- Email: [email protected]
Blockade of GRP receptors inhibits gastric emptying and gallbladder contraction but accelerates small intestinal transit
GASTROENTEROLOGY 2001;120:361–368
Blockade of GRP Receptors Inhibits Gastric Emptying and Gallbladder Contraction but Accelerates Small Intestinal Transit LUKAS P. DEGEN,* FUPING PENG,* ANNETTE COLLET,* LIVIO ROSSI,* SILVIA KETTERER,* YOLANDA SERRANO,* FINN LARSEN,‡ CHRISTOPH BEGLINGER,* and PIUS HILDEBRAND* *Division of Gastroenterology and Department of Research, University Hospital, Basel, Switzerland; and ‡Ipsen International Ltd., London, England
Background & Aims: This study was designed to characterize [D-F5Phe6D-Ala11]Bn(6-13)OMe (BIM26226) as a gastrin-releasing peptide (GRP)-preferring bombesin receptor antagonist and to determine whether GRP physiologically regulates gastrointestinal motility. Intravenous BIM26226 (5–500 g 䡠 kgⴚ1 䡠 hⴚ1) inhibits GRPinduced gallbladder contraction and plasma cholecystokinin (CCK) release in a dose-dependent fashion. Methods: Gastric emptying and small bowel transit of a solid meal were quantified using scintigraphy. Mealstimulated gallbladder contraction was measured by sonography in a 2-period crossover design. Results: Intravenous BIM26226 potently inhibited gastric lag time (114 ⴞ 7 vs. 41 ⴞ 6 minutes [control]) and gastric emptying rate (0.11 ⴞ 0.02%/min vs. 0.26 ⴞ 0.04%/ min [control]), whereas concomitant infusion of BIM26226 accelerated small bowel transit time (153 ⴞ 41 vs. 262 ⴞ 20 minutes [control]). A continuous liquid meal perfusion into the duodenum induced complete gallbladder contraction (t50%, 35 ⴞ 4 minutes), which BIM26226 inhibited significantly (t50%, 64 ⴞ 8 minutes). BIM26226 did not alter plasma CCK response, indicating that circulating CCK did not mediate these effects. Conclusions: These data show that BIM26226 is a potent antagonist of exogenous and endogenous GRP and suggest that GRP is a major physiologic regulator of gastric emptying, small bowel transit, and gallbladder contraction.
he amphibian peptide bombesin, as well as the related mammalian gastrin-releasing peptide (GRP) and neuromedin B (NMB), produce a wide range of pharmacologic responses in various tissues, including stimulation of exocrine glands, contraction of smooth muscles, release of hormones, and effects on the central nervous system.1,2 In humans, all currently known effects of bombesin-like peptides are mediated by the GRPpreferring bombesin receptor subtype, but effects on the central nervous system have not yet been characterized. However, it is not clear which of these effects are directly mediated via GRP receptors or indirectly by the release
T
of a variety of hormones or neurotransmitters. Studies measuring the effects of human GRP infusion show that concentrations termed physiologic stimulate gallbladder contraction and plasma cholecystokinin (CCK) release3,4; additionally, intravenous (IV) bombesin delays gastric emptying in humans.5 GRP is not a circulating hormone, but rather a neurotransmitter, thus explaining previous difficulties in defining the physiologic role of endogenous GRP through exogenous infusion of the peptide. Therefore, because it has not previously been possible to selectively inhibit binding of endogenous GRP to its receptor in target tissues, the role of endogenous GRP in the regulation of gastrointestinal motility in humans has not been determined to date. This mechanism could prove useful in examining the physiologic actions of this peptide in humans. Recently, several peptide analogues of bombesin have been synthesized by the group of David H. Coy (Tulane University, New Orleans, LA).6,7 These function as antagonists for the GRP-preferring bombesin receptor. Among them, [D-F5Phe6D-Ala11]Bn(6-13)OMe (BIM26226), an octapeptide analogue of bombesin, showed high selectivity and potency for the GRP receptor but was devoid of any agonist activity.8 In vitro studies have shown BIM26226 to be an effective inhibitor of GRP binding to rat pancreatic acini and AR4-2J cells (IC50, 5.5 and 3 nmol/L, respectively), with a 300-fold lower affinity for NMB receptors.8,9 BIM26226 has been examined for its in vitro binding to a total of 26 different receptors, among them all relevant brain-gut peptides; Ki values were ⬎1000 nmol/L for all nonbombesin receptors tested. In rats, BIM26226 inhibited bombesin-stimulated exocrine pancreatic secretion in vivo.10 These data,
Abbreviations used in this paper: GRP, gastrin-releasing peptide; NMB, neuromedin B. © 2001 by the American Gastroenterological Association 0016-5085/01/$35.00 doi:10.1053/gast.2001.21174
362
DEGEN ET AL.
as a whole,8 –10 show that BIM26226 is a specific GRP antagonist in vitro and in animals. The present investigation consisted of 3 separate studies. First, 8 subjects received placebo or 1 of 3 doses of BIM26226 together with graded doses of IV synthetic human GRP in a 4-period crossover fashion. The objective of this study was to show that BIM26226, administered as a continuous IV infusion, acts as a competitive antagonist of GRP-stimulated gallbladder contraction and plasma CCK release in humans; furthermore, the maximally effective dose of BIM26226 was to be established based on the inhibition of GRP-stimulated functions. Second, 8 subjects received this maximal dose of BIM26226 or saline (control) before and after ingestion of a labeled solid meal. The objective of this second study was to use the GRP-receptor antagonist as a tool to assess the physiologic effects of endogenous GRP on gastric emptying and small bowel transit time by scintigraphy. Finally, 6 subjects received a liquid meal continuously perfused into the duodenum, with or without BIM26226. The objectives of this last part of the study were to characterize the role of endogenous GRP on gallbladder contraction and plasma CCK release, independent of gastric emptying rates.
Materials and Methods Experimental Subjects Twenty-two healthy male volunteers, 21– 46 years old, were selected for the study. Volunteers were taking no medication before the study, and each subject was within 15% of his ideal body weight. Each subject had normal screening physical examination and laboratory test results, including urinalysis, complete blood count, serum chemistries, electrocardiography, and abdominal ultrasonography. Subjects had no history of significant illness or surgery. The study was approved by the Ethics Committee of the University Hospital of Basel. Written informed consent was obtained from each subject. All studies were conducted in the morning after an overnight (12-hour) fast.
Materials Synthetic human GRP (pyrogen-free and sterilized for IV use) was purchased from Novabiochem (Laufelfingen, Switzerland). Ensure was purchased from Abbott (Cham, Switzerland). BIM26226 (mol wt, 1075) was supplied by Ipsen International Ltd. (London, England) as lyophilized powder (chemical purity, 97%). An infusion rate of BIM26226 of 500 g 䡠 kg⫺1 䡠 h⫺1 induced 10- and 100-fold lower concentrations, respectively. Fifteen minutes after cessation of the infusion, the compound was no longer detectable, indicating a half-life of ⬍2 minutes. Based on plasma concentrations induced by the highest dose of BIM26226, it can be estimated that a ratio of agonist to antagonist at the receptor site in the
GASTROENTEROLOGY Vol. 120, No. 2
range of 1/1000 is necessary to inhibit the function by more than 97%.10
Effect of BIM26226 on Exogenous GRPStimulated Gallbladder Contraction and Plasma CCK Release Eight subjects participated in a 4-period, single-blind, placebo-controlled crossover study. On 4 days, separated by at least 3 days, subjects received either an IV infusion of saline (control) or 1 of 3 doses of BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1, equal to 4.56, 46.5, or 465 nmol 䡠 kg⫺1 䡠 h⫺1) for the duration of the experiment. Before the infusions began, subjects had to swallow a gastric tube that had been positioned under fluoroscopic control.11 Gastric acid secretion was aspirated to avoid endogenous stimulation of gallbladder contraction and CCK release through duodenal acid load. Furthermore, 3 indwelling catheters were placed in the arm veins, 2 for infusions and 1 for drawing blood at regular intervals to measure CCK plasma levels. Sixty minutes after the beginning of the BIM26226 infusion, GRP was administered in graded doses (10, 30, and 90 pmol 䡠 kg⫺1 䡠 h⫺1); each dose was administered throughout a 45-minute period.3
Measurement of Gallbladder Volumes Gallbladder contraction was assessed by high-resolution, real-time sonography (ALOKA SSD650) using a 3.5MHz probe on a sector scanner. Longitudinal sonograms of the gallbladder were recorded every 15 minutes to calculate the volume, assuming that the gallbladder shape approximated a solid whose sections are elliptic with the eccentricity depending on the eccentricity at the level of the maximal transverse section. These assumptions and the mathematical formula used to calculate the volume have been described previously.12,13 The key to this methodology is accurate estimation of the gallbladder volume. However, estimations of the gallbladder are dependent on some subjectivity by the observer. Gallbladder volumes were therefore calculated without knowledge of the subjects’ treatment status, i.e., saline or BIM26226. No gallstones, wall thickening, or other pathologic conditions were identified in any subject.
