc of the frag,ment-
SEPTEMBER
1984 VOL. 28 NO. 3
PROSTAGLANDINS
ed pancreas. The incubation was performed for 10 minutes at the temperature of 37OC as described previously /7/. The islets were separated from the exocrine tissue by three consecutive centrifugations, and were subsequently isolated using a Pasteur’s pipet. islets were incubated at The isolated pencreatic buffer pH temperature of 37 C in 20 ml of Krebs-Ringer 7.4, to which glucose and prostacyclin /The Upjohn Co., Kalamazoo, Mi, USA and the Wellcome Research Labs, The concentrations Beckenham, Kent, UK/ were added. of glucose were 2.8 mM/l /50 mg/dl/ and 16.7 mM/l /300 mg/dl/, while that of prostacyclin were 2.7 nM, 53.8 nM and 267 nM. In another experiment the islets were incubated in the presence of glucose at a concentration of 16.7 mmol/l with a carboprostacyclin derivative /Iloat concentrations of prost, Schering A.G., West Berlin/ 52 nM and 262 nM. Experiments were also performed in which the islets were incubated with 6-keto-PGFl, metabolite of prostacyclin, /Upjohn Co., USA/, a stable at a concentration of 270 nM. The samples for insulin level determinations were collected prior to and at 30, 60 and 90 minutes of incubation. The samples were freezed at -18OC, and then subjected to radioimmunological determinations of insulin with the use of INS-SET manufactured by OPiDI, Swierk near Warsaw. Student’s t-test was use in statistical analysis.Data are presented as mean 2 standard error. RESULTS The Krebs-Ringer buffer pH 8.5 in the amount of 20 )~l had no effect upon insulin secretion. Glucose stimulated release of insulin. Ninety minutes incubation of islets with glucose at the concentrations of 2.0 nM and 16.7 mM produced a two-fold and ten-fold increase in insulin, Prostacyclin at concentration of respectively. 2.7 nM showed tendency to decrease insulin release at both concentrations of glucose used. These changes, however, did not reach statistical significance. Prostacyclin at the concentration of 53.8 nM produced a dual response, i.e. a stimulation of insulin secretion by 30th minute and an inhibition by 90th minute. These results were significant when high,but not,low con267 nM centrations of glucose were used. Finally, solution of PG12 produced significant inhibition of insulin release /Table l/. When the islets were incubated nM glucose and 52 nM iloprost the significantly /p(O.OOl/ increased
290
in the presence of 16 insulin release was by 30th and 60th
SEPTEMBER1984 VOL. 28 NO. 3
$
W
b
p
7.9 20.3
34.0 +0.8
9.4 20.5
16.8 +2.3
30’
9.4a +0.6
12.7 _ +2.0
11.9 ~0.6
21.4 ,+ 1.5
60’
2.8
Release
/PG12/
Glucose
to glucose
7.4 +0.3
267.0
compared
6.9 +0.8
53.8
0.01
7.8 +0.3
2.7
ap<
8.9 +0.4
0’
Effect of Prostacyclin on the Glucose-Induced /Mean ,+ SEM/
Concentration /nM/
1.
0.0
PG12
Table
90’
Three Insulin
9.6a ~0.6
12.0 21.4
3.7 +0.9
21.4 21.6
mM
at of
8.6 20.3
5.9 +0.9
7.8 +0.3
9.7 ~0.4
0’
9.7 20.4
55.8b _ +2.1
9.9 +0.4 _
25.1 22.6
30’
Glucose
Concentrations by Pancreatic
10.9 +0.5 _
36.9 22.8
20.2 ~1.8
mM
Islets
34.7 22.0
60’
16.7
Rat
11. gb 20.8
40.0b 2 1.7
86.5 23.9
99.5 24.5
90’
PROSTAGLANDINS
minute, and significantly /p
produced no significant /Table 2/.
changes
in
262nM ILOPROST n= I4
L-l---i0 30 1. Fig. in 16.7 iloprost.
292
Insulin release mM glucose and
90
90
min
by pancreatic islets incubated different concenfrations.of
SEPTEMBER
1984 VOL. 28 NO. 3
2.
