INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE FIRST HORMONE EVER DISCOVERED (4), secretin is a 27-amino acid polypeptide and a member of the secretin/glucagon/vasoactive intestinal polypeptide (VIP) superfamily (see Refs. 12, 25 for review). Secretin-producing cells are located mainly in the mucosa of the upper small intestine in most species (6, 39). In addition, a small amount of secretin-like immunoreactivity (SLI) has been found in the gastric antral mucosa (13) and the brain (9, 35, 37). The major physiological roles of secretin are stimulation of exocrine pancreatic secretion of volume and bicarbonate (12, 25), inhibition of gastric emptying (22, 40), acid secretion, and release of gastrin (14, 42, 50). Secretin also exerts several other biological effects, including stimulation of duodenal bicarbonate secretion (21), inhibition of upper small intestinal motility and lower esophageal sphincter pressure (12), stimulation of hepatic bile flow (2, 23, 36), and electrolyte secretion (3, 32, 46).

Secretin is released by intraduodenal administration of acid, digested products of fat and protein, bile acid, and certain herbal extracts from the plant kingdom (10, 12, 25). It is now clear that the release of secretin by duodenal acidification in the rat is mediated via a secretin-releasing peptide (SRP) that is released into the upper small intestinal lumen (28). A secretin-releasing factor also exists in canine pancreatic juice (29). Study of secretin release from mucosal explants of canine duodenum has led to results indicating that the release of secretin is controlled by neurohormonal actions (34), whereas the release and action of SRP are mediated by a neural mechanism(s) (26). The neurotransmitter(s) or neuropeptide(s) that affects the action of SRP is not clearly defined at present. However, it is known that nerve fibers containing gastrin-releasing polypeptide (GRP)/bombesin (which are peptides of the same family and share COOH-terminal amino acid sequences; Ref. 17), pituitary adenylate cyclase-activating polypeptide (PACAP) (45), VIP (16), or galanin (38) innervate the mucosa of the upper small intestine. In view of this innervation and the presence of PACAP- and galanin-containing nerve fibers in the pancreatic islets and their potent stimulation and inhibition, respectively, of insulin release (38, 49), it is very likely that some or all of the above-mentioned neuropeptides may be involved in regulation of secretion from enteric endocrine cells such as the secretin-producing cells. Indeed, the release of secretin is stimulated by PACAP in conscious rats (24) and by GRP in pigs (18) and rats (27).

The cellular mechanism(s) through which the luminal stimulants SRP and neuropeptide(s) or neurotransmitter(s) elicits the release of secretin is not clear at present. Using secretin-producing cell-enriched mucosal cell preparations isolated from the upper small intestine of the dog (48) and the rat (11) or the mucosal explants of canine duodenum (33), it has been demonstrated that the release of secretin is regulated intracellularly by both Ca²⁺- and cAMP-dependent mechanisms. Because of their low yield and heterogeneity, it has been difficult to use these preparations to investigate further intracellular mechanisms of secretin release elicited by each of the luminal stimulants and the neuropeptides.

The purpose of the present study was to establish that the neuropeptides bombesin/GRP, VIP, and PACAP can act directly on secretin-producing cells to stimulate secretin secretion, using the murine intestinal neuroendocrine tumor cell line STC-1, which is known to express and secrete secretin (41, 47), as a model. We also explored whether, in addition to somatostatin, galanin can act as an inhibitory regulator of secretin secretion. To ascertain whether or not the responses of STC-1 cells to the tested neuropeptides resulted from malignant transformation, we also studied the effects of the neuropeptides on secretin release from secretin-producing cell-enriched preparations isolated from rat duodenal mucosa.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Materials. Rat secretin, PACAP-27, galanin, and VIP were synthesized by Dr. David Coy. Porcine GRP was synthesized by Dr. H. Yajima (Kyoto University, Kyoto, Japan). Bombesin and rat secretin were purchased from Peninsula Laboratories (Belmont, CA). Forskolin, 4-12-O-tetradecanoylphorbol 13-acetate (-TPA), diltiazem, IBMX, and other unspecified biochemicals were purchased from Sigma Chemical (St. Louis, MO). The Rp diastereoisomer of cAMPS (Rp-cAMPS) was obtained from Calbiochem (La Jolla, CA). Camostat [N,N-dimethylcarbamoylmethyl-p-(p-guanidinobenzoyloxy)phenylacetate methanesulfonate; a synthetic serine protease inhibitor] and plaunotol [(E,Z,E,)-7-hydroxymethyl-3,11,15-trimethyl-2,6,10,14-hexadecatetraen-1-ol; an antiulcer agent from a herbal extract] were obtained from Ono Pharmaceuticals and Sankyo Pharmaceuticals (both of Tokyo, Japan), respectively. Streptomycin, penicillin, and gentamicin sulfate were obtained from Flow Laboratories (McLean, VA). All the tissue culture wares and media were purchased either from GIBCO (Grand Island, NY) or Costar (Cambridge, MA). Radioreceptor assay kit for inositol-1,4,5-trisphosphate (IP₃) was purchased from DuPont-NEN (Boston, MA). The RIA kit for cAMP was obtained from Biomedical Technologies (Stoughton, MA).

