Effect of Gastric Inhibitory Polypeptide on Pentagastrin-Stimulated Acid

Secretion in Man V E R N O N M A X W E L L , BS, A R T H U R S H U L K E S , PhD, JOHN C. BROWN, PhD,

TRAVIS E. SOLOMON, MD, PhD, JOHN H. WALSH, MD, and MORTON I. GROSSMAN, MD, PhD

Eight male subjects were given pentagastrin by intravenous infusion in doses of 25, 74, 222, 667, and 2000 ng/kg/hr, each dose for 30 rain. On another day the same subjects were given the same doses of pentagastrin while gastric inhibitory polypeptide (GIP) was being infused intravenously in a dose of 2 txg/kg/hr. At the 222 ng/kg/hr dose of pentagastrin, acid output was significantly lower with GIP; at all other doses of pentagastrin, acid output did not differ significantly in tests with and without GIP. Pepsin output in the tests with and without GIP did not differ significantly at any dose of pentagastrin. Plasma concentration of GIP, measured by radioimmunoassay, showed a mean +- SE plateau level of 7.4 +_ 1.4 ng/ml during GIP infusion and 0.4 +- 0.1 ng/ml peak level after a stan- dard meal. We conclude that the increase in blood concentration Of GIP produced by feeding is probably inadequate to cause significant inhibition of gastric acid or pepsin secretion in man.

Using inhibition of acid secretion from Heidenhain pouches to follow purification, Brown and cowork- ers (1) isolated gastric inhibitory polypeptide (GIP), a 43-amino acid linear peptide, from porcine in- testinal mucosal extracts. In dogs with both gastric fistulas and Heidenhain pouches, during stimulation with pentagastrin, GIP strongly inhibited acid se- cretion from the Heidenhain pouch but had much less effect on acid secretion from the gastric fistula (2). In human subjects, a large dose of GIP given by rapid intravenous injection inhibited pentagastrin stimulated acid secretion (3).

The aim of the present study was to determine whether intravenous infusion of GIP inhibits pen-

Manuscript received June 8, 1979; revised manuscript re- ceived September 18, 1979; accepted September 28, 1979.

From the Center for Ulcer Research and Education VA Wads- worth Hospital Center, and UCLA School of Medidine, Los An- geles, California, USA; and Department of Physiology, Univer- sity of British Columbia, Vancouver, BC, Canada

Address for reprint requests: Dr. Morton I. Grossman, VA Wadsworth Hospital Center, Building 115, Room 115, Los Ange- les, California 90073.

tagastrin-stimulated acid secretion in human sub- jects. To assess the physiological role of GIP as an inhibitor of acid secretion, we compared blood GIP concentrat ions after exogenous infusion and after a meal.

MATERIALS AND METHODS

Eight male subjects, two normal and six asymptomatic duodenal ulcer patient s , were tested on two separate days. The mean age was 53 years, range 29-69 years. Af- ter an overnight fast, a radiopaque nasogastric tube (16 French) was fluoroscopically positioned with the tip in the middle,of the gastric antrum. Gastric juice was aspi- rated by a pump with pressure of 7-12 mm Hg below at- mospheric. The nasogastric tube was manually flushed with 10-20 ml of air and aspirated by hand at 5-rain inter- Vals to maintain patency. Residual juice was collected for 15 min and discarded, and then basal secretion was col- lected for three 10-rain periods. On one day, pentagastrin (Ayerst Laboratories, New York) was infused in graded doses of 0, 25, 74, 222,667, and 2000 ng/kg/hr. The doses were given in the order listed. Each dose was diluted with 0.15 M NaC1 and infused for 30 rain into a forearm vein by

Digestive Diseases and Sciences, Vol. 25, No. 2 (February 1980)

