J Gastroenterol 2001; 36:375–385

Effects of omeprazole and pirenzepine on enterochromaffin-like cells and parietal cells in rat stomach Akira Tari1,4, Yoshitada Kuruhara2, Yoshikazu Yonei3, Ryo Yamauchi4, Shiro Okahara4, Koji Sumii4, and Goro Kajiyama4 1

Department of Internal Medicine, Hiroshima Red Cross Hospital and Atomic-bomb Survivors Hospital, 1-9-6 Senda-machi, Naka-ku, Hiroshima 730-8619, Japan 2 Kawanishi Pharma Research Institute, Nippon Boehringer Ingelheim, Kawanishi, Japan 3 Department of Internal Medicine, Nippon Kokan Hospital, Kawasaki, Japan 4 First Department of Internal Medicine, Hiroshima University School of Medicine, Hiroshima, Japan

Editorial on page 436 Purpose. The purpose of this study was to investigate the mechanism of the regulation of histamine synthesis in enterochromaffin-like cells, chemically and structurally, by treatment with omeprazole and pirenzepine. Methods. The ultrastructures of enterochromaffin-like cells and parietal cells were examined in rats treated with oral omeprazole (20 mg/kg) or intraperitoneal pirenzepine (1 mg/kg) administration. Serum gastrin concentrations, mRNA levels of H1-K1-ATPase and histidine decarboxylase, and the fundic concentrations of somatostatin and histamine were determined. Results. Pirenzepine treatment suppressed omeprazoleinduced increases in serum gastrin levels and mRNA levels of H1-K1-ATPase and histidine decarboxylase. Pirenzepine also decreased omeprazole-induced increases of histamine concentration in fundic mucosa. Pirenzepine elevated somatostatin mRNA level, previously decreased by omeprazole treatment, in fundic mucosa. In the cytoplasm of enterochromaffin-like cells, omeprazole markedly reduced the numbers of vesicles and granules, but significantly increased their diameters, whereas pirenzepine treatment changed neither of these features. The densities and diameters of both vesicles and granules produced by treatment with omeprazole and pirenzepine were between those produced by treatment with omeprazole alone and pirenzepine alone. Conclusions. Omeprazole-induced hypergastrinemia and pirenzepine-induced somatostatin synthesis play important roles not only in histamine synthesis but also in ultrastructural changes in enterochromaffin-like cells.

Received: September 13, 2000 / Accepted: January 12, 2001 Reprint requests to: A. Tari

Key words: H1-K1-adenosine triphosphatase, histidine decarboxylase, histamine, somatostatin, enterochromaffin-like cell

Introduction Gastrin has two principal biological effects: stimulation of acid secretion from gastric parietal cells during and after meals, and stimulation of mucosal growth in the acid-secreting part of the stomach, especially in enterochromaffin-like (ECL) cells. Gastrin has also been reported to have trophic effects in the colon, duodenum, and pancreas.1 Early cultures of gastric and colorectal human adenocarcinomas have been shown to respond trophically to gastrin.2 The hypergastrinemia in peptic ulcer patients produced by long-term inhibition of gastric acid secretion with omeprazole causes problems due to the trophic effect of gastrin and specific increases in proton pump synthesis. In rats, long-term hypergastrinemia increases mucosal thickness and the ECL cell density in oxyntic mucosa, and results in the development of gastric carcinoids.3 ECL cells are present in the basal half of the gastric mucosa and are characterized ultrastructurally by the presence of numerous, fairly large, electron-lucent vesicles in the cytoplasm. In addition to the vesicles, a few electron-dense granules are observed.4 Local histamine is thought to play an important role in the physiological control of gastric acid secretion, and ECL cells may well represent the physiologically relevant local pool of gastric histamine.5 It was shown quite early that gastrin reduces the histamine concentration in the oxyntic mucosa and also activates histamine-forming enzyme.6 Recently, gastrin was shown to stimulate the release of histamine from isolated perfused stomach7 and from isolated ECL cells.8

