- Email: [email protected]
Sham feeding and pancreatic secretion
GASTROENTEROLOGY
1984:87:109-14
Sham Feeding and Pancreatic Secretion Evidence for Direct Vagal Stimulation Enzyme Output A. ANAGNOSTIDES, and P. N. MATON Gastroenterology England
Unit,
V. S. CHADWICK,
Department
of Medicine,
The cephalic phase of pancreatic secretion in humans was investigated using modified sham feeding and a duodenal perfusion system. Studies performed in 5 normal volunteers were designed so that trypsin and bicarbonate outputs during sham feeding, with or without pretreatment with atropine, were compared to “maximal” pancreatic secretory response to exogenous stimulation with caerulein and secretin. The role of gastric acid entry to the duodenum in mediating cephalic responses was assessed by a comparison between outputs observed when gastric aspiration (=80% efficient) was used alone and when acid entry was completely abolished by combining gastric aspiration with cimetidine pretreatment. To evaluate the role, if any, ofgut hormone release in the pancreatic secretory response to sham feeding, plasma gastrin and cholecystokinin concentrations were monitored throughout. Trypsin outputs during sham feeding were 31.9 * lo.45 kallikrein inactivator units per 30 min, equivalent to four times basal output and 92% of maximal, but were only 54% maximal in subjects pretreated with cimetidine. Atropine suppressed basal trypsin output and abolished the response to sham feeding (4.98 + 3.89 kallikrein inactivator units per 30 min). A modest increase in bicarbonate
Received January 4. 1983. Accepted February 3. 1984. Address requests for reprints to: P. N. Maton, M.D., Digestive Diseases Branch, Building 10, Room SC-103, National Institutes of Health, Bethesda, Maryland 20205. This work was supported by the Medical Research Council of Great Britain. P. N. M. was the British Society of Castroenterology Smith Kline 8r French Research Fellow. The authors thank J. Barr for technical assistance, Miss Ann Love for preparing the manuscript, Dr. V. Mutt for supplies of 99% pure CCK 33 and 39. E. R. Squibb & Sons, Inc., for CCK 8. and Dr. V. I,. W. Go for anti-CCK 8 antiserum. 8 1984 by the American Gastroenterological Association 0018-50851841$3,00
Royal
of
A. C. SELDEN, Postgraduate
Medical
School,
London,
secretion during sham feeding (3.30 * 1.97 mmoll30 min versus basal of 0.68 +- 0.74 mmoU30 min, p = 0.5) was not influenced by atropine but was abolished by cimetidine pretreatment. No significant changes in plasma gastrins were observed in these studies and plasma cholecystokinins remained undetectable throughout. We conclude that there is tonic vagal stimulation of trypsin secretion, and that sham feeding markedly increases trypsin output, which is augmented further by acid entry into the duodenum. There is no direct effect of cephalic stimulation on bicarbonate secretion or on gastrin or cholecystokinin release. The cephalic phase of pancreatic secretion was first described in dogs by Pavlov (1). Subsequent studies showed that stimulation of the vagus in anesthetized animals of several species caused secretion of enzymes or bicarbonate or a combination of these (2). The components of this vagal stimulation of pancreatic secretion have been best studied in the dog. In this species, sham feeding produces an enzyme-rich secretion that appears to be mediated both directly and indirectly through vagal release of antral gastrin (3), an important pancreatic secretagogue in the dog (4). The data available in humans are confusing. Two studies have shown an increase in both bicarbonate and enzyme secretion (5) and an increase in enzyme secretion alone (6). On the other hand, two further studies have shown no effect of sham feeding (7) and augmented enzyme secretion only when under the influence of background stimulation of secretin (8). In this study, we have examined the effect of sham Abbreviations used in this paper: CCK. cholecystokinin; high-performance liquid chromatography; KILJ. kallikrein vator units.
HPLC, inacti-
110 ANAGNOSTIDES ET AL.
GASTROENTEROLOGY
feeding on pancreatic secretion using a duodenal perfusion system and have assessed the magnitude of the cephalic phase by comparing the effects of sham feeding with the maximal pancreatic response to exogenous caerulein [a cholecystokinin (CCK) analogue] and secretin. To define the mechanisms of the responses to sham feeding, we (a) studied the effects of prior treatment with atropine and cimetidine and (b) measured plasma concentrations of gastrin and CCK. To measure CCK, we used a recently developed assay linked to high-performance liquid chromatography (HPLC). In this assay, fasting concentrations of CCKs are undetectable, but rise after a fat meal to 15 pmol each for CCK-8 and CCK33/39 (9).
Methods Subjects
and
Experimental
Design
Five fit, normal volunteers, 3 women and 2 men, aged 22-27 yr,were studied. Each subject underwent three duodenal perfusion studies on different days to determine: (a) “maximal” pancreatic secretion, stimulated by a combination of caerulein and secretin, (b) the response to sham feeding, and (c) the response to sham feeding after prior treatment with atropine. The order of the second and third studies was randomized. As the efficiency of gastric aspiration can never be lOO%, results might have been influenced by small amounts of acid entering the duodenum. To assess the magnitude of such influences, further studies were performed in 3 individuals who received cimetidine in sufficient dosage to suppress gastric acid secretion. In the basal and sham feeding periods, blood samples were taken for estimation of plasma CCK and gastrin concentrations. All volunteers gave their written informed consent to these studies. The protocol was approved by the Ethics Committee of the Royal Postgraduate Medical School.
Perfusion
System
After an overnight fast, a double-lumen duodenal tube with a mercury bag was positioned such that the infusion port was at the level of the ampulla of Vater and the aspiration ports were 14-17 cm distal as described previously (10). Gastric aspiration was performed through a second tube positioned in the antrum. The volunteers were seated in a comfortable chair and asked to expectorate their saliva throughout the remainder of the study. The duodenum was then perfused with a warmed (37°C) isoosmolar solution of 0.155 M sodium chloride, pH 7.0,containing 0.5g/L of polyethylene glycol (PEG) 4000 and 5 &i/L of [14C]PEG 4000 as nonabsorbable markers. Gastric and duodenal contents were aspirated using suction pumps (Air Shields, Inc., Hatboro, Philadelphia, Pa.). Tube patency was checked frequently. All gastric and duodenal aspirates were collected on ice, pooled every 10 min, and checked for the presence of food particles. Recovery of [14C]PEG 4000 was 49.5% k 4.3%
Vol.87. No. 1
(mean * SEM) from the duodenum and 2.5% + 1.1% from the stomach. From previous studies in our laboratory using the same perfusion system, we determined that the recovery of the gastric marker in gastric aspirates was 82.3% k 2.7% (lo), although no gastric markers were used in the present studies. During all of the present studies, duodenal aspirates remained >pH 7.0 throughout, suggesting that there was no gross entry of acid into the perfusion segment.
