Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
Glit, 1987, 28, 1291- 1295
Effect of luminal pH on the output of bicarbonate and
PGE2 by the normal human stomach
J R CRAMPTON, L C GIBBONS, AND W D W REES
Frotni the Department of Gastroenterology, Hope Hospital, University oJ Manchester School of Medicine,
Salford
The gastric output of bicarbonate and prostaglandin E2 has been calculated using a
perfusion technique before and after instillation of 100 mM hydrochloric acid into the stomach of
seven healthy volunteers. A significant increase in bicarbonate output occurred from 258±38
[tmol/3() min during the basal period to 531±86 [tmol/3() min after return of the intragastric pH to
neutral (p<()0()5). Prostaglandin E2 output also increased significantly from 410±136 pmol/30 min
to 1002±194 pmol/30 min (p<()-05). The changes were caused mainly by an increase in gastric
secretory volume with only non-significant increases in concentrations of bicarbonate and
prostaglandin E,. The results suggest that mechanisms exist to adjust the rate of gastric bicarbonate
secretion to the prevailing intraluminal pH and that this may occur through the release of
prostaglandin E,.
SUMMARY
The existence of an alkaline secretion by gastric and
duodenal mucosa of experimental ainimals and man
has now been established.' Evidence indicates that it
is able to create a pH gradient across the adherent
mucus gel and that this 'mucus-bicarbonate' barrier
may constitute a first line defence against acid
damage. In order for such a barrier to remain intact
under varying intraluminal conditions the rate of
bicarbonate secretion may alter according to luminal
pH. Evidence from animal experiments both in vitro
and ill vivo' indicates that this can occur and that the
response may be due to endogenous prostaglandins.
This study has examined whether gastric bicarbonate
secretion in man can be influenced by luminal acid.
before the study followed by 25 mg/h iv during
perfusion) to maintain an intragastric pH between 6
and 7. Resting gastric contents were aspirated before
perfusion and all saliva was removed with a dental
sucker for the duration of the experiment. Under
these conditions salivary contamination accounts for
only 3°/,% of measured gastric bicarbonate.' The
stomach was perfused with 3H-polyethylene glycol
(12-5 sOi/l in saline at pH 7-4; 2 ml/min) and the
duodenum with '4C-polyethylene glycol (12-5 [Ci/l,
in saline at pH 7-4; 2 ml/min). Aspiration sites were
Methods
MEASUREMENT OF GASTRIC BICARBONATE
SECRETION
This was according to a modification of a method
previously described.' After an overnight fast seven
healthy volunteers (aged 22-54, metan 32 yr)
swcallowed a multilumen polyethylene tube which
was positioned by fluoroscopy with its tip in the
second or third part of the duodenum (Fig. 1).
Ranitidine was administered ( 15t) mg orally one hour
AdLLrcss t'or corrcspondcilcc D)r W I) W Rcc.s lopc lospital (UtJnivcrsity of
Manclicstcr School of Mlcdicine). FI ccles Old Roidl. Sillford Mb Si 11)
RccciVCd for publiCf ion NI')1
M rch 1987
Fig. 1 Diagramn of the perfiusion method of measuremetnt oj
gastric bicarboniate secretion.
1,291
Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
1292
Cratnlptotn, Gibbotns, atnd Rees
positioned in antrum and distal duodenum. Samples
from the aspiration sites were collected by continuous suction at 10 min intervals for measurement
of marker concentration. One millilitre samples were
separated on ice and used for the measurement of
prostaglandin E2 concentration. In the middle of
each sampling period a separate I ml aliquot was
aspirated using a closed cooled syringe for immediate
determination of pH and pCO2.
pH4-
ANALYTICAI METHODS
Fig. 2 IntragastricpHandpCO. (mean+±SEM) during
experiments illustrating the effect of acid perfusion.
Concentrations of 3H-PEG and "4C-PEG in the
samples were measured by liquid scintillation counting (LKB Wallace 2000 Beta Counter, one ml
aspirate with 10 ml PCS scintillant) with quenching
correction by the external standards channels ratio
method. The pH and pCO2 of the samples were
measured immediately using a Corning 170 Blood
Gas Analyser where pH was greater than 6 and a
Radiometer Copenhagen pH meter 28 when the pH
was strongly acidic using the same calibration
buffers. Prostaglandin E2 concentrations were
measured by radioimmunoassay after extraction into
2 ml ethyl ether, evaporation and resuspension into
1 ml radioimmunoassay buffer. Pasteur antiserum
was used and the interassay coefficient of variation
for standards was 6%. In these experiments assays of
gastric samples at two fold dilutions gave similiar
results (103+9%, mean±SD, n=4). Additional
experiments using exogenous PGE2 addition
revealed a good correlation between amount added
and amounts measured (measured calculated as
101±17% added, mean±SD, n=8) suggesting very
low cross reactivity with potentially interfering substances. Furthermore the PGE2 antiserum used
showed selective reactivity with material cochromatographing with PGE2.