Cell culture. STC-1 cells were obtained from Dr. Seth Grant (Columbia University, NY) through Dr. Andrew Leiter (Tufts University, Boston, MA). The cells were maintained in DMEM containing 15% horse serum, 2.5% fetal bovine serum, streptomycin (100 µg/ml), penicillin (100 µU/ml), and gentamicin sulfate (50 µg/ml) in a humidified CO₂ incubator at 37°C. Cells (1 × 10⁶ cell/ml) were seeded into 12- or 24-well tissue culture plates (Costar) and cultured for 3-4 days until 80-90% confluency.

Preparation of rat secretin cell-enriched mucosal cells. Male Sprague-Dawley rats (250-300 g) fed ad libitum were used. After an overnight fasting with free access to drinking water, a deep anesthesia was induced in the rat with 50% urethan given intraperitoneally. After a midline laparotomy, a segment of upper small intestine (15 cm from the pylorus) was excised and flushed with an incomplete Krebs-Henseleit (KH) buffer (without Ca²⁺ and Mg²⁺) containing 1 mM dithiothreitol (DTT), 0.25% BSA, 10 mM HEPES, pH 7.4, and 2.5 mM EDTA (buffer A). The intestinal segment was cut open, and the mucosa was obtained by scraping with a glass slide. The mucosa was suspended in buffer A (5 vol of fresh wt) and incubated for 15 min at room temperature with occasional swirling. The mucosal suspension was then incubated with 0.008% collagenase (type I; Sigma Chemical) at 37°C for 10 min under 95% O₂-5% CO₂ with continuous gyration. Mucosal cells freed of connective tissue were collected by filtering through a nylon mesh and centrifuged at 500 g for 5 min at room temperature. The cells were washed three times by resuspension and centrifugation in KH buffer containing 2 mM L-glutamine, 10 mM glucose, 5 mM sodium pyruvate, 1 mM DTT, and 1% BSA plus 20 mM HEPES, pH 7.4 (buffer B). The final crude cell suspension was fractionated by centrifugation at 4°C and 875 g for 15 min in a discontinuous density gradient of 0%, 23%, 34%, and 55% Percoll (Pharmacia Biotech, Piscataway, NJ) in buffer B. Four cell layers were obtained at the interfaces of Percoll solutions of 0%/23% (fraction 1), 23%/34% (fraction 2), 34%/55% (fraction 3), and as a pellet in 55% Percoll (fraction 4), respectively. The cells in each fraction were collected, washed once with buffer B, and resuspended in the incubation medium to study secretin release. An aliquot of the cell suspension was used to determine the cell number. Another aliquot of the cell suspension was centrifuged to form a pellet and then homogenized in 0.1 N HCl. The homogenate was centrifuged, and the supernatant solution was lyophilized before reconstitution and determination of secretin content by a specific RIA.

Studies on secretin release from STC-1 cells. Monolayer cultures of STC-1 cells in 24-well plates were washed once with Earle's balanced salt solution (EBSS) containing 10 mM HEPES, pH 7.4, 5 mM sodium pyruvate, 2 mM L-glutamine, 0.01% soybean trypsin inhibitor (Sigma Chemical), and 0.2% BSA. The cells were then incubated in 0.3 ml of the same medium in the presence or absence of various test agents at 37°C for 60 min or an indicated time period. The incubation was stopped by cooling the cell culture plates on ice. The medium was removed for RIA of secretin. The cells were scraped with a rubber policeman and extracted in 1 ml of 0.1 N HCl and lyophilized before being reconstituted and assayed for secretin. The amount of SLI released to the medium was expressed as a percentage of the total cell content of SLI. The effect of a stimulant was assessed by comparing secretin release in its presence with the control and was expressed as the percent increase over the control. All data are presented as means ± SE averaged from the mean of duplicated wells of a specific number of experiments.

Study on secretin release from secretin-producing cells-enriched rat mucosal cells. The release of secretin from rat mucosal cells was studied with fraction 3, which contained the highest content of secretin. The cells were suspended in Hanks' balanced salt solution containing 5 mM sodium pyruvate, 2 mM L-glutamine, 10 mM HEPES, pH 7.4, and 0.2% BSA plus 0.01% soybean trypsin inhibitor at a cell concentration of 0.5-1 × 10⁶/ml. An aliquot of 0.5 ml of the cell suspension was placed in a glass test tube (16 × 100 mm), and a test agent was added. The tube was filled with 95% O₂-5% CO₂, stoppered, and then incubated at 37°C in a gyratory water bath for 1 h. The reaction was stopped by chilling and centrifuged at 4°C, and the supernatant solution was taken and stored at 20°C for secretin RIA. The cell pellets was extracted with 1 ml of 0.1 N HCl and lyophilized before secretin assay. Secretin release was expressed as a percentage of the total cell content.