270
0
nM
6-keto-PGFlu(
Table
11.3 +0.6
8.9 +o. 4
0’
10.1 20.7
17.1 +l.l
15.7 +l.l
15.1 21.2
0’
9.7 +0.4
90’
21.4 ~1.6
60’
mM
93.0 +4.2
42.2 +3.6 18.2 +1.6
90’
99.5 +4.5
mM
34.7 +2.0
60’
16.7
or
25.4 +2.6
30’
Glucose
Pancreatic Islets of Glucose with
21.4 +1.5
2.0
Release by Isolated of Two Concentrations /Mean ,+ SEM/
16.8 +2.3
30’
Glucose
Comparison of Insulin Incubated in Presence without 6-keto-PGFl&
PROSTAGLANDINS
DISCUSSION
Under the experimental conditions used, prostacyclin showed the whole spectrum of effects upon insulin ranging from stimulation to inhibition. These release, effects were concentration-related, and appeared to be specific for chemical structure of prostacyclin,because: 1/ they were mimicked by a stable analog of prostacyclin, iloprost; 2/ they could not be reproduced by a prostacyclin metabolite, 6-keto-PGFa . It is noteworthy that in rabbits infusion of 6-keto-PGFiti contrary to prostacyclin, did not change blood glucose levels /4/. It ight be interesting to compare the actions of with those of prostaglandin of the E series, prosta $ clin implicated in modulation of insulin secretion in adultFollowing in onset /type II/ diabetes mellitus /a/. vitro studies with PGEs, different investigators have sometimes formed opposite conclusions /for ref. see 8, 9/. When stimulation was provided by a sudden increment in glucose concentration, inhibition of acute insulin response has been usually observed /a/. However, secretion of insulin by perfused rat pancreas was not modified by either PGE2 /9/ or PG12 /lo/. More relevant to human pathophysiology, observations carried out in man revealed differences between PGE2 and PGI While both prostanoids evoked hyperglycemia when in ?’used intravenously /3, 4, ll/, only PGE2 diminished consistently glucose-induced insulin release. Prostacyclin repressed acute insulin response to glucose pulse in some non-diabetic patients, but in others it either increased it or did not affect it at all /3,12/. The lowest concentrations of PG12 used in experiments here reported approach those achieved in blood of patients receiving PG12 therapy /3_3/. These concentrations have not produced significant effects upon insulin release in rat isolated islets. Such result corresponds to clinical observations and indicate that in most patients treated with prostacyclin reasons other than inhibition of insulin release have to be seek for explanation of hyperglycemia. On the other hand, our results suggest that prostacyclin generated in islets /i, 2/ might exert locally an influence upon the insulin release. REFERENCES
1.
294
Kelly, thesis of the
and Laychock, and metabolism in rat. Prostaglandins
K.I.,
S.G. Prostaglandin isolated pancreatic =:759, 1981.
synislets
SEPTEMBER 1984 VOL. 28 NO. 3
\
PROSTAGLANDINS
2.
M. VanRollins, R. Striff, W. Fujimoto, and Metz, S., Lipoxygenase pathway in islet endoR. P. Robertson. crine cells. Oxidative metabolism of arachidonic acid promotes insulin release. J. Clin. Invest. 2: 1191, 1983.
3.
Szczeklik, A., R. Pieton, 3. Sieradzki, and R.Nitankowski. The effects of prostacyclin on glycemia and insulin release in man. Prostaglandins g:959, 1980.
4.
Dembiriska-Kied, A., E. Kostka-Trgbka, L. Grodzinska, A. tmuda, K. Bieron, A. Kedzior, W. Ochmanski, and Prostacyclin and blood glucose levels T. telazny. in humans and rabbits. Prostaglandins =:113, 1981.
5.
and R.J. Gryglewski. Stczeklik, A., Prostaglandins, and Cardiovascular Samuelsson and G.G. Neri Serneri, New York, in press.
6.
Lacy, P.E., and M. isolation of intact g:35, 1967.
7.
Sieradzki, J., H. Wanat, and A. Grtymalska. Studies on isolated Langerhans’ islets. I. Method of islets isolation and study of insulin secretion /in Polish/. Endokr. Pol. 31:403, 1980.
8.
Robertson, R.P. Prostaglandins, and diabetes mellitus. Ann.
9.
In: Platelets, System. /B. ads./ Raven Press,
Kostyanowsky. Method islets of Langerhans.
Rev.
for
the Diabetes
glucose homeostasis, Med. E:l, 1983.
Landgraf, R., and M.M.C. Landgraf-Leurs. The prostaglandin system and insulin release. Studies with the isolated perfused rat pancreas. Prostaglandins J7_: 599, 1979.