EXPERIMENTAL DESIGN
After an equilibration period of two hours basal
bicarbonate secretion was measured for 60 minutes.
The gastric perfusate was then changed to 3H-PEG in
100 mM HCl (100 mM HCl plus 50 mM NaCl to
maintain osmolality, at pH 16; 2 ml/min). The
acidified marker was perfused for 30 minutes after
which the original perfusate at pH 7.4 was recom-
menced. Aspiration continued throughout so that
bicarbonate secretion during the pre and post acid
periods could be calculated.
CALCULATION OF RESULTS
Effective gastric and duodenal volumes were calculated according to marker dilution and perfusion
rate. Duodenogastric reflux was calculated according
to the amount of "4C-PEG appearing in the gastric
aspirate. Free bicarbonate concentration was calcu-
T *100
87-
I.ApH
.~~~~~~~~~~~~~~80
65-
-60
PCO2
PCO2 -40 (mm Hg)
3
21
-20
HCI
60
120
180
Time (min)
lated according to the Hendersson-Hasselbach
equation and added to the CO2 bicarbonate (pCO2x
solubility constant (0.031)) to give a figure for total
bicarbonate concentration. Bicarbonate output was
calculated as the product of volume and concentration and was corrected for refluxed bicarbonate.
Prostaglandin E2 output likewise was the product of
prostaglandin E2 concentration and volume. All
outputs were expressed as mean±SEM per 30
minutes. Differences before and after acid perfusion
were assessed by a paired t-test for significance.
Results
GASTRIC pH AND
pCO2
Mean basal gastric pH under these circumstances was
6&40±0+ 10. During acid perfusion (Fig. 2) gastric pH
fell reaching a nadir after 30 minutes of 2 21+±0(14
and thereafter gradually rose as the infused acid was
cleared. The mean resting pCO2 was 35 9±3 1 mm
Hg which increased to a maximum of 88X8±10 8 mm
Hg during acid perfusion and fell again to levels close
to basal as the pH was restored to near neutral.
BICARBONATE CONCENTRATION
Basal bicarbonate concentration in the gastric
aspirate was 3 58±0 24 mmol/l. It was not possible to
measure accurately bicarbonate concentration when
the gastric pH fell below 5. At or below this pH,
virtually all the bicarbonate present is in the form of
CO2 which equilibrates rapidly with air within the
gastric lumen. This process is rapid at high levels of
PCO2 and thus causes inaccuracies of pCO, measurement. Under basal conditions where pCO2 is similar
to that occurring in plasma the losses are minimal.
Furthermore high values of pCO2 are inaccurately
measured by instruments calibrated for blood gas
analysis. In these experiments, however, bicarbonate
concentration can be measured once the pH returns
to near neutral. Bicarbonate concentration in the
aspirated samples was at its highest in the 30) minute
period after acid instillation (4.61+±0(28 mmol/l)
although the difference from basal was not statistically significant (p=0 3).
Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
Effect of luminalpH on the output of bicarbonate and PGE2 by the normal human stomach
1293
peak of 531±86 [tmol/30 min after restoration
of neutral pH (Fig. 4). Calculation of refluxed
bicarbonate indicated that there was no alteration in
duodenogastric reflux (basal refluxed bicarbonate
89±26 Rmol/30 min compared with 55±22 tmol/
min after acid).
Gastric
secretion
E2 OUTPUT
Prostaglandin E2 concentrations reached their highest in the 'post acid' period (20±5 pmol/ml compared
with 14±4 pmol/ml in the basal period) although the
difference failed to reach significance (p=0.2). When
expressed as an output, however, there was a significant increase in prostaglandin E2 output from a basal
of 411±136 pmol/30 min to 1002±194 pmol/30 min
(p<0-05) which declined as the pH was restored to
normal (Fig. 5).
PROSTAGLANDIN
volume
(ml)
30
60
90
120
150
180
Time (min)
Fig. 3 Calculated secretory volumes (mean±SEM) during
experiments.
SECRETORY VOLUMES
Basal secretory volume was calculated as 28-6±5-6
ml/30 min which increased to 75-5±17 5 ml/30 min
after acid perfusion (p<005) and thereafter declined
to levels close to basal (32.6±9.7 ml/30 min) (Fig. 3).