Determination Secretion
of Maximal
Pancreatic
After a 30-min equilibration period and a 30-min basal collection period, each subject received an i.v. infusion of pure synthetic caerulein (Farmitalia, Milan, Italy, in a dose of 75 pmol . kg-’ h-l) and secretin [GIH, Stockholm, Sweden, 1 clinical unit. kg-’ . h-’ (84 pmol . kg-’ . h-l)] in 0.155 mol/L sodium chloride containing 1% wtivol human serum albumin. This combination of peptides, which is known to maximally stimulate pancreatic secretion (ll), was continued for 40 min, the last 30 min of which were used for analysis.
Sham
Feeding
Studies
On two other days, each subject performed a sham feeding study with or without pretreatment with iv. atropine sulfate administered at the beginning of the equilibration period. Sufficient atropine was given to raise the heart rate to 90-100 beats/min (0.6-1.2 mg).If the heart rate subsequently fell significantly, a further dose of 0.3 mg was given. Three subjects underwent a further sham feeding study similar to the other studies but were given 600 mg of cimetidine orally about 90 min before the start of the equilibration period. This dose of cimetidine is sufficient to completely abolish gastric acid secretion (12), including secretion induced by sham feeding (13). All sham feeding studies followed the same protocol of a 30-min equilibration period, a 30-min basal period, and a 40-min sham feeding period, of which the last 30 min were used for analysis. Freshly prepared grilled ham sandwiches were used as the sham feeding stimulus. These were initially cooked outside the room and brought to the subjects who were immediately asked to chew the sandwiches and expectorate the masticated food. More sandwiches were cooked in the same room within the subjects’ view, the subjects’ having been instructed that they could chew as many sandwiches as they wished in the JO-min period, as long as they chewed continuously. Subjects were told that they could eat after the study was completed.
Chemical
Analyses
All chemical estimations were performed immediately. The volume and pH of the aspirates were measured, and in the duodenal samples, trypsin and bicarbonate (in samples kept under paraffin) were estimated using titra-
July 198~
SHAM FEEDING AND PANGKEATIC SECKETION
tion methods as previously described (10). Aliquots (500 ~1) of gastric and duodenal aspirates and perfusates were pipetted into scintillation fluid and [14C]PEG was measured in a p-counter using external standard quench correction. Duodenal outputs of bicarbonate and trypsin were then calculated using standard equations for recovery of PEG.
Radioimmunoassays Cholecystokinin 8 and CCK-33139 were assayed in plasma samples taken in the basal period and after 5, 10, 20, and 40 min of sham feeding. The forms of CCK were measured using a C-terminal radioimmunoassay (RIA) following prior concentration and separation of various forms of CCK from each other and from gastrins using reverse-phase HPLC as previously described (9). In this assay, fasting CCK concentrations are undetectable (<3 pmol/L for CCK-8 and <6 pmol/L for CCK-33139) but rise after a fat meal to -15 pmol each for the large and small forms. Blood samples (20 ml) were taken into heparin tubes containing 8000 IU of aprotinin (Trasylol, Bayer, London, U.K.) and centrifuged; the plasma was stored at -20°C until analysis. Plasma was then thawed and subjected to a preliminary HPLC step to remove protein and other interfering substances. Samples (10 ml) were diluted 1: 10 with 0.155 mol/L sodium chloride acidified to pH 2.1 with hydrochloric acid, and the resulting solution was centrifuged at 15,000 g for 30 min to remove cryoprecipitates. The supernatant was then applied to a reverse-phase HPLC column at 1 mlimin (Lichroprep RPl8 15-45, Merck, Darmstadt, Federal Republic of Germany) measuring 4.6 X of the sample, the column was 100 mm. After application washed through with 10 ml of 0.155 mol/L sodium chloride, and then peptides were eluted with 10 ml of acetonitrileiwater, 3 : 2. The eluate was collected into a siliconized glass tube and the acetonitrile was blown off under nitrogen. The resulting solution was then applied to a second high resolution gradient elution HPLC using a 4.6 x lOOmm column of Hypersil ODS. Elution of peptides was achieved using a gradient from 0.155 mol/L sodium chloride pH 2.1, to acetonitrileiwater 3:2. Fractions of 1 ml were collected and those fractions encompassing elution times of all CCKs were assayed using two RIAs-an antibody directed against the common carboxyl-terminal of CCKs and gastrins, and a relatively specific gastrin assay. This HPLC system reliably separates CCK-4, CCK-8, CCK-33139, gastrin 17, and gastrin 34; desulfated and oxidized forms of CCK do not cochromatograph with the biologically active forms. As gastrins and CCKs are sepa-
Table
1.
Secretory
Rates of Bicarbonate
During Various
111
rated by this system, the CCK assay is not a “difference” assay; the gastrin assay is simply run as a double check. Recoveries of CCK peptides added to plasma and processed through this two-step HPLC system are >96%. The carboxyl-terminal antiserum was raised in rabbits against CCK-8 conjugated to bovine serum albumin. Nonsulfated CCK-8 (E. R. Squibb & Sons, Inc., Princeton, N.J.) was used as a tracer. The CCK-8 was first purified on HPLC, iodinated using the chloramine T method, and the monoiodinated peptide was separated from the other reaction products by ion-exchange chromatography. The total volume of the assay was 1.8 ml, and contained 400 /*l of sample (equivalent to 4 ml of original plasma sample), 2000 counts per minute of iodinated CCK-8, antibody to a final dilution of 1 in 40,000, and 200 IU of aprotinin, all in 0.1 mol/L Verona1 buffer pH 8.2, containing 1% wtivol bovine serum albumin and 0.1% wtivol sodium azide. Standard and no-antibody tubes contained 400 ~1 of acetonitrilelwaterisodium chloride, 33 : 22 :45, in place of sample. Addition of tracer was delayed for 3 days and the assay was terminated after a further 3 days at 4°C. Separation of bound and free peptide was achieved by adding 500 ~1 of a charcoal suspension. Initial binding (0 standard] of tracer was 50%, and nonspecific binding ~5%. Standard curves of CCK-8, CCK33, and CCK-39 were parallel and the molar cross-reactivities of these peptides in the assay were CCK-8ICCK33/CCK-39, 1: 0.53 : 0.40. Molar cross-reactivities of other related peptides were nonsulfated CCK-8, 1.0; tetrin, 0.05; gastrin 17-1, 0.5; gastrin 34, 0.5. Secretin, bovine pancreatic polypeptide, vasoactive intestinal polypeptide, motilin, somatostatin, and gastric inhibitory peptide all showed
Statistics Differences between mean pancreatic secretion of trypsin and bicarbonate following different stimuli and differences in plasma gastrin concentrations were assessed using two-way analysis of variance, with differences between selected values assessed using paired Student’s ttests.