Effect of BIM26226 on Gastric Emptying and Small Bowel Transit Time of a Solid Meal In random order, 8 subjects received either an IV infusion of saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) for the duration of the experiment. Gastric emptying and small bowel transit were assessed by a noninvasive scintigraphic method developed and validated in recent years.14 –17 Polystyrene Amberlite 120-IR-Plus resin pellets (average diameter, 1 mm; range, 0.5–1.8 mm) were labeled with 100 Ci of 111In-Cl . The efficiency of the labeling is ⬎98%, as judged by 3 thin-layer chromatography.14 A capsule filled with approximately 0.5 g of pellets and coated with a layer of methacrylate was given to the fasting volunteers. As expected, the capsule
February 2001
dissolved in the ileocecal region and thereafter marked ileocecal transfer.18 External radioactive markers were placed over both anterior superior iliac spines. As soon as the radiolabeled capsule passed into the small bowel, IV infusion of either placebo or BIM26226 was begun. Thirty minutes later, a breakfast of 2 scrambled eggs, a slice of whole-wheat bread, and skim milk (35% protein, 52% carbohydrate, 13% fat, 219 kcal) was eaten within 10 minutes. The scrambled eggs were mixed and cooked with 1 mCi of 99mTc-labeled Amberlite 410 resin pellets (average diameter, 1 mm) to a firm consistency, thereby providing a solid medium. Four hours after breakfast, a standardized nonradiolabeled lunch was consumed. Gamma camera imaging started immediately after completion of ingestion of the radiolabeled breakfast with a large-field-of-view gamma camera of medium energy, parallel-hole collimator. Anterior and posterior images were taken with the subject in an erect position. Energy windows were 140 keV for the 99mTc counts and 245 keV for the 111In counts (each with 99mTc and 111In). The geometric means of the counts obtained from the anterior and posterior images were calculated for each region and then corrected for radionuclide decay. The downscatter of 111In into the 99mTc window was adjusted. Blood was drawn to determine plasma CCK concentrations at 15-minute intervals during the first 90 minutes and at 60-minute intervals thereafter.
Effect of BIM26226 on Duodenal MealStimulated Gallbladder Contraction On 2 different days and in random order, 6 subjects received either saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) at a rate of 3 mL/min to circumvent effects of BIM26226 on gastric emptying. This meal (540 mL) contained 72.1 g carbohydrates, 18.1 g fat, and 22.7 g proteins. Gallbladder contraction was measured as described above, and blood was drawn at intervals for CCK determinations.
Plasma CCK Determinations Plasma CCK concentrations were measured by radioimmunoassay as previously described.19 CCK determinations were done with antibody OAL656, which shows ⬍1% crossreactivity to sulfated and unsulfated forms of gastrin and does not bind to structurally unrelated peptides. Plasma levels of BIM26226 were determined by combined HPLC/MS (CEMAF SA, Poitiers, France).
Calculations and Statistical Analysis In the first study, with increasing doses of BIM26226, the effects of gallbladder contraction and plasma CCK concentrations were calculated as area under the curve and used for subsequent comparative statistics. Meal-stimulated gallbladder emptying data were fitted using Gauss–Newton nonlinear least-square regression.20 The half-emptying time (t50%) was defined as the period at which 50% emptied from the gallbladder. Gastric emptying was assessed by the gastric lag time and postlag emptying rate. The gastric lag time (minutes) is the time required for 10% of the radiolabeled contents to empty from the stomach.16 The gastric postlag emptying rate
EFFECTS OF GRP–RECEPTOR BLOCKADE
363
(%/min) is characterized as the slope estimated by the linear regression analysis of the data points from the first point beyond the lag time until 90% of the radiolabeled contents has emptied from the stomach.14 Small bowel transit time (minutes) was appraised by subtracting the time required for 10% of isotope to empty from the stomach from the time it took to enter the colon.15 The data were analyzed with an analysis of variance (ANOVA) model for either a 4- or a 2-period crossover design as appropriate. The normality and homogeneity of variance assumptions of the ANOVA model were tested by the Kolmogoroff–Smirnoff test and Levine test, respectively. In the case of significant differences, data were subsequently analyzed by a nonparametric test (Friedman). Differences between treatment groups were considered significant at P ⬍ 0.05 (2-tailed test).
Results Effect of BIM26226 on Exogenous GRPStimulated Gallbladder Contraction and Plasma CCK Release To determine whether BIM26226 is an effective antagonist of exogenous GRP, subjects received IV saline (control) or 1 of 3 doses of BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiments. In the control study (saline), infusion of graded doses of human GRP produced a dose-dependent gallbladder contraction with a maximum obtained at 90 pmol 䡠 kg⫺1 䡠 h⫺1 (Figure 1A). BIM26226 inhibited GRP-induced gallbladder contraction in a dose-dependent manner. Compared with placebo, all 3 doses of BIM26226 significantly (P ⬍ 0.05) inhibited GRP-stimulated gallbladder contraction. The highest dose of BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) completely abolished gallbladder contraction, and the initial volumes were between 23.0 ⫾ 2.8 and 23.7 ⫾ 1.6 mL, respectively, and did not differ among the 4 treatments (P ⬎ 0.05). The changes in plasma CCK levels in response to GRP infusion and the ability of BIM26226 to affect this increase are shown in Figure 1B. In the control study (saline), infusion of GRP produced a dose-dependent increase up to supraphysiologic plasma CCK levels. All 3 doses of BIM26226 completely inhibited CCK release (P ⬍ 0.05). Even the lowest dose (5 g 䡠 kg⫺1 䡠 h⫺1), which only partially antagonized gallbladder contraction, was able to keep plasma CCK at a basal level during maximal GRP stimulation. Effects of BIM26226 on Gastric Emptying and Small Bowel Transit Time of a Solid Meal The effects of BIM26226 on gastric emptying rates of a solid meal are shown in Figure 2A and Table 1.
364
DEGEN ET AL.
GASTROENTEROLOGY Vol. 120, No. 2
䡠 360 min after BIM26226 treatment, which was not different from control studies. Small bowel transit time amounted to 262 ⫾ 20 minutes in the control study (saline), whereas concomitant infusion of BIM26226 significantly accelerated this period to 153 ⫾ 41 minutes (Table 1). Effect of BIM26226 on Duodenal MealStimulated Gallbladder Contraction Intraduodenal perfusion of the test meal with saline (control) was followed by a time-dependent contraction of the gallbladder (Figure 3A) with a halfemptying time of 35 ⫾ 4 minutes. BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) induced significant inhibition of gallbladder contraction with a half-emptying time of 64 ⫾ 8 min-
Figure 1. Potency of BIM26226 as a GRP-receptor antagonist. (A ) Effect of saline (control) or BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1) on gallbladder contraction, induced by infusion of graded doses of human GRP (10, 30, or 90 pmol 䡠 kg⫺1 䡠 h⫺1). Gallbladder volumes are expressed as percentages of the basal volume. Values are means ⫾ SEM of 8 subjects. (B) Plasma CCK concentrations in the same studies. Results are expressed as means ⫾ SEM of 8 subjects. In a dose-dependent fashion, BIM26226 reduced GRP-stimulated gallbladder contraction and inhibited CCK release. The highest dose (500 g 䡠 kg⫺1 䡠 h⫺1) completely inhibited GRP-stimulated functions.
During the control study, the meal continuously emptied, with only 14% ⫾ 9% remaining in the stomach 6 hours after ingestion of the meal. The lag time was 41 ⫾ 6 minutes, and the gastric emptying rate was 0.26% ⫾ 0.04%/min, respectively. With BIM26226 infusion, 60% ⫾ 5% of the meal remained in the stomach after 6 hours. All parameters of gastric emptying of the test meal were significantly (P ⬍ 0.05) inhibited; gastric lag time was 114 ⫾ 29 minutes, and the gastric emptying rate was calculated as 0.11 ⫾ 0.02%/min (Table 1). In control studies, plasma CCK levels increased immediately after ingestion of the meal (Figure 2B). With BIM26226 infusion, the increase in plasma CCK concentrations was initially delayed and less pronounced but lasted longer. The area under the curve was 210 ⫾ 73 pmol 䡠 360 min with saline (control) and 191 ⫾ 44 pmol
Figure 2. Effects of BIM26226 on gastric emptying and plasma CCK release. (A ) Effect of BIM26226 on gastric emptying of a solid meal assessed by scintigraphy. Eight subjects received IV saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiment. Results are expressed as the percentage of meal contents remaining in the stomach. Values are means ⫾ SEM. BIM26226 significantly inhibited the lag phase as well as the emptying rate of the solid meal (P ⬍ 0.05) compared with saline (control). (B) Plasma CCK concentrations in the same studies. Results are expressed as means ⫾ SEM of 8 subjects. With BIM26226, integrated plasma CCK release did not differ from control values.
February 2001
EFFECTS OF GRP–RECEPTOR BLOCKADE
Table 1. Effects of BIM26226 on Postprandial Intestinal Motility Saline (control) Gastric lag time (min) Gastric emptying rate (%/min) Small bowel transit time (min) Gallbladder contraction t1/2 (min) aP
41 ⫾ 6
BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) 114 ⫾ 29a
0.26 ⫾ 0.04
0.11 ⫾ 0.02a
262 ⫾ 20
153 ⫾ 41a
35 ⫾ 4
64 ⫾ 8a
⬍ 0.05 vs. control.
utes (P ⬍ 0.05). In the control study, stimulated plasma CCK levels immediately increased 3-fold over basal levels and remained nearly constant for the duration of the experiment. There was no significant difference in the area under the plasma CCK curve between saline (control) and BIM26226-treated groups (Figure 3B).
