Unlike in rodents, CCK has not been established as a physiological regulator in avian exocrine pancreatic secretion. In the isolated duck pancreatic acini, 1 nM CCK was required for stimulation of amylase secretion, maximal effect being achieved at 10 nM; picomolar CCK was without effect. Vasoactive intestinal peptide (VIP)/pituitary adenylate cyclase activating peptide (PACAP) receptor (VPAC) agonists PACAP-38 and PACAP-27 (10-12-10-7 M) alone had no effect, but made picomolar CCK effective. VPAC agonist VIP 10-10-10-7 M stimulated amylase secretion marginally, but made CCK 10-12-10-10 M effective also. PACAP-27 and VIP both shifted the maximal CCK concentration from 10-8 to 10-9 M. This sensitizing effect was mimicked by forskolin. CCK dose dependently induced intracellular Ca2+ concentration ([Ca2+]i) oscillations. PACAP-38 (1 nM), PACAP-27 (1 nM), VIP (10 nM), or forskolin (10 μM) alone did not stimulate [Ca2+]i increase, neither did they modulate CCK (1 nM)-induced oscillations; but when they were added to cells simultaneously exposed to subthreshold CCK (10 pM), calcium spikes emerged. Amylase secretion induced by the simultaneous presence of 10 pM CCK and VPAC agonists was completely blocked by removing extracellular calcium, but the protein kinase C inhibitor staurosporine (1 μM) was without effect. CCK (10 nM)-induced secretion was inhibited by CCK1 receptor antagonist FK480 (1 μM). Gastrin from 10-12 to 10-6 M did not stimulate amylase secretion nor did it (100 nM) induce [Ca2+]i increase. The above data suggest that duck pancreatic acini possess both CCK1 and VPAC receptors; simultaneous activation of both is required for each to play a physiological role.
- amylase secretion
- calcium oscillation
- pancreatic acinar cell
in contrast to rodents, the physiological regulators for avian exocrine pancreatic secretion are rather poorly defined, especially for peptide regulation. On the one hand, the endogenous levels of the major exocrine pancreatic secretion regulators in mammals (including rodents), CCK and gastrin, seem to be similar in birds and in mammals. In the chicken, basal gastrin concentration in the plasma was found to be 50-100 pM and after feeding increases to 300-500 pM (16). Basal plasma CCK concentrations in the chicken were 5-10 pM, and increased to 20-40 pM after feeding (17, 30). Similarly, the plasma CCK concentration during fasting was found to be 3-6 pM in geese (46). These levels are rather similar to those in mammals. In situ studies indicated that both nerve and gut peptide appeared to play a significant role in avian exocrine pancreatic secretion. In situ pancreatic juice collection studies indicated that in turkey, vasoactive intestinal peptide (VIP) was very potent to stimulate exocrine pancreatic secretion (8). Direct intestinal HCl perfusion in turkey induced significant pancreatic protein and juice secretion (36). Feeding in chicken induced significant pancreatic juice secretion (29). Vagal stimulation induced significant pancreatic juice flow and protein output in anesthetized chicken and duck; protein secretion in the duck was particularly marked. CCK injection also induced increase in juice flow and protein output in chicken, and protein output in duck was particularly marked (21).
In vitro studies do not seem to confirm a major role for CCK in avian exocrine pancreatic secretion. In the isolated chicken pancreatic acini, CCK induced amylase secretion with a maximal concentration of 10-8 M (35), and this stimulation was inhibited by the CCK1 receptor antagonist devazepide (15). The maximal stimulatory concentration of CCK in isolated duck pancreatic acini was also found to be 10-8 M (19). This concentration is two orders of magnitude higher than that for rodents, which is 10-10 M (3, 34, 35). Taking into account the fact that CCK concentrations in the plasma are likely to be in the same range in birds and in rodents (17, 30, 46), whether CCK plays a major physiological role in avian exocrine secretion has become rather doubtful (14, 35).