10. Akpan,
J.O., M.C. Hurley, S. Pek, and W.E.M. Lands. The effects of prostaglandins on secretion of glucagon and insulin by the perfused rat pancreas. Canad. J. Biochem. 57:540, 1979.
11. Robertson,
R.P., and M. in E in defective insulin intolerance in diabetes %:747, 1977.
Chen. A role for prostaglandsecretion and carbohydrate mellitus. J. Clin. Invest.
12. Patrono,
C., F. Pugliese, G. Ciabattoni, S. Di Blasi, A. Pierucci, G.A. Cinotti, A. Maseri, and S.Chierchia, Prostacyclin does not affect insulin secretion in humans. Prostaglandins %:379, 1981.
SEPTEMBER
1984 VOL. 28 NO. 3
295
PROSTAGLANDINS
13. Cool<, P.J.,
C.G. Maidment, P. Dandone, A.A. Hutton, and I.M. James. The effect of intravenous epoprostenol /prostacyclin, PG12/ on cerebral blood flow and cardiac outout in man. St-. J. Clin.Pharmac. 2:707, 1983.
Editor: Richard Gryglewski
296
SEPTEMBER
1984 VOL. 28 NO. 3
EFFECTS OF PROSTACYCLIN AND ITS STABLE ANALOG, ILOPROST, UPON INSULIN SECRETION IN ISOLATED PANCREATIC ISLETS 3.
Sieradzki,
Institute of of Medicine,
H.
Wolan,
and A.
Szczeklik
Internal Medicine, Copernicus Kopernika 17, 31-501 Krakow,
Academy Poland
ABSTRACT We investigated the effects of prostacyclin in three concentrations of 2.7 nM, 53.8 nM and 267 nM on insulin release by isolated rat islets. The lowest concentration produced no significant effects. The middle concentration led to stimulation followed by inhibition of the reaction studied, while the highest concentration strongly depressed insulin release. These effects of prostacy.clin appeared to be specific, because they were mimicked by its stable analog, iloprost; but not by its metabolite, suggest that 6-keto-PGFla . The results in islets might exert locally prostacyclin generated an influence upon insulin release. INTRODUCTION There are two reasons to study the effects of prostacyclin on insulin secretion. Firstly, islets endocrine cells synthesize prostacyclin along with other cyclooxygenase products, e.g. PGE2 or PGF2& /l/. In these islets metabolism of arachidonic acid via the cyclooxygenass pathway reduces glucose-stimulated insulin while metabolism via lipoxygenase pathway has release, opposing effects /2/. Secondly, infusion of exogenous prostacyclin to man results in mild hyperglycemia /3,4/. While the hyperglycemic effects do no.t interfere with the usual therapeutic administration.of prostacyclin might constitute a risk in a nevertheless, /5/, they patient with poorly controlled diabetes. MATERIALS
AND METHOD
The experiments were made on male Wistar rats with the mean body we,ight approximating 180-220 g. The rats, were killed by decapitation. Pancrsatecfed murigran, tomy was performed and the pancreas was rinsed with the the temperature of which was O’C. Krebs-Ringer buffer, The islets were isolated according to Lacy and Kostyanowski /6/ using Serva /Heidelberg/ collagenase in the amount of approximately 10 mg per 3 cc of the frag,ment-
SEPTEMBER
1984 VOL. 28 NO. 3
PROSTAGLANDINS
ed pancreas. The incubation was performed for 10 minutes at the temperature of 37OC as described previously /7/. The islets were separated from the exocrine tissue by three consecutive centrifugations, and were subsequently isolated using a Pasteur’s pipet. islets were incubated at The isolated pencreatic buffer pH temperature of 37 C in 20 ml of Krebs-Ringer 7.4, to which glucose and prostacyclin /The Upjohn Co., Kalamazoo, Mi, USA and the Wellcome Research Labs, The concentrations Beckenham, Kent, UK/ were added. of glucose were 2.8 mM/l /50 mg/dl/ and 16.7 mM/l /300 mg/dl/, while that of prostacyclin were 2.7 nM, 53.8 nM and 267 nM. In another experiment the islets were incubated in the presence of glucose at a concentration of 16.7 mmol/l with a carboprostacyclin derivative /Iloat concentrations of prost, Schering A.G., West Berlin/ 52 nM and 262 nM. Experiments were also performed in which the islets were incubated with 6-keto-PGFl, metabolite of prostacyclin, /Upjohn Co., USA/, a stable at a concentration of 270 nM. The samples for insulin level determinations were collected prior to and at 30, 60 and 90 minutes of incubation. The samples were freezed at -18OC, and then subjected to radioimmunological determinations of insulin with the use of INS-SET manufactured by OPiDI, Swierk near Warsaw. Student’s t-test was use in statistical analysis.Data are presented as mean 2 standard error. RESULTS The Krebs-Ringer buffer pH 8.5 in the amount of 20 )~l had no effect upon insulin secretion. Glucose stimulated release of insulin. Ninety minutes incubation of islets with glucose at the concentrations of 2.0 nM and 16.7 mM produced a two-fold and ten-fold increase in insulin, Prostacyclin at concentration of respectively. 