Stimuli Sham Sham
Bicarbonate (mmoU30 Results
Basal
Maximal
Basal
0.68 c 0.74
15.53 t 3.16"
0.85 + 0.69
min)
expressed
as mean
‘_ SD. ’ p < 0.001,
b p = 0.05 vs. basal
secretion.
feeding 3.30 k 1.97b
Basal + atropine
feeding + atropine
0.31 2 0.24
1.21 _f 1.13
112
ANAGNOSTIDES
GASTROENTEROLOGY
ET AL.
Al Basal Maximal Basal
Basal
Sham Fed
Sham Fed
+ atropine
Figure
1. Trypsin output [mean 2 SD] in kallikrein inactivator units per 30 min in the basal state and after sham feeding with and without pretreatment with atropine. Trypsin outputs determined on another day in the basal state and during “maximal” stimulation (caerulein 75 pmol . kg-’ h-’ plus secretin 84 pmol kg-’ . h-‘) are shown for comparison.
Results Basal and Maximal
Pancreatic
Secretion
Basal secretion of bicarbonate and trypsin were measured on two occasions in each subject (Table 1 and Figure 1). The secretory rates were reproducible. In all subjects, there was a prompt rise in pancreatic enzyme secretion following i.v. caerulein and secretin, evident in the first lo-min period. Trypsin secretion rose 4.5-fold, bicarbonate secretion 2%fold (Table 1). These increases were both highly significant.
Sham Feeding
Table 2. Effect of Sham Feeding Sham feeding Sham feeding + atropine Results expressed
bicarbonate secretion (Table 1). Atropine also markedly reduced the trypsin response to sham feeding. Indeed, stimulated trypsin output following atropine was comparable with basal trypsin output in the studies performed without atropine (Figure 1). Three subjects repeated the sham feeding study after pretreatment with cimetidine. In these studies, there was no rise in bicarbonate secretion during sham feeding. Mean basal bicarbonate output was 0.25 mmol/30 min, and during sham feeding it was 0.23 mmol/30 min. However, a marked trypsin response was still observed. Mean basal trypsin outputs of 6.56 KIU/30 min rose to 18.1 KIUI30 min during sham feeding-54% of the maximal output in these 3 subjects. Hormonal
on Plasma
Gastrin
Responses
to Sham Feeding
Plasma gas&in concentrations. Basal plasma gastrin concentrations before sham feeding were 60.8 + 10.8 pmol/L. There was a tendency for gastrin concentrations to rise during sham feeding but this did not reach statistical significance and by the end of the sham feeding period, plasma levels were similar to basal values [Table 2). Pretreatment with atropine produced no significant changes in basal or sham feeding gastrin levels, as compared with the no-treatment studies. Again, there was a tendency for plasma gastrin concentrations to rise during the initial period of sham feeding with a fall by 40 min. Plasma cholecystokinin concentrations. Plasma concentrations of CCK-8 and CCK-33139 were undetectable (~3 pmol/L for CCK-8 and <6 pmol/L for CCK-33/39) in control periods and remained undetectable during sham feeding.
Studies
a In all subjects, sham feeding produced marked rise in trypsin output such that mean trypsin secretion of 31.9 ? lo.5 kallikrein inactivator units (KIU) per 30 min was 92% of that observed during maximal pancreatic secretion (Figure 1). In contrast, mean bicarbonate secretion was augmented to a modest extent (20% of maximum) and this rise was only just significant (p = 0.05)(Table 1). Pretreatment with atropine significantly reduced basal trypsin output (Figure 1) but had no effect on
Vol. 87, No. 1
Discussion These data demonstrate that in humans modified sham feeding stimulates substantial pancreatic secretion of trypsin, but has little or no effect on bicarbonate secretion. Circulating gastrins and CCKs were not implicated in the sham feeding responses, but the enzyme secretory response was completely inhibited by atropine. In assessing the current (and other) data on pancreatic secretion during sham feeding, the validity of
Concentrations
Basal
5 min
10 min
20 min
40 min
60.8 ? 10.8 65.7 2 7.2
65.3 ? 16.1 66.0 ? 6.4
83.5 -c 14.3 72.8 2 10.8
69.5 ? 15.0 74.8 + 7.8
60.5 ? 13.2 61.8 rt 7.4
as mean + SEM in picomoles
per liter.
July1984
the techniques used is critical. In particular, the central problem is that of entry of gastric acid into the duodenum. This could mask a bicarbonate response (acid neutralization), augment a bicarbonate response (via secretin release), or might affect trypsin output. That small amounts of acid did enter the duodenum in our studies, even with efficient gastric aspiration, is demonstrated by the differing results after cimetidine administration, whereby sham feeding-stimulated bicarbonate secretion was abolished and trypsin output increased to 55%, not 96%, of maximal output. Thus, acid entry into the duodenum stimulated a bicarbonate response and augmented the trypsin response. Two other studies have examined the effects of sham feeding under circumstances where no acid entered the duodenum. Novis et al. (6) studied achlorhydric patients and Preshaw et al. (3) studied dogs with gastric and pancreatic fistulas. In all three studies, sham feeding stimulated enzyme secretion but bicarbonate secretion was unchanged or only marginally increased. Although these data are of interest mechanistically, exclusion of acid from the duodenum is not physiologic. Comparing the data we obtained in subjects with and without pretreatment with cimetidine allows us to predict the likely effects of sham feeding in the normal state when acid entry into the duodenum is not restricted. Trypsin output will be stimulated maximally (of which 55% is a direct effect, and the rest is related to acid entering the duodenum] and bicarbonate output will also be increased-but only consequent upon acid entering the duodenum. Our results differ in several respects from those obtained by other workers. Novis et al. (6) in studying achlorhydric patients were unable to assess the magnitude of the pancreatic response as they did not use a perfusion system. Sarles et al. (5) also used an aspiration system but suggested that large increases in enzyme and bicarbonate outputs occurred during sham feeding-presumably because small amounts of acid entered the duodenum in their studies. Read et al. (7) and Defillipi et al. (8) failed to identify a pancreatic response to sham feeding. However, Read et al. (7) perfused the jejunum and failed to preequilibrate the duodenum, and Defillipi et al. (8) used a low perfusion rate-making aspiration of duodenal contents unreliable. When they increased duodenal fluid volume by pharmacologic doses of secretin, however, they did detect an increase in enzyme output response to sham feeding. The mechanisms whereby sham feeding stimulates pancreatic secretion are disputed-partly because of the differing results in various studies. Our data, demonstrating that sham feeding induces an enzyme-rich, bicarbonate-poor secretion, suggest