For CCK not to be a physiological mediator of exocrine pancreatic secretion would signify a drastic diversion in digestive physiology. To further clarify this question, we have therefore in this work examined whether it is possible that other peptides play a modulating role in CCK stimulation of avian exocrine pancreatic secretion. The duck was used because we previously found that in Peking duck pancreatic acini, CCK dose-response curve for amylase secretion was similar to that in chicken (19, 35). In addition, very little is known about the digestive physiology of this important economic animal. It was indeed found in this work that the cAMP-producing neurotransmitters PACAP-38, PACAP-27 and VIP significantly enhanced CCK-induced amylase secretion from isolated duck pancreatic acini, leading to a leftward expansion of the CCK dose-response curve, making the putative physiological CCK concentrations potent enough to induce sufficient amylase secretion. Furthermore, it was found that at higher (or pharmacological) CCK concentrations, this effect of PACAP and VIP was at a site downstream of calcium oscillations; but at lower (more physiological) CCK concentrations, the two converge to induce calcium oscillations. It was further established that the duck CCK receptor is likely to resemble the rodent CCK1 type, and the duck receptor for VIP/PACAP sensitization of CCK response is likely to be of the VPAC type. The simultaneous activation of both VPAC and CCK receptors is required for a physiological response.
Materials. CCK-8 (referred to as CCK in the text), PACAP-38, PACAP-27, VIP, forskolin, gastrin-1, α-amylase, and amylose azure were all purchased from Sigma-Aldrich (St. Louis, MO). Collagenase P was from Boehringer Mannheim (Mannheim, Germany) and fura 2-AM was from Molecular Probes (Eugene, OR). Cell-Tak was bought from Becton Dickinson (Bedford, MA). The CCK1 receptor antagonist FK480 (37) was provided by Fujisawa Pharmaceutical (Osaka, Japan).
Isolation of pancreatic acini. Peking ducks (Anas platynchos, domestica Linnaeus) with a body weight from 150 to 350 g (note that body weight up to 1 kg does not change the responses of the isolated pancreatic acini) were maintained in a 12:12-h light-dark cycle, with commercial duck food and tap water fed ad libitum, and killed by decapitation. The pancreas was excised and acini were prepared by collagenase digestion with a protocol similar to that for rat pancreatic acini preparation (4). Briefly, the pancreas was infiltrated with collagenase P-containing (5 mg in 5 ml) standard buffer. The pancreas was digested in this buffer for three sequential 5-min periods (with fresh collagenase each time) in a shaking water bath (120 cycles/min, 37°C). The digested tissue was then gently pipetted, filtered through a nylon mesh (150 mesh), and centrifuged in buffer containing bovine serum albumin 4% and washed three times. The acini so prepared were either left to recover for 30 min or were then loaded with fura 2-AM, both in a shaking water bath (37°C, 50 cycles per min). Rodent (male Sprague-Dawley rat 250-350 g, male Balb/c mouse 20-25 g, and male guinea pig 250-280 g) pancreatic acini were isolated similarly as before (4). The standard buffer used in this work had the following composition (in mM): 118 NaCl, 4.7 KCl, 2.5 CaCl2, 1.13 MgCl2, 1.0 NaH2PO42, 5.5 d-glucose, 10 HEPES, 2.0 l-glutamine, and 2 g/l bovine serum albumin, 2% minimum essential medium amino acids mixture (GIBCO BRL), 0.1 g/l soybean trypsin inhibitor, pH adjusted to 7.4 with 4 M NaOH and oxygenated with pure O2.
Amylase secretion. Recovered pancreatic acini were stimulated with indicated chemicals for 30 min in standard buffer in a shaking water bath (37°C, 50 cycles/min). Amylase released into the buffer was expressed as a percentage of total present before stimulation. Amylase was assayed with amylose azure as a substrate as described previously (31).
Measurement of intracellular calcium concentration. Isolated duck pancreatic acini were loaded with fura 2-AM (final concentration 10 μM) for 45 min, in a shaking water bath (37°C, 50 cycles/min). Fura 2-loaded acini were then attached to Sykes-Moore perfusion chambers, the bottom coverslips of which were previously treated with Cell-Tak. Acini attachment was allowed for 15 min, before the chambers were placed on the platform of an inverted fluorescence microscope (Olympus IX 70) coupled to an intracellular Ca2+ concentration ([Ca2+]i) measurement system (PTI). The attached acini were perfused constantly with perfusion buffer at 1 ml/min and stimulating chemicals were introduced by a change of the perfusion buffer, with the dead time corrected for in all recordings shown. The perfusion buffer used was the same as the standard buffer, with bovine serum albumin, minimum essential medium amino acids mixture, and soybean trypsin inhibitor omitted. The fura 2-loaded acinar cells were excited alternately at 340 and 380 nm (with a monochromater slit width set at 2 nm), and emission was measured at 510 nm with a band-pass filter setup (D510/40m). Fluorescence ratios of F340/F380 were then taken as indicative of [Ca2+]i changes and plotted against time as before (4, 5).