376

In rat parietal cells, the hypergastrinemia induced by multiple doses of omeprazole causes a significant increase in H1-K1-adenosine triphosphatase (ATPase) gene expression9 and in the number of morphologically fully stimulated parietal cells, in which the proton pump protein has moved into the apical membrane.10 In the rabbit, blocking H1-K1-ATPase by omeprazole enhances the degradation and macrophage-mediated elimination of parietal cells, and also causes an increase in parietal cell production.11 This effect of omeprazole is caused by the overstimulation of parietal cells by increased histamine release from enterochromaffin-like (ECL) cells,9 and causes vacuolation in approximately 27% of all parietal cells in guinea pigs.10 Therefore, when potent gastric acid secretion-inhibiting drugs are employed for long-term treatment, we recommend that serum gastrin levels be monitored to detect hypergastrinemia, to avoid overstimulation of both ECL cells and parietal cells. Pirenzepine, a selective muscarine receptor (M1) antagonist, binds to M3 receptors on the basolateral membrane of parietal cells. This activity suppresses the acid secretion stimulated by acetylcholine release from enteric nerve fibers following vagal stimulation or local reflexes through M3-receptors.12 Pirenzepine also suppresses gastrin release from G cells, directly or indirectly, by stimulating somatostatin secretion from D cells.13,14 In peptic ulcer patients, omeprazole-pirenzepine combination treatment significantly suppressed the marked hypergastrinemia induced by omeprazole treatment, and pirenzepine treatment seems to be useful in reducing the undesirable effects of omeprazole-induced hypergastrinemia.15 In addition, omeprazole-pirenzepine combination treatment effectively suppressed acid secretion, especially in patients in whom suppression by omeprazole alone was insufficient. Combination treatment with omeprazole and pirenzepine in rats significantly decreased gastrin mRNA level and significantly increased somatostatin mRNA level compared with the levels achieved by omeprazole treatment alone.14 These findings suggest that somatostatin plays a role in the inhibition of omeprazoleinduced hypergastrinemia produced by pirenzepine. The purpose of the present study was to examine the responses of ECL cells and parietal cells, in chemistry as well as in ultrastructure, to hypergastrinemia and to the change in somatostatin concentration of the fundic mucosa induced by treatment with omeprazole or pirenzepine in rats. Materials and methods Animal experimental procedure Male Wistar rats, weighing 220–250 g, received regular laboratory food ad libitum. Omeprazole was dissolved

A. Tari et al.: Regulation of histamine in rat ECL cells

in polyethylene glycol (PEG) 400 and NaHCO3 buffer solution to a final concentration of 5 mg/ml. Pirenzepine was dissolved in physiological saline to a final concentration of 0.25 mg/ml. Three types of experiments were conducted: (1) Oral doses of omeprazole 20 mg/kg plus vehicle (physiological saline) were administered once a day for 28 days, and the levels of acid secretion, serum gastrin, and mRNA were measured 3 h after the last dose. (2) Intraperitoneal injections of pirenzepine 1 mg/ kg plus vehicle (PEG 400 and NaHCO3 buffer solution) were administered once a day for 28 days, and the same parameters were measured 3 h after the last dose. (3) Both omeprazole and pirenzepine were administered once a day for 28 days, and the levels of acid secretion, serum gastrin, and mRNA were measured 3 h after the last doses of omeprazole and pirenzepine. Blood samples were collected immediately after rats were decapitated. The stomach was resected and then opened along the greater curvature. The intragastric contents were removed and rinsed with ice-cold 0.9% saline, and the oxyntic gland area was excised from the stomach. Gastric tissues were immediately frozen in liquid nitrogen and stored at 280°C until extraction. Gastrin radioimmunoassay Serum gastrin levels were determined using a Gastrin RIA-kit II (Dainabot, Tokyo, Japan). Somatostatin radioimmunoassay Somatostatin was extracted from fundic tissues and the concentration was determined by radioimmunoassay, as previously described.16 Histamine determination Histamine in the tissues was measured fluorimetrically, according to the methods by Yamatodani et al.,17 with some modifications. Tissues (100–500 mg wet weight.) were homogenized with 3 ml of 0.5 M perchloric acid in an ice bath. After centrifugation at 9000 g for 10 min, 800 µl of the supernatant was adjusted to pH 6 with 30% potassium hydroxide and stored at 240°C. For analysis, the frozen supernatants were centrifuged at 9000 g for 10 min, and 100 µl of the clean supernatants were injected directly onto an HPLC column (250 3 4.6 mm internal diameter [ID]) packed with a cation exchanger (TSK gel SP-2SW; 5 µmm; Tosoh, Tokyo, Japan). The column was eluted with 0.25 M potassium dihydrogen phosphate (pH 6.0) at a flow rate of 0.6 ml/min. The eluate from the column was mixed first with 0.05% ophthalaldehyde solution at a flow rate of 0.4 ml/min, and then a 1 M solution of sodium hydroxide was added (0.4 ml/min) to adjust the reaction mixture to pH 12.5. The solution was mixed in a reaction coil made of

