Department of Endocrinology, sjukhuset, Stockholm
Karolinska
STUDIES ON THE MECHANISM OF SOMATOSTATIN ACTION ON INSULIN RELEASE IN MAN II.
Comparison of the effects of somatostatin on insulin induced by glucose, glucagon and tolbutamide S.
Efendi\l=c'\,
R.
release
By Luft and A. Claro
ABSTRACT as small a dose as 70 \g=m\g given over a period of 90 min healthy subjects inhibited insulin release induced by glucose (500 mg/kg as a bolus \m=+-\20 mg/kg/min). This inhibition seemed to be of competitive nature since the effect was nearly overcome when the glucose dose was raised considerably. Somatostatin in nine subjects also inhibited insulin release induced by glucagon and tolbutamide, and this inhibition was of the same order of magnitude as that of glucose induced insulin release. Since all these insulinogogues enhance the accumulation of cyclic AMP in the \g=b\-cells, it is suggested that the adenylate cyclase-cyclic AMP system might be
Somatostatin in
to
seven
involved in the action of somatostatin. Somatostatin did not seem to interfere with the glycogenolytic effect of glucagon on the liver.
Somatostatin,
in addition to its growth hormone release inhibition action, also inhibits the release of insulin and glucagon from the islets of Langerhans (Alberti et al. 1973; Mortimer et al. 1974; Efendic et al. 1974; Efendic 8c Luft 1975a,b). We have recently demonstrated that, most likely, somatostatin is produced also in the pancreas (Luft et al. 1974). All these data suggest a physiological role for somatostatin in the regulation of insulin release. This implies that somatostatin would be the second physiological compound having a direct inhibiting action on insulin release, the catecholamines being the first.
we demonstrated that the action of somatostatin on insulin unlike that of catecholamines, is not mediated by the a-adrenergic release, 8c Luft 1975b). In an attempt to further characterize the receptors (Efendic of of mode action somatostatin in man we have investigated its effect on the glucose-insulin dose-relationship in normal subjects and, furthermore com¬ pared its effects on insulin release induced by glucose, glucagon and tolbu¬ tamide.
In
a
recent paper
MATERIAL AND METHODS Informed consent was obtained from 13 volunteers, 8 men and 5 women, aged 24-45 years. None of them was obese, all had a normal iv glucose tolerance (lkkos 8c Luft 1957) and a normal insulin response to glucose infusion (Cerasi 8c Luft 1967). All subjects had been on a free diet containing about 300 g of carbohydrate daily. Studies were begun early in the morning after an overnight fast. The tests were performed with the subjects in recumbent position after a short period of rest. Teflon catheters were inserted into a superficial brachial vein of both arms and kept patent with a slow infusion of saline.
Effect of somatostatin on glucose-insulin dose-response The study was performed on seven subjects. In three series
of experiments the effect of different glucose loadings on insulin release was evaluated. Glucose was admini¬ stered as a rapid iv injection, followed by a constant infusion through a Bowman pump for 60 min. The following doses of glucose were used: 1) 250 mg per kg body weight as a rapid injection and 10 mg per kg per min as infusion; 2) 500 and 20 mg, respectively; 3) 1000 and 40 mg, respectively. A 25% glucose solution was used for the rapid injection, while the glucose concentrations of the infusions were adjusted to fit the flow rates of the pump. Blood was drawn into heparinized tubes from the opposite brachial vein 10 min and immediately before the beginning of the glucose load and 5, 10, 20, 30, 40, 50, 60, 80, 100 and 120 min after the start of the glucose
injection. Somatostatin was given as infusion followed by
glucose
Effect of
somatostatin and tolbutamide The study studies. The
on
a
an
priming
iv
injection (0.5 Hg/kg)
30 min before the kg per 90 min.
infusion of somatostatin, 0.5 fig per
insulin release induced
by glucose, glucagon
was performed on nine subjects, the same subjects, participating in all glucose infusions were performed as above, with 500 mg per kg body weight as an iv priming dose and with 20 mg/kg per min for 60 min as an infusion. One mg of glucagon (Glucagon Lilly®) was injected over 1 min, and 1 g of tolbuta¬ mide (Rastinon® Hoechst) iv in a 20% solution over 3 min. In the glucagon and tolbutamide tests blood sampling was performed 10 min prior to and immediately before the injection, and 5, 10, 20, 30, 40, 50 and 60 min afterwards. Somatostatin was introduced 30 min before the insulinogogues as a bolus (1.0/;g/kg body weight), followed by a 90 min infusion at a constant rate of 2 [ag per kg per
90 min.
600
-,
-40
0 I
40
80
120 -40
g'ùcoie I
0
40
80
120 -40
0
40
80
I glucose
120 min
Fig. 1. glucose induced insulin release. Somatostatin was given glucose between 0 and 60 min. Three different doses of were glucose given (see Methods). Filled circles and solid lines denote control experi¬ ments, open circles and broken lines those with somatostatin. Mean + sem of 7 experi¬ ments. The significance of the paired differences between control and somatostatin experiments is denoted by: * P. In 150 ~"
!
EARLY
RESPONSE
g.