Statistical analysis. Data were presented as means ± SE. Student's paired t-test was used for comparison between test and control group, with a P < 0.05 taken as statistically significant. For multiple comparisons in Fig. 8, one-way ANOVA analysis was used.
Amylase secretion in isolated duck pancreatic acini induced by CCK and further sensitization by PACAP-27, PACAP-38, forskolin, and VIP. Figure 1 illustrates typical dose-response curves for amylase secretion in isolated rodent pancreatic acini. It is clear that picomolar CCK induced significant amylase secretion in all isolated rodent (rat, mouse, and guinea pig) pancreatic acini, and maximal-stimulating CCK concentration was 10-10 M.
As in rodents, CCK induced a concentration-dependent increase in amylase secretion from isolated duck pancreatic acini (Fig. 2A). But compared with rodent dose-response curves, the duck dose-response curve is rather narrow and the maximal concentration right-shifted by two orders of magnitude to 10-8 M (Fig. 2A) instead of the 10-10 M commonly observed in rodents (3, 34, 35, and Fig. 1). In control experiments, it was found that PACAP-27, PACAP-38 alone did not stimulate amylase secretion at all from 10-12 to 10-7 M. Forskolin alone from 10-10 to 10-5 M did not produce any amylase secretion either. VIP alone from 10-12 to 10-11 M had no effect on amylase secretion, and concentrations higher (10-10-10-7 M) induced only a minor increase (Fig. 3). However, addition of PACAP-27 (1 nM), forskolin (10 μM), or VIP (10 nM) resulted in a CCK dose-response curve shifted to the left by one order of magnitude, with the maximal concentration shifted from 10-8 M to 10-9 M (Fig. 2, B, D, E). CCK alone at 10-10 M induced no increase in amylase secretion, and a CCK concentration of 10-9 M was required to produce any significant increase in amylase secretion (Fig. 2A). But with the addition of PACAP-27 (1 nM), PACAP-38 (1 nM), forskolin (10 μM), or VIP (10 nM), this narrow range of stimulating concentrations (10-9-10-7 M) of CCK expanded considerably to encompass a range from 10-12 to 10-7 M. PACAP-38 was similar to VIP, PACAP-27, and forskolin in that it resulted in a leftward expansion of the CCK dose-response curve, but it also differed in that the maximal-stimulating CCK concentration remained at 10-8 M (Fig. 2C). In all cases, statistically significant amylase secretion was observed at 1 pM CCK, making CCK three orders of magnitude more potent than in the absence of PACAP-27, PACAP-38, forskolin, and VIP (Fig. 2, B-E).
Calcium oscillations induced by different concentrations of CCK, and lack of a modulating role by PACAP-27, PACAP-38, forskolin, and VIP in higher CCK concentration-induced calcium oscillations. In the freshly isolated duck pancreatic acini, CCK at 10 pM did not induce any increase in [Ca2+]i. Lower frequency calcium oscillations were observed when CCK concentration was raised to 100 pM. Whereas 1 nM CCK induced regular high-frequency calcium oscillations, 10 nM induced a plateau increase (Fig. 4). Therefore, as in rodent pancreatic acinar cells, duck pancreatic acinar cells also showed a concentration-dependent transformation in the pattern of calcium oscillations, the threshold concentration being 100 pM and standard oscillation-inducing concentration being 1 nM. It is important to bear in mind that CCK at 10 nM (a concentration inducing maximal amylase secretion in the duck) induced plateau calcium increase. This is consistent with the rodent situation, where CCK at 100 pM induced plateau calcium increase, resulting in maximal secretion of amylase (3, 20, 26).
Because CCK-induced amylase secretion was significantly sensitized by the addition of PACAP-38, PACAP-27, forskolin, and VIP (Fig. 2), the question as to whether CCK-induced calcium oscillations were also modulated by PACAP-38, PACAP-27, forskolin, and VIP arises naturally. We therefore examined the possible effect of these factors on CCK-induced calcium oscillations. In each of these experiments, regular calcium oscillations were induced by 1 nM CCK, then on top of CCK was added PACAP-27, PACAP-38, forskolin, and VIP. Significantly, marked enhancing effect on amylase secretion notwithstanding (Fig. 2), the addition of PACAP-27 (1 nM), PACAP-38 (1 nM), forskolin (10 μM), or VIP (10 nM) had no effect at all on CCK (1 nM)-induced calcium oscillations in the isolated duck pancreatic acini (Fig. 5). In separate experiments, it was found that PACAP-27, PACAP-38, forskolin, and VIP alone did not induce any changes in [Ca2+]i either (Fig. 6).