365
Several mechanisms whereby bombesin-like peptides can influence intestinal motility are possible: (1) GRP could directly activate its receptor on smooth muscle cells; (2) GRP could activate cholinergic mechanisms (this explanation is based on the observation that GRP is found mainly in mucosal, submucosal, and myenteric plexuses21–23); (3) bombesin-like peptides are known to release a variety of other regulatory factors, such as CCK, somatostatin, glucagon-like peptide 1, gastrin, motilin, or neurotensin, which in turn could mediate smooth muscle contraction24,25; and (4) it might be possible that the effects are of central (vagal) origin because bombesinlike immunoreactivity has been demonstrated in the human brain. Whether BIM26226 crosses the blood– brain barrier is not known, but GRP, CCK, peptide YY,
Adverse Effects No adverse effects were observed or symptoms reported during or after BIM26226 administration.
Discussion In this study we show that BIM26226 is a potent GRP receptor antagonist in humans. BIM26226 dosedependently reduced gallbladder contraction stimulated by exogenous GRP and inhibited plasma CCK release. In additional unpublished studies, we have also established that BIM26226 acts as a competitive antagonist for a variety of exogenous GRP-stimulated gut functions, including gastric acid secretion, exocrine pancreatic secretory responses, and hormone release (gastrin, pancreatic polypeptide), thus providing experimental evidence for the potency and efficacy of the compound to antagonize the function of GRP-preferring bombesin receptors in humans. This potency has previously been shown in vitro and in animals.8,9 Using BIM26226 as a specific tool to assess the consequences of GRP-receptor blockade during endogenous GRP stimulation by oral meal ingestion, it was shown to (1) dramatically decrease gastric emptying, (2) inhibit gallbladder contraction, and (3) accelerate small bowel transit time in humans. Postprandial plasma CCK release from the duodenal mucosa was reduced most likely because of delayed gastric emptying. However, BIM26226 decreased gallbladder contraction without exerting any effect on plasma CCK release. These results represent the first human data on the physiologic role that endogenous GRP exerts on gastrointestinal motility using a potent and specific GRP-receptor antagonist.
Figure 3. Effects of BIM26226 on gallbladder contraction and plasma CCK release. (A ) Effect of BIM26226 on gallbladder contraction induced by duodenal perfusion of a liquid meal. Six subjects received IV saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiment. Results are expressed as the percentage of basal (premeal) volumes. Values are means ⫾ SEM. BIM26226 significantly (P ⬎ 0.05) reduced gallbladder contraction compared with saline (control). (B) Effect of BIM26226 on plasma CCK concentrations during the same studies. Results are expressed as means ⫾ SEM of 6 subjects. Plasma CCK levels, with and without BIM26226, did not significantly differ, indicating that the effect of BIM26226 on gallbladder contraction was not mediated by circulating CCK.
366
DEGEN ET AL.
and gastrin did not cross the blood– brain barrier in an in vitro model using porcine brain capillary endothelial cells ( Ju¨rgen Drewe and Christoph Beglinger, unpublished observations). Which of these different mechanisms are involved in mediating the effects on gastric emptying, small bowel transit, and gallbladder contraction can be answered only partially by the results of the present study. Only circulating hormones could be analyzed and interpreted in this context; all other mechanisms exert their effects in a paracrine or neuroendocrine manner. Their GRP receptors might be blocked by BIM26226, but there is no parameter that would allow the quantification of these effects in vivo, and we must therefore rely on in vitro studies or animal experiments to test these hypotheses. The extent of this inhibiting effect of GRP-receptor blockade on gastric emptying was unexpected from previous in vivo studies in animals or humans. It has been shown that infusion of bombesin inhibits gastric emptying in humans.5 More recently, bombesin and GRP (10⫺9 to 10⫺6 mol/L) have been shown to stimulate isolated muscle strips of the human stomach in vitro; bombesin stimulated both circular and longitudinal preparations from all regions of the stomach, and these effects were unaltered by atropine or tetrodotoxin.26,27 Additional potential mechanisms of action are based on in vitro data from animals. GRP has been isolated from canine antral muscle, but in the rat, GRP was found predominantly in the mucosal plexus of the gastric fundus and antrum.23 Bombesin increased the frequency and amplitude of contractions in smooth muscle strips from canine stomach; the action was myogenic in the circular muscle but was blocked by atropine in the longitudinal muscle.28 The contraction of rat stomach strips induced by bombesin-like peptides could not be antagonized by a variety of neural inhibitors.29 On the other hand, GRP nerves were found in the porcine antrum, and these nerves mediate vagal signals that are responsible for motor functions and hormone release.30 One of the hormonal candidates for a regulatory function of GRP stimulation is CCK, which mediates various gastrointestinal motor functions, including inhibition of gastric emptying in dogs, rats, and humans.31 In the present study, CCK plasma levels were unchanged during intraduodenal meal stimulation with BIM26226, implying that CCK cannot account for the delay in gastric emptying. The same is true for gastrin, which binds with high affinity to CCK-B receptors; again, postprandial plasma gastrin levels were unchanged with and without BIM26226 (data not shown). Circulating somatostatin, glucagon-like peptide 1, motilin, and neurotensin, all
GASTROENTEROLOGY Vol. 120, No. 2
further potential inhibitors of gastric emptying, were not analyzed in the present study. In the isolated porcine stomach, however, vagus-stimulated somatostatin secretion in the antrum was strongly inhibited by a GRPreceptor antagonist.32 Furthermore, results obtained from preparations of rat stomach indicated that GRPstimulated antral somatostatin release, which was sensitive to neural inhibitors, could mediate the effects on gastric emptying.33 In the context of gastric acid secretion studies, we determined the quantity of local somatostatin messenger RNA in biopsy specimens from the gastric antrum and corpus by Northern blotting, but we were not able to show a significant change with BIM26226 administration ( John Calam and Pius Hildebrand, unpublished results). Which pathways mediate the effects on gastric motility remains to be determined, but it is clear that blockade of endogenously stimulated GRP receptors led to an almost complete cessation of gastric emptying. This is in keeping with findings in dogs, in which administration of a monoclonal antibombesin antibody led to increased gastric retention.34 The data imply a major regulatory function of GRP in gastric emptying in humans. There is ample evidence that bombesin-like peptides exert some effects on human small bowel motility. Isolated smooth muscle cells from longitudinal and circular smooth muscle layers of the human jejunum have been shown to be contracted by bombesin-like peptides in picomolar concentrations; the contractile response was not affected by muscarinic, opioid, CCK, or serotonin antagonists, indicating that bombesin-like peptides directly cause intestinal smooth muscle contraction.26 In vitro studies with human duodenal muscle strips showed that longitudinally cut strips were excitable by GRP, whereas circularly cut strips showed no response.35 However, high doses of bombesin have been shown to inhibit basal mechanical activity of the human duodenum and jejunum.36 However, the response of endogenous GRP on postprandial small bowel transit time can be estimated only by using a GRP-receptor antagonist. The acceleration of small bowel transit time by BIM26226 indicated that, among the regulators of motility, the postprandial effect of GRP alone seems to prolong small bowel transit time. Both exogenous bombesin and GRP cause contraction of the human gallbladder.3,37 We and others have previously suggested that this effect is largely mediated through CCK by showing that the specific CCK-A– receptor antagonist loxiglumide completely inhibits GRP-stimulated gallbladder contraction.38 In addition, GRP receptors have been identified on human gallblad-
February 2001
der tissue using immunohistochemical techniques.39 On the other hand, in most species bombesin has a weak direct stimulatory effect on gallbladder contraction in vitro. The present results suggest that both exogenously and endogenously GRP-stimulated gallbladder contraction is at least partially independent from circulating plasma CCK. First, gallbladder contraction stimulated by increasing doses of GRP was only partially inhibited by the lower 2 doses of BIM26226, which in turn completely abolished plasma CCK release (Figure 1). Second, gallbladder contraction was significantly reduced during duodenal meal stimulation with concomitant infusion of the GRP-receptor antagonist, whereas CCK plasma levels remained unaffected during the same time frame. Taken together, these data suggest that GRP is a physiologic, peptidergic regulator of gallbladder motility. However, gallbladder volumes, as measured by ultrasonography, are the result of pressure gradients produced by the detrusor muscle of the gallbladder itself, the sphincter of Oddi, and hepatic bile production. Thus, in vitro studies with isolated human smooth muscle tissues and smooth muscle cells must be undertaken to gain insight into the regulatory mechanisms involved. Several studies have suggested a regulatory role for GRP in controlling CCK release because IV infusion of bombesin-like peptides induces increased CCK plasma levels.3,22 In this study with continuous duodenal perfusion of a liquid meal, we did not observe a difference in plasma CCK secretion with or without BIM26226. Therefore, CCK release after IV application of bombesinlike peptides represents a pharmacologic rather than a physiologic event. Unfortunately, neural release of CCK in response to GRP, either in the central nervous system or in the periphery, has not yet been studied. In conclusion, we show that BIM26226 is a potent antagonist of GRP based on the inhibition of exogenous GRP-stimulated gallbladder contraction and plasma CCK release. Additionally, our studies show the usefulness of BIM26226 as a probe of the physiologic activity of endogenous GRP, revealing (1) a major role of this peptide in gastric emptying, (2) a significant role in small bowel transit and in gallbladder contraction, and (3) a possibly limited role of GRP in regulation of plasma CCK secretion.
EFFECTS OF GRP–RECEPTOR BLOCKADE
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
19.