HPLC. The molecular form of SLI released in the medium and present in the STC-1 cells was determined by reverse-phase HPLC on a Varian MCH-10 column (4.6 × 300 mm) after extraction with Sep-Pak C₁₈ cartridges, using the same chromatographic solvents as described previously (44). However, a modified gradient (see Fig. 1) was used to resolve porcine and rat secretins from SLI (murine secretin) of STC-1 cells.

RIA of secretin. RIA of secretin was carried out as described previously (10), except for the use of a rabbit anti-secretin serum raised against rat secretin, ¹²⁵I-labeled rat secretin as the tracer, and unlabeled rat secretin as the standard. The anti-rat secretin serum was raised by the same method as described previously (10) and used at a final dilution of 1:5 × 10⁶. ¹²⁵I-labeled rat secretin was prepared and purified as described previously for ¹²⁵I-labeled porcine secretin (10).

Measurement of IP₃ production in STC-1 cells. Monolayer cultures of STC-1 cells were preincubated with a vehicle or PACAP or galanin in EBSS medium for 15 min at 37°C. Bombesin (10 nM) was then added, and the cells were further incubated for 0-300 s. The reaction was stopped by placing the tissue culture plate on ice, removing the medium immediately, and then adding 0.5 ml of ice-cold 16.6% TCA. The cell was scraped with a rubber policeman and transferred in a microcentrifuge tube, vortexed vigorously and allowed to stand at 4°C for 1 h. After centrifugation at 15,000 g for 1 min at 4°C, 300 µl of the supernatant solution were extracted with 800 µl of 1,1,2-trichloro-1,2,2-trifluoroethane/tri-n-octylamine (3:1, vol/vol). The IP₃ content in the upper aqueous phase was then determined by the radioreceptor assay according to the procedure provided by the manufacturer of the assay kit.

Measurement of intracellular cAMP. The intracellular cAMP content in STC-1 cells was measured by RIA, as described previously (7). Briefly, STC-1 cells were first preincubated with a vehicle or various test agents for 10 min, then PACAP (5 × 10⁸ M) was added and the cells were further incubated for 2 min. Intracellular cAMP was extracted with 10% TCA. After removing TCA by extraction with ether, we then assayed the cAMP content by RIA according to the manufacturer's suggested protocol.

Statistical analysis. Statistical analyses were performed by using ANOVA for single-factor experiments with multiple treatments, followed by Dunnett's post hoc analysis for comparing treatment means with the same control, using a Systat software program (Systat, Evanston, IL) as described previously (8). When only two means were compared, Student's paired t-test was used. A difference of P < 0.05 between two means was regarded as significant.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

SLI in STC-1 cells. The concentration of SLI in STC-1 cells was 205.9 ± 9.6 fmol/mg cell protein (n = 18). Reverse-phase HPLC of the cell extract as shown in Fig. 1 indicated that there was a major peak of SLI stored in STC-1 cells, which had a retention time of 99 min, that was somewhat different from those of porcine secretin (93 min) and rat secretin (95 min). The SLI released from STC-1 cells to the medium on stimulation with 50 mM KCl also exhibited a major peak at 99 min (Fig. 1). The minor peak eluted at 105 min could be a degraded product or a precursor molecule but could not be identified due to the lack of appropriate standards.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Results of reverse-phase HPLC of secretin-like immunoreactivity (SLI) present in the medium and cell extract of STC-1 cells. Chromatogram of SLI in the cell extract (solid line) and the medium collected after stimulation with KCl (dotted line) exhibited a major molecular form eluted at 99 min, which was different from the retention time of porcine and rat secretins (pSec and rSec, respectively) (indicated by arrows).

SLI in rat secretin-producing cell-enriched cell preparation. Cell yield and SLI content in the mucosal cell fractions isolated before and after Percoll density gradient centrifugation are summarized in Table 1. Secretin-containing cells were most enriched in the fraction located at the interface between 34% (density = 1.043 g/ml) and 55% (density = 1.056 g/ml) of Percoll (fraction 3) in the Percoll density gradient, increasing the content of SLI by threefold from 402 ± 48 to 1,341 ± 107 fmol/10⁶ cells. Meanwhile, the viability of the cells as judged by the trypan blue exclusion method increased from 61.8% to 95%. The total number of cells of fraction 3 from the small intestines of five rats was 4.1 × 10⁶. The presence of a considerable amount of SLI in fractions 1 and 2 suggested that secretin-producing cells are polydispersed with respect to cell density.

Effect of various agents on SLI release from STC-1 cells. The results summarized in Table 2 indicate that the release of SLI from STC-1 cells was stimulated significantly by dibutyryl-cAMP (data not shown), forskolin, the protein kinase C (PKC)-activating phorbol ester -TPA, the Ca²⁺ ionophore A-23187, and depolarization with KCl. The stimulatory effect of forskolin was enhanced by IBMX (0.5 mM), but the latter alone did not significantly stimulate the release of SLI. Among the pharmacological agents studied, KCl was the most potent stimulant. Luminal stimulants such as camostat, sodium oleate, and plaunotol also significantly stimulated the release of SLI. Thus the responses of STC-1 cells to these agents were very similar to those reported previously in secretin-producing cell-enriched preparations isolated from the mucosae of canine duodenum (48) and rat upper small intestine (11).