Sensitizing effect by PACAP-27, PACAP-38, forskolin, and VIP on intracellular calcium at lower CCK concentrations. Although no effect of the VIP/PACAP agonists was observed on 1 nM CCK-induced calcium oscillations, they did seem to have an obvious sensitizing effect on lower CCK concentrations such as 10 pM. As shown in Figs. 4 and 7A, CCK 10 pM alone did not induce any calcium increases in any of the cells examined (albeit the cells were fully functional as indicated by their response to CCK 1 nM, see Fig. 7Aa). But the simultaneous presence of PACAP-27 (1 nM), PACAP-38 (1 nM), forskolin (10 μM), or VIP (10 nM), and 10 pM CCK led to the emergence of distinct calcium spikes in a large number of cells examined, the responding numbers being 6 of 9, 5 of 9, 7 of 10, and 6 of 10 experiments, respectively (Fig. 7A). In the remainder of the cells, no calcium spikes were observed, although the cells were fully functional as confirmed by their normal response to CCK 1 nM (Fig. 7B).
Mutual dependence for amylase secretion of the VIP/PACAP and CCK systems and their independence of protein kinase C. From Fig. 7, it is obvious that calcium oscillations may be responsible for amylase secretion detected at lower CCK concentrations when VIP/PACAP were also present, as shown in Fig. 2. To further confirm a role of cytosolic calcium increase, calcium was removed from the extracellular medium (with omission of added calcium chloride, and addition of EGTA 0.5 mM). As shown in Fig. 8, this maneuver completely obliterated amylase secretion in each case (compare columns 3-6 with columns 7-10). But when 1 μM staurosporine was added, amylase secretion remained unchanged (compare columns 7-10 with columns 11-14). These experiments indicate that at lower CCK concentrations (such as 10 pM), the VIP/PACAP R193 and CCK signaling pathways converge to lead to the emergence of calcium oscillations. And these calcium oscillations are likely to be responsible for the observed amylase secretion.
Inhibition by CCK1 receptor antagonist FK480 of CCK-induced amylase secretion and lack of effect by gastrin on amylase secretion and [Ca2+]i increase. To investigate the type of CCK receptor involved, the effect of CCK1 receptor antagonist FK480 on CCK-induced amylase secretion was investigated (Fig. 9). It was found that FK480 (1 μM) inhibited markedly amylase secretion induced by 10 nM CCK. This inhibitory effect by FK480 was dose dependent, with detectable inhibition observed starting at an FK480 concentration of 10 nM (not shown). Furthermore, it was found that gastrin in the concentration range from 10-12 to 10-6 M had no effect on amylase secretion in the isolated duck pancreatic acini (Fig. 10A). Gastrin at the supraphysiological concentration of 100 nM also did not induce any increase in [Ca2+]i (Fig. 10B). These suggested that in duck pancreatic acini, CCK effects were mediated by CCK1 instead of CCK2 receptors.
Several laboratories have investigated the innervation of the avian exocrine pancreas. Mensah-Brown and Pallet (32) found that Houbara Bustard (Chlamydotis undulata) exocrine pancreas was innervated by VIP, galanin, neuropeptide Y, and by tyrosine hydroxylase positive terminals. VIP and galanin innervation of the chicken exocrine pancreas has been documented (22, 23). Vaillant et al. (42) identified VIP innervation in turkey pancreas. CGRP innervation has also been identified in the chicken (9). The fact that feeding induces release of both CCK (17, 30) and gastrin (16) would probably also indicate their positive role in avian exocrine pancreatic secretion. The presence of VIP innervation in avian exocrine pancreas provides a direct line of evidence that VIP or VIP-like receptors are likely to have some physiological role for exocrine secretion.