0163-2116/80/0200-0113503.00/1 �9 1980 Digestive Disease Systems, Inc. 113

MAXWELL ET AL

a syringe infusion pump (Harvard Apparatus, Millis, Massachusetts) at a rate of 0.53 ml/min. On the other day, GIP was added to each dose of pentagastrin in an amount sufficient to give a dose of 2/xg/kg/hr. Immediately before each test, the required amount of GIP was dissolved in water for injection containing 1 mg/ml of human plasma albumin. The solution was passed through a 0.22-/zm fil- ter (Swinnex-25 SXGS 025, Millipore Co., Bedford, Mas- sachusetts) which had first been thoroughly washed with water containing 1 mg/ml albumin. Samples of the in- fusion fluid containing GIP were taken from the infusion tubing at the end of the test. The GIP concentration as measured by radioimmunoassay did not differ significant- ly from the calculated value. The order of tests with and without GIP was randomized. Acid concentration was de- termined by titration of 0.2 ml of gastric juice with 0.2 M NaOH to pH 7.0 on an automatic titrator (Radiometer, Copenhagen, Denmark). Pepsin was measured by the method of Berstad (4). Before and at 30-rain intervals dur- ing both tests, venous blood samples for measurement of plasma GIP by radioimmunoassay were taken thi'ough an indwelling needle in a vein in the arm oppositeto the one in which the infusion was given. 1000 KIU of aprotinin (FBA Pharmaceuticals, New York, New York) was add- ed to each ml of blood. On a third test day the subjects ate a meal consisting of 2 hard boiled eggs, 2 slices of but- tered toast, and 8 ounces of orange juice (53 g carbohy- drate, 20 g protein, 22 g fat. 495 kcal). Blood samples were taken before and at 15-rain intervals for 2 hr after the meal.

This study was approved by the Committee on Human Studies of the VA Wadsworth Hospital Center (Decem- ber 13. 1977).

GrIP RadioimmUnoassay. Antisera to GIP were raised in rabbits immunized in our laboratory with pure porcine GIP conjugated to keyhole limpet hemocyanin (Calbio- chem) by the carbodiimide method (5). The conjugate was emulsified with complete Freund's adjuvant and admims- tered by multiple intradermal injections and boosted twice at intervals of six weeks, The equivalent of 15 nmol of GIP was given in each injection. One rabbit (7831) pro- duced significant antibodies after the second boost, and this antiserum was used in the present study at a final dilution of 1:7000.

The same porcine GIP used for immunization was used also as the radioimmunoassay standard and stored at -70~ in 100-/zl portions containing 1/zgJml GIP in 0.2 M acetic acid with 0.5~A (w/v) bovine albumin and aprotinin. 200 KIU/ml. A new vial Was used for each assay. No loss in potency of GIP solutions stored under these conditions was apparent as determined by inhibitory potency in ra- dioimmunoassay standard curves performed on different occasions with the same antibody. GIP was labeled by the chloramine T method (6) and immediately purified on a G10 Sephadex column (1 • 10 cm) eluted with 0.2 M ace- tic acid. 5% Plasmanate (Cutter Laboratory), and aproti- nin (200 KIU/ml). The label was further purified on CM- Sephadex (1 • 20 cm) equilibrated in 0.05 M ammonium acetate, pH 5.0, containing 5% (w/v) Plasmanate, and aprotinin (200 KIU/ml). An elution gradient was pro- duced with 1.0 M ammonium acetate using a 50-ml Eden- meyer flask as the mixing chamber. Two thousand counts

per minute of label were added to each assay tube and there was no significant inhibition of initial binding up tO 10,000 cpm. The IDa0 was generally 70 fmol/ml. Secretin, glucagon, and VIP did not cross-react with the antibody at concentrations several orders of magnitude higher than the GIP standard. Synthetic fragments of GIP were not available for studies of the specificity Of this antibody. However, acid extracts of human and dog duodenal and jejunal mucosa contained immunoreactive GIP which eluted from G50 gel filtration columns With elution vol- umes similar to pure porcine GIP. The actual potency of human and dog GIP in the assay cannot be determined until pure peptides are available.

The assay diluent was 0.04 Mphosphate buffer, pH 6.5, containing 0.005% sodium azide, 10% (w/v) Plasmanate, and 500 KIU/ml of aprotinin. Incubation volume was 1 ml and consisted of 200/zl plasma, 100/zl antibody, and 700 /zl diluent. Incubation time was 48 hr at 4~ For assay of plasma the standard curve contained 200/xl of charcoal- stripped human plasma. The free and antibody-bound la- beled GIP were separated by dextran-coated charcoal containing 10% bovine plasma. Nonspecific binding in the absence of added antibody Was less than 10%.