Effects of neuropeptides on SLI release. To test the hypothesis that neuropeptides participate in regulation of secretin release, we studied the effect of bombesin/GRP and PACAP-27 on the release of SLI from both STC-1 cells and the isolated secretin-producing cell-enriched preparation. Because the elevation of cellular cAMP level is known to elicit secretin secretion (11, 48), we used the response of these cells to forskolin (30 µM) plus IBMX (0.5 mM) as a positive control. Thus, if a cell preparation, particularly the mucosal cells, failed to respond to this mode of stimulation, it was unlikely to respond to a neuropeptide. As shown in Fig. 2, incubation of STC-1 cells with a submaximal concentration of GRP (10 nM), bombesin (10 nM), or PACAP (50 nM) at 37°C for 60 min resulted in a significant increase in SLI release over basal secretion by 116 ± 13%, 105 ± 11%, and 65 ± 14%, respectively. These increases were somewhat smaller than that of the positive control (238 ± 11%; data not shown) of forskolin plus IBMX. Similarly, all three peptides stimulated the release of SLI from rat mucosal cell suspension with significant increases of 43 ± 15% (by GRP), 39 ± 14% (by bombesin), and 43 ± 10% (by PACAP-27) over basal. The effects of these neuropeptides were comparable to that of the positive control in the mucosal cell preparation (increased by 69 ± 4%; data not shown).

View larger version (32K): [in this window] [in a new window]   Fig. 2.   Effect of bombesin, gastrin-releasing polypeptide (GRP), or pituitary adenylate cyclase-activating polypeptide-27 (PACAP-27) on secretin release from STC-1 and mucosal secretin-producing cell-enriched preparation isolated from rat upper small intestine. Bombesin (10 nM), GRP (10 nM), and PACAP-27 (50 nM) significantly stimulated secretin release from both STC-1 cells and rat mucosal cells after 60 min of incubation. Data (in %stimulation over corresponding controls) are means ± SE of 4 experiments. Controls were 2.27 ± 0.25% and 3.67 ± 0.66% of the total cellular SLI content for STC-1 and mucosal cells, respectively.

Time-dependent effects of neuropeptides on SLI release. The stimulatory effects of bombesin, GRP, and PACAP-27 on SLI release from STC-1 cells were time dependent. As shown in Fig. 3, there was a small basal secretion of SLI (0.74 ± 0.21% of total secretin content, n = 10) over a period of 120 min at 37°C. In the presence of bombesin (10 nM) or GRP (10 nM), the release of SLI increased rapidly, reaching 1% of the cellular content within 15 min, and continuously increased at a reduced rate, reaching 4% at 120 min. In the presence of PACAP-27 (50 nM), there was a smaller but linear increase in the release of SLI over a period of 60 min, reaching 2% of total cellular content. In the subsequent studies, the cells were incubated with these stimulants for 60 min at 37°C.

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Time-dependent effects of GRP, bombesin, and PACAP-27 on secretion of SLI from STC-1 cells. Bombesin (10 nM), GRP (10 nM), or PACAP-27 (50 nM) continuously stimulated release of SLI from STC-1 cells over the basal secretion for a 120-min period. Data represent average of 5 experiments.

Concentration-dependent effects of bombesin, GRP, PACAP-27, and VIP. The stimulatory effects of these neuropeptides on the release of SLI from STC-1 cells were also concentration dependent. As shown in Fig. 4, incubation of STC-1 cells with either GRP or bombesin at 1 nM and higher concentrations resulted in a significant concentration-dependent increase in SLI secretion that reached a plateau at 10 nM. In both cases, the EC₅₀ of bombesin or GRP was ~1 nM. In contrast, higher concentrations of PACAP-27 were required to stimulate SLI release from STC-1 cells, with a threshold dose of 18 nM, reaching the same extent of stimulation maximally attained by bombesin and GRP (100% increase over control) at 180 nM. Because PACAP-27 at concentrations above 180 nM started to exhibit cross reaction with RIA of secretin, we did not determine the maximal effect of PACAP-27. Synthetic VIP appeared to be more potent than PACAP-27, starting to stimulate secretin release as low as 0.1 nM (52 ± 12% over basal) and reaching a maximal stimulation at 100 nM (570 ± 78% over basal) with an EC₅₀ of 10 nM (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Concentration-dependent effect of GRP, bombesin, and PACAP-27 on secretion of SLI from STC-1 cells. Bombesin and GRP stimulated release of SLI from STC-1 cells concentration dependently with an EC50 of 1 nM and a maximal stimulation attained at 10 nM. PACAP also stimulated release of SLI concentration dependently but at higher concentrations than the other peptides. Data represent average of 4 experiments.