SHAM FEEDING AND PANCREATIC SECRETION 113
that the response is similar to that seen after infusions of pure CCK-33 and CCK-8 (14,15). Atropine suppressed both basal and stimulated enzyme outputs suggesting that there is vagal tone in the fasting state maintaining enzyme secretion, and that sham feeding exerts its effects directly or indirectly via muscarinic cholinergic receptors. Such would not seem to be the case for bicarbonate secretion. Various postulates have been proposed to account for the effects of this vagal stimulation. The effect could be a direct one on the pancreas or indirect via neurogenic gastrin or CCK release. In the present studies, plasma gastrins did not rise significantly during sham feeding. Furthermore, atropinization virtually abolished the effects of sham feeding on trypsin secretion, but had no significant effect on plasma gastrin concentrations. We, therefore, think it unlikely that gastrin is important in the cephalic phase of pancreatic secretion in humans. Similarly, we found no evidence of release of CCK into the circulation during sham feeding-plasma concentrations of CCKs being undetectable throughout. It could be that our assay is not sensitive enough to detect the small amounts of CCK released. However, our assay is sufficiently sensitive to detect physiologic concentrations of circulating CCKs. After a fat meal, we found that plasma CCK-8 and CCK33139 concentrations each rise to -15 pmol/L (9). Furthermore, when we infused CCK-8 in sufficient amounts to produce trypsin secretion rates comparable with those in the present study (50%-90% of maximum), we were able to detect elevations in (15,161.Thus, if the plasma CCK concentrations trypsin secretion rates observed in the present study were entirely due to circulating CCKs, then our assay should detect them. We cannot exclude the possibility that vagal stimulation caused by sham feeding releases minute amounts of CCK into the circulation. but if that is the case, it is unlikely that CCK alone is sufficient to account for the pancreatic response. We conclude from these studies that central vagal stimulation of the pancreas is an important controlling mechanism for pancreatic enzyme secretion in humans, and that this effect is probably mediated directly through muscarinic receptors on acinar cells. Such a mechanism is consistent with the results obtained in isolated guinea pig acinar cells where both cholinergic agents and CCKs are potent stimulators of amylase secretion but act at different receptors, and atropine blocks the effect of cholinergic agents but not those of CCK peptides (17). These studies may have clinical implications for understanding pancreatic function in health and disease. For example, in celiac disease (18)and after various types of gastric surgery (19.20), there is an
114
ANAGNOSTIDES
ET AL
inadequate pancreatic response to intraluminal stimuli or liquid meals. However, under the more physiologic conditions of eating a solid, appetizing meal, the cephalic phase could ameliorate the effects of the blunted luminal responses. These data also provide an explanation for the observations that the most powerful luminal stimulants of pancreatic enzyme secretion are not fats and proteins but their digestion products [2). Cephalic stimulation of enzyme secretion would allow the undigested thyme entering the duodenum to be broken down and the resulting digestion products could then maintain pancreatic secretion by luminal activity as long as the nutrients remained unabsorbed.
References 1. Pavlov IP. The work of the digestive
2.
3.
4.
5. 6. 7.
8. 9.
glands. Translated by Thompson WT. London: Griffin, 1902. Meyer JH. Control of pancreatic secretion. In: Johnson LR, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:821-9. Preshaw RM, Cooke AR, Grossman MI. Sham feeding and pancreatic secretion in the dog. Gastroenterology 1966;50: 171-8. Stening GF, Grossman MI. Gastrin related peptides as stimulants of pancreatic and gastric secretion. Am J Physiol 1969;217:262-6. Sarles H, Dani G, Prezelin G, Souville C, Figarella C. Cephalic phase of pancreatic secretion in man. Gut 1968;9:214-21. Novis BH, Bank S, Marks IN. The cephalic phase of pancreatic secretion in man. Stand J Gastroenterol 1971;6:417-22. Read NW, Cooper K, Fordtran JS. Effect of modified sham feeding on jejunal transport and pancreatic and biliary secretion in man. Am J Physiol 1978;234:E417-20. Defillipi C, Solomon T, Valenzuela JE. Pancreatic response to sham feeding in humans. Digestion 1982;23:2!7-23. Maton PN, Selden AC, Chadwick VS. Measurement of large
GASTROENTEROLOGY
Vol. 87. No. 1
and small forms of cholecystokinin in human plasma using high pressure liquid chromatography and radioimmunoassay. Regul Pept 1982;4:251-60. 10. Bjornsson OG, Maton PN, Fletcher DR, Chadwick VS. Effects of duodenal perfusion with sodium taurocholate on biliary and pancreatic secretion in man. Eur J Clin Invest 1982;12:97-105. 11. Ribet A, Tournout R, Vayss N. Use of caerulein with submaximal doses of secretin as a test of pancreatic function in man. Gut 1976;17:431-4. 12. Pounder RE, Williams JG, Hunt RH, Vincent SH, MiltonThompson GJ, Misiewicz JJ. The effects of oral cimetidine on food stimulated gastric acid secretion and 24 hour intragastric acidity. In: Burland WL, Simkins MA, eds. Proceedings of the Second International Symposium on Histamine HZ-Receptor Antagonists. Amsterdam: Excerpta Medica, 1977:188-226. on gastric 13. Schoon IM, Olbe L. Inhibitory effect of cimetidine acid secretion vagally activated by physiological means in duodenal ulcer subjects. Gut 1978;19:27-31. 14. Debas HT, Grossman MI. Pure cholecystokinin: pancreatic protein and bicarbonate response. Digestion 1973;9:469-81. AA, Maton PN, Selden AC, Chadwick VS. 15. Anagnostides Effects of graded doses of cholecystokinin octapeptide (CCK 8) on pancreatic enzyme secretion in man. Clin Sci 1982;63: 21P. 16. Maton PN, Selden AC, Fitzpatrick ML, Chadwick VS. Infusion of cholecystokinin octapeptide in man: relation between plasma cholecystokinin concentrations and gallbladder emptying rates. Eur J Clin Invest 1984;14:37-41. 17. Gardner JD, Jensen RT. Regulation of pancreatic enzyme secretion in vitro. In: Johnson LR, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:831-71. 18. Di Magno EP, Go VLW, Summerskill WHJ. Impaired cholecystokinin-pancreozymin secretion, intraluminal digestion and maldigestion of fat in sprue. Gastroenterology 1972; 63:25-32. 19. MacGregor I, Parent J, Meyer JH. Gastric emptying of liquid meals and biliary secretion after subtotal gastrectomy or truncal vagotomy and pyloroplasty in man. Gastroenterology 1977;72:195-205. 20. Malagelada JR, Go VLW, Summerskill WHJ. Altered pancreatic and biliary function after vagotomy. Gastroenterology 1974;66:22-7.
