- Email: [email protected]
The Effect of Vasoactive Intestinal Polypeptide on Gastric Acid Secretion Is Predominantly Mediated by Somatostatin
GASTROENTEROLOGY 1991;100:1195-1200
The Effect of Vasoactive Intestinal Polypeptide on Gastric Acid Secretion Is Predominantly Mediated by Somatostatin MITCHELL 1. SCHUBERT Department of Medicine, Medical College of Virginia, and McGuire Veterans Administration Hospital, Richmond, Virginia
The mechanism of action of vasoactive intestinal polypeptide on gastric acid secretion was examined in the isolated, luminally perfused mouse stomach. Vasoactive intestinal polypeptide caused a weak, transient increase in basal and histamine-stimulated acid secretion and a sustained increase in somatostatin secretion. The sustained increase in somatostatin despite return of acid to basal levels indicated that somatostatin secretion was a direct response to vasoactive intestinal polypeptide and not mediated by intraluminal acidification. The increase in somatostatin secretion was partiy responsible for the weak, transient nature of the acid response since incubation with pertussis toxin, which is known to block the inhibitory effect of exogenous and endogenous somatostatin, converted the acid response to a sustained increase throughout the period of stimulation. The inhibitory influence of somatostatin was confirmed with selective vasoactive intestinal polypeptide antagonists. The antagonists inhibited vasoactive intestinal polypeptideinduced somatostatin secretion but caused a sustained increase in acid secretion. The pattern of response implied that somatostatin secretion was more sensitive than acid secretion to vasoactive intestinal polypeptide and vasoactive intestinal polypeptide antagonists and that suppression of somatostatin eliminated the main inhibitory influence on acid secretion. In addition, both vasoactive intestinal polypeptide antagonists inhibited basal somatostatin secretion, implying that input from tonically active vasoactive intestinal polypeptide neurons is responsible, at least in part, for basal somatostatin secretion. asoactive intestinal polypeptide (VIP) is present V in nerve fibers of gastric mucosa its presence in nerve fibers of fundic mucosa suggests that it (1,2);
might be involved in the regulation of mucosal secre-
tory function. Vasoactive intestinal polypeptide stimulates gastric somatostatin secretion in the isolated perfused rat stomach (3,4) and in cultures of somatostatin cells derived from canine fundic mucosa (5). In suspensions enriched with parietal cells derived from canine fundic mucosa (5), VIP in the presence of phosphodiesterase inhibitors weakly stimulates 14e-aminopyrine accumulation, an index of parietal cell secretion. The augmentation induced by phosphodiesterase inhibitors suggests that VIP acts by stimulating an increase in cyclic adenosine monophosphate (AMP) levels. Vasoactive intestinal polypeptide is known to increase cyclic AMP in gastric mucosal glands (6), although it is not known whether this increase occurs in one or all types of mucosal cells. The net effect of VIP on acid secretion in intact mucosa in vitro or in vivo depends on the balance between direct stimulation of acid secretion and concurrent release of inhibitors such as somatostatin in proximity of parietal cells. In vivo, VIP has been variously reported to stimulate acid secretion in cats (7), inhibit acid secretion in dogs (8), and have no effect in humans (g). In the present study, an isolated luminally perfused mouse stomach, a preparation that retains intact neural and paracrine pathways but eliminates the influence of gastrin, was used to examine the mechanism of action of VIP on acid secretion and the participation of VIP neurons in the regulation of acid secretion. The results indicate that VIP exerts a direct stimulatory and an indirect inhibitory effect on acid secretion. The inhibitory effect is mediated by release of somatostatin and can be reversed by treatment of the preparation with pertussis toxin. Selective VIP Abbreviation used in this paper: TTX, tetrodotoxin. © 1991 by the American Gastroenterological Association 0016'5085/91/$3.00
1196
MITCHELL L. SCHUBERT
antagonists strongly inhibited basal and stimulated somatostatin secretion implying that VIP neurons are important determinants of gastric somatostatin secretion.
Materials and Methods
Materials Vasoactive intestinal polypeptide, VIP[10-28], and [4CI-D-Phe",Leu 17 jVIP were purchased from Bachem, Torrance, CA; histamine dihydrochloride, atropine sulfate, indomethacin, and tetrodotoxin (TTX) from Sigma Chemical, St. Louis, MO; cimetidine from Smith Kline & French Laboratories, Philadelphia, PA; and pertussis toxin from List Biological Laboratories, Campbell, CA.
Animals Male albino mice weighing 25-40 g were deprived of food overnight but allowed free access to water containing 10% glucose. The animals were anesthetized with 20% urethan (0.25 mL/50 g body wt) injected IP.
[solation and Luminal Perfusion of the Stomach Mouse stomachs were isolated according to the method of Bunce and Parsons (10,11). The stomach was cannulated at its pyloric and esophageal ends with polyethylene tubes (PE 160). Fifteen milliliters of saline was injected via the proximal tube to flush out the remaining contents. The stomach was incubated in 20 mL of a serosal solution with the following composition (in mmol/L): NaCI, 115.4; NaHC0 3 , 24.3; KCI, 4.5; MgSO., 2.4; KHzPO., 1.2; CaCI2 , 0.65; and dextrose, 31. The lumen of the stomach was perfused at the rate of 1 mL/min with a solution of the following composition (in mmol/L): NaCI, 140; KCl, 4.5; MgSO., 2.4; CaCl z, 1.3; and dextrose, 31. The serosal solution was gassed with 95% OJ5% CO z and the luminal perfusate with 100% Oz. Drugs were added to the serosal solution, and both solutions were maintained at 37°C.
Experimental Design The experimental design consisted of a 30-minute equilibration period followed by a 90-180-minute sampling period. The sampling period consisted of the following sequence: (a) a 30-minute control basal period; (b) a 30minute period during which VIP (1 fLmol/L), VIP[10-28j (1 fLmol/L), or [4CI-D-Phe 6 ,Leu17jVIP (1 fLmol/L) alone or in combination was added to the serosal solution; and (c) a 3D-minute control period without additions. For experiments involving pertussis toxin, pertussis toxin (125 ng/mL) was added to the serosal solution 60 minutes before addition of VIP. In separate experiments, atropine (1 fLmol/L), cimetidine (10 fLmol/L), indomethacin (1 fLmol/L), or TTX (5 fLmol/L) was added to the serosal solution by itself for 30 minutes and with VIP for a further 30 minutes.