Synergistic effect of PACAP-27 added together with other secretagogues. In the presence of IBMX, PACAP-27 and VIP were shown to increase cAMP production in STC-1 cells (7) and their stimulatory effect was mediated mainly through the adenylate cyclase signal transduction pathway. Thus we studied whether PACAP would have a synergistic effect with other secretagogues that stimulated SLI release via different signal transduction pathways. As shown in Fig. 5, PACAP-27 (50 nM) in combination with IBMX (0.5 mM), bombesin (10 nM), KCl (50 mM), or -TPA (0.1 µM) produced a significantly greater stimulation of SLI release from STC-1 cells than the sum of the individual effects of PACAP and the other secretagogue studied in the same experiments, thereby suggesting a synergistic effect. A similar synergism between VIP and each of these agents was also observed (data not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 5.   Synergistic effect of PACAP-27 with other secretagogues on SLI release from STC-1 cells. Incubation of STC-1 cells in the presence of PACAP (50 nM) together with either IBMX (0.5 mM), KCl (50 mM), bombesin (10 nM), or 4-12-O-tetradecanoylphorbol 13-acetate (-TPA; 0.1 µM) resulted in greater stimulation of SLI secretion (hatched bars) than the sum (solid bars) of the effect of PACAP alone (crosshatched bars) and that of the other secretagogue alone (open bars) that was determined in the same experiment, indicating a synergistic effect. Data are means ± SE of 9 experiments. * P < 0.05, ** P < 0.01, significant difference between secretagogue + PACAP and PACAP alone.

Inhibition by galanin of secretagogue-stimulated SLI secretion. As shown in Fig. 6, rat galanin alone had no effect on basal secretion of SLI from STC-1 cells but inhibited KCl-, bombesin-, and PACAP-stimulated SLI secretion in a concentration-dependent manner. A small but significant inhibition by galanin of -TPA-stimulated secretion was also observed. Pretreatment of the cells with pertussis toxin (PTx) prevented inhibition by galanin of SLI secretion stimulated by bombesin (Fig. 7, top), PACAP (Fig. 7, middle), or KCl (Fig. 7, bottom). In contrast, PTx alone had no effect on both basal and secretagogue-stimulated secretions.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Concentration-dependent effect of rat galanin on secretagogue-stimulated SLI release from STC-1 cells. STC-1 cells were incubated in the absence (control) or presence of KCl (25 mM), -TPA (100 nM), bombesin (10 nM), or PACAP (100 nM) with varying doses of galanin (0 or 1010-107 M) for 1 h, and amount of SLI released was then determined. Galanin significantly inhibited each secretagogue-stimulated SLI release (at 0 galanin concentration) at concentrations indicated.

View larger version (41K): [in this window] [in a new window]   Fig. 7.   Effect of pertussis toxin (PTx) on inhibition by galanin (Gal) of secretagogue-stimulated SLI release. STC-1 cells were preincubated in the absence or presence of 100 ng/ml PTx for 10 h and then incubated with bombesin (10 nM; top), PACAP (100 nM; middle), or KCl (25 mM; bottom), either alone or in combination with galanin (10 nM), for 1 h. The amounts of SLI released under each set of conditions were determined and compared. Galanin significantly inhibited secretagogue-stimulated SLI release, and this effect was reversed by pretreatment of STC-1 cells with PTx. * P < 0.05, ** P < 0.01 vs. control.

Effects of galanin and PACAP on bombesin-stimulated IP₃ production. Bombesin has been shown to elevate intracellular Ca²⁺ concentration in STC-1 cells (35). This effect is most likely due to activation of a phophatidylinositol phosphate-specific phospholipase C pathway. Therefore, we determined the effect of bombesin on IP₃ production in STC-1 cells. In addition, we also determined whether or not inhibition by galanin or potentiation by PACAP on bombesin-stimulated secretin secretion was mediated by changes in bombesin-stimulated IP₃ production. As shown in Fig. 8, incubation of STC-1 cells with bombesin (10 nM) resulted in a time-dependent increase of cellular IP₃ content, reaching a maximal value of 559 ± 120 fmol/mg cell protein at 30 s that was in equivalent to a 14.8-fold increase over the basal value (27 ± 4 fmol/mg cell protein). The IP₃ level declined subsequently but remained elevated over basal for 2 min. Preincubation of the cells with galanin (10 nM) for 15 min significantly reduced the stimulatory effect of bombesin on IP₃ production by 58% (233 ± 13 fmol/mg cell protein at 30 s). Galanin alone had no significant effect on basal IP₃ production. On the other hand, preincubation of the cells with PACAP (50 nM) did not affect basal or bombesin-stimulated IP₃ production (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 8.   Effect of galanin on bombesin-stimulated inositol-1,4,5-trisphosphate (IP3) production from STC-1 cells. STC-1 cells were preincubated with or without 10 nM galanin before incubation in the absence or presence of 5 nM bombesin and determination of cellular IP3 content as described in METHODS. , Basal; , galanin alone; , bombesin alone; , bombesin after preincubation with galanin. Galanin by itself had no effect on basal IP3 formation but significantly inhibited bombesin-stimulated IP3 formation.

