Oral PGE2 inhibits gastric acid secretion in man

PROSTAGLANDINS ORAL PGE 2 INHIBITS GASTRIC ACID SECRETION IN MAN. Befrits R., Johansson C. Departments of Medicine and Clinical Chemistry Karolinska Hospital 104 01 Stockholm Sweden

ABSIRACI

The effect of oral prostaglandin E 2 (PGE 2) on gastric acid secretion was.examined in healthy subjects. The gastric secretion was stimulated by a modified shamfeeding procedure. Each subject underwent one control test and three tests with intragastrically administered graded doses of PGEp: 0.5, 1.0 and 2.0 mg. Oral P G L significantly suppressed the peak and total acid response z to vagal stimulation. The total acid output in controls was 27.5 + 3.2 mmol/90 min and 20.8 + 2.8, 15.8 + 2.2 (p
In 1967 i.v. administration of prostaglandins of the E-type were found to inhibit the gastric acid in the dog (1) and later in man (2,3). Continuous exposure of the gastric mucosa for PGE 2 in the rat was followed by a reduction of the acid secretion rate (4), but investigations in man with oral natural prostaglandins failed to demonstrate any acid inhibition (5,6). From this reason and in the absence of any proven effect of cyclo-oxygenase blockers, Bennett et al suggested that endogenous PGE_ was an improbable modulator of the gastric acid secretion (7). Thi~ seemed unlikely to us, since PGE_z is a major metabolite of arachidonic acid metabolism in the human gastric mucosa (8,9). We therefore undertook this study in which the gastric acid secretion has been examined in response to modified shamfeeding after graded doses of oral PGE 2.

JANUARY 1985 VOL. 29 NO. 1

143

PROSTAGLANDINS

METHODS

Four tests were made in each of five healthy male volunteers aged 22 to 29 years and of normal body weight (range 70 to 82, mean 79 kg). The experiments were made after an overnight fast on separate days and in random order, with placebo and with 0.5, 1.0 and 2.0 mg PGE?. Crystalline PGE 2 was kindly supplied as a gift from the Upjohn Ltd~ Canada and was Kept at -20 ° . Immediately before use the doses were dissolved Jn 95% w/v ethanol and diluted to 10 ml with isotonic saline. The same amount of ethanol dJiuted Jn isotonic saline was given as placebo. The study was approved by the Ethical Committee at the KarolJnska Hospital and the subjects gave informed consent to participate. Experimental procedure The subject was intubated with a Ryle's tube (Portex 14 FG) to which a thinner tube (Portex NT/3/80) had been attached with its end 10 cm proximal to the tip of the wider tube. With the subject on his left side in a semi-recumbent position and connected to a salivary suction apparatus, the residual gastric contents were aspirated and the stomach rinsed with 150 m] distilled water. A catheter was placed in an antecubitai vein for blood sampling at intervals. During the test the stomach was perfused with a mannitol solution (Nannitol, 150 g • L -1, ACO, Sweden) dJlutedwith distilled water, which contained 35 ug - m1-1 vitamin B12 as a marker (Cycobemin, Sigma, USA). The pH of the solution is 6.4, the osmolality 260 mosm, and it has no buffering capacity. It was infused through the proximal tube by a pump (Infusomat, Braun, Germany) giving 10 ml • min -1. Gastric contents were collected through the distal tube by siphonage and gentle manual suction in 15 min periods. After 3 x 15 min coIlections of the basal gastric secretion the tubes were clamped, and PGE 2 or placebo was given intragastrically. Fortyfive minutes thereafter the gastric residuum was removed and the perfusion and gastric collections were continued during and after shamfeeding

(rig 1). Modified shamfeeding was made by the "chew-and-spit"-method, which has been validated against adequate shamfeeding in duodenal ulcer patients by Stenquist et al. (10). Subjects were served buttered toast, fried ham and eggs to chew and spit out during 15 min. They were instructed to avoid swallowing the food and to rinse the mouth with orange juice between morsels. The orange juice contained polyethylene glycol (PEG) 4000 as a marker of the meal.