References 1. Sunday ME, Kaplan LM, Motoyama E, Chin WW, Spindel ER. Biology of disease. Gastrin-releasing peptide (mammalian bombesin) gene expression in health and disease. Lab Invest 1988; 59:5–24. 2. Walsh JH. Bombesin-like peptides. In: Schultz SG, Makhlouf GM, Rauner BB, eds. Handbook of physiology. Section 6. The gastro-
20. 21. 22.
367
intestinal system. Volume 2. Neural and endocrine biology. Bethesda, MD: American Physiological Society, 1989:587– 610. Hildebrand P, Werth B, Beglinger C, Delco F, Jansen JB, Lamers CB, Gyr K. Human gastrin-releasing peptide: biological potency in humans. Regul Pept 1991;36:423– 433. Lundell L, Cantor P, Sjovall M, Rehfeld JF, Olbe L. Factors influencing the release of cholecystokinin induced by gastrin-releasing peptide in man. Scand J Gastroenterol 1991;26:544 –550. Scarpignato C, Micali B, Vitulo F, Zimbaro G, Bertaccini G. Inhibition of gastric emptying by bombesin in man. Digestion 1982; 23:128 –131. Jensen RT, Coy DH. Progress in the development of potent bombesin receptor antagonists. Trends Pharmacol Sci 1991;12:13– 19. Heinz-Erian P, Coy DH, Tamura M, Jones SW, Gardner JD, Jensen RT. [D-Phe12]bombesin analogues: a new class of bombesin receptor antagonists. Am J Physiol 1987;252:G439 –G442. Coy DH, Mungan Z, Rossowski WJ, Cheng BL, Lin JT, Mrozinski JE, Jensen RT. Development of a potent bombesin receptor antagonist with prolonged in vivo inhibitory activity on bombesinstimulated amylase and protein release in the rat. Peptides 1992;13:775–781. Dietrich JB, Hildebrand P, Jeker LB, Pansky A, Eberle AN, Beglinger C. Effects of BIM26226, a potent and specific bombesin receptor antagonist, on amylase release and binding of bombesin-like peptides to AR4-2J cells. Regul Pept 1994;53: 165–173. Varga G, Reidelberger RD, Liehr RM, Bussjaeger LJ, Coy DH, Solomon TE. Effects of potent bombesin antagonist on exocrine pancreatic secretion in rats. Peptides 1991;12:493– 497. Beglinger C, Fried M, Whitehouse I, Jansen JB, Lamers CB, Gyr K. Pancreatic enzyme response to a liquid meal and to hormonal stimulation. Correlation with plasma secretin and cholecystokinin levels. J Clin Invest 1985;75:1471–1476. Everson GT, Braverman DZ, Johnson ML, Kern F Jr. A critical evaluation of real-time ultrasonography for the study of gallbladder volume and contraction. Gastroenterology 1980;79:40 – 46. Beglinger C, Hildebrand P, Adler G, Werth B, Luo H, Delco F, Gyr K. Postprandial control of gallbladder contraction and exocrine pancreatic secretion in man. Eur J Clin Invest 1992;22:827– 834. Camilleri M, Colemont LJ, Phillips SF, Brown ML, Thomforde GM, Chapman N, Zinsmeister AR. Human gastric emptying and colonic filling of solids characterized by a new method. Am J Physiol 1989;257:G284 –G290. Greydanus MP, Camilleri M, Colemont LJ, Phillips SF, Brown ML, Thomforde GM. Ileocolonic transfer of solid chyme in small intestinal neuropathies and myopathies. Gastroenterology 1990; 99:158 –164. Camilleri M, Zinsmeister AR, Greydanus MP, Brown ML, Proano M. Towards a less costly but accurate test of gastric emptying and small bowel transit. Dig Dis Sci 1991;36:609 – 615. Degen LP, Phillips SF. Variability of gastrointestinal transit in healthy women and men. Gut 1996;39:299 –305. Proano M, Camilleri M, Phillips SF, Thomforde GM, Brown ML, Tucker RL. Unprepared human colon does not discriminate between solids and liquids. Am J Physiol 1991;260:G13–G16. Hildebrand P, Ensinck JW, Ketterer S, Delco F, Mossi S, Bangerter U, Beglinger C. Effect of a cholecystokinin antagonist on meal-stimulated insulin and pancreatic polypeptide release in humans. J Clin Endocrinol Metab 1991;72:1123–1129. Elashoff JD, Reedy TJ, Meyer JH. Analysis of gastric emptying data. Gastroenterology 1982;83:1306 –1312. Price J, Penman E, Wass JA, Rees LH. Bombesin-like immunoreactivity in human gastrointestinal tract. Regul Pept 1984;9:1–10. Dockray GJ, Vaillant C, Walsh JH. The neuronal origin of bom-
368
23. 24.
25.
26.
27.
28.
29.
30.
31.
32.
DEGEN ET AL.
besin-like immunoreactivity in the rat gastrointestinal tract. Neuroscience 1979;4:1561–1568. Walsh JH, Wong HC, Dockray GJ. Bombesin-like peptides in mammals. Fed Proc 1979;38:2315–2319. McDonald TJ, Ghatei MA, Bloom SR, Track NS, Radziuk J, Dupre J, Mutt V. A qualitative comparison of canine plasma gastroenteropancreatic hormone response to bombesin and the porcine gastrin-releasing peptide (GRP). Regul Pept 1981;2:293–304. Ghatei MA, Jung RT, Stevenson JC, Hillyard CJ, Adrian TE, Lee YC, Christofides ND, Sarson DL, Mashiter K, MacIntyre I, Bloom SR. Bombesin: action on gut hormones and calcium in man. J Clin Endocrinol Metab 1982;54:980 –985. Pogrzeba B, Mandrek K, Ludtke FE, Lepsien G, Golenhofen K. Contractile action of gastrin-releasing peptide on isolated preparations of human gastroduodenal muscle. Dig Dis 1991;9:354 – 359. Ludtke FE, Pogrzeba B, Mandrek K, Lepsien G, Neufang T, Golenhofen K. Bombesin—the most stimulating peptide of human gastric smooth muscle. Dig Dis 1991;9:371–381. Mayer EA, Elashoff J, Walsh JH. Characterization of bombesin effects on canine gastric muscle. Am J Physiol 1982;243:G141– G147. Girard F, Bachelard H, St-Pierre S, Rioux F. The contractile effect of bombesin, gastrin releasing peptide and various fragments in the rat stomach strip. Eur J Pharmacol 1984;102:489 – 497. Holst JJ, Knuhtsen S, Orskov C, Skak-Nielsen T, Poulsen SS, Nielsen OV. GRP-producing nerves control antral somatostatin and gastrin secretion in pigs. Am J Physiol 1987;253:G767– G774. Fried M, Erlacher U, Schwizer W, Lo¨chner C, Koerfer J, Beglinger C, Jansen JB, Lamers CB, Harder F, Bischof-Delaloye A, Stalder GA, Rovati L. Role of cholecystokinin in the regulation of gastric emptying and pancreatic enzyme secretion in humans. Studies with the cholecystokinin-receptor antagonist loxiglumide. Gastroenterology 1991;101:503–511. Holst JJ, Harling H, Messell T, Coy DH. Identification of the neurotransmitter/neuromodulator functions of the neuropeptide gastrin-releasing peptide in the porcine antrum, using the antagonist (Leu13-psi-CH2 NH-Leu14)-bombesin. Scand J Gastroenterol 1990;25:89 –96.
GASTROENTEROLOGY Vol. 120, No. 2
33. Martindale R, Kauffman GL, Levin S, Walsh JH, Yamada T. Differential regulation of gastrin and somatostatin secretion from isolated perfused rat stomachs. Gastroenterology 1982;83: 240 –244. 34. Kovacs TO, Cuttitta F, Maxwell V, Hamel D, Walsh JH. The effects of a monoclonal antibody to bombesin on gastric function in dogs (abstr). Gastroenterology 1986;90:1502. 35. Micheletti R, Grider JR, Makhlouf GM. Identification of bombesin receptors on isolated muscle cells from human intestine. Regul Pept 1988;21:219 –226. 36. Corazziari E, Torsoli A, Melchiorri P, Delle Fave G, Habib FI. Effect of bombesin on the mechanical activity of the human duodenum and jejunum. Rendic Gastroenterol 1974;6:55–56. 37. Corazziari E, Torsoli A, Melchiorri P, Delle Fave G. Effect of bombesin on human gallbladder emptying. Rendic Gastroenterol 1974;6:52–55. 38. Hildebrand P, Drewe J, Luo H, Ketterer S, Gyr K, Beglinger C. Role of cholecystokinin in mediating GRP-stimulated gastric, biliary and pancreatic functions in man. Regul Pept 1992;22;41:119 – 129. 39. Tsutsumi Y, Nagura H, Watanabe K, Yanaihara N. Methods in laboratory investigation. Immunoreactivity of N-terminal fragment of gastrin-releasing peptide as histochemical marker for pancreatobiliary duct-type cells. Lab Invest 1984;50:94 –100.
Received June 12, 2000. Accepted September 22, 2000. Address requests for reprints to: Pius Hildebrand, M.D., Department of Research, University Hospital Basel, Hebelstrasse 20, CH-4031 Switzerland; e-mail: [email protected]; fax: (41) 61-2652350. Supported in part by grant no. 3200-040604.94 from the Swiss National Science Foundation and by a research grant from Ipsen International Ltd. (London, England). P. Hildebrand received funding support from the Sandoz Foundation. The authors thank Gerdien Gamboni for expert technical assistance and Kathleen A. Bucher for editorial assistance.