Effect of galanin and bombesin on PACAP-stimulated cAMP production. We have shown previously that PACAP and VIP stimulated cAMP production in STC-1 cells (7). Therefore, we explored whether or not the effect of galanin to inhibit or that of bombesin to enhance PACAP-stimulated secretin release was mediated via inhibition or enhancement of cAMP production, respectively. STC-1 cells were preincubated with or without 10 nM galanin or bombesin for 15 min before addition of PACAP (50 nM) and IBMX and a further incubation of 2 min at 37°C that resulted in maximal cAMP production (7). As shown in Fig. 9, PACAP stimulated cAMP production by 168% in the presence of IBMX. Preincubation of the cells with galanin resulted in a significant reduction in PACAP-stimulated cAMP production (down to 54% over the control) that corresponded to a 68% inhibition. Galanin alone had no effect on basal cAMP production. As reported previously (7), bombesin did not affect basal or PACAP-stimulated cAMP production (data not shown).

View larger version (26K): [in this window] [in a new window]   Fig. 9.   Effect of galanin on PACAP-stimulated cAMP production in STC-1 cells. STC-1 cells were preincubated with or without 10 nM galanin before incubation in the absence or presence of 50 nM PACAP or 0.25 mM IBMX or both and then cAMP was determined as described in METHODS. Galanin by itself had no effect on basal cAMP formation but significantly inhibited PACAP-stimulated cAMP formation detected in the presence of IBMX.

Effect of EGTA, diltiazem, and Rp-cAMPS on bombesin- and PACAP-stimulated secretin release and their synergism. To determine the involvement of extracellular Ca²⁺ and protein kinase A (PKA) activity on bombesin- and PACAP-stimulated secretin release, the effect of the Ca²⁺ chelator EGTA (1 mM), the L-type voltage-gated Ca²⁺ channel blocker diltiazem (10 µM), and the PKA-specific inhibitor Rp-cAMPS (100 µM) were studied. As shown in Fig. 10, 5 nM bombesin-stimulated secretin release was reduced in the presence of EGTA, decreasing from 436 ± 58% to 194 ± 27% over basal with an average inhibition of 59 ± 2% (P < 0.01, n = 4). The inhibited secretin release, however, remained significantly higher than the basal secretion (P < 0.01). EGTA also reduced 50 nM PACAP-stimulated secretin release by 75 ± 13% (P < 0.01, n = 4), decreasing from 49.0 ± 5% to 10 ± 6% over basal secretion and thus no longer being significantly different from basal secretion. As shown in Fig. 10, diltiazem inhibited bombesin-stimulated secretin release by 64 ± 5% (P < 0.01, n = 11), decreasing from 461 ± 6% to 172 ± 40% over basal secretion. Again, diltiazem-inhibited secretion remained significantly greater than basal secretion (P < 0.05). Diltiazem also inhibited PACAP-stimulated secretin secretion by 55 ± 11% (P < 0.05, n = 11), reducing it from 52 ± 9% to 23.4 ± 7% over basal. As shown in Fig. 11, incubation of STC-1 cells with 100 µM Rp-cAMPS resulted in a significant reduction of PACAP-stimulated secretin release from 69 ± 16% to 26 ± 10% over basal (P < 0.01, n = 9). Rp-cAMPS also reduced bombesin-stimulated secretin release by 30 ± 6% (P < 0.05, n = 9), decreasing from 339 ± 50% to 239 ± 44% over the basal secretion. Interestingly, none of the above-mentioned agents prevented potentiation between PACAP and bombesin, despite their inhibition of the effect of each peptide individually. Thus, in the presence of EGTA, bombesin plus PACAP stimulated secretin release by 539 ± 119% over the control (with EGTA alone) that was significantly greater than the sum of the individual effects of the two peptides that was 207 ± 37% over the same control (P < 0.01, n = 3). In the presence of diltiazem, bombesin plus PACAP increased secretin release by 513 ± 229% over control (with diltiazem alone), which was significantly greater than 170 ± 52% over the control produced by the sum of the individual effects of these two peptides with diltiazem. (P < 0.05, n = 3). In the presence of Rp-cAMPS, the effect of bombesin plus PACAP was 528 ± 107% over the control (with Rp-cAMPS alone), which was significantly greater than the sum of the individual effects of bombesin and PACAP, namely 268 ± 51% over control (P < 0.05, n = 9).

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Effect of EGTA and diltiazem on PACAP- and bombesin-stimulated secretin release. STC-1 cells were incubated with or without PACAP (50 nM) or bombesin (5 nM) in the absence or presence of either 1 mM EGTA (left) or 10 µM diltiazem (right) to determine secretin release as described in METHODS. Both EGTA and diltiazem significantly reduced PACAP- or bombesin-stimulated SLI release.