144

JANUARY 1985 VOL. 29 NO. 1

PROSTAGLANDINS

Chemical analyses and calculations The volumes of the gastric recoveries were measured and the acidity was determined by automatic titration of I - 2 ml samples with 0 . 1 M NaOH to pH 7 kRadJometer, Copenhagen). Gastric recoveries were analysed for concentrations of chloride by titration with AgNO 3 and of vitamin B19 (11) and PEG (12). Concentrations of sodium ~nd potassium were'aetermJned using flame photometry in gastric collections from controls and in series with 2.0 mg PGE 2. Plasma levels of immunoreactive gastrin were determined with RIA using antibodies kindly supplied by J. Rehfelt, Denmark. AII analyses were made in duplicate. The correct volume for each 15-min period was calculated from the infused amount of marker divided by the marker concentration in the gastric collections. The corrected volume multiplied by concentrations of H+, K+, Na+ equals the output during 15 min. The corrected volume subtracted by the infused volume equals the volume of the gastric secretion, provided that the reflux of duodenal contents into the stomach i s negligible. The peak output or volume is the sum of the two highest consecutive 15 min periods a f t e r starting the shamfeeding in the control experiments and the identical half hour Jn tests with active drugs. The t o t a l secretory response denotes the output or secreted volume during 1.5 hrs following the start of the shamfeeding. Means of corrected values are given X SEM Jf not otherwise stated. Student's t - t e s t for paired differences was used to test significance. RESULTS An average 88 + 3, 89 + 1, 91 + 3 and 84 + 3 p e r c e n t o f the marker

was recovered Tn gastrTc colle~tJons from--control series and tests with 0.5, 1.0, 2.0 mg PGE2 respectively. Oral PGE2 45 min prior to stimulation suppressed the gastric acid secretory response to modified shamfeedJng in healthy subjects (fig I). The reduction by 0.5 mg oral PGE? was slight and insignificant. Both 1.0 and 2.0 mg PGE2 induced ~JgnifJcant inhibitions of the peak and total gastric acid responses (table I). The inhibitory effect of the two higher doses did not differ. Both 1.0 and 2.0 mg suppressed the volumes secreted during 1.5 h after start of the shamfeeding but only the lower dose reduced the peak volume. The gastric chlorides were reduced in parallel to the acid suppression ( t a b l e 1 ). Gastric outputs of sodium and potassium in response to modified shamfeedJng were significantly reduced by 2.0 mg PGE (table 1). 2

JANUARY 1985 VOL. 29 NO. l

145

PROSTAGLANDINS

i i.g.Irng PGE2 /placebo Acid output mmol/15rain

T 6

,'

J ~Lr3 I J-I * L "r-I

2

"I

I

-X.

J. }t

~'1 *

0

i

0

Fig 1.

146

controls

60

i

i

I

120

I

I lmg PGE 2 I I ,

180 rain.

Design of the modified shamfeeding (MSF) experiments. The gastric output (mmol/15 min) was suppressed by 1 mg PGE 2 given intragastrically (i,g.) 45 min prior to start of the shamfeeding. Mean + SEM; n:5. * P-level for difference from controls < 0.05.

JANUARY 1985 V O L 29 NO. 1

PROSTAGLANDINS

(.-

~

• t'M

bJ (...3 03 o..

v

+~

O

c;

4-1 £'4

,--4 ('0

t'M

I

I

c;

+

4-1

E E

v

:'

{:3 4-1 t.m (',4

(=3

I

I

+1

t.~

£'4

03 03 E

.E ,--i 4.~ o .,-{ > 3: "(:3

~k

03 03

©

[.4

¢J

co

+

O Z

E

",,,D

t'M

~

c~ +l

c;+ 1

i

i

+

O E E

Z

C3

-I-I

'43

I

C3

I

+1

O',

'q3

",,,D

03 " (o

t.P I:)

03 ,.-I O

I-

C3

4..)

Z 03

+

o (J

~O Z

,;,

,;,

~,

';,

@,J

6(3 122)

~ CO

122)

(=3 l'.,'h {J'3

I

O

I

~ r..j

O

E E

:

.J ,~ )-C3 I.-.-

D..

~O ~O 4~

+l

+l

+l

~

t=-

~-.

~

[:~t

,=; +1

d+ t

;,

,43

C'.4

I~

03

o3

C3 O~

r~ v ,~

+t

O rj

Ol.--

D.. t--"m

~0

03 --~

(../3

.q

403 "I- 0J

V

Z ~'-~r

E © ,-4 I O ,-~ E (.J E

•

•

°

t403 f.) r'03 f-4 03

O. 03 4J .~

°~ ,-.I m

0 E

°° c"

0

0

.~

O

"~ rm

03 U'I m ~ +l o v D,,. r" [u 03 030303

-r"

E E

+1

+l

+t

+1 > 03 ,--I

u3 ~J

E

E

.e-03 ,...-i ..o

f--4 O +

U3

~ c"

~A

JANUARY 1985 VOL. 29 NO. 1

E

E r~ ~J U3

03

u'~

U3

C • O '~3 r

,,i (-~ o_

E

O

E

C3

147

PROSTAGLANDINS

There was a slight but significant release of gastrin in the control I experiments; the elevation amounted to an average ~.8 + 0.6 pg • ml- • min -1 (p0.025) during 30 min following start of the sh~mfeeding procedure. Although all three doses of PGE 2 gave a late suppression of the gastrin levels, the early release pattern varied in a non dose-related way. Paradoxically the dose of 2.0 mg PGE 2 appeared to give a similar elevation as in controls (2.4 + 1.1 pg" ml -I. min -I, p
(fig 2).