A. Tari et al.: Regulation of histamine in rat ECL cells

polytetrafluoroethylene tubing (7 m 3 0.33 mm [ID]) at 50 6 0.2°C, and then 4% (v/v) sulfuric acid was added (0.4 ml/min), with a short reaction coil (1 m 3 0.25 mm [ID]) being used for mixing. The pH of final reaction mixture was 3.0. The fluorescence intensity was measured at 450 nm, with excitation at 360 nm, in a spectrofluorometer (model F-1000; Hitachi, Tokyo, Japan) equipped with a 12-µl square flow cell and a chromatographic data processor (System Instruments, Tokyo, Japan). The fluorescence intensity was directly proportional to the histamine concentration in a range of 0.1– 40 pmol of authentic histamine in the diluted eluate. Recovery of histamine was 95%–100%, and the coefficient of variation was 0.78% (n 5 47). RNA preparation Total RNA was isolated by the acid guanidiniumphenol-chloroform method,18 with some modifications. Fundic tissues were homogenized in a mixture of denaturing solution (4 M guanidinium thiocyanate, 25 mM sodium citrate, pH 7.0, 0.5% N-laurylsarcosine, and 0.1 M 2-mercaptoethanol), 2 M sodium acetate, pH 4.0, and water-saturated phenol, using a Polytron (Kinematica, Luzern, Switzerland), and purified with a chloroform-isoamyl alcohol mixture (24 : 1) and isopropanol. The concentration and quality of RNA samples were determined by measuring their optical densities at 260 and 280 nm, and by staining with ethidium bromide after agarose gel electrophoresis. Hybridization probes Complementary RNA probes were labeled according to the method of Melton et al.19 H1-K1-ATPase α-subunit cRNA and histidine decarboxylase cRNA were synthesized from a riboprobe plasmid pGEM-3 containing a rat H1-K1-ATPase cDNA insert (kindly provided by Gary Shull20), using T7 RNA polymerase, and from a pBluescriptSK 2 containing a rat histidine decarboxylase cDNA insert,21 (kindly provided by Rod Dimaline), using T3 RNA polymerase, respectively. Rat somatostatin cDNA (pSP65cSST; a gift of Dr. Jack E. Dixon) and human β-actin cDNA (Wako Pure Chemical Industries, Osaka, Japan) were labeled by a random primer extension method (Megaprime DNA labelling system; Amersham Pharmacia Biotech, Buckinghamshire, UK). Northern blot analysis Total RNA in each sample was denatured with glyoxal and separated on 1.4% agarose gel by electrophoresis. The RNA samples were then transferred to nylon membranes (Hybond-N1; Amersham, Arlington Heights, IL, USA) by electroblotting. Membranes were irradiated