View larger version (28K): [in this window] [in a new window]   Fig. 11.   Effect of Rp-cAMPS on PACAP- and bombesin-stimulated secretin release. STC-1 cells were incubated with PACAP (50 nM) or bombesin (5 nM) in the absence or presence of 100 µM Rp-cAMPS to determine secretin release as described in METHODS. Rp-cAMPS significantly inhibited PACAP- and bombesin-stimulated secretin release. * P < 0.05 vs. Rp-cAMPS. ** P < 0.01 vs. Rp-cAMPS.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Study of the cellular mechanisms of secretin secretion from intestinal endocrine cells has been very limited. The results of previous studies using secretin-producing cell-enriched mucosal cell preparations isolated from dog duodenum (48) and rat upper small intestine (11) have indicated that secretin release from secretin-producing cells is mediated by Ca²⁺- and cAMP-dependent mechanisms and can be elicited by activation of PKC. However, due to the low yield and heterogeneity of the intestinal secretin-producing cell preparation, it is difficult to identify the natural secretagogue that triggers the release of these second messengers or PKC activation. The discovery of the more homogeneous secretin-expressing tumor cell lines (41, 47) thus may provide better models for studying the cellular mechanism of secretin secretion. In the present study, we have tested the murine intestinal neuroendocrine tumor cell line STC-1 for use as a model to study secretin release. Our results have indicated that the release of SLI from STC-1 cells can be elicited by various pharmacological stimulants, including forskolin, KCl, -TPA, and A-23187, and by luminal secretagogues, including sodium oleate, camostat, and plaunotol. All of these agents have been shown to stimulate the release of secretin either from secretin-producing cell-enriched mucosal preparations (11, 48) or in intact animals (10, 12, 25). The observed stimulation of SLI release by forskolin, which activates adenylate cyclase, by A-23187 and KCl, which increase Ca²⁺ influx, and by -TPA, which activates PKC activity, suggested that regulation of secretin secretion in STC-1 cells, similar to that in the mucosal secretin-producing cells, is mediated via Ca²⁺- and cAMP-dependent mechanisms and through modulation of PKC activity. The responses of STC-1 cells to sodium oleate, camostat, and plaunotol suggested that these luminal stimulants can act directly on secretin-producing cells and it will be possible to use these cells for studying the mechanism of action by these secretagogues. The observed stimulatory effects of bombesin, GRP, VIP, and PACAP on SLI secretion suggested that STC-1 cells are also suitable for studying receptor-mediated secretin release. Indeed, it has been shown that STC-1 cells contain GRP/bombesin receptors (43). Furthermore, the observation that these neuropeptides also stimulated secretin release from the mucosal secretin-producing cell preparation suggested that the stimulation by these neuropeptides is an intrinsic property of secretin-producing cells rather than a result of tumorigenic transformation of STC-1 cells. Therefore, it may be concluded that the STC-1 cell line is a useful model for studying the cellular mechanism of secretin secretion, despite the fact that this cell line also secrets smaller amounts of CCK (8), GIP (5), GLP-1 (1), glucagon, and insulin (41).

Previous studies have provided evidence indicating that secretion of secretin may be subject to neurohormonal regulation. Thus, in cultured explants of canine duodenal mucosa, the release of SLI was shown to be stimulated by -aminobutyric acid that was potentiated by carbamyl choline but inhibited by somatostatin and Met-enkephalin (34). In rats, the acid-induced release of secretin is mediated by an SRP (28). It has been shown that the release and action of SRP are dependent on the vagal afferent pathways that are independent of cholinergic tone (26). The noncholinergic neural factor or factors may include a neuropeptide(s) that acts directly on the secretin-producing cells. Indeed, it has been shown recently that intravenous infusion of GRP in pigs (18) and rats (27) and of PACAP (24) in rats resulted in elevation of plasma secretin concentration. Because both GRP- and PACAP-containing neurons and nerve fibers exist in the myenteric and submucous plexus of the upper small intestine (17, 45), it is very likely that the above-mentioned effects of GRP and PACAP on the release of secretin may mimic neurocrine actions of these peptides on the mucosal secretin-producing cells. Similarly, VIP-containing nerve fibers are known to innervate the intestinal mucosa (16) and may have a similar effect. To test if these neuropeptides can act directly on secretin-producing cells, we studied the effects of GRP, bombesin, and PACAP-27 on secretin secretion from both STC-1 cells and a secretin-producing cell-enriched mucosal cell preparation isolated from rat upper small intestine. Our results have demonstrated clearly that all three peptides can stimulate secretin secretion from the mucosal cell preparation. This observation suggests that all three peptides can act directly on mucosal cells. However, due to heterogeneity in the mucosal cell preparation, it is not certain at present whether these peptides acted directly on secretin-producing cells or through the release of SRP and/or another yet unidentified mediator(s). On the other hand, the observation that all three peptides stimulated secretion of secretin from STC-1 cells in a concentration- and time-dependent manner suggests that they definitely can act directly on secretin-producing cells. We also observed that VIP, which has a high sequence homology with PACAP and often shares receptors with the latter peptide, also potently stimulated the release of secretin in STC-1 cells. Therefore, it is very likely that GRP/bombesin and PACAP/VIP participate in regulation of secretin-producing cell function. In addition, the release of SLI stimulated by these neuropeptides and other secretagogues was inhibited by galanin in a concentration-dependent manner. The effect of galanin appears to be mediated through a PTx-sensitive G protein-coupled receptor. Moreover, its inhibitory action appeared to be mediated through inhibition of bombesin- and PACAP-stimulated production of second messengers. Thus galanin, which is also abundant in the enteric nervous system (38), appears to be another candidate neuropeptide involved in the inhibitory regulation of secretin secretion. Therefore, it is likely that the release of secretin in vivo under various physiological conditions may be the net effect of the concerted actions of these neuropeptides and luminal stimulants. This question should be an important and challenging problem for future studies.