2.7 nM showed tendency to decrease insulin release at both concentrations of glucose used. These changes, however, did not reach statistical significance. Prostacyclin at the concentration of 53.8 nM produced a dual response, i.e. a stimulation of insulin secretion by 30th minute and an inhibition by 90th minute. These results were significant when high,but not,low con267 nM centrations of glucose were used. Finally, solution of PG12 produced significant inhibition of insulin release /Table l/. When the islets were incubated nM glucose and 52 nM iloprost the significantly /p(O.OOl/ increased
290
in the presence of 16 insulin release was by 30th and 60th
SEPTEMBER1984 VOL. 28 NO. 3
$
W
b
p
7.9 20.3
34.0 +0.8
9.4 20.5
16.8 +2.3
30’
9.4a +0.6
12.7 _ +2.0
11.9 ~0.6
21.4 ,+ 1.5
60’
2.8
Release
/PG12/
Glucose
to glucose
7.4 +0.3
267.0
compared
6.9 +0.8
53.8
0.01
7.8 +0.3
2.7
ap<
8.9 +0.4
0’
Effect of Prostacyclin on the Glucose-Induced /Mean ,+ SEM/
Concentration /nM/
1.
0.0
PG12
Table
90’
Three Insulin
9.6a ~0.6
12.0 21.4
3.7 +0.9
21.4 21.6
mM
at of
8.6 20.3
5.9 +0.9
7.8 +0.3
9.7 ~0.4
0’
9.7 20.4
55.8b _ +2.1
9.9 +0.4 _
25.1 22.6
30’
Glucose
Concentrations by Pancreatic
10.9 +0.5 _
36.9 22.8
20.2 ~1.8
mM
Islets
34.7 22.0
60’
16.7
Rat
11. gb 20.8
40.0b 2 1.7
86.5 23.9
99.5 24.5
90’
PROSTAGLANDINS
minute, and significantly /p
produced no significant /Table 2/.
changes
in
262nM ILOPROST n= I4
L-l---i0 30 1. Fig. in 16.7 iloprost.
292
Insulin release mM glucose and
90
90
min
by pancreatic islets incubated different concenfrations.of
SEPTEMBER
1984 VOL. 28 NO. 3
2.
270
0
nM
6-keto-PGFlu(
Table
11.3 +0.6
8.9 +o. 4
0’
10.1 20.7
17.1 +l.l
15.7 +l.l
15.1 21.2
0’
9.7 +0.4
90’
21.4 ~1.6
60’
mM
93.0 +4.2
42.2 +3.6 18.2 +1.6
90’
99.5 +4.5
mM
34.7 +2.0
60’
16.7
or
25.4 +2.6
30’
Glucose
Pancreatic Islets of Glucose with
21.4 +1.5
2.0
Release by Isolated of Two Concentrations /Mean ,+ SEM/
16.8 +2.3
30’
Glucose
Comparison of Insulin Incubated in Presence without 6-keto-PGFl&
PROSTAGLANDINS
DISCUSSION
Under the experimental conditions used, prostacyclin showed the whole spectrum of effects upon insulin ranging from stimulation to inhibition. These release, effects were concentration-related, and appeared to be specific for chemical structure of prostacyclin,because: 1/ they were mimicked by a stable analog of prostacyclin, iloprost; 2/ they could not be reproduced by a prostacyclin metabolite, 6-keto-PGFa . It is noteworthy that in rabbits infusion of 6-keto-PGFiti contrary to prostacyclin, did not change blood glucose levels /4/. It ight be interesting to compare the actions of with those of prostaglandin of the E series, prosta $ clin implicated in modulation of insulin secretion in adultFollowing in onset /type II/ diabetes mellitus /a/. vitro studies with PGEs, different investigators have sometimes formed opposite conclusions /for ref. see 8, 9/. When stimulation was provided by a sudden increment in glucose concentration, inhibition of acute insulin response has been usually observed /a/. However, secretion of insulin by perfused rat pancreas was not modified by either PGE2 /9/ or PG12 /lo/. More relevant to human pathophysiology, observations carried out in man revealed differences between PGE2 and PGI While both prostanoids evoked hyperglycemia when in ?’used intravenously /3, 4, ll/, only PGE2 diminished consistently glucose-induced insulin release. Prostacyclin repressed acute insulin response to glucose pulse in some non-diabetic patients, but in others it either increased it or did not affect it at all /3,12/. The lowest concentrations of PG12 used in experiments here reported approach those achieved in blood of patients receiving PG12 therapy /3_3/. These concentrations have not produced significant effects upon insulin release in rat isolated islets. Such result corresponds to clinical observations and indicate that in most patients treated with prostacyclin reasons other than inhibition of insulin release have to be seek for explanation of hyperglycemia. On the other hand, our results suggest that prostacyclin generated in islets /i, 2/ might exert locally an influence upon the insulin release. REFERENCES
1.