1984:87:109-14
Sham Feeding and Pancreatic Secretion Evidence for Direct Vagal Stimulation Enzyme Output A. ANAGNOSTIDES, and P. N. MATON Gastroenterology England
Unit,
V. S. CHADWICK,
Department
of Medicine,
The cephalic phase of pancreatic secretion in humans was investigated using modified sham feeding and a duodenal perfusion system. Studies performed in 5 normal volunteers were designed so that trypsin and bicarbonate outputs during sham feeding, with or without pretreatment with atropine, were compared to “maximal” pancreatic secretory response to exogenous stimulation with caerulein and secretin. The role of gastric acid entry to the duodenum in mediating cephalic responses was assessed by a comparison between outputs observed when gastric aspiration (=80% efficient) was used alone and when acid entry was completely abolished by combining gastric aspiration with cimetidine pretreatment. To evaluate the role, if any, ofgut hormone release in the pancreatic secretory response to sham feeding, plasma gastrin and cholecystokinin concentrations were monitored throughout. Trypsin outputs during sham feeding were 31.9 * lo.45 kallikrein inactivator units per 30 min, equivalent to four times basal output and 92% of maximal, but were only 54% maximal in subjects pretreated with cimetidine. Atropine suppressed basal trypsin output and abolished the response to sham feeding (4.98 + 3.89 kallikrein inactivator units per 30 min). A modest increase in bicarbonate
Received January 4. 1983. Accepted February 3. 1984. Address requests for reprints to: P. N. Maton, M.D., Digestive Diseases Branch, Building 10, Room SC-103, National Institutes of Health, Bethesda, Maryland 20205. This work was supported by the Medical Research Council of Great Britain. P. N. M. was the British Society of Castroenterology Smith Kline 8r French Research Fellow. The authors thank J. Barr for technical assistance, Miss Ann Love for preparing the manuscript, Dr. V. Mutt for supplies of 99% pure CCK 33 and 39. E. R. Squibb & Sons, Inc., for CCK 8. and Dr. V. I,. W. Go for anti-CCK 8 antiserum. 8 1984 by the American Gastroenterological Association 0018-50851841$3,00
Royal
of
A. C. SELDEN, Postgraduate
Medical
School,
London,
secretion during sham feeding (3.30 * 1.97 mmoll30 min versus basal of 0.68 +- 0.74 mmoU30 min, p = 0.5) was not influenced by atropine but was abolished by cimetidine pretreatment. No significant changes in plasma gastrins were observed in these studies and plasma cholecystokinins remained undetectable throughout. We conclude that there is tonic vagal stimulation of trypsin secretion, and that sham feeding markedly increases trypsin output, which is augmented further by acid entry into the duodenum. There is no direct effect of cephalic stimulation on bicarbonate secretion or on gastrin or cholecystokinin release. The cephalic phase of pancreatic secretion was first described in dogs by Pavlov (1). Subsequent studies showed that stimulation of the vagus in anesthetized animals of several species caused secretion of enzymes or bicarbonate or a combination of these (2). The components of this vagal stimulation of pancreatic secretion have been best studied in the dog. In this species, sham feeding produces an enzyme-rich secretion that appears to be mediated both directly and indirectly through vagal release of antral gastrin (3), an important pancreatic secretagogue in the dog (4). The data available in humans are confusing. Two studies have shown an increase in both bicarbonate and enzyme secretion (5) and an increase in enzyme secretion alone (6). On the other hand, two further studies have shown no effect of sham feeding (7) and augmented enzyme secretion only when under the influence of background stimulation of secretin (8). In this study, we have examined the effect of sham Abbreviations used in this paper: CCK. cholecystokinin; high-performance liquid chromatography; KILJ. kallikrein vator units.
HPLC, inacti-
110 ANAGNOSTIDES ET AL.
GASTROENTEROLOGY
feeding on pancreatic secretion using a duodenal perfusion system and have assessed the magnitude of the cephalic phase by comparing the effects of sham feeding with the maximal pancreatic response to exogenous caerulein [a cholecystokinin (CCK) analogue] and secretin. To define the mechanisms of the responses to sham feeding, we (a) studied the effects of prior treatment with atropine and cimetidine and (b) measured plasma concentrations of gastrin and CCK. To measure CCK, we used a recently developed assay linked to high-performance liquid chromatography (HPLC). In this assay, fasting concentrations of CCKs are undetectable, but rise after a fat meal to 15 pmol each for CCK-8 and CCK33/39 (9).
Methods Subjects
and
Experimental
Design
Five fit, normal volunteers, 3 women and 2 men, aged 22-27 yr,were studied. Each subject underwent three duodenal perfusion studies on different days to determine: (a) “maximal” pancreatic secretion, stimulated by a combination of caerulein and secretin, (b) the response to sham feeding, and (c) the response to sham feeding after prior treatment with atropine. The order of the second and third studies was randomized. As the efficiency of gastric aspiration can never be lOO%, results might have been influenced by small amounts of acid entering the duodenum. To assess the magnitude of such influences, further studies were performed in 3 individuals who received cimetidine in sufficient dosage to suppress gastric acid secretion. In the basal and sham feeding periods, blood samples were taken for estimation of plasma CCK and gastrin concentrations. All volunteers gave their written informed consent to these studies. The protocol was approved by the Ethics Committee of the Royal Postgraduate Medical School.
Perfusion
System
After an overnight fast, a double-lumen duodenal tube with a mercury bag was positioned such that the infusion port was at the level of the ampulla of Vater and the aspiration ports were 14-17 cm distal as described previously (10). Gastric aspiration was performed through a second tube positioned in the antrum. The volunteers were seated in a comfortable chair and asked to expectorate their saliva throughout the remainder of the study. The duodenum was then perfused with a warmed (37°C) isoosmolar solution of 0.155 M sodium chloride, pH 7.0,containing 0.5g/L of polyethylene glycol (PEG) 4000 and 5 &i/L of [14C]PEG 4000 as nonabsorbable markers. Gastric and duodenal contents were aspirated using suction pumps (Air Shields, Inc., Hatboro, Philadelphia, Pa.). Tube patency was checked frequently. All gastric and duodenal aspirates were collected on ice, pooled every 10 min, and checked for the presence of food particles. Recovery of [14C]PEG 4000 was 49.5% k 4.3%
Vol.87. No. 1
(mean * SEM) from the duodenum and 2.5% + 1.1% from the stomach. From previous studies in our laboratory using the same perfusion system, we determined that the recovery of the gastric marker in gastric aspirates was 82.3% k 2.7% (lo), although no gastric markers were used in the present studies. During all of the present studies, duodenal aspirates remained >pH 7.0 throughout, suggesting that there was no gross entry of acid into the perfusion segment.