Another significant observation of the present study is that there is a synergistic effect between PACAP and other secretagogues on secretion of SLI from STC-1 cells. This effect appears to involve a synergistic effect of different signal cascades. The effect of PACAP on secretin release was accompanied by production of cAMP, substantially enhanced by the phosphodiesterase inhibitor IBMX, and inhibited by the PKA-specific inhibitor Rp-cAMPS, thereby strongly suggesting the involvement of the cAMP/PKA cascade. In addition, maintenance of intracellular Ca²⁺ homeostasis through influx of extracellular Ca²⁺ appears to be as important because both EGTA and diltiazem inhibited PACAP-stimulated secretin release. However, these observations should not be regarded as indicating that PACAP also stimulates secretin secretion via activation of Ca²⁺ influx through Ca²⁺ channels, including the L-type Ca²⁺ channels, although this could not be ruled out without further study. It should be noted that several subtypes of adenylate cyclase are known to be either activated or inhibited by Ca²⁺ or Ca²⁺/calmodulin (15), and that hormone secretion is a process of regulated exocytosis that is known to involve an Mg-ATP-dependent priming step before a final step of Ca²⁺-dependent secretion (19, 20). Because we were unable to observe an elevation of intracellular Ca²⁺ concentration by PACAP in STC-1 cells (data not shown), PACAP-stimulated secretin release would be very likely to be dependent on the extracellular Ca²⁺ source for this final step of exocytosis, thereby explaining its inhibition by both EGTA and diltiazem. In a previous study (7), we demonstrated that PACAP or VIP was unable to maintain a significantly elevated intracellular cAMP level in STC-1 cells unless IBMX was present, thereby suggesting a high level of endogenous phosphodiesterase activity in these cells. It is possible that a small elevated steady-state level of cAMP stimulated by PACAP is sufficient to activate a certain subtype(s) of PKA in STC-1 cells to elicit a small but significant stimulation of secretin release. Bombesin has been shown to cause elevation of intracellular Ca²⁺ concentration that was in part sensitive to diltiazem (43), and KCl activates voltage-gated Ca²⁺ channels (30) in STC-1 cells, thereby suggesting involvement of a Ca²⁺-dependent cascade in their action on secretin release. Our observation that bombesin stimulated IP₃ production suggests that this peptide also acts via a receptor coupled to a phosphatidylinositol phosphate-specific phospholipase C signal cascade that includes a mobilization of intracellular Ca²⁺ in response to the production of IP₃ and activation of PKC. In the present study, we observed that bombesin and PACAP potentiated each other on secretin release without a synergistic effect on production of either IP₃ or cAMP. Therefore, their potentiation effect represents synergism between their signal cascades downstream of the production of these second messengers. We also observed that synergism between these two peptides was not affected by EGTA and diltiazem, or by the PKA-specific inhibitor Rp-cAMPS. These observations appear to suggest that synergism between these two peptides is not mediated through influx of extracellular Ca²⁺ or the activation of PKA activity, although it cannot be ruled out at present that a small residual PKA activity is sufficient to potentiate with bombesin-activated signal molecule(s). The observation of a small but significant inhibition of bombesin-stimulated secretin release by Rp-cAMPS appears to indicate that the endogenous PKA activity participates in interaction with the bombesin-elicited signal cascade and thus supports this point of view. Nevertheless, the effector(s) activated by PACAP to involve in potentiation with bombesin remains to be identified. On the other hand, the observation that -TPA, a well-known activator of PKC, also potentiates with PACAP to stimulate secretin release suggests that synergism between bombesin and PACAP may be in part mediated by PKC. In view of the finding that bombesin still stimulated secretin release significantly in the presence of EGTA or diltiazem and elevated intracellular Ca²⁺ concentration in the presence of diltiazem (43), it is very likely that Ca²⁺ released from the intracellular Ca²⁺ store by IP₃ elevated by bombesin also participates in the synergistic effect with PACAP. Thus the mechanism of potentiation between bombesin and PACAP cannot be completely deduced at present and remains an interesting problem for future study. Nevertheless, this mode of potentiation appeared to be different from that recently reported for potentiation between substance P and VIP in pituitary lactotroph cells (31). In this system, VIP, though ineffective alone, enhanced IP₃ production and translocation of a few subtypes of PKC stimulated by substance P. Substance P, which had no effect by itself on cAMP production, also enhanced cAMP formation stimulated by VIP. Thus synergistic cross talk between Ca²⁺ and cAMP signal cascades appears to be quite diversified and varies among different cell types. We believe that synergism between neuropeptides with different signal cascades may be an important characteristic of neuropeptide regulation of secretin secretion. Thus it is tempting to speculate that GRP/bombesin and PACAP and possibly VIP may act as coregulators of secretin secretion under yet to be identified physiological conditions, whereas under other conditions, these neuropeptides may act synergistically with luminal stimulants, such as nutrients, pharmacological agents, and the putative SRP. We believe these are new and important questions regarding regulation of gut endocrine cells that merit further study.