377

on a UV transilluminator for 5 min. Blots were preincubated for 6 h at 37°C in sealed plastic bags containing hybridization buffer (5 3 sodium chloride sodium phosphate ethylene diamine tetraacetic acid [EDTA; SSPE], pH 7.4, 5 3 Denhardt’s solution, 50% deionized formamide, 0.9% sodium dodecyl sulfate [SDS], 200 mg/ml salmon sperm DNA, and 100 µg/ml yeast RNA), and then hybridized with 32P-labeled H1K1-ATPase α-subunit or histidine decarboxylase cRNA probe for 16 h at 55°C and with 32P-labeled somatostatin or β-actin cDNA probe for 24 h at 37°C. Blots were washed with two changes of 2 3 SSPE/0.1% SDS for 30 min each, and a final wash in 0.1 3 SSPE/0.5% SDS at 65°C. The membranes were exposed to X-ray film (Kodak XAR-5) for 1–3 days at 280°C. The autoradiograms were analyzed by scanning densitometry with a dual-wavelength flying-spot scanner (Shimadzu, Kyoto, Japan). The scan was linear with added RNA over the range observed in our experiments, thus allowing quantitation of mRNA levels by this method. Slot-blot analysis To obviate possible artifacts due to mRNA degradation, the Northern blot technique was supplemented by the use of slot blots. Total RNA in duplicate was denatured by formaldehyde and applied to nylon membranes, using a slot blot procedure (Schleicher and Schuell, Keen, NH, USA). The baked membranes were prehybridized and then hybridized with 32P-labeled probes as described above. Membranes were then washed in 0.1 3 SSPE/1.0% SDS at 65°C and exposed to X-ray film (Kodak XRP) for 16 h at 280°C. The exposed films were developed and the autoradiograms were quantitated by densitometry as above. Electron microscopy Rat fundic tissue was processed into 1-mm3 sections and fixed in 2.5% glutaraldehyde diluted with cacodylate buffer (pH 7.4) at 4°C for 2 h and then postfixed in 2% osmium tetroxide at 4°C for 2 h. The sections were then dehydrated with ethanol and embedded in epoxy resin (Epon 812; TAAB, Aldermaston, UK). Ultrathin sections were prepared and double-stained with uranium acetate and lead citrate. The sections were observed with a transmission electron microscope (CM-30; Philips, Eindhoven, The Netherlands, and HS-9; Hitachi, Tokyo, Japan) at an accelerating voltage of 150 kV. For semiquantitative morphometry, images were taken of all parietal cells. Those cells with normal mitochondria and with the nucleus and apical surface present were used for analysis. A minimum of 101 cells were analyzed for each experimental condition. The classification was based on a method that has been de-

378

A. Tari et al.: Regulation of histamine in rat ECL cells

Fig. 1. A Intragastric pH and B serum gastrin concentration after treatment with oral vehicle and intraperitoneal physiological saline (control), oral omeprazole and intraperitoneal physiological saline (OPZ), oral omeprazole and intraperitoneal pirenzepine (OPZ 1 PZP), and oral vehicle and intraperitoneal pirenzepine (PZP) for 4 weeks. Values for results are expressed as means 6 SE of four to six rats. **P , 0.01 and ***P , 0.001 compared with the control group, and ††P , 0.01 and †††P , 0.001 compared with the OPZ group