Determination Secretion
of Maximal
Pancreatic
After a 30-min equilibration period and a 30-min basal collection period, each subject received an i.v. infusion of pure synthetic caerulein (Farmitalia, Milan, Italy, in a dose of 75 pmol . kg-’ h-l) and secretin [GIH, Stockholm, Sweden, 1 clinical unit. kg-’ . h-’ (84 pmol . kg-’ . h-l)] in 0.155 mol/L sodium chloride containing 1% wtivol human serum albumin. This combination of peptides, which is known to maximally stimulate pancreatic secretion (ll), was continued for 40 min, the last 30 min of which were used for analysis.
Sham
Feeding
Studies
On two other days, each subject performed a sham feeding study with or without pretreatment with iv. atropine sulfate administered at the beginning of the equilibration period. Sufficient atropine was given to raise the heart rate to 90-100 beats/min (0.6-1.2 mg).If the heart rate subsequently fell significantly, a further dose of 0.3 mg was given. Three subjects underwent a further sham feeding study similar to the other studies but were given 600 mg of cimetidine orally about 90 min before the start of the equilibration period. This dose of cimetidine is sufficient to completely abolish gastric acid secretion (12), including secretion induced by sham feeding (13). All sham feeding studies followed the same protocol of a 30-min equilibration period, a 30-min basal period, and a 40-min sham feeding period, of which the last 30 min were used for analysis. Freshly prepared grilled ham sandwiches were used as the sham feeding stimulus. These were initially cooked outside the room and brought to the subjects who were immediately asked to chew the sandwiches and expectorate the masticated food. More sandwiches were cooked in the same room within the subjects’ view, the subjects’ having been instructed that they could chew as many sandwiches as they wished in the JO-min period, as long as they chewed continuously. Subjects were told that they could eat after the study was completed.
Chemical
Analyses
All chemical estimations were performed immediately. The volume and pH of the aspirates were measured, and in the duodenal samples, trypsin and bicarbonate (in samples kept under paraffin) were estimated using titra-
July 198~
SHAM FEEDING AND PANGKEATIC SECKETION
tion methods as previously described (10). Aliquots (500 ~1) of gastric and duodenal aspirates and perfusates were pipetted into scintillation fluid and [14C]PEG was measured in a p-counter using external standard quench correction. Duodenal outputs of bicarbonate and trypsin were then calculated using standard equations for recovery of PEG.
Radioimmunoassays Cholecystokinin 8 and CCK-33139 were assayed in plasma samples taken in the basal period and after 5, 10, 20, and 40 min of sham feeding. The forms of CCK were measured using a C-terminal radioimmunoassay (RIA) following prior concentration and separation of various forms of CCK from each other and from gastrins using reverse-phase HPLC as previously described (9). In this assay, fasting CCK concentrations are undetectable (<3 pmol/L for CCK-8 and <6 pmol/L for CCK-33139) but rise after a fat meal to -15 pmol each for the large and small forms. Blood samples (20 ml) were taken into heparin tubes containing 8000 IU of aprotinin (Trasylol, Bayer, London, U.K.) and centrifuged; the plasma was stored at -20°C until analysis. Plasma was then thawed and subjected to a preliminary HPLC step to remove protein and other interfering substances. Samples (10 ml) were diluted 1: 10 with 0.155 mol/L sodium chloride acidified to pH 2.1 with hydrochloric acid, and the resulting solution was centrifuged at 15,000 g for 30 min to remove cryoprecipitates. The supernatant was then applied to a reverse-phase HPLC column at 1 mlimin (Lichroprep RPl8 15-45, Merck, Darmstadt, Federal Republic of Germany) measuring 4.6 X of the sample, the column was 100 mm. After application washed through with 10 ml of 0.155 mol/L sodium chloride, and then peptides were eluted with 10 ml of acetonitrileiwater, 3 : 2. The eluate was collected into a siliconized glass tube and the acetonitrile was blown off under nitrogen. The resulting solution was then applied to a second high resolution gradient elution HPLC using a 4.6 x lOOmm column of Hypersil ODS. Elution of peptides was achieved using a gradient from 0.155 mol/L sodium chloride pH 2.1, to acetonitrileiwater 3:2. Fractions of 1 ml were collected and those fractions encompassing elution times of all CCKs were assayed using two RIAs-an antibody directed against the common carboxyl-terminal of CCKs and gastrins, and a relatively specific gastrin assay. This HPLC system reliably separates CCK-4, CCK-8, CCK-33139, gastrin 17, and gastrin 34; desulfated and oxidized forms of CCK do not cochromatograph with the biologically active forms. As gastrins and CCKs are sepa-
Table
1.
Secretory
Rates of Bicarbonate
During Various
111
rated by this system, the CCK assay is not a “difference” assay; the gastrin assay is simply run as a double check. Recoveries of CCK peptides added to plasma and processed through this two-step HPLC system are >96%. The carboxyl-terminal antiserum was raised in rabbits against CCK-8 conjugated to bovine serum albumin. Nonsulfated CCK-8 (E. R. Squibb & Sons, Inc., Princeton, N.J.) was used as a tracer. The CCK-8 was first purified on HPLC, iodinated using the chloramine T method, and the monoiodinated peptide was separated from the other reaction products by ion-exchange chromatography. The total volume of the assay was 1.8 ml, and contained 400 /*l of sample (equivalent to 4 ml of original plasma sample), 2000 counts per minute of iodinated CCK-8, antibody to a final dilution of 1 in 40,000, and 200 IU of aprotinin, all in 0.1 mol/L Verona1 buffer pH 8.2, containing 1% wtivol bovine serum albumin and 0.1% wtivol sodium azide. Standard and no-antibody tubes contained 400 ~1 of acetonitrilelwaterisodium chloride, 33 : 22 :45, in place of sample. Addition of tracer was delayed for 3 days and the assay was terminated after a further 3 days at 4°C. Separation of bound and free peptide was achieved by adding 500 ~1 of a charcoal suspension. Initial binding (0 standard] of tracer was 50%, and nonspecific binding ~5%. Standard curves of CCK-8, CCK33, and CCK-39 were parallel and the molar cross-reactivities of these peptides in the assay were CCK-8ICCK33/CCK-39, 1: 0.53 : 0.40. Molar cross-reactivities of other related peptides were nonsulfated CCK-8, 1.0; tetrin, 0.05; gastrin 17-1, 0.5; gastrin 34, 0.5. Secretin, bovine pancreatic polypeptide, vasoactive intestinal polypeptide, motilin, somatostatin, and gastric inhibitory peptide all showed
Statistics Differences between mean pancreatic secretion of trypsin and bicarbonate following different stimuli and differences in plasma gastrin concentrations were assessed using two-way analysis of variance, with differences between selected values assessed using paired Student’s ttests.