ZX P-GAS pg.m1-1 32 1 0' -1

i

X/ , , \ ,

\,

--,..,

,

i5min periods

-2 -3

Fig 2.

148

Changes of p-gastrin in response to modified shamfeeding in control tests and after administration of three doses of PGE 2 intragastrically. Mean; n:5.

JANUARY 1985 VOL. 29 NO. 1

PROSTAGLANDINS DISCUSSION The present study clearly demonstrates that oral PGE 2 inhibits, the gastric acid secretion in response to modified shamf~eding in man. 1.0 and 2.0 mg oral PGE 2 produced a signicifant approximately 40% inhibition, whereas the-reduction by 0.5 mg was marginal and insignificant. The observation agrees with earlier Jn vJvo experiments in the rat in which gastric perfusion of PGE I (-~) or PGE 2 (4) suppressed the basal secretion and the acid response to histamine, pentagastrin and electrical vagal stimulation. In contrast, later investigations in the dog (14) and in man (5,6) have failed to show acid suppressive effects of topically or orally administered natural PGs. This failure has usually been attributed to a supposed degradation of natural PGs in the gastric lumen (5,6). However, the observation that oral PGE I without acid antisecretory effects nevertheless produced diarrhoea in healthy subjects (5) argues against a rapid digestive inactivation of the compound. In addition, recent in vitro experiments showed that added PGE9 is stable in freshly withdrawn gastric juice; further more, PGE 2 but ~ot its metabolites could be identified in human gastric contents (15). An alternative reason for the absent acid suppressive properties of orally administered natural PGs in the dog and in man is that tested doses have been inadequate. In man, Horton found oral 0.7 mg PGE 1 to be ineffective whereas 2.8 mg apparently increased the acid response to pentagastrin (5). Karim reported that oral PGE 2 at 2.5 to 4.0 mg was ineffective to suppress the human basal secretion but re-evaluation of his resuIts show that the acid secretion rate is consistently higher'after active drug compared to controls (6). h- I Also, in man during hJstamJnestimulation i.v. PGA I at 5.25 mg • was less acid suppressive than half of this dose (16). Thus, it seems from these experiments that very high doses appear to be less effective to suppress acid and may even have a paradoxicai stimulatory effect. This phenomenon may account for the failure in the present study to gel a dose-dependent acid inhibition by orai PGE?; oral 2 mg PGE 7 was found no more effective than I mg. Interestingiy; shamfee6ing after the larger dose produced higher plasma gastrin levels compared to the two lower test doses. The observed inhibition of the acid response to shamfeedin~ by oral PGE 2 in man revives the hypothesis that endogenously produced •prostaglandins Lake part in gastric acid regulation (7). A number of studies exist in support of such a modulatory rSle. Infusion of arachidonic acid to the dog (17) and rat (18) suppressed the acid secretion. The inhibition was prevented by blockers of endogenous PG-biosynthesis. In man, PGE 2 is a major metabolite of arachidonic acid metabolism in tissue preparations of the gastric mucosa (8) and can be recovered from human gastric contents (19).

JANUARY 1985 VOL. 29 NO. 1

149

PROSTAGLANDINS Several studies have failed to demonstrate effects of blockers of PG-biosynthesJs on the gastric acid secretion in the dog and Jn man (7, 20, 21) but other studies in the rat (22) and recently in man (23) have shown an enhancement of the basal and stimulated gastric acid secretory response. A study from our group suggests that endogenous PGs participate in inhibitory regulatory mechanisms of gastric acid secretion (24). It is therefore not surprising that natural prostaglandJns do affect the gastric acid secretion after oral administration, such as recently reported (25) and demonstrated in the present study. ACKNOWLEDGEMENTS This work was supported by the Swedish Medical Research Council 19X-00580 and grants from the Karolinska Institute. REFERENCES I.

Robert, A., N@zamis, 3.E and Phillips, 3.P. Inhibition of gastric secretion by prostaglandins. Am J Dig Dis, 12: 10731076, 1 9 6 7 .

2.

Classen, M., Koch, H., Bickhardt, J., Topf, G. and Demling, L. The effect of prostaglandin E I on the pentagastrin-stimulated gastric secretion inman. Digestion, 4: 333-344, 1971.

3.

Newman, A', de Moraes-Filho, J.P.P., Philippakos, D. and Misiewicz, J.J. Tbe effect of intravenous infusions of prostaglandins E 2 and F2 on human gastric function. GUI, 16: 272-276, 1975.

4.

Banerjee, A.K., Phiilips, J.~and Winning, W.M. E-type prostaglandins and gastric acid secretion in ~he rat. Nature (New Bio!) , 238: 177-179, 1972.

5.