Blockade of GRP Receptors Inhibits Gastric Emptying and Gallbladder Contraction but Accelerates Small Intestinal Transit LUKAS P. DEGEN,* FUPING PENG,* ANNETTE COLLET,* LIVIO ROSSI,* SILVIA KETTERER,* YOLANDA SERRANO,* FINN LARSEN,‡ CHRISTOPH BEGLINGER,* and PIUS HILDEBRAND* *Division of Gastroenterology and Department of Research, University Hospital, Basel, Switzerland; and ‡Ipsen International Ltd., London, England
Background & Aims: This study was designed to characterize [D-F5Phe6D-Ala11]Bn(6-13)OMe (BIM26226) as a gastrin-releasing peptide (GRP)-preferring bombesin receptor antagonist and to determine whether GRP physiologically regulates gastrointestinal motility. Intravenous BIM26226 (5–500 g 䡠 kgⴚ1 䡠 hⴚ1) inhibits GRPinduced gallbladder contraction and plasma cholecystokinin (CCK) release in a dose-dependent fashion. Methods: Gastric emptying and small bowel transit of a solid meal were quantified using scintigraphy. Mealstimulated gallbladder contraction was measured by sonography in a 2-period crossover design. Results: Intravenous BIM26226 potently inhibited gastric lag time (114 ⴞ 7 vs. 41 ⴞ 6 minutes [control]) and gastric emptying rate (0.11 ⴞ 0.02%/min vs. 0.26 ⴞ 0.04%/ min [control]), whereas concomitant infusion of BIM26226 accelerated small bowel transit time (153 ⴞ 41 vs. 262 ⴞ 20 minutes [control]). A continuous liquid meal perfusion into the duodenum induced complete gallbladder contraction (t50%, 35 ⴞ 4 minutes), which BIM26226 inhibited significantly (t50%, 64 ⴞ 8 minutes). BIM26226 did not alter plasma CCK response, indicating that circulating CCK did not mediate these effects. Conclusions: These data show that BIM26226 is a potent antagonist of exogenous and endogenous GRP and suggest that GRP is a major physiologic regulator of gastric emptying, small bowel transit, and gallbladder contraction.
he amphibian peptide bombesin, as well as the related mammalian gastrin-releasing peptide (GRP) and neuromedin B (NMB), produce a wide range of pharmacologic responses in various tissues, including stimulation of exocrine glands, contraction of smooth muscles, release of hormones, and effects on the central nervous system.1,2 In humans, all currently known effects of bombesin-like peptides are mediated by the GRPpreferring bombesin receptor subtype, but effects on the central nervous system have not yet been characterized. However, it is not clear which of these effects are directly mediated via GRP receptors or indirectly by the release
T
of a variety of hormones or neurotransmitters. Studies measuring the effects of human GRP infusion show that concentrations termed physiologic stimulate gallbladder contraction and plasma cholecystokinin (CCK) release3,4; additionally, intravenous (IV) bombesin delays gastric emptying in humans.5 GRP is not a circulating hormone, but rather a neurotransmitter, thus explaining previous difficulties in defining the physiologic role of endogenous GRP through exogenous infusion of the peptide. Therefore, because it has not previously been possible to selectively inhibit binding of endogenous GRP to its receptor in target tissues, the role of endogenous GRP in the regulation of gastrointestinal motility in humans has not been determined to date. This mechanism could prove useful in examining the physiologic actions of this peptide in humans. Recently, several peptide analogues of bombesin have been synthesized by the group of David H. Coy (Tulane University, New Orleans, LA).6,7 These function as antagonists for the GRP-preferring bombesin receptor. Among them, [D-F5Phe6D-Ala11]Bn(6-13)OMe (BIM26226), an octapeptide analogue of bombesin, showed high selectivity and potency for the GRP receptor but was devoid of any agonist activity.8 In vitro studies have shown BIM26226 to be an effective inhibitor of GRP binding to rat pancreatic acini and AR4-2J cells (IC50, 5.5 and 3 nmol/L, respectively), with a 300-fold lower affinity for NMB receptors.8,9 BIM26226 has been examined for its in vitro binding to a total of 26 different receptors, among them all relevant brain-gut peptides; Ki values were ⬎1000 nmol/L for all nonbombesin receptors tested. In rats, BIM26226 inhibited bombesin-stimulated exocrine pancreatic secretion in vivo.10 These data,
Abbreviations used in this paper: GRP, gastrin-releasing peptide; NMB, neuromedin B. © 2001 by the American Gastroenterological Association 0016-5085/01/$35.00 doi:10.1053/gast.2001.21174
362
DEGEN ET AL.
as a whole,8 –10 show that BIM26226 is a specific GRP antagonist in vitro and in animals. The present investigation consisted of 3 separate studies. First, 8 subjects received placebo or 1 of 3 doses of BIM26226 together with graded doses of IV synthetic human GRP in a 4-period crossover fashion. The objective of this study was to show that BIM26226, administered as a continuous IV infusion, acts as a competitive antagonist of GRP-stimulated gallbladder contraction and plasma CCK release in humans; furthermore, the maximally effective dose of BIM26226 was to be established based on the inhibition of GRP-stimulated functions. Second, 8 subjects received this maximal dose of BIM26226 or saline (control) before and after ingestion of a labeled solid meal. The objective of this second study was to use the GRP-receptor antagonist as a tool to assess the physiologic effects of endogenous GRP on gastric emptying and small bowel transit time by scintigraphy. Finally, 6 subjects received a liquid meal continuously perfused into the duodenum, with or without BIM26226. The objectives of this last part of the study were to characterize the role of endogenous GRP on gallbladder contraction and plasma CCK release, independent of gastric emptying rates.
Materials and Methods Experimental Subjects Twenty-two healthy male volunteers, 21– 46 years old, were selected for the study. Volunteers were taking no medication before the study, and each subject was within 15% of his ideal body weight. Each subject had normal screening physical examination and laboratory test results, including urinalysis, complete blood count, serum chemistries, electrocardiography, and abdominal ultrasonography. Subjects had no history of significant illness or surgery. The study was approved by the Ethics Committee of the University Hospital of Basel. Written informed consent was obtained from each subject. All studies were conducted in the morning after an overnight (12-hour) fast.
Materials Synthetic human GRP (pyrogen-free and sterilized for IV use) was purchased from Novabiochem (Laufelfingen, Switzerland). Ensure was purchased from Abbott (Cham, Switzerland). BIM26226 (mol wt, 1075) was supplied by Ipsen International Ltd. (London, England) as lyophilized powder (chemical purity, 97%). An infusion rate of BIM26226 of 500 g 䡠 kg⫺1 䡠 h⫺1 induced 10- and 100-fold lower concentrations, respectively. Fifteen minutes after cessation of the infusion, the compound was no longer detectable, indicating a half-life of ⬍2 minutes. Based on plasma concentrations induced by the highest dose of BIM26226, it can be estimated that a ratio of agonist to antagonist at the receptor site in the
GASTROENTEROLOGY Vol. 120, No. 2
range of 1/1000 is necessary to inhibit the function by more than 97%.10
Effect of BIM26226 on Exogenous GRPStimulated Gallbladder Contraction and Plasma CCK Release Eight subjects participated in a 4-period, single-blind, placebo-controlled crossover study. On 4 days, separated by at least 3 days, subjects received either an IV infusion of saline (control) or 1 of 3 doses of BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1, equal to 4.56, 46.5, or 465 nmol 䡠 kg⫺1 䡠 h⫺1) for the duration of the experiment. Before the infusions began, subjects had to swallow a gastric tube that had been positioned under fluoroscopic control.11 Gastric acid secretion was aspirated to avoid endogenous stimulation of gallbladder contraction and CCK release through duodenal acid load. Furthermore, 3 indwelling catheters were placed in the arm veins, 2 for infusions and 1 for drawing blood at regular intervals to measure CCK plasma levels. Sixty minutes after the beginning of the BIM26226 infusion, GRP was administered in graded doses (10, 30, and 90 pmol 䡠 kg⫺1 䡠 h⫺1); each dose was administered throughout a 45-minute period.3
Measurement of Gallbladder Volumes Gallbladder contraction was assessed by high-resolution, real-time sonography (ALOKA SSD650) using a 3.5MHz probe on a sector scanner. Longitudinal sonograms of the gallbladder were recorded every 15 minutes to calculate the volume, assuming that the gallbladder shape approximated a solid whose sections are elliptic with the eccentricity depending on the eccentricity at the level of the maximal transverse section. These assumptions and the mathematical formula used to calculate the volume have been described previously.12,13 The key to this methodology is accurate estimation of the gallbladder volume. However, estimations of the gallbladder are dependent on some subjectivity by the observer. Gallbladder volumes were therefore calculated without knowledge of the subjects’ treatment status, i.e., saline or BIM26226. No gallstones, wall thickening, or other pathologic conditions were identified in any subject.