GASTROENTEROLOGY Vol. 100, No.5
For experiments with histamine, the following sequence was used: (a) a 3D-minute control basal period; (b) a 60-minute period during which pertussis toxin (125 ng/mL) was added to the serosal solution; (c) a 3D-minute period during which histamine (10 fLmol/L) was added to the serosal solution; (d) a 30-minute period during which VIP (1 fLmol/L) was added to the serosal solution; and (e) a 3D-minute control period. Control experiments were performed in which histamine alone or in the presence of pertussis toxin was added for 30- or 60-minute periods. One-milliliter samples of the luminal effluent were obtained at I-minute intervals during the experimental periods and at 5-minute intervals during the control basal periods. The samples were divided into 0.5-mL aliquots for immediate measurement of pH and titratable acid and then stored at - 20°C for subsequent measurement of somatostatin by radioimmunoassay. Acid concentration was measured by titration to pH 7.4 with O.OlN NaOH using an automatic titrator (Radiometer, Copenhagen).
Radioimmunoassay Somatostatin was measured in duplicate by radioimmunoassay as described in detail previously (12,13). Somatostatin antibody 1001 (final dilution, 1:66,000), a gift from Dr. Tadataka Yamada, University of Michigan, was used. [, 25 I]Tyr'-somatostatin was purchased from New England Nuclear, Boston, MA. The IC so of the assay was 85 ± 16 (mean ± SD; n = 10 assays). Interassay and intraassay coefficients of variability were 9.1 % and 8.2%, respectively. The limit of detection was 5 pg/mL of sample.
Data Analysis Acid secretion was expressed as the change (Le., increment or decrement) in nanomoles per minute or in percent from the basal level during the 10 minutes immediately preceding the experimental period. Somatostatin secretion was expressed in similar manner in picograms per minute or in percent from the preceding basal level. Significance of difference for acid or somatostatin secretion was tested against the preceding controlltlvel or against synchronous values determined in separate control experiments using Student's t test. All values are given as means ± SE.
Results
Basal Acid and Somatostatin Secretion Basal acid secretion was uniform from one experimental series to another (mean ± SE, 91.1 ± 0.9 nmol/min) and reverted to basal control levels after cessation of the experimental period (mean ± SE, 86.5 ± 1.0 nmol/min). Basal somatostatin secretion was also uniform from one experimental series to another (mean ± SE, 23.2 ± 1.4 pg/min) and reverted to basal control levels after cessation of the experimental period (mean ± SE, 23.7 ± 1.3 pg/min).
VIP REGULATION OF ACID AND SOMATOSTATIN
May 1991
Effect of Vasoactive Intestinal Polypeptide on Histamine-Stimulated Acid and Somatostatin Secretion
+
180 170
..
160
C
0
150
QI
... ~ v_
Qlrt!
"'''' crt! ._ m
140
~O
130
.........
....CS;;€! rt!
~
E
120
'"
110
0
100
.~-- -----------------~
90
r-6
O-L
c:~
1197
120
0._ rt!
....Qlrt!'"
A submaximal concentration of histamine (10 J..l.mol/L) elicited an increase in acid secretion of 20.8 ± 4.9 nmol/min (25% ± 6%) (P < 0.01; n = 6) and a corresponding increase in somatostatin secretion of 10.3 ± 1.1 pg/min (44% ± 6%) above basal level (P < 0.001; n = 6) (Figure 2). As shown previously (14), the increase in somatostatin secretion was mediated by the increase in intraluminal acidity . Vasoactive intestinal polypeptide (1 J..l.mol/L) transiently augmented the acid response to 10 J..l.mol/L histamine from 20.8 ± 4.9 nmol/min (25% ± 6%) with histamine alone to 55.0 ± 16.6 nmol/min (61% ± 12%)forhistamineplusVIP(P < 0.05 for the difference between the two responses) (Figure 2). The return of acid secretion to the level observed with histamine alone coincided with an increase of somatostatin secretion from 10.3 ± 1.1 pg/min (44% ± 6%) with histamine alone to 27.9 ± 1.7
tim 110
QI .....
",0
Histamine 10JlM
+~ :J:~
300
90 O-L
~~~0--~~10--~-2~0--~-3~0~--4~0--~/~0
-10
Minutes
0
:;::;
...
01
Figure 1. Effect of 1 f,LmollL VIP alone (e) or after pertussis toxin (PTX) (0) on basal acid and somatostatin secretion in isolated mouse stomach. Pertussis toxin (125 ng/mL) was present for 60 minutes before and 30 minutes after addition of VIP. Dotted lines indicate the levels of basal secretion. Data are means :t SE of 5-6 experiments.
260
c
V'I'"
c'" ._ al
180
E~
140
....~o
'"E 0
V'I
100 60
Effect of Vasoactive Intestinal Polypeptide on Basal Acid and Somatostatin Secretion Addition of VIP (1 J..l.mol/L) to the serosal solution for a period of 30 minutes caused a transient 15-minute increase in basal acid secretion with a peak of 17.4 ± 4.3 nmol/min (21% ± 6%) (P < 0.01; n = 8) (Figure 1). The return of acid secretion to basal level in the second 15-mjnute period coincided with a peak increase in somatostatin secretion of 7.8 ± 1.9 pg/min (63% ± 19%) (P < 0.02; n = 6) (Figure 1). A 10 times lower concentration of VIP (0.1 J..l.mol/L) produced about 50% of the acid response obtained with 1 J..l.mol/L VIP (7.5 ± 1.3 nmol/min or 10% ± 2%; P < 0.01; n = 6). A concentration of 0.01 J..l.mol/L VIP appeared to be threshold for both acid and somatostatin secretion. Neither TTX (5 J..l.mol/L), atropine (1 J..l.mol/L), cimetidine (10 J..l.mol/L), nor indomethacin (1 J..l.mol/L) had any significant effect on VIP-stimulated acid or somatostatin secretion.
220
-
~ v_ Ol'"
01:. 200 180
c 160 o==-
.'"'" +' 01", b CD
01V'l0
+~ :r:~
140
120
------------------~
100 80 01:.
I
I ~~L-~lo~~~~~~~~do--~~--~~0~-7~~--7/~0 0
-10
Minutes
Figure 2. Effect of 1 f,LmollL VIP on acid and somatostatin secretion in response to 10 f,Lm.ollL histamine alone (e) or after pertussis toxin (PTX) (0). Pertussis toxin (125 ng/mL) was present for 60 minutes before and 60 minutes after stimulation with histamine. Dotted lines indicate the levels of basal secretion. Data are means :t SE of 6-11 experiments.
1198
MITCHELL 1. SCHUBERT
pg/min (121% ± 15%) with histamine plus VIP (P < 0.001 for the difference between the two responses) (Figure 2).