Stimuli Sham Sham
Bicarbonate (mmoU30 Results
Basal
Maximal
Basal
0.68 c 0.74
15.53 t 3.16"
0.85 + 0.69
min)
expressed
as mean
‘_ SD. ’ p < 0.001,
b p = 0.05 vs. basal
secretion.
feeding 3.30 k 1.97b
Basal + atropine
feeding + atropine
0.31 2 0.24
1.21 _f 1.13
112
ANAGNOSTIDES
GASTROENTEROLOGY
ET AL.
Al Basal Maximal Basal
Basal
Sham Fed
Sham Fed
+ atropine
Figure
1. Trypsin output [mean 2 SD] in kallikrein inactivator units per 30 min in the basal state and after sham feeding with and without pretreatment with atropine. Trypsin outputs determined on another day in the basal state and during “maximal” stimulation (caerulein 75 pmol . kg-’ h-’ plus secretin 84 pmol kg-’ . h-‘) are shown for comparison.
Results Basal and Maximal
Pancreatic
Secretion
Basal secretion of bicarbonate and trypsin were measured on two occasions in each subject (Table 1 and Figure 1). The secretory rates were reproducible. In all subjects, there was a prompt rise in pancreatic enzyme secretion following i.v. caerulein and secretin, evident in the first lo-min period. Trypsin secretion rose 4.5-fold, bicarbonate secretion 2%fold (Table 1). These increases were both highly significant.
Sham Feeding
Table 2. Effect of Sham Feeding Sham feeding Sham feeding + atropine Results expressed
bicarbonate secretion (Table 1). Atropine also markedly reduced the trypsin response to sham feeding. Indeed, stimulated trypsin output following atropine was comparable with basal trypsin output in the studies performed without atropine (Figure 1). Three subjects repeated the sham feeding study after pretreatment with cimetidine. In these studies, there was no rise in bicarbonate secretion during sham feeding. Mean basal bicarbonate output was 0.25 mmol/30 min, and during sham feeding it was 0.23 mmol/30 min. However, a marked trypsin response was still observed. Mean basal trypsin outputs of 6.56 KIU/30 min rose to 18.1 KIUI30 min during sham feeding-54% of the maximal output in these 3 subjects. Hormonal
on Plasma
Gastrin
Responses
to Sham Feeding
Plasma gas&in concentrations. Basal plasma gastrin concentrations before sham feeding were 60.8 + 10.8 pmol/L. There was a tendency for gastrin concentrations to rise during sham feeding but this did not reach statistical significance and by the end of the sham feeding period, plasma levels were similar to basal values [Table 2). Pretreatment with atropine produced no significant changes in basal or sham feeding gastrin levels, as compared with the no-treatment studies. Again, there was a tendency for plasma gastrin concentrations to rise during the initial period of sham feeding with a fall by 40 min. Plasma cholecystokinin concentrations. Plasma concentrations of CCK-8 and CCK-33139 were undetectable (~3 pmol/L for CCK-8 and <6 pmol/L for CCK-33/39) in control periods and remained undetectable during sham feeding.
Studies
a In all subjects, sham feeding produced marked rise in trypsin output such that mean trypsin secretion of 31.9 ? lo.5 kallikrein inactivator units (KIU) per 30 min was 92% of that observed during maximal pancreatic secretion (Figure 1). In contrast, mean bicarbonate secretion was augmented to a modest extent (20% of maximum) and this rise was only just significant (p = 0.05)(Table 1). Pretreatment with atropine significantly reduced basal trypsin output (Figure 1) but had no effect on
Vol. 87, No. 1
Discussion These data demonstrate that in humans modified sham feeding stimulates substantial pancreatic secretion of trypsin, but has little or no effect on bicarbonate secretion. Circulating gastrins and CCKs were not implicated in the sham feeding responses, but the enzyme secretory response was completely inhibited by atropine. In assessing the current (and other) data on pancreatic secretion during sham feeding, the validity of
Concentrations
Basal
5 min
10 min
20 min
40 min
60.8 ? 10.8 65.7 2 7.2
65.3 ? 16.1 66.0 ? 6.4
83.5 -c 14.3 72.8 2 10.8
69.5 ? 15.0 74.8 + 7.8
60.5 ? 13.2 61.8 rt 7.4
as mean + SEM in picomoles
per liter.
July1984
the techniques used is critical. In particular, the central problem is that of entry of gastric acid into the duodenum. This could mask a bicarbonate response (acid neutralization), augment a bicarbonate response (via secretin release), or might affect trypsin output. That small amounts of acid did enter the duodenum in our studies, even with efficient gastric aspiration, is demonstrated by the differing results after cimetidine administration, whereby sham feeding-stimulated bicarbonate secretion was abolished and trypsin output increased to 55%, not 96%, of maximal output. Thus, acid entry into the duodenum stimulated a bicarbonate response and augmented the trypsin response. Two other studies have examined the effects of sham feeding under circumstances where no acid entered the duodenum. Novis et al. (6) studied achlorhydric patients and Preshaw et al. (3) studied dogs with gastric and pancreatic fistulas. In all three studies, sham feeding stimulated enzyme secretion but bicarbonate secretion was unchanged or only marginally increased. Although these data are of interest mechanistically, exclusion of acid from the duodenum is not physiologic. Comparing the data we obtained in subjects with and without pretreatment with cimetidine allows us to predict the likely effects of sham feeding in the normal state when acid entry into the duodenum is not restricted. Trypsin output will be stimulated maximally (of which 55% is a direct effect, and the rest is related to acid entering the duodenum] and bicarbonate output will also be increased-but only consequent upon acid entering the duodenum. Our results differ in several respects from those obtained by other workers. Novis et al. (6) in studying achlorhydric patients were unable to assess the magnitude of the pancreatic response as they did not use a perfusion system. Sarles et al. (5) also used an aspiration system but suggested that large increases in enzyme and bicarbonate outputs occurred during sham feeding-presumably because small amounts of acid entered the duodenum in their studies. Read et al. (7) and Defillipi et al. (8) failed to identify a pancreatic response to sham feeding. However, Read et al. (7) perfused the jejunum and failed to preequilibrate the duodenum, and Defillipi et al. (8) used a low perfusion rate-making aspiration of duodenal contents unreliable. When they increased duodenal fluid volume by pharmacologic doses of secretin, however, they did detect an increase in enzyme output response to sham feeding. The mechanisms whereby sham feeding stimulates pancreatic secretion are disputed-partly because of the differing results in various studies. Our data, demonstrating that sham feeding induces an enzyme-rich, bicarbonate-poor secretion, suggest
