The Effect on Gastrin Secretion of Agents which Increase the Intracellular Concentration of 3',5'-Adenosine Monophosphate J. R. HAYES* AND R. H. WILLIAMS Division of Endocrinology, Washington 98195
Department
of Medicine,
University
of Washington,
Seattle,
but each enhanced arginine-stimulated gastrin release. A biphasic pattern of gastrin release in response to arginine was seen in all experiments. The studies emphasize the close functional similarity between the antral G cells and the B cells of the pancreatic islet. (Endocrinology 97: 1210, 1975)
ABSTRACT. An in vitro technique that allows study of gastrin secretion from isolated pieces of rat gastric antrum was used to study the effect on gastrin release of agents known to increase intracellular cAMP levels. Isuprel (10- 5 M) ? PGE, (10" 5 M), theophylline (IO^M), and dibutyryl cAMP (5 x 10~4M) did not affect gastrin release when used alone,
T
HE G cell of the antrum ex- an in vitro technique was employed that hibits morphological and histochem- allows the study of gastrin release into ical similarity with the B cell of the pan- perifusion medium from isolated pieces of creatic islet so that both may be classified gastric antrum. in the APUD series (1). In addition, there is Materials and Methods evidence to suggest that these cells share a common embryologic. origin (2). Extensive study of the B cell has established that Perifusion Technique glucose-stimulated insulin secretion is Male Wistar rats weighing 150-200 g were potentiated by agents which increase used in all experiments. The rats were killed by the intracellular concentration of 3',5'- decapitation and small pieces of tissue 5 mm in adenosine monophosphate (cAMP) (3,4). width and approximately 0.8 g wet wt were The present study investigates the possibil- removed from the antrum of the stomach. The ity that similar mechanisms may be in- pylorus marked the distal border of each piece. volved in the control of gastrin ,'secretion. Each piece was washed briefly and placed on a We have investigated the effect pn gastrin wire grid in a perfusion chamber with a capacity release of isoproterenol (Isuprel), prosta- of 0.7 ml (Fig. 1). The basic buffer used was glandin Ej (PGEj), theophylline, and di- synthetic interstitial fluid (SIF) (11) containing 1 butyryl cAMP, all of which are known to mg/ml glucose. The buffer, which was gassed continually with 95% O :5% CO2, maintained a elevate the intracellular level of cAMP pH of 7.35-7.45 and 37 2C throughout the exper(5,6). These agents have also been shown iment. A peristaltic pump passed buffer conto influence gastric parietal cells (7-10). tained in reservoir flasks through the perifusion Since gastrin release may be altered by chamber at a flow rate of 0.4 ml/min. A threechanges in gastric pH produced by parietal way tap at the base of the perifusion chamber cell secretion, it was felt that the interpreta- allowed the perifusion medium to be changed tion of in vivo experiments on gastrin easily. The effluent was collected in glass tubes release would be difficult. In these studies in a fraction collector and stored at - 2 0 C until Received March 5, 1975. * Present address: Department of Medicine, Queen's University, Belfast, Northern Ireland. Isuprel, theophylline and dibutyryl cAMP were obtained from the Sigma Co. PGE, was obtained from the Upjohn Company.
assayed. Preliminary experiments established that gastrin release from the antrum of fed animals was greater and more reproducible than that from animals fasted 18 hours. Similar observations have been made regarding insulin release from perifused pancreatic pieces (12). In all experi-
1210
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 07:02 For personal use only. No other uses without permission. . All rights reserved.