Effect of Pertussis Toxin on Histamine-Stimulated and Vasoactive Intestinal Polypeptide-Stimulated Acid Secretion Pertussis toxin, previously shown to block the inhibitory effect of somatostatin on acid secretion (15-17), was used to determine whether acid-induced and VIP-induced increase in somatostatin secretion attenuated the acid response. Preincubation of the stomach with pertussis toxin converted the transient 15-minute increase in acid secretion induced by VIP (1 J..Lmol/L) alone to a sustained increase which persisted throughout the entire 3D-minute period. During the second 15minute period, the increase in acid secretion was 11.0 ± 0.8 nmol/min (14% ± 1%) (P < 0.001; n = 5) (Figure 1). Preincubation with pertussis toxin augmented significantly the acid response to 10 J..LmollL histamine from 20.8 ± 4.9 nmol/min (25% ± 6%) with histamine alone to 61.9 ± 11.3 nmol/min (61% ± 10%) (n = 11) after incubation with pertussis toxin (P < 0.01 for the difference between the two responses) (Figure 2). The increase in histaminestimulated acid secretion was attributed to blockade of the inhibitory effect of ambient somatostatin. Somatostatin secretion increased significantly coincidentally with acid secretion from 44% ± 6% to 80% ± 8% (n = 9) (P < 0.01 for the difference between the two responses) (Figure 2). The increase in somatostatin secretion was attributed to the increase in acid secretion after incubation with pertussis toxin. Addition of VIP in pertussis toxin-treated preparations induced a sustained increase in histaminestimulated acid secretion that persisted throughout the entire 3D-minute period (Figure 2). During the latter 15-minute period, the acid response induced by histamine plus VIP after incubation with pertussis toxin [82.0 ± 13.6 nmol/min (80% ± 11%)] was significantly higher than the response induced by histamine plus VIP without prior treatment with pertussis toxin (P < 0.001; n = 11) (Figure 2).
Effect of Vasoactive Intestinal Polypeptide Antagonists on Basal and Vasoactive Intestinal Polypeptide-Stimulated Acid and Somatostatin Secretion Addition of selective VIP antagonists inhibited basal somatostatin secretion: VIP[10-28] (1 J..Lmol/L) caused a decrease of 8.9 ± 1.5 pg/min (33% ± 3%)
GASTROENTEROLOGY Vol. 100, No.5
(P
< 0.001; n = 7) and [4CI-D-Phe 6 ,Leu 17 ]VIP (1
J..Lmol/L) a decrease of 15.7 ± 2.3 pg/min (37% ± 3%) below basal level (P < 0.001; n = 7) (Figure 3). Acid secretion decreased slightly (7%-9% below basal levels) (Figure 3). At the same concentrations, both VIP antagonists abolished VIP-stimulated somatostatin secretion and caused a sustained increase in VIP-stimulated acid secretion that persisted throughout the entire 30minute period [29.4 ± 4.3 nmol/min (36% ± 5%); P < 0.001; n = 6 with VIP[10-28] and 28.8 ± 5.5 nmol/min (27% ± 3%); P < 0.001; n = 6 with [4CI-DPhe6 ,Leu 17 ]VIP] (Figure 4). Discussion This study shows that VIP has a direct stimulatory effect on acid and somatostatin secretion in the isolated luminally perfused mouse stomach. Compared with the effect of other acid secretagogues, such as histamine (14), the effect of VIP is both small (about 20% of maximal histamine-stimulated secretion) and
110
t
VIP Antagonist 1JiM
,
------------------~
1
c
0 ''::;
...
OJ ~ uOJ <0 VI'"
...
.:: ~
...
90
80
~o
o~
<0
~
E
70
0 VI
60
[4CI-D-Phe6.Leu17) VIP ...-/ 0-( 110 c~
0._ <0
....OJ '"<0 bCIJ
OJ-
VlO
+';f!.
90
:J:~
80
o-L _IO~..J.....-...J.O-.l..--I0'-1---1-2-':~-.l..--=i'::-0- - - 1 _4-,:IO:--L--f/ ~5
Minutes
Figure 3. Effect of 1 j..Lmol!L VIP[1D-28] (e) or 1 j..Lmol!L [4CI-DPhe6 ,Leu"]VIP (0) on basal acid and somatostatin secretion. Dotted lines indicate the levels of basal secretion. Data are means ± SE of 5-7 experiments.
VIP REGULATION OF ACID AND SOMATOSTATIN
May 1991
.
VIP (l11M) + VIP Antagonist (l11M)
~
180 170 160
§
1 I
150
I
'';:;
~::::-
~,'
.~~ 130
........
C;~
....!1l
E o
\
\~
\~ \
120
~
VI
VIP Alo",
/
140
CU!1l
VI'"
.:go
,L.r
110
+
100
",
VIP[1~2~~t 4
90 80
01. 150
140
/' + VIP [10-28]
130
+,
c~
0._ !1l
....~~'"
v ....
[4CI-D-Phe 6 , Leu'7] VIP
\i
120
~o
+ ~
:r:~
\
110
" T VIP Alone
't(
100
90
01.
'--'- - - ' - - - - ' - - ' - - - - - ' - - - - - - - ' - - - - t / I----"
-10
0
10
20
30
40
60
Minutes
Figure 4. Effect of VIP alone (0) or in combination witb 1 ~mol!L VlP[1D-28] (e) or 1 ~mol!L [4CI-D-Phe 6 ,Leu17]VlP (0) on acid and somatostatin secretion. Dotted lines indicate tbe levels of basal secretion. Data are means ± SE of 6-8 experiments.