Effect of BIM26226 on Gastric Emptying and Small Bowel Transit Time of a Solid Meal In random order, 8 subjects received either an IV infusion of saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) for the duration of the experiment. Gastric emptying and small bowel transit were assessed by a noninvasive scintigraphic method developed and validated in recent years.14 –17 Polystyrene Amberlite 120-IR-Plus resin pellets (average diameter, 1 mm; range, 0.5–1.8 mm) were labeled with 100 Ci of 111In-Cl . The efficiency of the labeling is ⬎98%, as judged by 3 thin-layer chromatography.14 A capsule filled with approximately 0.5 g of pellets and coated with a layer of methacrylate was given to the fasting volunteers. As expected, the capsule
February 2001
dissolved in the ileocecal region and thereafter marked ileocecal transfer.18 External radioactive markers were placed over both anterior superior iliac spines. As soon as the radiolabeled capsule passed into the small bowel, IV infusion of either placebo or BIM26226 was begun. Thirty minutes later, a breakfast of 2 scrambled eggs, a slice of whole-wheat bread, and skim milk (35% protein, 52% carbohydrate, 13% fat, 219 kcal) was eaten within 10 minutes. The scrambled eggs were mixed and cooked with 1 mCi of 99mTc-labeled Amberlite 410 resin pellets (average diameter, 1 mm) to a firm consistency, thereby providing a solid medium. Four hours after breakfast, a standardized nonradiolabeled lunch was consumed. Gamma camera imaging started immediately after completion of ingestion of the radiolabeled breakfast with a large-field-of-view gamma camera of medium energy, parallel-hole collimator. Anterior and posterior images were taken with the subject in an erect position. Energy windows were 140 keV for the 99mTc counts and 245 keV for the 111In counts (each with 99mTc and 111In). The geometric means of the counts obtained from the anterior and posterior images were calculated for each region and then corrected for radionuclide decay. The downscatter of 111In into the 99mTc window was adjusted. Blood was drawn to determine plasma CCK concentrations at 15-minute intervals during the first 90 minutes and at 60-minute intervals thereafter.
Effect of BIM26226 on Duodenal MealStimulated Gallbladder Contraction On 2 different days and in random order, 6 subjects received either saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) at a rate of 3 mL/min to circumvent effects of BIM26226 on gastric emptying. This meal (540 mL) contained 72.1 g carbohydrates, 18.1 g fat, and 22.7 g proteins. Gallbladder contraction was measured as described above, and blood was drawn at intervals for CCK determinations.
Plasma CCK Determinations Plasma CCK concentrations were measured by radioimmunoassay as previously described.19 CCK determinations were done with antibody OAL656, which shows ⬍1% crossreactivity to sulfated and unsulfated forms of gastrin and does not bind to structurally unrelated peptides. Plasma levels of BIM26226 were determined by combined HPLC/MS (CEMAF SA, Poitiers, France).
Calculations and Statistical Analysis In the first study, with increasing doses of BIM26226, the effects of gallbladder contraction and plasma CCK concentrations were calculated as area under the curve and used for subsequent comparative statistics. Meal-stimulated gallbladder emptying data were fitted using Gauss–Newton nonlinear least-square regression.20 The half-emptying time (t50%) was defined as the period at which 50% emptied from the gallbladder. Gastric emptying was assessed by the gastric lag time and postlag emptying rate. The gastric lag time (minutes) is the time required for 10% of the radiolabeled contents to empty from the stomach.16 The gastric postlag emptying rate
EFFECTS OF GRP–RECEPTOR BLOCKADE
363
(%/min) is characterized as the slope estimated by the linear regression analysis of the data points from the first point beyond the lag time until 90% of the radiolabeled contents has emptied from the stomach.14 Small bowel transit time (minutes) was appraised by subtracting the time required for 10% of isotope to empty from the stomach from the time it took to enter the colon.15 The data were analyzed with an analysis of variance (ANOVA) model for either a 4- or a 2-period crossover design as appropriate. The normality and homogeneity of variance assumptions of the ANOVA model were tested by the Kolmogoroff–Smirnoff test and Levine test, respectively. In the case of significant differences, data were subsequently analyzed by a nonparametric test (Friedman). Differences between treatment groups were considered significant at P ⬍ 0.05 (2-tailed test).
Results Effect of BIM26226 on Exogenous GRPStimulated Gallbladder Contraction and Plasma CCK Release To determine whether BIM26226 is an effective antagonist of exogenous GRP, subjects received IV saline (control) or 1 of 3 doses of BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiments. In the control study (saline), infusion of graded doses of human GRP produced a dose-dependent gallbladder contraction with a maximum obtained at 90 pmol 䡠 kg⫺1 䡠 h⫺1 (Figure 1A). BIM26226 inhibited GRP-induced gallbladder contraction in a dose-dependent manner. Compared with placebo, all 3 doses of BIM26226 significantly (P ⬍ 0.05) inhibited GRP-stimulated gallbladder contraction. The highest dose of BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) completely abolished gallbladder contraction, and the initial volumes were between 23.0 ⫾ 2.8 and 23.7 ⫾ 1.6 mL, respectively, and did not differ among the 4 treatments (P ⬎ 0.05). The changes in plasma CCK levels in response to GRP infusion and the ability of BIM26226 to affect this increase are shown in Figure 1B. In the control study (saline), infusion of GRP produced a dose-dependent increase up to supraphysiologic plasma CCK levels. All 3 doses of BIM26226 completely inhibited CCK release (P ⬍ 0.05). Even the lowest dose (5 g 䡠 kg⫺1 䡠 h⫺1), which only partially antagonized gallbladder contraction, was able to keep plasma CCK at a basal level during maximal GRP stimulation. Effects of BIM26226 on Gastric Emptying and Small Bowel Transit Time of a Solid Meal The effects of BIM26226 on gastric emptying rates of a solid meal are shown in Figure 2A and Table 1.
364
DEGEN ET AL.
GASTROENTEROLOGY Vol. 120, No. 2
䡠 360 min after BIM26226 treatment, which was not different from control studies. Small bowel transit time amounted to 262 ⫾ 20 minutes in the control study (saline), whereas concomitant infusion of BIM26226 significantly accelerated this period to 153 ⫾ 41 minutes (Table 1). Effect of BIM26226 on Duodenal MealStimulated Gallbladder Contraction Intraduodenal perfusion of the test meal with saline (control) was followed by a time-dependent contraction of the gallbladder (Figure 3A) with a halfemptying time of 35 ⫾ 4 minutes. BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) induced significant inhibition of gallbladder contraction with a half-emptying time of 64 ⫾ 8 min-
Figure 1. Potency of BIM26226 as a GRP-receptor antagonist. (A ) Effect of saline (control) or BIM26226 (5, 50, or 500 g 䡠 kg⫺1 䡠 h⫺1) on gallbladder contraction, induced by infusion of graded doses of human GRP (10, 30, or 90 pmol 䡠 kg⫺1 䡠 h⫺1). Gallbladder volumes are expressed as percentages of the basal volume. Values are means ⫾ SEM of 8 subjects. (B) Plasma CCK concentrations in the same studies. Results are expressed as means ⫾ SEM of 8 subjects. In a dose-dependent fashion, BIM26226 reduced GRP-stimulated gallbladder contraction and inhibited CCK release. The highest dose (500 g 䡠 kg⫺1 䡠 h⫺1) completely inhibited GRP-stimulated functions.
During the control study, the meal continuously emptied, with only 14% ⫾ 9% remaining in the stomach 6 hours after ingestion of the meal. The lag time was 41 ⫾ 6 minutes, and the gastric emptying rate was 0.26% ⫾ 0.04%/min, respectively. With BIM26226 infusion, 60% ⫾ 5% of the meal remained in the stomach after 6 hours. All parameters of gastric emptying of the test meal were significantly (P ⬍ 0.05) inhibited; gastric lag time was 114 ⫾ 29 minutes, and the gastric emptying rate was calculated as 0.11 ⫾ 0.02%/min (Table 1). In control studies, plasma CCK levels increased immediately after ingestion of the meal (Figure 2B). With BIM26226 infusion, the increase in plasma CCK concentrations was initially delayed and less pronounced but lasted longer. The area under the curve was 210 ⫾ 73 pmol 䡠 360 min with saline (control) and 191 ⫾ 44 pmol
Figure 2. Effects of BIM26226 on gastric emptying and plasma CCK release. (A ) Effect of BIM26226 on gastric emptying of a solid meal assessed by scintigraphy. Eight subjects received IV saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiment. Results are expressed as the percentage of meal contents remaining in the stomach. Values are means ⫾ SEM. BIM26226 significantly inhibited the lag phase as well as the emptying rate of the solid meal (P ⬍ 0.05) compared with saline (control). (B) Plasma CCK concentrations in the same studies. Results are expressed as means ⫾ SEM of 8 subjects. With BIM26226, integrated plasma CCK release did not differ from control values.
February 2001
EFFECTS OF GRP–RECEPTOR BLOCKADE
Table 1. Effects of BIM26226 on Postprandial Intestinal Motility Saline (control) Gastric lag time (min) Gastric emptying rate (%/min) Small bowel transit time (min) Gallbladder contraction t1/2 (min) aP
41 ⫾ 6
BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) 114 ⫾ 29a
0.26 ⫾ 0.04
0.11 ⫾ 0.02a
262 ⫾ 20
153 ⫾ 41a
35 ⫾ 4
64 ⫾ 8a
⬍ 0.05 vs. control.
utes (P ⬍ 0.05). In the control study, stimulated plasma CCK levels immediately increased 3-fold over basal levels and remained nearly constant for the duration of the experiment. There was no significant difference in the area under the plasma CCK curve between saline (control) and BIM26226-treated groups (Figure 3B).