In the freshly isolated duck pancreatic acini, CCK alone induced amylase secretion, with a maximal concentration of 10-8 M (Fig. 2). This is in sharp contrast with rodent pancreatic acini, which has a maximal CCK concentration of 10-10 M (3, 34, 35, and Fig. 1 in this work). But in the presence of cAMP-mobilizing compounds PACAP-27, PACAP-38 and VIP, CCK potency in the duck was dramatically increased, resulting in a leftward expansion of the CCK dose-response curve; now CCK in the picomolar range induced significant amylase secretion. Therefore, in the presence of endogenous PACAP and VIP, CCK became physiologically relevant. The well-established innervation of VIP (22, 23, 32, 42) and the presence of PACAP in avian pancreas (33) lend support to a role for VIP/PACAP receptors in avian exocrine pancreatic secretion.
It is noteworthy that PACAP-27, PACAP-38 from 10-12 to 10-7 M, forskolin from 10-10 to 10-5 M, and VIP from 10-12 to 10-11 M did not stimulate amylase secretion. VIP at higher concentrations (10-10-10-7 M) induced only a minor increase (Fig. 3). These data are different from those obtained in rodents. In guinea pig pancreatic acini, PACAP-38 alone has a dose-dependent stimulating effect on amylase secretion, with significant stimulation from 1 nM; on top of that, PACAP-38 also shifted CCK dose-response curve to the left by one order of magnitude (28). In rat pancreatic acini, PACAP-38 and VIP alone each stimulated amylase secretion; both enhanced CCK-induced amylase secretion without any shift in maximal CCK-stimulating concentration (27).
Neither PACAP nor VIP alone had any significant effect on [Ca2+]i (Fig. 6). PACAP had no effect on amylase secretion, whereas VIP had only a minor effect on amylase secretion at concentrations 10-10-10-7 M (Fig. 3). The effect of forskolin (Figs. 2, 3, 5, and 6) implies that the cAMP pathway predominates in PACAP- and VIP-mediated sensitization of CCK. Because PACAP (PACAP-27, PACAP-38) and VIP had rather similar potency in enhancing CCK-induced amylase secretion in duck pancreatic acini, the VPAC1/VPAC2 receptors, rather than the PAC1 receptors, were probably involved. The reason is that at the PAC1 receptors, PACAP-27 and PACAP-38 are known to be 1,000-fold more potent than VIP, whereas at the VPAC1/VPAC2 receptors, the binding affinities for PACAP-27, PACAP-38 and VIP are rather similar (43), and both VPAC1 and VPAC2 are known to work through the cAMP pathway rather than activating phospholipase C (7, 43). But definitive conclusion will have to await detailed ligand-binding studies. Although PACAP-38 and PACAP-27 both sensitized CCK-induced amylase secretion (Fig. 2, B and C), PACAP-27 shifted the maximal stimulating CCK concentration from 10-8 to 10-9 M, whereas with PACAP-38 the maximal concentration remained at 10-8 M (Fig. 2, B and C). It may be noted at this point that PACAP-38 stimulates phospholipase C-dependent calcium increase in AR4-2J cells via the PAC1 receptor, and normal rat pancreatic acini possess only the VPAC1/2 type (1, 38).
As would be expected, calcium oscillations induced by CCK alone showed a concentration-dependent transformation of the oscillation patterns. The minimum effective concentration in inducing calcium oscillations was ∼100 pM (Fig. 4). The optimal concentration for oscillation was 1 nM, whereas 10 nM induced a plateau increase. Although the concentration ranges are quite different, the general pattern of oscillation transformation on increasing CCK concentrations is similar to that in rodents such as the rat (3, 20, 26).
The CCK receptor involved in duck pancreatic acinar cell secretion was likely to be the CCK1 type, because gastrin from 10-12 to 10-6 M was completely without effect on amylase secretion, and gastrin at a concentration as high as 100 nM induced no calcium increase at all. In addition, the CCK1 receptor antagonist FK480 inhibited significantly CCK-induced amylase secretion.
The above suggest that VIP/PACAP as neurotransmitter or neuromodulator (11) and CCK as a gut hormone, are mutually dependent to execute a physiological function in duck exocrine pancreatic secretion. Interestingly, recent work by Logsdon and colleagues (25) indicates that human pancreatic acini do not effectively respond to CCK stimulation, possibly due to a substantially reduced level of CCK receptor expression. A lack of the CCK1 receptor function in amylase secretion is also suggested in cyanomolgus monkey due to low levels of receptor expression (24). Judging from the complete lack of amylase secretion by CCK stimulation (up to 100 nM) in human pancreatic acini (25), human pancreatic acini even in the presence of VPAC may not respond to CCK stimulation. Therefore, functionally, birds seem more closely related to rodents than to primates in terms of direct CCK stimulation of pancreatic secretion.