Marker recovery ranged between 32 and 86% (mean
62%) and was consistent for individuals.
BICARBONATE OUTPUT
Gastric bicarbonate output increased in the 'post
acid' period from a basal of 258±38 Rmol/30 min to a
Discussion
In order to provide effective mucosal protection the
mucus-bicarbonate barrier needs to be maintained
in a dynamic equilibrium with luminal acid and
pepsin so that the pH gradient is preserved and the
epithelium undamaged. The mucus gel layer, an
important constituent of the barrier, is under a
state of continued erosion by luminal pepsin6 and
replenishment occurs by fresh production or secretion of preformed mucus.7 Bicarbonate secretion into
the mucus gel is continually neutralised by luminal
acid creating a pH gradient' and steady state secre-
E3Gastric HCO
3Reflux HCO 3
I* P<0-05
Acid
PGE2
(pmol/30 min)
HCO3
output
0
30
60
90
120
150
180
Time (min)
Fig. 4 Calculated total gastric bicarbonate output
(mean ±SEM, hatched bar) before and after acid perfusion
with amount due to refluxed bicarbonate (mean ±SEM,
stippled bar) indicated.
Time (min)
Fig. 5 Calculated PGE, output (mean ±SEM) for each
period of the experiments.
Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
(Crampton Gibotis, atnd Rees
1294
tion in able to maintain the gradient under most
luminal conditions. Evidence from in vitro experiments indicates that under acidic conditions where
luminal pH falls below 1-4, the gradient may be
overwhelmed.' It is likely, however, that in vivo
secretion of bicarbonate is modulated by various
neurohormonal control mechanisms"' which may
increase bicarbonate output in accordance with
increased requirements. One mechanism which may
be of importance in maintaining the integrity of the
mucus-bicarbonate barrier is a response of epithelial
bicarbonate secretion to luminal acid. Experiments
in dogs with Heidenhain pouches indicate that
bicarbonate secretion by the pouch is increased
dramatically by the instillation of acid into the main
stomach."' In addition in vitro experiments on
isolated mucosa using a paired chamber with a
common serosal side solution show that acidification
of one mucosa is capable of increasing bicarbonate
secretion by the other.'" In man a response of
proximal duodenal bicarbonate secretion to luminal
acid has been demonstrated" and this has been
shown to be attenuated in patients with duodenal
ulcer'4 but no studies to date have examined the effect
of luminal pH on gastric bicarbonate secretion.
Because this response may play an important role in
the dynamics of gastric mucosal protection we have
examined the effect of acid on bicarbonate output by
the normal human stomach.
Our results indicate that acid perfusion produces
an increase in gastric secretory volume. The secretion
contains more bicarbonate than in the resting state
and the output is significantly increased. The
response represents a doubling of the basal rate even
though in these experiments the luminal pH fell only
to 2. It is possible that higher increments may be seen
in the presence of lower pH levels which are frequently achieved in vivo under normal conditions.
Although bicarbonate output could not be measured
during acid perfusion the increase in volume occurred while the pH was low and that combined with the
high pCO2 levels suggests that bicarbonate secretion
was actually increased during acid perfusion.
Furthermore the increase in secretory volume may
itself be important as mathematical analysis of a
model of the mucus-bicarbonate barrier indicates
volume flux as a vital component of the equilibrium. "
In these experiments the measured change in bicarbonate output observed occurred on return of the pH
to near neutral. Although an apparently late
response it is important to appreciate that return to
neutrality occurred in the experimental situation due
to aspiration of acid, which may be very different to
normal gastric acid disposal which is likely to be
slower.
There are a number of possible mechanisms for the
response to intraluminal acid observed in this study.
Gastric bicarbonate secretion may be stimulated by
cholinergic agents, c(cIIlc;uium.' cGM P,I E and F-type
prost,aglandins,Is CCK', and pancreatic glucagon."
Simillrly inhibition of bicarbonate secretion by indomethacin"' implies that enidogenous prostaglandins
are involved in regulating biasal secretion. Prostaglandin E, analogues have been shown to increase
gastric epithelial bicarbonate secretion in l'itro.`' In
high doses in man this effect can also be demiionstrated.22 In lower doses, however, prostaglandin EF
analogues do not stimulate bicarbonate secretion in
man but do prevent the inhibition induced by aspirin
or taurocholate indicating that endogenous prostaglandins are implicated in modulating basal secretion.'" Animal experiments have suggested that the
response to luminal acid occurs via local release of
prostaglandins.24 A recent study has reported that
acid instillation into the human stomach increases
luminal prostaglandin E2 output` and clearly the
similar effect shown in our study may be responsible
for the secretory response observed.