294
Kelly, thesis of the
and Laychock, and metabolism in rat. Prostaglandins
K.I.,
S.G. Prostaglandin isolated pancreatic =:759, 1981.
synislets
SEPTEMBER 1984 VOL. 28 NO. 3
\
PROSTAGLANDINS
2.
M. VanRollins, R. Striff, W. Fujimoto, and Metz, S., Lipoxygenase pathway in islet endoR. P. Robertson. crine cells. Oxidative metabolism of arachidonic acid promotes insulin release. J. Clin. Invest. 2: 1191, 1983.
3.
Szczeklik, A., R. Pieton, 3. Sieradzki, and R.Nitankowski. The effects of prostacyclin on glycemia and insulin release in man. Prostaglandins g:959, 1980.
4.
Dembiriska-Kied, A., E. Kostka-Trgbka, L. Grodzinska, A. tmuda, K. Bieron, A. Kedzior, W. Ochmanski, and Prostacyclin and blood glucose levels T. telazny. in humans and rabbits. Prostaglandins =:113, 1981.
5.
and R.J. Gryglewski. Stczeklik, A., Prostaglandins, and Cardiovascular Samuelsson and G.G. Neri Serneri, New York, in press.
6.
Lacy, P.E., and M. isolation of intact g:35, 1967.
7.
Sieradzki, J., H. Wanat, and A. Grtymalska. Studies on isolated Langerhans’ islets. I. Method of islets isolation and study of insulin secretion /in Polish/. Endokr. Pol. 31:403, 1980.
8.
Robertson, R.P. Prostaglandins, and diabetes mellitus. Ann.
9.
In: Platelets, System. /B. ads./ Raven Press,
Kostyanowsky. Method islets of Langerhans.
Rev.
for
the Diabetes
glucose homeostasis, Med. E:l, 1983.
Landgraf, R., and M.M.C. Landgraf-Leurs. The prostaglandin system and insulin release. Studies with the isolated perfused rat pancreas. Prostaglandins J7_: 599, 1979.
10. Akpan,
J.O., M.C. Hurley, S. Pek, and W.E.M. Lands. The effects of prostaglandins on secretion of glucagon and insulin by the perfused rat pancreas. Canad. J. Biochem. 57:540, 1979.
11. Robertson,
R.P., and M. in E in defective insulin intolerance in diabetes %:747, 1977.
Chen. A role for prostaglandsecretion and carbohydrate mellitus. J. Clin. Invest.
12. Patrono,
C., F. Pugliese, G. Ciabattoni, S. Di Blasi, A. Pierucci, G.A. Cinotti, A. Maseri, and S.Chierchia, Prostacyclin does not affect insulin secretion in humans. Prostaglandins %:379, 1981.
SEPTEMBER
1984 VOL. 28 NO. 3
295
PROSTAGLANDINS
13. Cool<, P.J.,
C.G. Maidment, P. Dandone, A.A. Hutton, and I.M. James. The effect of intravenous epoprostenol /prostacyclin, PG12/ on cerebral blood flow and cardiac outout in man. St-. J. Clin.Pharmac. 2:707, 1983.
Editor: Richard Gryglewski
296
SEPTEMBER
1984 VOL. 28 NO. 3