All samples were measured in one assay. The intra- assay coefficient of variation for ten estimations was 5.6%.

RESULTS

At only one dose of pentagastr in (222 ng/kg/hr), was acid output significantly lower with GIP in- fusion than during the control Study (Figure 1). At all other doses of pentagastr in acid output was not significantly different in the studies with and with- out GIP. The sum of the responses to all doses of pentagastr in was not significantly different with and without GIP infusion. The effect of GIP on the re- sponse to pentagastr in was not significantly dif- ferent in the two normal subjects as compared with the six duodenal ulcer patients.

Pepsin output did not differ significantly in the tests with and without GIP at any dose of pen- tagastrin (Figure 2).

Plasma GIP concentra t ion did not change signifi- cantly during infusion of pentagastr in alone (Figure 3). Mean p lasma GIP concentrat ion increased dur- ing infusion of GIP reaching a plateau of 7.4 ng/ml at 1 hr and remaining at that level during the remain- der of the infusion (Figure 3).

The mean peak GIP response to the meal was on- ly about 0.4 ng/ml, much lower than the mean plateau response to infusion of GIP.

DISCUSSION

The present s tudy shows that infusion of exoge- nous GIP produced negligible inhibition o f pen-

114 Digestive Diseases and Sciences, Vol. 25, No. 2 (February 1980)

GIP AND ACID SECRETION IN MAN

16-

12-

8- 0o9

4"

0'

GIP, 2~gkg-lh -1 I

Opentagastrin+GIP . | �9 pentagastrin 1 .~ n=

B o 74 2228 72 PENTAGASTRIN

ngkg-lh-1 Fig 1. Acid output in response to graded doses of pentagastrin with and without a continuous infusion of GIP. B stands for ba- sal.

tagastrin-stimulated acid secretion in human sub- jects even though the dose of GIP used produced plateau blood concentrations of GIP much greater than those seen after eating a small mixed meal. Al- though higher blood levels have been seen in other studies after feeding meals conta in ing larger amounts of glucose and fat (1), these are still well below the concentrations produced by the infusion of exogenous GIP in the present study. Using the same radioimmunoassay used in the present study, Varner et al (7) found that intraduodenal fat in a dose sufficient to cause marked inhibition of acid secretion produced a peak plasma GIP level of 0.82 ng/ml. This finding suggests that GIP does not play an important role as an inhibitor of gastric acid se- cretion in man. Since studies in dogs have shown that GIP is much more effective as an inhibitor of gastrin-stimulated acid secretion in vagally dener- vated Heidenhain pouches than in vagally in- nervated gastric fistulas (2), it will be of interest to determine the effectiveness of GIP in vagotomized human subjects.

Cleator and Gourlay (3) gave 2/-~g/kg of GIP as a rapid intravenous injection and found significant in- hibition of pentagastrin-stimulated acid secretion in human subjects. Such rapid injection of GIP would produce even higher blood concentrations of GIP than those seen in the present study with continu- ous infusion of GIP.

Infusion of fat into the duodenum inhibits gastric acid secretion and releases GIP (1) but the extent to

1200 I- 13_ =~. aoo O.E

~- 400 I..iJ o..

GIP, 2~g kg-lh-1 ] �9 pentagastrin+GlP �9 pentagastrin 1

0 25 74 222 667 2000

PENTAGASTRIN ngkg-lh-1

Fig 2. Pepsin output in response to graded doses of pentagastrin with and without a continuous infusion of GIP. B stands for ba- sal.