STUDIES IN GASTRIN SECRETION ments reported here, tissues were taken from animals fasted for 3-4 hours, at which time the stomach was empty. It was also found that gastrin secretion did not depend on the weight of the antral tissue within the narrow range of weights used in these experiments. In the initial period of perifusion there was a marked release of gastrin which rapidly decreased to lower stable levels in 15 minutes; thereafter, gastrin release remained constant for up to 90 minutes. In the experiments described below, perifusion began when the tissue was placed in the chamber and consisted of a 30-min stabilization period followed by an experimental period of 65 min. The latter consisted of a 20-min control period, a 30-min period when the tissue was exposed to the test substance, and a further 15-min post-stimulation period. Effluent was collected over 5-min intervals except during the initial 10 min of stimulation, when samples were collected over 2Vfe min. Gastrin release was studied under the following experimental protocols: (a) the effect of Isuprel (10~ 5 M), P G E J (10~ 5 M), theophylline (10~ 4 M), and dibutyryl cAMP (5 x 10" 4 M) alone; (b) the effect of these agents on arginine-stimulated gastrin secretion. Arginine has been shown to stimulate gastrin release when given intravenously to patients with chronic pancreatitis (13) and was used in the present experiments at a concentration of 0.5 g/100 ml. Preliminary experiments had established that this dose produced approximately half-maximal stimulation. Each group consisted of 5 experiments. Assay Gastrin levels were measured using a sensitive and specific radioimmunoassay (14). The assay
1211
to fraction collector
perifusion chamber
37°C
S if ••-Glucose + Stimulus
Sif + Glucose
FIG. 1. Diagram of perifusion apparatus. SIF is synthetic interstitial fluid. uses antibody raised in rabbits to synthetic human gastrin I. The standard (human type synthetic 68/4399) was obtained from the Medical Research Council. 125I labeled synthetic human gastrin with a specific activity of 700—900 jtiCi/ug was prepared using a modification of the chloramine T method of Hunter and Greenwood (15). The separation procedure used dextrancoated charcoal (16). The precision (SD) of the assay was determined by calculating the difference between replicate determinations in a
TABLE 1. Effect of Isuprel, PGEi, theophylline, and DB cAMP in gastrin release" Gastrin (pg/ml) at stated time intervals (min) 0-10
10-20
20-25
25-30
30-35
35-40
45-50
50-55
55-60
60-65
5
54 3
50 3
60 10
65 10
50 5
52 4
45 3
45 3
40 6
PGE, 10"5M SEM
60 8
52 6
60 10
60 6
58 5
58 5
57 5
55 4
50 4
40 7 52 5
Theophylline 10"4M
52 6
52 4
65 10
70 12
44 3
48 5
50 4
52 6
57 6
45 10
50 7
52 8
57 9
52 3 42 10
50 4
60 7
66 5
60 3
62 6
Isuprel 10" M SEM
SEM
DB cAMP 5 x 10-"M SEM
Compounds were added to the perifusion fluid from 20 through 50 min (underlined time periods).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 07:02 For personal use only. No other uses without permission. . All rights reserved.
Endo < 1975 Vol 97 . No 5
HAYES AND WILLIAMS
1212
Arginine 0.5q/100ml
tit
t
150-
50-
10
20
30
40
Minutes
the levels were again significantly elevated over the last 20 minutes of stimulation. Following withdrawal of the stimulus, gastrin levels returned to basal concentrations. The effects of Isuprel, PGEj, theophylline, and dibutyryl cAMP on argininestimulated gastrin release are shown in Fig. 3. Basal gastrin output was calculated from the area under the curve over the 20minute control period. The gastrin response to each agent was estimated from the area under the curve and above the projected baseline over the period of stimu60 70 lation. The basal output and the gastrin response in each group were compared
p
Department
of Medicine,
University
of Washington,
Seattle,
but each enhanced arginine-stimulated gastrin release. A biphasic pattern of gastrin release in response to arginine was seen in all experiments. The studies emphasize the close functional similarity between the antral G cells and the B cells of the pancreatic islet. (Endocrinology 97: 1210, 1975)
ABSTRACT. An in vitro technique that allows study of gastrin secretion from isolated pieces of rat gastric antrum was used to study the effect on gastrin release of agents known to increase intracellular cAMP levels. Isuprel (10- 5 M) ? PGE, (10" 5 M), theophylline (IO^M), and dibutyryl cAMP (5 x 10~4M) did not affect gastrin release when used alone,
T
HE G cell of the antrum ex- an in vitro technique was employed that hibits morphological and histochem- allows the study of gastrin release into ical similarity with the B cell of the pan- perifusion medium from isolated pieces of creatic islet so that both may be classified gastric antrum. in the APUD series (1). In addition, there is Materials and Methods evidence to suggest that these cells share a common embryologic. origin (2). Extensive study of the B cell has established that Perifusion Technique glucose-stimulated insulin secretion is Male Wistar rats weighing 150-200 g were potentiated by agents which increase used in all experiments. The rats were killed by the intracellular concentration of 3',5'- decapitation and small pieces of tissue 5 mm in adenosine monophosphate (cAMP) (3,4). width and approximately 0.8 g wet wt were The present study investigates the possibil- removed from the antrum of the stomach. The ity that similar mechanisms may be in- pylorus marked the distal border of each piece. volved in the control of gastrin ,'secretion. Each piece was washed briefly and placed on a We have investigated the effect pn gastrin wire grid in a perfusion chamber with a capacity release of isoproterenol (Isuprel), prosta- of 0.7 ml (Fig. 1). The basic buffer used was glandin Ej (PGEj), theophylline, and di- synthetic interstitial fluid (SIF) (11) containing 1 butyryl cAMP, all of which are known to mg/ml glucose. The buffer, which was gassed continually with 95% O :5% CO2, maintained a elevate the intracellular level of cAMP pH of 7.35-7.45 and 37 2C throughout the exper(5,6). These agents have also been shown iment. A peristaltic pump passed buffer conto influence gastric parietal cells (7-10). tained in reservoir flasks through the perifusion Since gastrin release may be altered by chamber at a flow rate of 0.4 ml/min. A threechanges in gastric pH produced by parietal way tap at the base of the perifusion chamber cell secretion, it was felt that the interpreta- allowed the perifusion medium to be changed tion of in vivo experiments on gastrin easily. The effluent was collected in glass tubes release would be difficult. In these studies in a fraction collector and stored at - 2 0 C until Received March 5, 1975. * Present address: Department of Medicine, Queen's University, Belfast, Northern Ireland. Isuprel, theophylline and dibutyryl cAMP were obtained from the Sigma Co. PGE, was obtained from the Upjohn Company.