transient. The transient nature of acid secretion probably represents an inhibitory effect of concurrent stimulation of somatostatin secretion. Compared with acid, somatostatin secretion was sustained throughout the period of stimulation with VIP. The sustained nature of this effect despite the return of acid secretion to control levels implies that the VIP-induced somatostatin secretion was not a consequence of luminal acidification. We had previously shown that potent acid secretagogues, such as
1199
histamine, which do not have a direct effect on the somatostatin cell, cause a substantial increase in somatostatin secretion as a result of luminal acidification (14). The effect of VIP on somatostatin and acid secretion was not affected by atropine, cimetidine, or the axonal blocker TTX, implying that it was not mediated by release of endogenous neurotransmitters or histamine. These direct stimulatory effects of VIP on acid and somatostatin secretion confirm results obtained in isolated suspensions of parietal cells and cultures of somatostatin cells derived from the fundus of the canine stomach (5). The possibility of functional coupling between the responses of different cells cannot be determined in preparations enriched in one cell type (Le., parietal or somatostatin cells). In the isolated, acid-secreting luminally perfused mouse stomach that retains intact paracrine and neural mechanisms (11,14,16), the acid and somatostatin responses appear to be functionally coupled. This was mostly evident in the reciprocal inhibitory effect of somatostatin on acid secretion and provides an explanation for the transient nature of the acid secretory response to VIP. This interpretation is supported by the fact that elimination of this inhibitory effect by preincubation with pertussis toxin converts the transient secretion induced by VIP to a sustained secretion throughout the period of stimulation. This effect was observed whether VIP was given by itself or in combination with histamine. In previous studies we had shown that pertussis toxin blocks the inhibitory effect of exogenous and endogenous somatostatin on acid secretion in this preparation when the stimulus of acid secretion (e.g., histamine) acts via cyclic AMP but not when it acts via other intracellular messengers (e.g., gastrin) (15-17). The effect of VIP in various tissues (18), including gastric tissues (6), is mediated via cyclic AMP and would thus be susceptible to pertussis toxin. Further support for the presence of a reciprocal inhibitory influence of somatostatin on VIP-stimulated acid secretion was obtained by the use of selective VIP antagonists (19). The antagonists abolished the somatostatin response to VIP but augmented acid secretion, which became sustained throughout the period of observation. The results imply (a) that somatostatin secretion is much more sensitive to stimulation by VIP and to blockade by selective VIP antagonists than acid secretion and (b) that the concurrent increase in somatostatin secretion is an important cause of the transient, partial nature of the acid response to VIP. The results also suggest that parietal cells are relatively insensitive to VIP and VIP antagonists. An important aspect of the present study is that in the absence of exogenous VIP, both VIP antagonists
1200
MITCHELL 1. SCHUBERT
significantly inhibited basal somatostatin secretion. This implies that VIP neurons in the stomach are tonically active and that input from these neurons is responsible, at least in part, for basal somatostatin secretion. References 1. Larsson 11, Fahrenkrug J, Schaffalitsky DE, Muckadell 0, Sundler F, Hakanson R, Rehfeld JF. Localization of vasoactive intestinal peptide (VIP) to central and peripheral neurons. Proc Natl Acad Sci USA 1978;73:3197-3200. 2. Schultzberg M, Dreyfus CF, Gershon MD, Hokfelt T, Elde RP, Nilsson G, Said SI, Goldstein M. VIP-, enkephalin-, substance P-, and somatostatin-like immunoreactivity in neurons intrinsic to the intestine: immunohistochemical evidence from organotypic tissue cultures. Brain Res 1978;155:239-248. 3. Chiba T, Taminato T, Kadowaki S, Abe H, Chihara K, Seino Y, Matsukura S, Fujita T. Effects of glucagon, secretin, and vasoactive intestinal peptide on gastric somatostatin and gastrin release from isolated perfused rat stomach. Gastroenterology 1980;79:67-71. 4. Saffouri B, Duvall JW, Arimura A, Makhlouf GM. Effects of vasoactive intestinal peptide and secretin on gastrin and somatostatin secretion from the perfused rat stomach. Gastroenterology 1984;86:839-842. 5. Chiba T, Park J, Yamada T. Glucagon and vasoactive intestinal polypeptide stimulate somatostatin secretion from isolated canine fundic mucosal cell cultures (abstr). Gastroenterology 1985;88:1348. 6. Dupont C, Gespach C, Chenut B, Rosselin G. Regulation of vasoactive intestinal peptide of cyclic AMP accumulation in gastric epithelial cells. FEBS Lett 1980;113:25-28. 7. Vagne M, Konturek SJ, Chayvialle JA. Effect of vasoactive intestinal peptide on gastric secretion in the cat. Gastroenterology 1982;83:250-255. 8. Makhlouf GM, Zfass AM, Said SI, Schebalin M. Effects of synthetic vasoactive intestinal peptide (VIP), secretin and their partial sequences on gastric secretion. Proc Soc Exp Bioi Med 1978;157:565-568. 9. Holm-Bentzen M, Christiansen J, Kirkegaard P, Olson PS, Petersen B, Fahrenkrug J. The effect of vasoactive intestinal
GASTROENTEROLOGY Vol. 100, No.5
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
polypeptide on meal-stimulated gastric acid secretion in man. Scand J GastroenteroI1983;18:659-661. Bunce KT, Parsons ME. A quantitative study of metiamide, a histamine-H 2 -antagonist, on the isolated whole rat stomach. J Physiol Lond 1976;258:453-465. Schubert ML, Edwards NF, Arimura A, Makhlouf GM. Paracrine regulation of gastric acid secretion by fundic somatostatin. Am J PhysioI1987;252:G485-G490. Schubert ML, Saffouri B, Walsh JH, Makhlouf GM. Inhibition of neurally mediated gastrin secretion by bombesin antiserum. Am J PhysioI1985;248:G546-G562. Yamada T, Marshak D, Basinger S, WalshJH, Morley J, Steel W. Somatostatin-like immunoreactivity in the retina. Proc Natl Acad Sci USA 1980;77:1691-1695. Schubert ML, Edwards NF, Makhlouf GM. Regulation of gastric somatostatin secretion in the mouse by luminal acidity: a local feedback mechanism. Gastroenterology 1988;94:317-322. Park J, Chiba T, Yamada T. Mechanisms for direct inhibition of canine gastric parietal cells by somatostatin. J Bioi Chern 1987;262:14190-14196. Schubert ML, Hightower J. Inhibition of acid secretion by bombesin is partly mediated by release of fundic somatostatin. Gastroenterology 1989;97:561-567. Schubert ML, Hightower J, Makhlouf GM. Linkage between somatostatin and acid secretion: evidence from use of pertussis toxin. AmJ PhysioI1989;256:G418-G422. Amiranoff B, Rosselin G. VIP receptors and control of cyclic AMP production. In: Said SI, ed. Vasoactive intestinal polypeptides. New York: Raven, 1982:307-322. Pando I SJ, Dharmsathaphorn K, Schoeffield MS, Vale W, Rivier J. Vasoactive intestinal peptide receptor antagonist [4CI-DPhe',Leu 17 ]VIP. Am JPhysioI1986;250:G553-G557.
Received April 3, 1990. Accepted October 2,1990. Address requests for reprints to: Mitchell 1. Schubert, M.D., McGuire Veterans Administration Hospital, Division of Gastroenterology; Code I11N, 1201 Broad Rock Boulevard, Richmond, Virginia 23249. This work was supported by the Veterans Administration Medical Research Fund and the American Gastroenterological Association Industry Research Scholar Award.