SHAM FEEDING AND PANCREATIC SECRETION 113
that the response is similar to that seen after infusions of pure CCK-33 and CCK-8 (14,15). Atropine suppressed both basal and stimulated enzyme outputs suggesting that there is vagal tone in the fasting state maintaining enzyme secretion, and that sham feeding exerts its effects directly or indirectly via muscarinic cholinergic receptors. Such would not seem to be the case for bicarbonate secretion. Various postulates have been proposed to account for the effects of this vagal stimulation. The effect could be a direct one on the pancreas or indirect via neurogenic gastrin or CCK release. In the present studies, plasma gastrins did not rise significantly during sham feeding. Furthermore, atropinization virtually abolished the effects of sham feeding on trypsin secretion, but had no significant effect on plasma gastrin concentrations. We, therefore, think it unlikely that gastrin is important in the cephalic phase of pancreatic secretion in humans. Similarly, we found no evidence of release of CCK into the circulation during sham feeding-plasma concentrations of CCKs being undetectable throughout. It could be that our assay is not sensitive enough to detect the small amounts of CCK released. However, our assay is sufficiently sensitive to detect physiologic concentrations of circulating CCKs. After a fat meal, we found that plasma CCK-8 and CCK33139 concentrations each rise to -15 pmol/L (9). Furthermore, when we infused CCK-8 in sufficient amounts to produce trypsin secretion rates comparable with those in the present study (50%-90% of maximum), we were able to detect elevations in (15,161.Thus, if the plasma CCK concentrations trypsin secretion rates observed in the present study were entirely due to circulating CCKs, then our assay should detect them. We cannot exclude the possibility that vagal stimulation caused by sham feeding releases minute amounts of CCK into the circulation. but if that is the case, it is unlikely that CCK alone is sufficient to account for the pancreatic response. We conclude from these studies that central vagal stimulation of the pancreas is an important controlling mechanism for pancreatic enzyme secretion in humans, and that this effect is probably mediated directly through muscarinic receptors on acinar cells. Such a mechanism is consistent with the results obtained in isolated guinea pig acinar cells where both cholinergic agents and CCKs are potent stimulators of amylase secretion but act at different receptors, and atropine blocks the effect of cholinergic agents but not those of CCK peptides (17). These studies may have clinical implications for understanding pancreatic function in health and disease. For example, in celiac disease (18)and after various types of gastric surgery (19.20), there is an
114
ANAGNOSTIDES
ET AL
inadequate pancreatic response to intraluminal stimuli or liquid meals. However, under the more physiologic conditions of eating a solid, appetizing meal, the cephalic phase could ameliorate the effects of the blunted luminal responses. These data also provide an explanation for the observations that the most powerful luminal stimulants of pancreatic enzyme secretion are not fats and proteins but their digestion products [2). Cephalic stimulation of enzyme secretion would allow the undigested thyme entering the duodenum to be broken down and the resulting digestion products could then maintain pancreatic secretion by luminal activity as long as the nutrients remained unabsorbed.
References 1. Pavlov IP. The work of the digestive
2.
3.
4.
5. 6. 7.
8. 9.
glands. Translated by Thompson WT. London: Griffin, 1902. Meyer JH. Control of pancreatic secretion. In: Johnson LR, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:821-9. Preshaw RM, Cooke AR, Grossman MI. Sham feeding and pancreatic secretion in the dog. Gastroenterology 1966;50: 171-8. Stening GF, Grossman MI. Gastrin related peptides as stimulants of pancreatic and gastric secretion. Am J Physiol 1969;217:262-6. Sarles H, Dani G, Prezelin G, Souville C, Figarella C. Cephalic phase of pancreatic secretion in man. Gut 1968;9:214-21. Novis BH, Bank S, Marks IN. The cephalic phase of pancreatic secretion in man. Stand J Gastroenterol 1971;6:417-22. Read NW, Cooper K, Fordtran JS. Effect of modified sham feeding on jejunal transport and pancreatic and biliary secretion in man. Am J Physiol 1978;234:E417-20. Defillipi C, Solomon T, Valenzuela JE. Pancreatic response to sham feeding in humans. Digestion 1982;23:2!7-23. Maton PN, Selden AC, Chadwick VS. Measurement of large
GASTROENTEROLOGY
Vol. 87. No. 1
and small forms of cholecystokinin in human plasma using high pressure liquid chromatography and radioimmunoassay. Regul Pept 1982;4:251-60. 10. Bjornsson OG, Maton PN, Fletcher DR, Chadwick VS. Effects of duodenal perfusion with sodium taurocholate on biliary and pancreatic secretion in man. Eur J Clin Invest 1982;12:97-105. 11. Ribet A, Tournout R, Vayss N. Use of caerulein with submaximal doses of secretin as a test of pancreatic function in man. Gut 1976;17:431-4. 12. Pounder RE, Williams JG, Hunt RH, Vincent SH, MiltonThompson GJ, Misiewicz JJ. The effects of oral cimetidine on food stimulated gastric acid secretion and 24 hour intragastric acidity. In: Burland WL, Simkins MA, eds. Proceedings of the Second International Symposium on Histamine HZ-Receptor Antagonists. Amsterdam: Excerpta Medica, 1977:188-226. on gastric 13. Schoon IM, Olbe L. Inhibitory effect of cimetidine acid secretion vagally activated by physiological means in duodenal ulcer subjects. Gut 1978;19:27-31. 14. Debas HT, Grossman MI. Pure cholecystokinin: pancreatic protein and bicarbonate response. Digestion 1973;9:469-81. AA, Maton PN, Selden AC, Chadwick VS. 15. Anagnostides Effects of graded doses of cholecystokinin octapeptide (CCK 8) on pancreatic enzyme secretion in man. Clin Sci 1982;63: 21P. 16. Maton PN, Selden AC, Fitzpatrick ML, Chadwick VS. Infusion of cholecystokinin octapeptide in man: relation between plasma cholecystokinin concentrations and gallbladder emptying rates. Eur J Clin Invest 1984;14:37-41. 17. Gardner JD, Jensen RT. Regulation of pancreatic enzyme secretion in vitro. In: Johnson LR, ed. Physiology of the gastrointestinal tract. New York: Raven, 1981:831-71. 18. Di Magno EP, Go VLW, Summerskill WHJ. Impaired cholecystokinin-pancreozymin secretion, intraluminal digestion and maldigestion of fat in sprue. Gastroenterology 1972; 63:25-32. 19. MacGregor I, Parent J, Meyer JH. Gastric emptying of liquid meals and biliary secretion after subtotal gastrectomy or truncal vagotomy and pyloroplasty in man. Gastroenterology 1977;72:195-205. 20. Malagelada JR, Go VLW, Summerskill WHJ. Altered pancreatic and biliary function after vagotomy. Gastroenterology 1974;66:22-7.