/"II f7l
T/
OJ
I
«H
LATE RESPONSE
f
/H
100-
/
1
/
/]
v_
B
6
-
•£ 5"
/
t
/£
^
Í'
E
'
Ä
3
£ 2 '
i
i 0J
i-1-1-1
100
200
400
-1
i-1-1-1
800 100
200
400
800
glucose mg/100ml Modification of the
TOTAL RESPONSE
0 -I
/ I?"
/
X
i-1-1-1
100
200
400
800
glucose (mgx min xmi')
Fig. 2. glucose-insulin dose-response relationship by
somatostatin.
Early
response corresponds to the 10-min values, late response to the 60-min ones of Fig. 1. Total response relates the 0-60 min integrated incremental plasma insulin values to the integrated blood glucose concentration. Symbols as for Fig. 1.
1
400
o o
300 3
200
100
|
150
a.
c
100
50
H
h-1-r
-40 -20
0
20 40 60 80 100 120
somatostatin
Fig.
3.
Effect of somatostatin on glucose induced insulin release (for doses, see Methods). Filled circles and solid lines represent control experiments, open circles and broken lines experiments with somatostatin. Mean ± sem of 9 experiments. The significance of the paired differences between control and somatostatin experiments are denoted by: * P < 0.05, ** P < 0.01 and **» P < 0.001. -
=
=
glucagon experiments, the same dose of somatostatin reduced the insulin response by about 45 % (Fig. 4, Table 1). Somatostatin seemed not to modify the glycogenolytic effect of glucagon since the increase in blood glucose was even more pronounced in the experiments where glucagon was combined with the
somatostatin. The effect of somatostatin on tolbutamide induced insulin release was similar to that described for the glucagon experiments, the inhibition amounting to
(Fig. 5, Table 1). It is obvious from Table 1 that the response to the
49%
insulinogogues
used
bO
C
+1
u
Ci
Ö o 00
I
IT) tÛ ÎO r-I OcoOîi—
—i n CM if) r- »O
+1
o
co
^H
CO CO Ol 'í CO Q
CO
CO
O »O OÍ
r-
w
(N (N
+1
Ci
+1 -tí
s.
Ü
*-o»joolo»oooiooooo
OlOOt--CN^eoO
+1
OlCOQOtMtûQtOOOfj
a
o
ho
03
CJ
3
Ü r-M-}QOOOOOCOOO»-000
oOOO(MCNCOOl(H^O ,_,
_|
_
,_,
CM
+1
(M
ho
+1
G
U
3
ooooooooo ifjCTj'HOOCMCN'-iN'-i
cooiNtDHinooco ,—i i—(i—i
Karolinska
STUDIES ON THE MECHANISM OF SOMATOSTATIN ACTION ON INSULIN RELEASE IN MAN II.
Comparison of the effects of somatostatin on insulin induced by glucose, glucagon and tolbutamide S.
Efendi\l=c'\,
R.
release
By Luft and A. Claro
ABSTRACT as small a dose as 70 \g=m\g given over a period of 90 min healthy subjects inhibited insulin release induced by glucose (500 mg/kg as a bolus \m=+-\20 mg/kg/min). This inhibition seemed to be of competitive nature since the effect was nearly overcome when the glucose dose was raised considerably. Somatostatin in nine subjects also inhibited insulin release induced by glucagon and tolbutamide, and this inhibition was of the same order of magnitude as that of glucose induced insulin release. Since all these insulinogogues enhance the accumulation of cyclic AMP in the \g=b\-cells, it is suggested that the adenylate cyclase-cyclic AMP system might be
Somatostatin in
to
seven
involved in the action of somatostatin. Somatostatin did not seem to interfere with the glycogenolytic effect of glucagon on the liver.
Somatostatin,
in addition to its growth hormone release inhibition action, also inhibits the release of insulin and glucagon from the islets of Langerhans (Alberti et al. 1973; Mortimer et al. 1974; Efendic et al. 1974; Efendic 8c Luft 1975a,b). We have recently demonstrated that, most likely, somatostatin is produced also in the pancreas (Luft et al. 1974). All these data suggest a physiological role for somatostatin in the regulation of insulin release. This implies that somatostatin would be the second physiological compound having a direct inhibiting action on insulin release, the catecholamines being the first.
we demonstrated that the action of somatostatin on insulin unlike that of catecholamines, is not mediated by the a-adrenergic release, 8c Luft 1975b). In an attempt to further characterize the receptors (Efendic of of mode action somatostatin in man we have investigated its effect on the glucose-insulin dose-relationship in normal subjects and, furthermore com¬ pared its effects on insulin release induced by glucose, glucagon and tolbu¬ tamide.
In
a
recent paper
MATERIAL AND METHODS Informed consent was obtained from 13 volunteers, 8 men and 5 women, aged 24-45 years. None of them was obese, all had a normal iv glucose tolerance (lkkos 8c Luft 1957) and a normal insulin response to glucose infusion (Cerasi 8c Luft 1967). All subjects had been on a free diet containing about 300 g of carbohydrate daily. Studies were begun early in the morning after an overnight fast. The tests were performed with the subjects in recumbent position after a short period of rest. Teflon catheters were inserted into a superficial brachial vein of both arms and kept patent with a slow infusion of saline.