The Effect of Vasoactive Intestinal Polypeptide on Gastric Acid Secretion Is Predominantly Mediated by Somatostatin MITCHELL 1. SCHUBERT Department of Medicine, Medical College of Virginia, and McGuire Veterans Administration Hospital, Richmond, Virginia
The mechanism of action of vasoactive intestinal polypeptide on gastric acid secretion was examined in the isolated, luminally perfused mouse stomach. Vasoactive intestinal polypeptide caused a weak, transient increase in basal and histamine-stimulated acid secretion and a sustained increase in somatostatin secretion. The sustained increase in somatostatin despite return of acid to basal levels indicated that somatostatin secretion was a direct response to vasoactive intestinal polypeptide and not mediated by intraluminal acidification. The increase in somatostatin secretion was partiy responsible for the weak, transient nature of the acid response since incubation with pertussis toxin, which is known to block the inhibitory effect of exogenous and endogenous somatostatin, converted the acid response to a sustained increase throughout the period of stimulation. The inhibitory influence of somatostatin was confirmed with selective vasoactive intestinal polypeptide antagonists. The antagonists inhibited vasoactive intestinal polypeptideinduced somatostatin secretion but caused a sustained increase in acid secretion. The pattern of response implied that somatostatin secretion was more sensitive than acid secretion to vasoactive intestinal polypeptide and vasoactive intestinal polypeptide antagonists and that suppression of somatostatin eliminated the main inhibitory influence on acid secretion. In addition, both vasoactive intestinal polypeptide antagonists inhibited basal somatostatin secretion, implying that input from tonically active vasoactive intestinal polypeptide neurons is responsible, at least in part, for basal somatostatin secretion. asoactive intestinal polypeptide (VIP) is present V in nerve fibers of gastric mucosa its presence in nerve fibers of fundic mucosa suggests that it (1,2);
might be involved in the regulation of mucosal secre-
tory function. Vasoactive intestinal polypeptide stimulates gastric somatostatin secretion in the isolated perfused rat stomach (3,4) and in cultures of somatostatin cells derived from canine fundic mucosa (5). In suspensions enriched with parietal cells derived from canine fundic mucosa (5), VIP in the presence of phosphodiesterase inhibitors weakly stimulates 14e-aminopyrine accumulation, an index of parietal cell secretion. The augmentation induced by phosphodiesterase inhibitors suggests that VIP acts by stimulating an increase in cyclic adenosine monophosphate (AMP) levels. Vasoactive intestinal polypeptide is known to increase cyclic AMP in gastric mucosal glands (6), although it is not known whether this increase occurs in one or all types of mucosal cells. The net effect of VIP on acid secretion in intact mucosa in vitro or in vivo depends on the balance between direct stimulation of acid secretion and concurrent release of inhibitors such as somatostatin in proximity of parietal cells. In vivo, VIP has been variously reported to stimulate acid secretion in cats (7), inhibit acid secretion in dogs (8), and have no effect in humans (g). In the present study, an isolated luminally perfused mouse stomach, a preparation that retains intact neural and paracrine pathways but eliminates the influence of gastrin, was used to examine the mechanism of action of VIP on acid secretion and the participation of VIP neurons in the regulation of acid secretion. The results indicate that VIP exerts a direct stimulatory and an indirect inhibitory effect on acid secretion. The inhibitory effect is mediated by release of somatostatin and can be reversed by treatment of the preparation with pertussis toxin. Selective VIP Abbreviation used in this paper: TTX, tetrodotoxin. © 1991 by the American Gastroenterological Association 0016'5085/91/$3.00
1196
MITCHELL L. SCHUBERT
antagonists strongly inhibited basal and stimulated somatostatin secretion implying that VIP neurons are important determinants of gastric somatostatin secretion.
Materials and Methods
Materials Vasoactive intestinal polypeptide, VIP[10-28], and [4CI-D-Phe",Leu 17 jVIP were purchased from Bachem, Torrance, CA; histamine dihydrochloride, atropine sulfate, indomethacin, and tetrodotoxin (TTX) from Sigma Chemical, St. Louis, MO; cimetidine from Smith Kline & French Laboratories, Philadelphia, PA; and pertussis toxin from List Biological Laboratories, Campbell, CA.
Animals Male albino mice weighing 25-40 g were deprived of food overnight but allowed free access to water containing 10% glucose. The animals were anesthetized with 20% urethan (0.25 mL/50 g body wt) injected IP.
[solation and Luminal Perfusion of the Stomach Mouse stomachs were isolated according to the method of Bunce and Parsons (10,11). The stomach was cannulated at its pyloric and esophageal ends with polyethylene tubes (PE 160). Fifteen milliliters of saline was injected via the proximal tube to flush out the remaining contents. The stomach was incubated in 20 mL of a serosal solution with the following composition (in mmol/L): NaCI, 115.4; NaHC0 3 , 24.3; KCI, 4.5; MgSO., 2.4; KHzPO., 1.2; CaCI2 , 0.65; and dextrose, 31. The lumen of the stomach was perfused at the rate of 1 mL/min with a solution of the following composition (in mmol/L): NaCI, 140; KCl, 4.5; MgSO., 2.4; CaCl z, 1.3; and dextrose, 31. The serosal solution was gassed with 95% OJ5% CO z and the luminal perfusate with 100% Oz. Drugs were added to the serosal solution, and both solutions were maintained at 37°C.
Experimental Design The experimental design consisted of a 30-minute equilibration period followed by a 90-180-minute sampling period. The sampling period consisted of the following sequence: (a) a 30-minute control basal period; (b) a 30minute period during which VIP (1 fLmol/L), VIP[10-28j (1 fLmol/L), or [4CI-D-Phe 6 ,Leu17jVIP (1 fLmol/L) alone or in combination was added to the serosal solution; and (c) a 3D-minute control period without additions. For experiments involving pertussis toxin, pertussis toxin (125 ng/mL) was added to the serosal solution 60 minutes before addition of VIP. In separate experiments, atropine (1 fLmol/L), cimetidine (10 fLmol/L), indomethacin (1 fLmol/L), or TTX (5 fLmol/L) was added to the serosal solution by itself for 30 minutes and with VIP for a further 30 minutes.