365
Several mechanisms whereby bombesin-like peptides can influence intestinal motility are possible: (1) GRP could directly activate its receptor on smooth muscle cells; (2) GRP could activate cholinergic mechanisms (this explanation is based on the observation that GRP is found mainly in mucosal, submucosal, and myenteric plexuses21–23); (3) bombesin-like peptides are known to release a variety of other regulatory factors, such as CCK, somatostatin, glucagon-like peptide 1, gastrin, motilin, or neurotensin, which in turn could mediate smooth muscle contraction24,25; and (4) it might be possible that the effects are of central (vagal) origin because bombesinlike immunoreactivity has been demonstrated in the human brain. Whether BIM26226 crosses the blood– brain barrier is not known, but GRP, CCK, peptide YY,
Adverse Effects No adverse effects were observed or symptoms reported during or after BIM26226 administration.
Discussion In this study we show that BIM26226 is a potent GRP receptor antagonist in humans. BIM26226 dosedependently reduced gallbladder contraction stimulated by exogenous GRP and inhibited plasma CCK release. In additional unpublished studies, we have also established that BIM26226 acts as a competitive antagonist for a variety of exogenous GRP-stimulated gut functions, including gastric acid secretion, exocrine pancreatic secretory responses, and hormone release (gastrin, pancreatic polypeptide), thus providing experimental evidence for the potency and efficacy of the compound to antagonize the function of GRP-preferring bombesin receptors in humans. This potency has previously been shown in vitro and in animals.8,9 Using BIM26226 as a specific tool to assess the consequences of GRP-receptor blockade during endogenous GRP stimulation by oral meal ingestion, it was shown to (1) dramatically decrease gastric emptying, (2) inhibit gallbladder contraction, and (3) accelerate small bowel transit time in humans. Postprandial plasma CCK release from the duodenal mucosa was reduced most likely because of delayed gastric emptying. However, BIM26226 decreased gallbladder contraction without exerting any effect on plasma CCK release. These results represent the first human data on the physiologic role that endogenous GRP exerts on gastrointestinal motility using a potent and specific GRP-receptor antagonist.
Figure 3. Effects of BIM26226 on gallbladder contraction and plasma CCK release. (A ) Effect of BIM26226 on gallbladder contraction induced by duodenal perfusion of a liquid meal. Six subjects received IV saline (control) or BIM26226 (500 g 䡠 kg⫺1 䡠 h⫺1) throughout the experiment. Results are expressed as the percentage of basal (premeal) volumes. Values are means ⫾ SEM. BIM26226 significantly (P ⬎ 0.05) reduced gallbladder contraction compared with saline (control). (B) Effect of BIM26226 on plasma CCK concentrations during the same studies. Results are expressed as means ⫾ SEM of 6 subjects. Plasma CCK levels, with and without BIM26226, did not significantly differ, indicating that the effect of BIM26226 on gallbladder contraction was not mediated by circulating CCK.
366
DEGEN ET AL.
and gastrin did not cross the blood– brain barrier in an in vitro model using porcine brain capillary endothelial cells ( Ju¨rgen Drewe and Christoph Beglinger, unpublished observations). Which of these different mechanisms are involved in mediating the effects on gastric emptying, small bowel transit, and gallbladder contraction can be answered only partially by the results of the present study. Only circulating hormones could be analyzed and interpreted in this context; all other mechanisms exert their effects in a paracrine or neuroendocrine manner. Their GRP receptors might be blocked by BIM26226, but there is no parameter that would allow the quantification of these effects in vivo, and we must therefore rely on in vitro studies or animal experiments to test these hypotheses. The extent of this inhibiting effect of GRP-receptor blockade on gastric emptying was unexpected from previous in vivo studies in animals or humans. It has been shown that infusion of bombesin inhibits gastric emptying in humans.5 More recently, bombesin and GRP (10⫺9 to 10⫺6 mol/L) have been shown to stimulate isolated muscle strips of the human stomach in vitro; bombesin stimulated both circular and longitudinal preparations from all regions of the stomach, and these effects were unaltered by atropine or tetrodotoxin.26,27 Additional potential mechanisms of action are based on in vitro data from animals. GRP has been isolated from canine antral muscle, but in the rat, GRP was found predominantly in the mucosal plexus of the gastric fundus and antrum.23 Bombesin increased the frequency and amplitude of contractions in smooth muscle strips from canine stomach; the action was myogenic in the circular muscle but was blocked by atropine in the longitudinal muscle.28 The contraction of rat stomach strips induced by bombesin-like peptides could not be antagonized by a variety of neural inhibitors.29 On the other hand, GRP nerves were found in the porcine antrum, and these nerves mediate vagal signals that are responsible for motor functions and hormone release.30 One of the hormonal candidates for a regulatory function of GRP stimulation is CCK, which mediates various gastrointestinal motor functions, including inhibition of gastric emptying in dogs, rats, and humans.31 In the present study, CCK plasma levels were unchanged during intraduodenal meal stimulation with BIM26226, implying that CCK cannot account for the delay in gastric emptying. The same is true for gastrin, which binds with high affinity to CCK-B receptors; again, postprandial plasma gastrin levels were unchanged with and without BIM26226 (data not shown). Circulating somatostatin, glucagon-like peptide 1, motilin, and neurotensin, all
GASTROENTEROLOGY Vol. 120, No. 2
further potential inhibitors of gastric emptying, were not analyzed in the present study. In the isolated porcine stomach, however, vagus-stimulated somatostatin secretion in the antrum was strongly inhibited by a GRPreceptor antagonist.32 Furthermore, results obtained from preparations of rat stomach indicated that GRPstimulated antral somatostatin release, which was sensitive to neural inhibitors, could mediate the effects on gastric emptying.33 In the context of gastric acid secretion studies, we determined the quantity of local somatostatin messenger RNA in biopsy specimens from the gastric antrum and corpus by Northern blotting, but we were not able to show a significant change with BIM26226 administration ( John Calam and Pius Hildebrand, unpublished results). Which pathways mediate the effects on gastric motility remains to be determined, but it is clear that blockade of endogenously stimulated GRP receptors led to an almost complete cessation of gastric emptying. This is in keeping with findings in dogs, in which administration of a monoclonal antibombesin antibody led to increased gastric retention.34 The data imply a major regulatory function of GRP in gastric emptying in humans. There is ample evidence that bombesin-like peptides exert some effects on human small bowel motility. Isolated smooth muscle cells from longitudinal and circular smooth muscle layers of the human jejunum have been shown to be contracted by bombesin-like peptides in picomolar concentrations; the contractile response was not affected by muscarinic, opioid, CCK, or serotonin antagonists, indicating that bombesin-like peptides directly cause intestinal smooth muscle contraction.26 In vitro studies with human duodenal muscle strips showed that longitudinally cut strips were excitable by GRP, whereas circularly cut strips showed no response.35 However, high doses of bombesin have been shown to inhibit basal mechanical activity of the human duodenum and jejunum.36 However, the response of endogenous GRP on postprandial small bowel transit time can be estimated only by using a GRP-receptor antagonist. The acceleration of small bowel transit time by BIM26226 indicated that, among the regulators of motility, the postprandial effect of GRP alone seems to prolong small bowel transit time. Both exogenous bombesin and GRP cause contraction of the human gallbladder.3,37 We and others have previously suggested that this effect is largely mediated through CCK by showing that the specific CCK-A– receptor antagonist loxiglumide completely inhibits GRP-stimulated gallbladder contraction.38 In addition, GRP receptors have been identified on human gallblad-
February 2001
der tissue using immunohistochemical techniques.39 On the other hand, in most species bombesin has a weak direct stimulatory effect on gallbladder contraction in vitro. The present results suggest that both exogenously and endogenously GRP-stimulated gallbladder contraction is at least partially independent from circulating plasma CCK. First, gallbladder contraction stimulated by increasing doses of GRP was only partially inhibited by the lower 2 doses of BIM26226, which in turn completely abolished plasma CCK release (Figure 1). Second, gallbladder contraction was significantly reduced during duodenal meal stimulation with concomitant infusion of the GRP-receptor antagonist, whereas CCK plasma levels remained unaffected during the same time frame. Taken together, these data suggest that GRP is a physiologic, peptidergic regulator of gallbladder motility. However, gallbladder volumes, as measured by ultrasonography, are the result of pressure gradients produced by the detrusor muscle of the gallbladder itself, the sphincter of Oddi, and hepatic bile production. Thus, in vitro studies with isolated human smooth muscle tissues and smooth muscle cells must be undertaken to gain insight into the regulatory mechanisms involved. Several studies have suggested a regulatory role for GRP in controlling CCK release because IV infusion of bombesin-like peptides induces increased CCK plasma levels.3,22 In this study with continuous duodenal perfusion of a liquid meal, we did not observe a difference in plasma CCK secretion with or without BIM26226. Therefore, CCK release after IV application of bombesinlike peptides represents a pharmacologic rather than a physiologic event. Unfortunately, neural release of CCK in response to GRP, either in the central nervous system or in the periphery, has not yet been studied. In conclusion, we show that BIM26226 is a potent antagonist of GRP based on the inhibition of exogenous GRP-stimulated gallbladder contraction and plasma CCK release. Additionally, our studies show the usefulness of BIM26226 as a probe of the physiologic activity of endogenous GRP, revealing (1) a major role of this peptide in gastric emptying, (2) a significant role in small bowel transit and in gallbladder contraction, and (3) a possibly limited role of GRP in regulation of plasma CCK secretion.
EFFECTS OF GRP–RECEPTOR BLOCKADE
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17. 18.
19.