From the present work, some very interesting conclusions could be drawn as to the exact site of action on amylase secretion of the cAMP-mediated sensitization of PACAP-27, PACAP-38, VIP, and forskolin on CCK function. Yule and colleagues (18, 39) investigated the effect of cAMP signaling pathway on CCK-induced calcium oscillations in mouse pancreatic acinar cells and found phosphorylation of inositol 1,4,5-trisphosphate receptors (IP3R) essential for such an effect. But judging from Fig. 5, it is obvious that IP3R is unlikely to be the site of modulation in duck pancreatic acinar cells when CCK concentration was high (at 1 nM), because PACAP-27, PACAP-38, VIP, and forskolin had no effect on CCK (1 nM)-induced calcium oscillations. Therefore, their site(s) of action must be downstream of calcium increases.
Recently, Yoshimura et al. (45) suggested that cAMP sensitizes induced amylase secretion by priming zymogen granule for exocytosis in rat parotid acinar cell. By expanding the readily releasable (primed) pool of zymogen granules, a sensitizing effect of cAMP mobilizing compounds on calcium-mobilizing secretagogue-induced amylase secretion was achieved. The VPAC receptors in duck pancreatic acini may also execute this sensitizing effect for CCK receptors by a zymogen granule-priming effect. This priming effect could be due to cAMP-dependent regulation of the molecular motor protein kinesin (41) or direct cAMP-dependent regulation of the exocytotic SNAP receptor (SNARE) proteins such as VAMP2 (10, 12, 13). Other proteins playing a role in the final step of exocytosis such as Rab3D are also a potential site to effect zymogen granule priming (2, 44).
The mechanisms involved with lower or physiological concentrations of CCK may be quite different. Figure 7 indicated that a significant sensitizing effect of PACAP-38, PACAP-27, forskolin, or VIP toward low concentrations of CCK existed. The simultaneous presence of both VIP/PACAP and CCK signaling produced sparse calcium spikes in a majority of cells examined (Fig. 7A), whereas no calcium increases were ever seen when either was present alone (Figs. 4A, 6, 7Aa). This indicated that under circumstances resembling in vivo conditions, VIP/PACAP and CCK signaling converge to produce calcium spikes, ultimately leading to exocytosis and amylase secretion. This ensures that only when both signaling systems are activated at the same time will amylase secretion occur. Activation of either alone will have no effect. The dependence of amylase secretion on calcium was further confirmed by experiments shown in Fig. 8. Removal of extracellular calcium completely eliminated the above mentioned amylase secretion, but the protein kinase C inhibitor staurosporine was completely without effect. The cAMP signaling pathway should eventually function through the calcium-dependent exocytotic machinery is supported by recent reports indicating that IP3 receptors exist in the form of a macromolecular complex with both protein kinase A and protein phosphatases as integral components (6, 40). In this complex, IP3R phosphorylation by protein kinase A increases the sensitivity of IP3R toward IP3; therefore cells become more sensitive to external stimulation. The data we presented in this work are a direct demonstration at the physiological level of this newly defined molecular principle.
In conclusion, our data suggest that VIP/PACAP and CCK are mutually dependent to have a full physiological role in digestive enzyme secretion in the duck exocrine pancreas, and the two signaling pathways converge to produce spiking increases in intracellular calcium concentration, which may be due to intramacromolecular complex phosphorylation of the IP3 receptor by protein kinase A, which in turn is an integral component of the macromolecular complex. At higher or pharmacological CCK concentrations, this convergence dissociates and then VIP/PACAP signaling may primarily play the role of priming the zymogen granules for exocytosis and the resultant digestive enzyme secretion.
This work was supported by grants from The Natural Science Foundation of China (Nos. 39825112, 30070286) and by a PhD student-training grant from the Chinese Ministry of Education.
We are grateful to Fujisawa Pharmaceutical (Osaka, Japan) for making available to us the CCK1 receptor antagonist FK480. We thank Prof. C. J. Niu (Institute of Ecology, Beijing Normal University) for advice on ANOVA analysis.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2004 the American Physiological Society