which the release of GIP accounts for the inhibition of acid secretion is uncertain. Infusion of exoge- nous GIP in doses that produce blood concentra- tions greater than those seen after meals markedly inhibited acid secretion from Heidenhain pouches but only slightly inhibited secretion from gastric fis- tulas (2). This suggests that GIP may play a signifi- cant role in the inhibition of acid secretion by fat in the intestine in the case of the Heidenhain pouch but not in the case of the gastric fistula. Since fat in the intestine inhibits acid secretion from the gastric fistula at least as effectively as from the Heidenhain

l GIP, 2~gkg-lh-1 Qpentagastrin+GIP

~Em ~O-<~" 11 t6 ] "pentagastrinn=

<,=4

B 0 25 74 222 667 2000 PENTAGASTRIN

ng kg hr-1 Fig 3. Plasma GIP concentrations during infusion of pentagastrin alone and pentagastrin plus GIP. B stands for basal.

Digestive Diseases and Sciences, Vol. 25, No. 2 (February 1980) 1 15

M A X W E L L ET A L

p o u c h (8), this sugges t s tha t n e r v o u s f ac to r s o r an

as y e t un iden t i f i ed h o r m o n e p l a y a m a j o r ro le in in-

h ib i t ion o f ac id s e c r e t i o n f rom the i n n e r v a t e d s tom-

ach . In a d d i t i o n to inh ib i t ion o f a c id s ec re t i on , the o th-

e r m a j o r b io log i ca l a c t i o n o f G I P is e n h a n c e m e n t o f g l u c o s e - i n d u c e d r e l ea se o f insul in (1). S ince in-

fus ion o f e x o g e n o u s G I P at d o s e s w h i c h p r o d u c e b l o o d c o n c e n t r a t i o n s c o m p a r a b l e to t h o s e seen af- t e r ea t ing a l a rge m i x e d mea l c a u s e d e n h a n c e m e n t

o f g l u c o s e - i n d u c e d insul in r e l e a se (9), this a c t i on o f G I P m a y be r e g a r d e d as phys io log i ca l .

ACKNOWLEDGMENTS

This study was supported by a Veterans Administra- tion Senior Medical Investigatorship (M.I.G.) and Re- search Associateship (T.E.S.), by NIAMDD grant 17328 to CURE and by grants from the Medical Research Coun- cil of Canada and the Vancouver Foundation to J.C.B. Arthur Shulkes was supported by the Rowden White Foundation, Universi ty of Melbourne, Australia. We thank Ruth Abercrombie for preparing the figures, and Kuwa Chou for typing the manuscript.

REFERENCES

1. Brown JC, Dryburgh JR, Ross SA, Dupre J: Identification and actions of gastric inhibitory polypeptide. Recent Prog Horm Res 31:487-532, 1975

2. Soon-Shiong P, Debas HT, Brown JC: Cholinergic inhibition of gastric inhibitory polypeptide (GIP) action. Gastroenterol- ogy (in press)

3. Cleator IGM, Gourlay RH: Release of immunoreactive gas- tric inhibitory polypeptide (IR-GIP) by oral ingestion of food substances. Am J Surg 130:128-135, 1975

4. Berstad A: A modified hemoglobin substrate method for the estimation of pepsin in gastric juice. Scand J Gastroenterol 5:343-348, 1970

5. Goodfriend TL, Levine L, Fasman GD: Antibodies to bra- dykinin and angiotensin: a use of carbodiimides in immunolo- gy. Science 144:1344-1346, 1964

6. Hunter WM, Greenwood FC: Preparation of iodine-131-1a - belled human growth hormone of high specific activity. Na- ture 194:495-496, 1962

7. Varner AA, Isenberg JI, Elashoff JD, Lamers CB, Shulkes AA: Intravenous fat inhibits amino acid stimulated gastric acid secretion in man. Gastroenterology 76: 1264, 1979

8. Debas HT, Yamagishi T: Gastric inhibitory polypeptide (GIP) is not the primary mediator of the enterogastrone action of fat. Gastroenterology 74:1118, 1978

9. Andersen DK, Elahi D, Brown JC, et al: Oral glucose aug- mentation of insulin secretion. Interactions of gastric inhib- itory polypeptide with ambient glucose and insulin levels. J Clin Invest 62:152-161, 1978

116 Digestive Diseases and Sciences, Vol. 25, No. 2 (February 1980)