References 1. Sunday ME, Kaplan LM, Motoyama E, Chin WW, Spindel ER. Biology of disease. Gastrin-releasing peptide (mammalian bombesin) gene expression in health and disease. Lab Invest 1988; 59:5–24. 2. Walsh JH. Bombesin-like peptides. In: Schultz SG, Makhlouf GM, Rauner BB, eds. Handbook of physiology. Section 6. The gastro-
20. 21. 22.
367
intestinal system. Volume 2. Neural and endocrine biology. Bethesda, MD: American Physiological Society, 1989:587– 610. Hildebrand P, Werth B, Beglinger C, Delco F, Jansen JB, Lamers CB, Gyr K. Human gastrin-releasing peptide: biological potency in humans. Regul Pept 1991;36:423– 433. Lundell L, Cantor P, Sjovall M, Rehfeld JF, Olbe L. Factors influencing the release of cholecystokinin induced by gastrin-releasing peptide in man. Scand J Gastroenterol 1991;26:544 –550. Scarpignato C, Micali B, Vitulo F, Zimbaro G, Bertaccini G. Inhibition of gastric emptying by bombesin in man. Digestion 1982; 23:128 –131. Jensen RT, Coy DH. Progress in the development of potent bombesin receptor antagonists. Trends Pharmacol Sci 1991;12:13– 19. Heinz-Erian P, Coy DH, Tamura M, Jones SW, Gardner JD, Jensen RT. [D-Phe12]bombesin analogues: a new class of bombesin receptor antagonists. Am J Physiol 1987;252:G439 –G442. Coy DH, Mungan Z, Rossowski WJ, Cheng BL, Lin JT, Mrozinski JE, Jensen RT. Development of a potent bombesin receptor antagonist with prolonged in vivo inhibitory activity on bombesinstimulated amylase and protein release in the rat. Peptides 1992;13:775–781. Dietrich JB, Hildebrand P, Jeker LB, Pansky A, Eberle AN, Beglinger C. Effects of BIM26226, a potent and specific bombesin receptor antagonist, on amylase release and binding of bombesin-like peptides to AR4-2J cells. Regul Pept 1994;53: 165–173. Varga G, Reidelberger RD, Liehr RM, Bussjaeger LJ, Coy DH, Solomon TE. Effects of potent bombesin antagonist on exocrine pancreatic secretion in rats. Peptides 1991;12:493– 497. Beglinger C, Fried M, Whitehouse I, Jansen JB, Lamers CB, Gyr K. Pancreatic enzyme response to a liquid meal and to hormonal stimulation. Correlation with plasma secretin and cholecystokinin levels. J Clin Invest 1985;75:1471–1476. Everson GT, Braverman DZ, Johnson ML, Kern F Jr. A critical evaluation of real-time ultrasonography for the study of gallbladder volume and contraction. Gastroenterology 1980;79:40 – 46. Beglinger C, Hildebrand P, Adler G, Werth B, Luo H, Delco F, Gyr K. Postprandial control of gallbladder contraction and exocrine pancreatic secretion in man. Eur J Clin Invest 1992;22:827– 834. Camilleri M, Colemont LJ, Phillips SF, Brown ML, Thomforde GM, Chapman N, Zinsmeister AR. Human gastric emptying and colonic filling of solids characterized by a new method. Am J Physiol 1989;257:G284 –G290. Greydanus MP, Camilleri M, Colemont LJ, Phillips SF, Brown ML, Thomforde GM. Ileocolonic transfer of solid chyme in small intestinal neuropathies and myopathies. Gastroenterology 1990; 99:158 –164. Camilleri M, Zinsmeister AR, Greydanus MP, Brown ML, Proano M. Towards a less costly but accurate test of gastric emptying and small bowel transit. Dig Dis Sci 1991;36:609 – 615. Degen LP, Phillips SF. Variability of gastrointestinal transit in healthy women and men. Gut 1996;39:299 –305. Proano M, Camilleri M, Phillips SF, Thomforde GM, Brown ML, Tucker RL. Unprepared human colon does not discriminate between solids and liquids. Am J Physiol 1991;260:G13–G16. Hildebrand P, Ensinck JW, Ketterer S, Delco F, Mossi S, Bangerter U, Beglinger C. Effect of a cholecystokinin antagonist on meal-stimulated insulin and pancreatic polypeptide release in humans. J Clin Endocrinol Metab 1991;72:1123–1129. Elashoff JD, Reedy TJ, Meyer JH. Analysis of gastric emptying data. Gastroenterology 1982;83:1306 –1312. Price J, Penman E, Wass JA, Rees LH. Bombesin-like immunoreactivity in human gastrointestinal tract. Regul Pept 1984;9:1–10. Dockray GJ, Vaillant C, Walsh JH. The neuronal origin of bom-
368
23. 24.
25.
26.
27.
28.
29.
30.
31.
32.
DEGEN ET AL.
besin-like immunoreactivity in the rat gastrointestinal tract. Neuroscience 1979;4:1561–1568. Walsh JH, Wong HC, Dockray GJ. Bombesin-like peptides in mammals. Fed Proc 1979;38:2315–2319. McDonald TJ, Ghatei MA, Bloom SR, Track NS, Radziuk J, Dupre J, Mutt V. A qualitative comparison of canine plasma gastroenteropancreatic hormone response to bombesin and the porcine gastrin-releasing peptide (GRP). Regul Pept 1981;2:293–304. Ghatei MA, Jung RT, Stevenson JC, Hillyard CJ, Adrian TE, Lee YC, Christofides ND, Sarson DL, Mashiter K, MacIntyre I, Bloom SR. Bombesin: action on gut hormones and calcium in man. J Clin Endocrinol Metab 1982;54:980 –985. Pogrzeba B, Mandrek K, Ludtke FE, Lepsien G, Golenhofen K. Contractile action of gastrin-releasing peptide on isolated preparations of human gastroduodenal muscle. Dig Dis 1991;9:354 – 359. Ludtke FE, Pogrzeba B, Mandrek K, Lepsien G, Neufang T, Golenhofen K. Bombesin—the most stimulating peptide of human gastric smooth muscle. Dig Dis 1991;9:371–381. Mayer EA, Elashoff J, Walsh JH. Characterization of bombesin effects on canine gastric muscle. Am J Physiol 1982;243:G141– G147. Girard F, Bachelard H, St-Pierre S, Rioux F. The contractile effect of bombesin, gastrin releasing peptide and various fragments in the rat stomach strip. Eur J Pharmacol 1984;102:489 – 497. Holst JJ, Knuhtsen S, Orskov C, Skak-Nielsen T, Poulsen SS, Nielsen OV. GRP-producing nerves control antral somatostatin and gastrin secretion in pigs. Am J Physiol 1987;253:G767– G774. Fried M, Erlacher U, Schwizer W, Lo¨chner C, Koerfer J, Beglinger C, Jansen JB, Lamers CB, Harder F, Bischof-Delaloye A, Stalder GA, Rovati L. Role of cholecystokinin in the regulation of gastric emptying and pancreatic enzyme secretion in humans. Studies with the cholecystokinin-receptor antagonist loxiglumide. Gastroenterology 1991;101:503–511. Holst JJ, Harling H, Messell T, Coy DH. Identification of the neurotransmitter/neuromodulator functions of the neuropeptide gastrin-releasing peptide in the porcine antrum, using the antagonist (Leu13-psi-CH2 NH-Leu14)-bombesin. Scand J Gastroenterol 1990;25:89 –96.
GASTROENTEROLOGY Vol. 120, No. 2
33. Martindale R, Kauffman GL, Levin S, Walsh JH, Yamada T. Differential regulation of gastrin and somatostatin secretion from isolated perfused rat stomachs. Gastroenterology 1982;83: 240 –244. 34. Kovacs TO, Cuttitta F, Maxwell V, Hamel D, Walsh JH. The effects of a monoclonal antibody to bombesin on gastric function in dogs (abstr). Gastroenterology 1986;90:1502. 35. Micheletti R, Grider JR, Makhlouf GM. Identification of bombesin receptors on isolated muscle cells from human intestine. Regul Pept 1988;21:219 –226. 36. Corazziari E, Torsoli A, Melchiorri P, Delle Fave G, Habib FI. Effect of bombesin on the mechanical activity of the human duodenum and jejunum. Rendic Gastroenterol 1974;6:55–56. 37. Corazziari E, Torsoli A, Melchiorri P, Delle Fave G. Effect of bombesin on human gallbladder emptying. Rendic Gastroenterol 1974;6:52–55. 38. Hildebrand P, Drewe J, Luo H, Ketterer S, Gyr K, Beglinger C. Role of cholecystokinin in mediating GRP-stimulated gastric, biliary and pancreatic functions in man. Regul Pept 1992;22;41:119 – 129. 39. Tsutsumi Y, Nagura H, Watanabe K, Yanaihara N. Methods in laboratory investigation. Immunoreactivity of N-terminal fragment of gastrin-releasing peptide as histochemical marker for pancreatobiliary duct-type cells. Lab Invest 1984;50:94 –100.
Received June 12, 2000. Accepted September 22, 2000. Address requests for reprints to: Pius Hildebrand, M.D., Department of Research, University Hospital Basel, Hebelstrasse 20, CH-4031 Switzerland; e-mail: [email protected]; fax: (41) 61-2652350. Supported in part by grant no. 3200-040604.94 from the Swiss National Science Foundation and by a research grant from Ipsen International Ltd. (London, England). P. Hildebrand received funding support from the Sandoz Foundation. The authors thank Gerdien Gamboni for expert technical assistance and Kathleen A. Bucher for editorial assistance.