Horton, E.W., Main, I.H.M,,Thompson, CIJ. and Wright, P.M. Effect of orally administe£ed prostaglandin E I on gastric secretion and gastrointestinal motility in man. GUT, 9: 655-658, 1968.

6.

Karim, S.M.M., Carter, D.C., Bhana, D. and Ganesan, P.A. Effect of orally administered prostaglandin E 2 and its 15-methyl analogues on gastric secretion. #tit Ned J, 1: 143-146, 1973.

7.

Bennett, A., Stamford,' I.F. and ~nger, W.G. Prostaglandin E 2 and gastrip acid secretion in man. J Phy~iol, 229: 349-360, 1973 . . . .

8.

Peskar, ~.M. On the synthesis of prostaglandins by human gastric mucosa and its modification by drugs. Biochim et Biophys A e t a , 487: 307-314, 1977.

150

JANffARY 1985 VOL. 29 NO. 1

PROSTAGLANDINS

9.

Aly, A., 3ohansson, C. and Green, K. Bioconversion of arachJdonic acid in the human gastrointestinal tract. Biochem Ned, in press, 1984.

10.

Stenquist, B., Knutson, U. and Olbe, L. Gastric acid responses to adequate and modified shamfeeding and to insulin hypoglycemia in duodenal ulcer patients. Scand J Gastroent, 13: 357-362, 1978.

11.

Larss6n, L.E.I. Non-labelled vitamin B12 as a dilution indicator in gastrointestinal research. J Clin ]nvest, 18: 535-539, 1966.

12.

Hyd6n, S. The recovery of polyethylene glycol after passage through the digestive tract. Inst Anim Physiol, 22: 411-424. 1955.

13.

Shaw, 3.E. and Ramwell, P.W. in rats by prostaglandin E I. p 55-64, 1967.

14.

Robert, A., Schultz, J.R., Nezamis, J.E. and Lancaster, C. Gastric antisecretory and antiulcer properties of PGE~, 15-methyl PGE2 and 16,16 dimethyl PGE 2. Gastroenterology, 70: ~359-370, 1976.

15.

A]y, A., Fitzpatrick, F-A., Green, K., Johansson, C. and ffynalda, F. Gastric luminal contents of PGE 2 dosing pentagastrin and cholJnergJc stimulation Jn man. Measurement by gas chromatography-mass spectrometry. J ClJn Invest, submitted,

16.

Wilson, D.E., Phillips, C. and Levine, R.A. Inhibition of gastric secretion in man by prostaglandin A I. Gastroenterology, 61: 201-206, 1971.

17.

Conoiiy, M.E., Bieck, P.R., Payne, N.A. Adkins, B. and Oates, J.A. Effect of the prostaglandin precursor, arachidonic acid, on histamine stimulated gastric secretion in the conscious dog, and observation on the effect of inhibiting endogenous prostaglandin synthesis. GUT, 18: 429-437, 1977.

18.

Frame, H.H. and Main, I.H.N. Effects of arachidonic a~id on rat gastric acid secretion in response to different secretogogues; inhibition of these effects by indometacin. Br 3 Pharmac, 69: 171-178, 1980.

19.

Aly, A., Green, K. and Johansson, C. Prostaglandin E 2 in basal gastric secretion and during stimuIation of muscarinic reseptors in man. Measurements with gas chromatography-mass spectrometry. Acta Phys Scand, In press, 1984.

JANUARY 1985 VOL. 29 NO. 1

Inhibition of gastric secretion Worcester Foundation for Exp Biol,

151

PROSTAGLANDINS 20.

Nicoloff,

21.

Winship, D.H. and Bernhard, G.C. Basal and histamine-stimulated human gastric acid secretion. Gastroenterology, 58: 762-765, 1970.

22.

Main, I.H.M. and W h i t t l e , B.J.R. I n v e s t i g a t i o n of the vasod i l a t o r and a n t i s e c r e t o r y r o l e of prostaglandins in the rat g a s t r i c mucosa by use of n o n - s t e r o i d a l anti-inflammatory drugs. Br J Pharmac, 53: 217-22~, 1975.

23.

Levine, R.A. and Schwartzel, E.H. Effect of indomethacJn on basal and histamine stimulated human gastric acid secretion. GUJ, 25: 718-722, 1984.

24.

BefrJts, R., Samuelsson, K. and Johansson, C. Gastric acid inhibition by antral acidification mediated by endogenous prostaglandins. Scand J Gastroent, Jn press, 1984.

25.

Reele, S.B. and Bohan, D. Oral antJsecretory activity of prostaglandin E 2 in man. Dig Dis ScJ, 29: 390-393, 1984.

D.M.

Editor: B. Whittle

152

Indomethacin.

Arch Surg, 97: 809-815,

Received:7-13-84

1968.

Accepted:10-29-84

JANUARY 1985 VOL. 29 NO. 1