The increased output of prostaglandin E observed
in these experiments may be important both in
relation to the bicarbonate response and to other
mechanisms of mucosal protection. Prostaglandin E,
is capable of increasing mucus gel thickness" and
release,7 increasing mucosal blood flow," affecting
epithelial cell surface hydrophobicity,>" and cellular
restitution after injury,"' all of which are likely to be
important in mucosal protection. The prostaglandin
E2 output rose in our studies as the pl1 fell and
declined again as neutrality was restored indiciating
the close relationship between pH and secretory
response by gastric mucosa.
In conclusion these experiments indicate the
presence of an autoregulatory mechanism which may
be capable of maintaining the gastric mucusbicarbonate barrier and mucosal integrity under
varying luminal acidity. Studies to evaluate the
integrity of this response in patients with peptic ulcer
disease would clearly be of interest.
The authors wish to thank Ms J Rostron for secretarial assistance and the Department of Medical
Illustration for preparing the figures. Dr Crampton is
in receipt of a North West Regional Health Authority
Grant.
References
I Flemstr6m G. Gastric secretion of hicarboniatc. In:
Johnson RL, Christensen J, Grossman MI, Jacobson
ED, Shultz SD, eds. P/hysiology of the GI tract. Vol 1.
New York: Raven Press, 1981: 603-14.
2 Rees WDW, Turnberg LA. Mechanisms of' giastric
mucosal protection: a role for the 'mucus-bicarbonate'
barrier. C.lin Sci 1982; 62: 343-8.
Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
Effect of liuminalpH on the output of bicarbonate and PGE, by the normal huiiman stomach
3 Heylings JR, Garner A, Flemstr6$m G. Regulation of
gastroduodenal HCO 3 by luminal acid in the frog in
vitro. Am JPhysiol 1984;246 (Gastrointest Liver Physiol
9): G235-42.
4 Garner A, Hurst BC. Alkaline secretion by the canine
Heidenhain pouch in response to endogenous acid,
some gastrointestinal hormones and prostaglandins. In:
Gati T, Szollar LG, Ungvary G, eds. Advatnces in
phyfsiological science. Vol 12. Nutrition, digestion andt
mnetabolism. Oxford: Pergamon, 1981.
5 Rees WDW, Botham D, Turnberg LA. A demonstration of bicarbonate production by the normal human
stomach in vivo. Dig Di.s Sci 1982; 27: 961-6.
6 Pearson J, Allen A, Venables C. Gastric mucus; isolation and polymeric structure of the undegraded glycoprotein; its breakdown by pepsin. Gastroenterology
1980; 78: 709-15.
7 Zalewsky C, Moody F. Mechanisms of mucus release in
exposed canine gastric mucosa. Gastroenterology 1979;
77: 719-29.
8 Bahari HMM, Ross IN, Turnberg LA. Demonstration
of a pH gradient across the mucus layer on the surface of
human gastric mucosa in vitro. Gut 1982; 23: 513-6.
9 Ross IN, Bahari HMM, Turnberg LA. The pH gradient
across mucus adherent to rat fundic mucosa in vivo and
the effect of potentially damaging agents. Gastroenterology 1981; 81: 713-8.
1 () Flemstr6m G, Heylings J, Garner A. Gastric and
duodenal HCO3 transport in vitro: effects of hormones
and local transmitters. Am J Physiol 1982; 242: (Gastrointest Liver Physiol 5): G I 00-1 0.
11 Garner A, Hurst BC. Gastric bicarbonate secretion and
mucosal cell loss in the dog. In: Harmon JW, ed. Basic
mechanisms of gastrointestinal mucosal cell injury and
protection. Baltimore: Williams and Wilkins, 1981.
12 Heylings JR. A technique for studying the influence of
luminal acid on frog gastric and duodenal HCO-3
transport in vitro. J Physiol 1983; 342: 4-5.
13 Isenberg JI, Hogan DL, Koss MA, Selling JA. Human
duodenal mucosal bicarbonate secretion. Evidence for
basal secretion and stimulation by hydrochloric acid and
a synthetic prostaglandin El analogue. Gastroenterology
1986; 91: 370-8.
14 Isenberg JI, Selling JA, Hogan DL, Thomas FJ, Koss
MA. Duodenal ulcer patients have impaired proximal
duodenal mucosal bicarbonate secretion [Abstracti.