GASTROENTEROLOGY Vol. 100, No.5
For experiments with histamine, the following sequence was used: (a) a 3D-minute control basal period; (b) a 60-minute period during which pertussis toxin (125 ng/mL) was added to the serosal solution; (c) a 3D-minute period during which histamine (10 fLmol/L) was added to the serosal solution; (d) a 30-minute period during which VIP (1 fLmol/L) was added to the serosal solution; and (e) a 3D-minute control period. Control experiments were performed in which histamine alone or in the presence of pertussis toxin was added for 30- or 60-minute periods. One-milliliter samples of the luminal effluent were obtained at I-minute intervals during the experimental periods and at 5-minute intervals during the control basal periods. The samples were divided into 0.5-mL aliquots for immediate measurement of pH and titratable acid and then stored at - 20°C for subsequent measurement of somatostatin by radioimmunoassay. Acid concentration was measured by titration to pH 7.4 with O.OlN NaOH using an automatic titrator (Radiometer, Copenhagen).
Radioimmunoassay Somatostatin was measured in duplicate by radioimmunoassay as described in detail previously (12,13). Somatostatin antibody 1001 (final dilution, 1:66,000), a gift from Dr. Tadataka Yamada, University of Michigan, was used. [, 25 I]Tyr'-somatostatin was purchased from New England Nuclear, Boston, MA. The IC so of the assay was 85 ± 16 (mean ± SD; n = 10 assays). Interassay and intraassay coefficients of variability were 9.1 % and 8.2%, respectively. The limit of detection was 5 pg/mL of sample.
Data Analysis Acid secretion was expressed as the change (Le., increment or decrement) in nanomoles per minute or in percent from the basal level during the 10 minutes immediately preceding the experimental period. Somatostatin secretion was expressed in similar manner in picograms per minute or in percent from the preceding basal level. Significance of difference for acid or somatostatin secretion was tested against the preceding controlltlvel or against synchronous values determined in separate control experiments using Student's t test. All values are given as means ± SE.
Results
Basal Acid and Somatostatin Secretion Basal acid secretion was uniform from one experimental series to another (mean ± SE, 91.1 ± 0.9 nmol/min) and reverted to basal control levels after cessation of the experimental period (mean ± SE, 86.5 ± 1.0 nmol/min). Basal somatostatin secretion was also uniform from one experimental series to another (mean ± SE, 23.2 ± 1.4 pg/min) and reverted to basal control levels after cessation of the experimental period (mean ± SE, 23.7 ± 1.3 pg/min).
VIP REGULATION OF ACID AND SOMATOSTATIN
May 1991
Effect of Vasoactive Intestinal Polypeptide on Histamine-Stimulated Acid and Somatostatin Secretion
+
180 170
..
160
C
0
150
QI
... ~ v_
Qlrt!
"'''' crt! ._ m
140
~O
130
.........
....CS;;€! rt!
~
E
120
'"
110
0
100
.~-- -----------------~
90
r-6
O-L
c:~
1197
120
0._ rt!
....Qlrt!'"
A submaximal concentration of histamine (10 J..l.mol/L) elicited an increase in acid secretion of 20.8 ± 4.9 nmol/min (25% ± 6%) (P < 0.01; n = 6) and a corresponding increase in somatostatin secretion of 10.3 ± 1.1 pg/min (44% ± 6%) above basal level (P < 0.001; n = 6) (Figure 2). As shown previously (14), the increase in somatostatin secretion was mediated by the increase in intraluminal acidity . Vasoactive intestinal polypeptide (1 J..l.mol/L) transiently augmented the acid response to 10 J..l.mol/L histamine from 20.8 ± 4.9 nmol/min (25% ± 6%) with histamine alone to 55.0 ± 16.6 nmol/min (61% ± 12%)forhistamineplusVIP(P < 0.05 for the difference between the two responses) (Figure 2). The return of acid secretion to the level observed with histamine alone coincided with an increase of somatostatin secretion from 10.3 ± 1.1 pg/min (44% ± 6%) with histamine alone to 27.9 ± 1.7
tim 110
QI .....
",0
Histamine 10JlM
+~ :J:~
300
90 O-L
~~~0--~~10--~-2~0--~-3~0~--4~0--~/~0
-10
Minutes
0
:;::;
...
01
Figure 1. Effect of 1 f,LmollL VIP alone (e) or after pertussis toxin (PTX) (0) on basal acid and somatostatin secretion in isolated mouse stomach. Pertussis toxin (125 ng/mL) was present for 60 minutes before and 30 minutes after addition of VIP. Dotted lines indicate the levels of basal secretion. Data are means :t SE of 5-6 experiments.
260
c
V'I'"
c'" ._ al
180
E~
140
....~o
'"E 0
V'I
100 60
Effect of Vasoactive Intestinal Polypeptide on Basal Acid and Somatostatin Secretion Addition of VIP (1 J..l.mol/L) to the serosal solution for a period of 30 minutes caused a transient 15-minute increase in basal acid secretion with a peak of 17.4 ± 4.3 nmol/min (21% ± 6%) (P < 0.01; n = 8) (Figure 1). The return of acid secretion to basal level in the second 15-mjnute period coincided with a peak increase in somatostatin secretion of 7.8 ± 1.9 pg/min (63% ± 19%) (P < 0.02; n = 6) (Figure 1). A 10 times lower concentration of VIP (0.1 J..l.mol/L) produced about 50% of the acid response obtained with 1 J..l.mol/L VIP (7.5 ± 1.3 nmol/min or 10% ± 2%; P < 0.01; n = 6). A concentration of 0.01 J..l.mol/L VIP appeared to be threshold for both acid and somatostatin secretion. Neither TTX (5 J..l.mol/L), atropine (1 J..l.mol/L), cimetidine (10 J..l.mol/L), nor indomethacin (1 J..l.mol/L) had any significant effect on VIP-stimulated acid or somatostatin secretion.
220
-
~ v_ Ol'"
01:. 200 180
c 160 o==-
.'"'" +' 01", b CD
01V'l0
+~ :r:~
140
120
------------------~
100 80 01:.
I
I ~~L-~lo~~~~~~~~do--~~--~~0~-7~~--7/~0 0
-10
Minutes
Figure 2. Effect of 1 f,LmollL VIP on acid and somatostatin secretion in response to 10 f,Lm.ollL histamine alone (e) or after pertussis toxin (PTX) (0). Pertussis toxin (125 ng/mL) was present for 60 minutes before and 60 minutes after stimulation with histamine. Dotted lines indicate the levels of basal secretion. Data are means :t SE of 6-11 experiments.
1198
MITCHELL 1. SCHUBERT
pg/min (121% ± 15%) with histamine plus VIP (P < 0.001 for the difference between the two responses) (Figure 2).