Effect of somatostatin on glucose-insulin dose-response The study was performed on seven subjects. In three series
of experiments the effect of different glucose loadings on insulin release was evaluated. Glucose was admini¬ stered as a rapid iv injection, followed by a constant infusion through a Bowman pump for 60 min. The following doses of glucose were used: 1) 250 mg per kg body weight as a rapid injection and 10 mg per kg per min as infusion; 2) 500 and 20 mg, respectively; 3) 1000 and 40 mg, respectively. A 25% glucose solution was used for the rapid injection, while the glucose concentrations of the infusions were adjusted to fit the flow rates of the pump. Blood was drawn into heparinized tubes from the opposite brachial vein 10 min and immediately before the beginning of the glucose load and 5, 10, 20, 30, 40, 50, 60, 80, 100 and 120 min after the start of the glucose
injection. Somatostatin was given as infusion followed by
glucose
Effect of
somatostatin and tolbutamide The study studies. The
on
a
an
priming
iv
injection (0.5 Hg/kg)
30 min before the kg per 90 min.
infusion of somatostatin, 0.5 fig per
insulin release induced
by glucose, glucagon
was performed on nine subjects, the same subjects, participating in all glucose infusions were performed as above, with 500 mg per kg body weight as an iv priming dose and with 20 mg/kg per min for 60 min as an infusion. One mg of glucagon (Glucagon Lilly®) was injected over 1 min, and 1 g of tolbuta¬ mide (Rastinon® Hoechst) iv in a 20% solution over 3 min. In the glucagon and tolbutamide tests blood sampling was performed 10 min prior to and immediately before the injection, and 5, 10, 20, 30, 40, 50 and 60 min afterwards. Somatostatin was introduced 30 min before the insulinogogues as a bolus (1.0/;g/kg body weight), followed by a 90 min infusion at a constant rate of 2 [ag per kg per
90 min.
600
-,
-40
0 I
40
80
120 -40
g'ùcoie I
0
40
80
120 -40
0
40
80
I glucose
120 min
Fig. 1. glucose induced insulin release. Somatostatin was given glucose between 0 and 60 min. Three different doses of were glucose given (see Methods). Filled circles and solid lines denote control experi¬ ments, open circles and broken lines those with somatostatin. Mean + sem of 7 experi¬ ments. The significance of the paired differences between control and somatostatin experiments is denoted by: * P. In 150 ~"
!
EARLY
RESPONSE
g.
/"II f7l
T/
OJ
I
«H
LATE RESPONSE
f
/H
100-
/
1
/
/]
v_
B
6
-
•£ 5"
/
t
/£
^
Í'
E
'
Ä
3
£ 2 '
i
i 0J
i-1-1-1
100
200
400
-1
i-1-1-1
800 100
200
400
800
glucose mg/100ml Modification of the
TOTAL RESPONSE
0 -I
/ I?"
/
X
i-1-1-1
100
200
400
800
glucose (mgx min xmi')
Fig. 2. glucose-insulin dose-response relationship by
somatostatin.
Early
response corresponds to the 10-min values, late response to the 60-min ones of Fig. 1. Total response relates the 0-60 min integrated incremental plasma insulin values to the integrated blood glucose concentration. Symbols as for Fig. 1.
1
400
o o
300 3
200
100
|
150
a.
c
100
50
H
h-1-r
-40 -20
0
20 40 60 80 100 120
somatostatin
Fig.
3.
Effect of somatostatin on glucose induced insulin release (for doses, see Methods). Filled circles and solid lines represent control experiments, open circles and broken lines experiments with somatostatin. Mean ± sem of 9 experiments. The significance of the paired differences between control and somatostatin experiments are denoted by: * P < 0.05, ** P < 0.01 and **» P < 0.001. -
=
=
glucagon experiments, the same dose of somatostatin reduced the insulin response by about 45 % (Fig. 4, Table 1). Somatostatin seemed not to modify the glycogenolytic effect of glucagon since the increase in blood glucose was even more pronounced in the experiments where glucagon was combined with the
somatostatin. The effect of somatostatin on tolbutamide induced insulin release was similar to that described for the glucagon experiments, the inhibition amounting to
(Fig. 5, Table 1). It is obvious from Table 1 that the response to the
49%
insulinogogues
used
bO
C
+1
u
Ci
Ö o 00
I
IT) tÛ ÎO r-I OcoOîi—
—i n CM if) r- »O
+1
o
co
^H
CO CO Ol 'í CO Q
CO
CO
O »O OÍ
r-
w
(N (N
+1
Ci
+1 -tí
s.
Ü
*-o»joolo»oooiooooo
OlOOt--CN^eoO
+1
OlCOQOtMtûQtOOOfj
a
o
ho
03
CJ
3
Ü r-M-}QOOOOOCOOO»-000
oOOO(MCNCOOl(H^O ,_,
_|
_
,_,
CM
+1
(M
ho
+1
G
U
3
ooooooooo ifjCTj'HOOCMCN'-iN'-i
cooiNtDHinooco ,—i i—(i—i