Gastroenterology 1986; 90: 1472.
15 Engel E, Peskoff A, Kauffman GL, Grossman MI.
Analysis of hydrogen ion concentration in the gastric gel
mucus layer. Am J Physiol 1984; 10: G321-38.
16 Flemstr6m G. Active alkalinisation by amphibian
gastric fundic mucosa in vitro. Am J Physiol (Endocrinol
Metab Gastrointest Phvsiol 1977; 2: E1-12.
17 Flemstr6m G, Garner A. Stimulation of gastric acid and
bicarbonate secretions by calcium in guinea pig stomach
1295
and amphibian isolated mucosa. Acta Physiol Scand
1980; 110: 4 19-26.
18 Smeeton L, Hurst B, Allen A, Garner A. Gastric and
duodenal HCO 3 transport in vivo: influence of prostaglandins. Am J Physiol 1983; 245 (Gastrointest Liver
Physiol 8): G75 1-9.
19 Konturek SJ, Bilski J, Tasler J, Lashiewicz J. Gut
hormones in stimulation of gastroduodenal alkaline
secretion in conscious dogs. Am J Physiol 1985; 248:
(Gastroiniest Liver Ph vsiol 11): G687-91.
20 Rees WDW, Gibbons LC, Turnberg LA. Effects of nonsteroidal anti-inflammatory drugs and prostaglandins on
alkali secretion by rabbit gastric fundus in vitro. Gut
1983; 24: 784-9.
21 Garner A, Flemstr6m G, Heylings JR. Effects of antiinflammatory agents and prostaglandins on acid and
bicarbonate secretion in the amphibian isolated mucosa.
Gastroenterology 1979; 77: 451-7.
22 Johansson C, Aly A, Nilsson E, Flemstr6m G. Stimulation of gastric HCO, secretion by E, prostaglandins in
man. In: Samuelsson B, Paoletti R, Ramwell P, eds.
Advances in prostaglandin, thromboxane and leuikotriene research. Vol 12. New York: Raven Press, 1983.
23 Rees WDW, Gibbons LC, Warhurst G, Turnberg LA.
Prostaglandin E. prevents inhibition of human gastric
bicarbonate secretion by taurocholate and aspirin.
Gastroenterology 1983; 84: 1282.
24 Garner A, Flemstr6m G, Heylings JR, Hurst BC,
Hampson SE, Hughes CL. Factors regulating basal
HCO3 transport by duodcnal epithelia. In: Allen, et al,
eds. Mechanisms of mucosal protection in the upper
gastrointestinal tract. New York: Raven Press, 1984:
153-8.
25 Aly A, Green K, Johansson C. Acid instillation
increases gastric luminal prostaglandin E. output in
man. Acta Med Scand 1985; 218: 505-10.
26 Bickel M, Kauffman G. Gastric gel mucus thickness:
effect of distension and 16,16-dimethyl prostaglandin
E2 and carbenoxolone. Gastroenterology 1981; 80:
770-5.
27 McQueen S, Hulton D, Allen A, Garner A. Gastric and
duodenal surface mucus gel thickness in rat: effects of
prostaglandins and damaging agents. Am J Physiol
1982; 245: (Gastrointest Liver Physiol 8): G388-93.
28 Kauffman GL. The role of prostaglandins in the regulation of gastric mucosal blood flow. Prostaglandins 1981;
21: 33-8.
29 Lichtenberger LM, Richards JE, Hills BA. Effect of
16,16-dimethyl prostaglandin E. on surface hydrophobicity of aspirin treated canine gastric mucosa.
Gastroenterology 1985; 88: 308-14.
30 Tarnawski A, Hollander D, Stachura J, Krauser WJ,
Gergely H. Prostaglandin protection of the gastric
mucosa against alcohol injury - a dynamic time-related
process. Gastroenterology 1985; 88: 334-52.
Downloaded from http://gut.bmj.com/ on July 31, 2017 - Published by group.bmj.com
Effect of luminal pH on the output of
bicarbonate and PGE2 by the normal
human stomach.
J R Crampton, L C Gibbons and W D Rees
Gut 1987 28: 1291-1295
doi: 10.1136/gut.28.10.1291
Updated information and services can be found at:
http://gut.bmj.com/content/28/10/1291
These include:
Email alerting
service
Topic
Collections
Receive free email alerts when new articles cite this article.
Sign up in the box at the top right corner of the online
article.
Articles on similar topics can be found in the following
collections
Stomach and duodenum (1689)
Notes
To request permissions go to:
http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:
http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:
http://group.bmj.com/subscribe/