Effect of Pertussis Toxin on Histamine-Stimulated and Vasoactive Intestinal Polypeptide-Stimulated Acid Secretion Pertussis toxin, previously shown to block the inhibitory effect of somatostatin on acid secretion (15-17), was used to determine whether acid-induced and VIP-induced increase in somatostatin secretion attenuated the acid response. Preincubation of the stomach with pertussis toxin converted the transient 15-minute increase in acid secretion induced by VIP (1 J..Lmol/L) alone to a sustained increase which persisted throughout the entire 3D-minute period. During the second 15minute period, the increase in acid secretion was 11.0 ± 0.8 nmol/min (14% ± 1%) (P < 0.001; n = 5) (Figure 1). Preincubation with pertussis toxin augmented significantly the acid response to 10 J..LmollL histamine from 20.8 ± 4.9 nmol/min (25% ± 6%) with histamine alone to 61.9 ± 11.3 nmol/min (61% ± 10%) (n = 11) after incubation with pertussis toxin (P < 0.01 for the difference between the two responses) (Figure 2). The increase in histaminestimulated acid secretion was attributed to blockade of the inhibitory effect of ambient somatostatin. Somatostatin secretion increased significantly coincidentally with acid secretion from 44% ± 6% to 80% ± 8% (n = 9) (P < 0.01 for the difference between the two responses) (Figure 2). The increase in somatostatin secretion was attributed to the increase in acid secretion after incubation with pertussis toxin. Addition of VIP in pertussis toxin-treated preparations induced a sustained increase in histaminestimulated acid secretion that persisted throughout the entire 3D-minute period (Figure 2). During the latter 15-minute period, the acid response induced by histamine plus VIP after incubation with pertussis toxin [82.0 ± 13.6 nmol/min (80% ± 11%)] was significantly higher than the response induced by histamine plus VIP without prior treatment with pertussis toxin (P < 0.001; n = 11) (Figure 2).
Effect of Vasoactive Intestinal Polypeptide Antagonists on Basal and Vasoactive Intestinal Polypeptide-Stimulated Acid and Somatostatin Secretion Addition of selective VIP antagonists inhibited basal somatostatin secretion: VIP[10-28] (1 J..Lmol/L) caused a decrease of 8.9 ± 1.5 pg/min (33% ± 3%)
GASTROENTEROLOGY Vol. 100, No.5
(P
< 0.001; n = 7) and [4CI-D-Phe 6 ,Leu 17 ]VIP (1
J..Lmol/L) a decrease of 15.7 ± 2.3 pg/min (37% ± 3%) below basal level (P < 0.001; n = 7) (Figure 3). Acid secretion decreased slightly (7%-9% below basal levels) (Figure 3). At the same concentrations, both VIP antagonists abolished VIP-stimulated somatostatin secretion and caused a sustained increase in VIP-stimulated acid secretion that persisted throughout the entire 30minute period [29.4 ± 4.3 nmol/min (36% ± 5%); P < 0.001; n = 6 with VIP[10-28] and 28.8 ± 5.5 nmol/min (27% ± 3%); P < 0.001; n = 6 with [4CI-DPhe6 ,Leu 17 ]VIP] (Figure 4). Discussion This study shows that VIP has a direct stimulatory effect on acid and somatostatin secretion in the isolated luminally perfused mouse stomach. Compared with the effect of other acid secretagogues, such as histamine (14), the effect of VIP is both small (about 20% of maximal histamine-stimulated secretion) and
110
t
VIP Antagonist 1JiM
,
------------------~
1
c
0 ''::;
...
OJ ~ uOJ <0 VI'"
...
.:: ~
...
90
80
~o
o~
<0
~
E
70
0 VI
60
[4CI-D-Phe6.Leu17) VIP ...-/ 0-( 110 c~
0._ <0
....OJ '"<0 bCIJ
OJ-
VlO
+';f!.
90
:J:~
80
o-L _IO~..J.....-...J.O-.l..--I0'-1---1-2-':~-.l..--=i'::-0- - - 1 _4-,:IO:--L--f/ ~5
Minutes
Figure 3. Effect of 1 j..Lmol!L VIP[1D-28] (e) or 1 j..Lmol!L [4CI-DPhe6 ,Leu"]VIP (0) on basal acid and somatostatin secretion. Dotted lines indicate the levels of basal secretion. Data are means ± SE of 5-7 experiments.
VIP REGULATION OF ACID AND SOMATOSTATIN
May 1991
.
VIP (l11M) + VIP Antagonist (l11M)
~
180 170 160
§
1 I
150
I
'';:;
~::::-
~,'
.~~ 130
........
C;~
....!1l
E o
\
\~
\~ \
120
~
VI
VIP Alo",
/
140
CU!1l
VI'"
.:go
,L.r
110
+
100
",
VIP[1~2~~t 4
90 80
01. 150
140
/' + VIP [10-28]
130
+,
c~
0._ !1l
....~~'"
v ....
[4CI-D-Phe 6 , Leu'7] VIP
\i
120
~o
+ ~
:r:~
\
110
" T VIP Alone
't(
100
90
01.
'--'- - - ' - - - - ' - - ' - - - - - ' - - - - - - - ' - - - - t / I----"
-10
0
10
20
30
40
60
Minutes
Figure 4. Effect of VIP alone (0) or in combination witb 1 ~mol!L VlP[1D-28] (e) or 1 ~mol!L [4CI-D-Phe 6 ,Leu17]VlP (0) on acid and somatostatin secretion. Dotted lines indicate tbe levels of basal secretion. Data are means ± SE of 6-8 experiments.