assayed. Preliminary experiments established that gastrin release from the antrum of fed animals was greater and more reproducible than that from animals fasted 18 hours. Similar observations have been made regarding insulin release from perifused pancreatic pieces (12). In all experi-
1210
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 07:02 For personal use only. No other uses without permission. . All rights reserved.
STUDIES IN GASTRIN SECRETION ments reported here, tissues were taken from animals fasted for 3-4 hours, at which time the stomach was empty. It was also found that gastrin secretion did not depend on the weight of the antral tissue within the narrow range of weights used in these experiments. In the initial period of perifusion there was a marked release of gastrin which rapidly decreased to lower stable levels in 15 minutes; thereafter, gastrin release remained constant for up to 90 minutes. In the experiments described below, perifusion began when the tissue was placed in the chamber and consisted of a 30-min stabilization period followed by an experimental period of 65 min. The latter consisted of a 20-min control period, a 30-min period when the tissue was exposed to the test substance, and a further 15-min post-stimulation period. Effluent was collected over 5-min intervals except during the initial 10 min of stimulation, when samples were collected over 2Vfe min. Gastrin release was studied under the following experimental protocols: (a) the effect of Isuprel (10~ 5 M), P G E J (10~ 5 M), theophylline (10~ 4 M), and dibutyryl cAMP (5 x 10" 4 M) alone; (b) the effect of these agents on arginine-stimulated gastrin secretion. Arginine has been shown to stimulate gastrin release when given intravenously to patients with chronic pancreatitis (13) and was used in the present experiments at a concentration of 0.5 g/100 ml. Preliminary experiments had established that this dose produced approximately half-maximal stimulation. Each group consisted of 5 experiments. Assay Gastrin levels were measured using a sensitive and specific radioimmunoassay (14). The assay
1211
to fraction collector
perifusion chamber
37°C
S if ••-Glucose + Stimulus
Sif + Glucose
FIG. 1. Diagram of perifusion apparatus. SIF is synthetic interstitial fluid. uses antibody raised in rabbits to synthetic human gastrin I. The standard (human type synthetic 68/4399) was obtained from the Medical Research Council. 125I labeled synthetic human gastrin with a specific activity of 700—900 jtiCi/ug was prepared using a modification of the chloramine T method of Hunter and Greenwood (15). The separation procedure used dextrancoated charcoal (16). The precision (SD) of the assay was determined by calculating the difference between replicate determinations in a
TABLE 1. Effect of Isuprel, PGEi, theophylline, and DB cAMP in gastrin release" Gastrin (pg/ml) at stated time intervals (min) 0-10
10-20
20-25
25-30
30-35
35-40
45-50
50-55
55-60
60-65
5
54 3
50 3
60 10
65 10
50 5
52 4
45 3
45 3
40 6
PGE, 10"5M SEM
60 8
52 6
60 10
60 6
58 5
58 5
57 5
55 4
50 4
40 7 52 5
Theophylline 10"4M
52 6
52 4
65 10
70 12
44 3
48 5
50 4
52 6
57 6
45 10
50 7
52 8
57 9
52 3 42 10
50 4
60 7
66 5
60 3
62 6
Isuprel 10" M SEM
SEM
DB cAMP 5 x 10-"M SEM
Compounds were added to the perifusion fluid from 20 through 50 min (underlined time periods).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 23 November 2015. at 07:02 For personal use only. No other uses without permission. . All rights reserved.
Endo < 1975 Vol 97 . No 5
HAYES AND WILLIAMS
1212
Arginine 0.5q/100ml
tit
t
150-
50-
10
20
30
40
Minutes
the levels were again significantly elevated over the last 20 minutes of stimulation. Following withdrawal of the stimulus, gastrin levels returned to basal concentrations. The effects of Isuprel, PGEj, theophylline, and dibutyryl cAMP on argininestimulated gastrin release are shown in Fig. 3. Basal gastrin output was calculated from the area under the curve over the 20minute control period. The gastrin response to each agent was estimated from the area under the curve and above the projected baseline over the period of stimu60 70 lation. The basal output and the gastrin response in each group were compared
p