transient. The transient nature of acid secretion probably represents an inhibitory effect of concurrent stimulation of somatostatin secretion. Compared with acid, somatostatin secretion was sustained throughout the period of stimulation with VIP. The sustained nature of this effect despite the return of acid secretion to control levels implies that the VIP-induced somatostatin secretion was not a consequence of luminal acidification. We had previously shown that potent acid secretagogues, such as
1199
histamine, which do not have a direct effect on the somatostatin cell, cause a substantial increase in somatostatin secretion as a result of luminal acidification (14). The effect of VIP on somatostatin and acid secretion was not affected by atropine, cimetidine, or the axonal blocker TTX, implying that it was not mediated by release of endogenous neurotransmitters or histamine. These direct stimulatory effects of VIP on acid and somatostatin secretion confirm results obtained in isolated suspensions of parietal cells and cultures of somatostatin cells derived from the fundus of the canine stomach (5). The possibility of functional coupling between the responses of different cells cannot be determined in preparations enriched in one cell type (Le., parietal or somatostatin cells). In the isolated, acid-secreting luminally perfused mouse stomach that retains intact paracrine and neural mechanisms (11,14,16), the acid and somatostatin responses appear to be functionally coupled. This was mostly evident in the reciprocal inhibitory effect of somatostatin on acid secretion and provides an explanation for the transient nature of the acid secretory response to VIP. This interpretation is supported by the fact that elimination of this inhibitory effect by preincubation with pertussis toxin converts the transient secretion induced by VIP to a sustained secretion throughout the period of stimulation. This effect was observed whether VIP was given by itself or in combination with histamine. In previous studies we had shown that pertussis toxin blocks the inhibitory effect of exogenous and endogenous somatostatin on acid secretion in this preparation when the stimulus of acid secretion (e.g., histamine) acts via cyclic AMP but not when it acts via other intracellular messengers (e.g., gastrin) (15-17). The effect of VIP in various tissues (18), including gastric tissues (6), is mediated via cyclic AMP and would thus be susceptible to pertussis toxin. Further support for the presence of a reciprocal inhibitory influence of somatostatin on VIP-stimulated acid secretion was obtained by the use of selective VIP antagonists (19). The antagonists abolished the somatostatin response to VIP but augmented acid secretion, which became sustained throughout the period of observation. The results imply (a) that somatostatin secretion is much more sensitive to stimulation by VIP and to blockade by selective VIP antagonists than acid secretion and (b) that the concurrent increase in somatostatin secretion is an important cause of the transient, partial nature of the acid response to VIP. The results also suggest that parietal cells are relatively insensitive to VIP and VIP antagonists. An important aspect of the present study is that in the absence of exogenous VIP, both VIP antagonists
1200
MITCHELL 1. SCHUBERT
significantly inhibited basal somatostatin secretion. This implies that VIP neurons in the stomach are tonically active and that input from these neurons is responsible, at least in part, for basal somatostatin secretion. References 1. Larsson 11, Fahrenkrug J, Schaffalitsky DE, Muckadell 0, Sundler F, Hakanson R, Rehfeld JF. Localization of vasoactive intestinal peptide (VIP) to central and peripheral neurons. Proc Natl Acad Sci USA 1978;73:3197-3200. 2. Schultzberg M, Dreyfus CF, Gershon MD, Hokfelt T, Elde RP, Nilsson G, Said SI, Goldstein M. VIP-, enkephalin-, substance P-, and somatostatin-like immunoreactivity in neurons intrinsic to the intestine: immunohistochemical evidence from organotypic tissue cultures. Brain Res 1978;155:239-248. 3. Chiba T, Taminato T, Kadowaki S, Abe H, Chihara K, Seino Y, Matsukura S, Fujita T. Effects of glucagon, secretin, and vasoactive intestinal peptide on gastric somatostatin and gastrin release from isolated perfused rat stomach. Gastroenterology 1980;79:67-71. 4. Saffouri B, Duvall JW, Arimura A, Makhlouf GM. Effects of vasoactive intestinal peptide and secretin on gastrin and somatostatin secretion from the perfused rat stomach. Gastroenterology 1984;86:839-842. 5. Chiba T, Park J, Yamada T. Glucagon and vasoactive intestinal polypeptide stimulate somatostatin secretion from isolated canine fundic mucosal cell cultures (abstr). Gastroenterology 1985;88:1348. 6. Dupont C, Gespach C, Chenut B, Rosselin G. Regulation of vasoactive intestinal peptide of cyclic AMP accumulation in gastric epithelial cells. FEBS Lett 1980;113:25-28. 7. Vagne M, Konturek SJ, Chayvialle JA. Effect of vasoactive intestinal peptide on gastric secretion in the cat. Gastroenterology 1982;83:250-255. 8. Makhlouf GM, Zfass AM, Said SI, Schebalin M. Effects of synthetic vasoactive intestinal peptide (VIP), secretin and their partial sequences on gastric secretion. Proc Soc Exp Bioi Med 1978;157:565-568. 9. Holm-Bentzen M, Christiansen J, Kirkegaard P, Olson PS, Petersen B, Fahrenkrug J. The effect of vasoactive intestinal
GASTROENTEROLOGY Vol. 100, No.5
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
polypeptide on meal-stimulated gastric acid secretion in man. Scand J GastroenteroI1983;18:659-661. Bunce KT, Parsons ME. A quantitative study of metiamide, a histamine-H 2 -antagonist, on the isolated whole rat stomach. J Physiol Lond 1976;258:453-465. Schubert ML, Edwards NF, Arimura A, Makhlouf GM. Paracrine regulation of gastric acid secretion by fundic somatostatin. Am J PhysioI1987;252:G485-G490. Schubert ML, Saffouri B, Walsh JH, Makhlouf GM. Inhibition of neurally mediated gastrin secretion by bombesin antiserum. Am J PhysioI1985;248:G546-G562. Yamada T, Marshak D, Basinger S, WalshJH, Morley J, Steel W. Somatostatin-like immunoreactivity in the retina. Proc Natl Acad Sci USA 1980;77:1691-1695. Schubert ML, Edwards NF, Makhlouf GM. Regulation of gastric somatostatin secretion in the mouse by luminal acidity: a local feedback mechanism. Gastroenterology 1988;94:317-322. Park J, Chiba T, Yamada T. Mechanisms for direct inhibition of canine gastric parietal cells by somatostatin. J Bioi Chern 1987;262:14190-14196. Schubert ML, Hightower J. Inhibition of acid secretion by bombesin is partly mediated by release of fundic somatostatin. Gastroenterology 1989;97:561-567. Schubert ML, Hightower J, Makhlouf GM. Linkage between somatostatin and acid secretion: evidence from use of pertussis toxin. AmJ PhysioI1989;256:G418-G422. Amiranoff B, Rosselin G. VIP receptors and control of cyclic AMP production. In: Said SI, ed. Vasoactive intestinal polypeptides. New York: Raven, 1982:307-322. Pando I SJ, Dharmsathaphorn K, Schoeffield MS, Vale W, Rivier J. Vasoactive intestinal peptide receptor antagonist [4CI-DPhe',Leu 17 ]VIP. Am JPhysioI1986;250:G553-G557.
Received April 3, 1990. Accepted October 2,1990. Address requests for reprints to: Mitchell 1. Schubert, M.D., McGuire Veterans Administration Hospital, Division of Gastroenterology; Code I11N, 1201 Broad Rock Boulevard, Richmond, Virginia 23249. This work was supported by the Veterans Administration Medical Research Fund and the American Gastroenterological Association Industry Research Scholar Award.