scribed previously.22 The cells were classified into three categories: (1) resting cells, which had small secretory canaliculi and numerous tubulovesicles; (2) partially stimulated cells, which had expanded secretory canaliculi and also had significant numbers of tubulovesicles; these were decreased in number compared with those in resting cells; and (3) fully stimulated cells, which had large, expanded secretory canaliculi and few tubulovesicles. The ultrastructure of ECL cells was examined. ECL cells can be identified by the presence of numerous vesicles and occasional granules in the cytoplasm.4 The granules are defined as membrane-enclosed organelles displaying an electron-dense core and a thin electronlucent halo between the membrane and the dense core; the diameter of the dense core represents at least 50% of the diameter of the entire organelle.23 The vesicles are defined as membrane-enclosed organelles without a dense core, or possessing a small, often eccentrically located dense core, the diameter of the dense core being only a fraction of the diameter of the organelle.23 ECL profiles were photographed and used for morphometry only when the nucleus was included. The diameters and numbers of cytoplasmic granules and vesicles of ECL cells were quantitated with an image analyzer (SQ3100F; Photoron). The cytoplasmic granule/vesicle profile size was established by measuring the profile diameter of all such organelles in all available ECL cell profiles from each group. Intragastric pH measurement Intragastric pH was measured using a pH meter (TOA Electronic, Tokyo, Japan) by placing the electrode directly on the fundic mucosa.

Statistical analysis Values for results are presented as means 6 SEM. Statistical analysis of data was made by unpaired Mann-Whitney U-test (two-tailed) to detect significant differences. The χ2 test was used for analysis of the morphology of the cells in each set of animal experiments.

Results Effect of pirenzepine alone Treatment of rats with pirenzepine resulted in significant elevation of intragastric pH, to 4.5 (Fig. 1A). At 3 h after the administration of the last pirenzepine dose there was a significant increase in serum gastrin level to 620.5 6 58.2 pg/ml (1.5-fold the control level) (Fig. 1B). Fundic histamine concentration in control rats was 19.75 6 1.14 µg/g wet wt. Pirenzepine significantly increased histamine concentration in fundic mucosa, to 1.8-fold of the control level (36.03 6 4.40 µg/g wet wt) (Fig. 2A). The increased level of intragastric pH caused by pirenzepine treatment was significantly smaller than that caused by omeprazole treatment and that caused by combination treatment with omeprazole plus pirenzepine. The increased concentrations of both serum gastrin and fundic histamine caused by pirenzepine treatment were significantly smaller than those caused by omeprazole treatment. Pirenzepine treatment did not alter the concentration of somatostatin-like immunoreactivity in fundic mucosa (Fig. 2B). As shown in the Northern blots in Fig. 3A–D, there was no change in the level of H1-K1-ATPase mRNA (Fig. 3A), histidine decarboxylase mRNA (Fig. 3B), or

A. Tari et al.: Regulation of histamine in rat ECL cells

379

Fig. 2A,B. Concentrations of A histamine and B somatostatin-like immunoreactivity in fundic tissues after 4-week treatment with oral vehicle and intraperitoneal physiological saline (control), oral omeprazole and intraperitoneal physiological saline (OPZ), oral omeprazole and intraperitoneal pirenzepine (OPZ 1 PZP), and oral vehicle and intraperitoneal pirenzepine (PZP). Specimens were sampled 3 h after the last dose. Values represent means 6 SE of six rats. *P , 0.05; **P , 0.01; and ***P , 0.001 compared with the control group; †P , 0.05; ††P , 0.01; and †††P , 0.001 compared with the OPZ group, and §§P , 0.01 compared with the OPZ 1 PZP group

Fig. 3A–D. Northern blot analysis of A H1-K1-ATPase α-subunit mRNA, B histidine decarboxylase mRNA, C somatostatin mRNA, and D β-actin mRNA. Total fundic RNA (20 µg) from different animals was fractionated by electrophoresis on 1.4% agarose gels, electroblotted onto a nylon membrane, and hybridized with either H1-K1-ATPase, histidine decarboxylase, somatostatin, or β-actin probes, as described in the text. Autoradiograms were quantitated in a linear range by scanning densitometry. Arbitrary units used in this experiment are shown below lanes. Positions of 28S and 18S rRNA are indicated. O.D., optical density

380

A. Tari et al.: Regulation of histamine in rat ECL cells

Fig. 4A–D. Fundic levels of A H1-K1-ATPase mRNA, B histidine decarboxylase mRNA, C somatostatin mRNA, and D β-actin mRNA after treatment with oral vehicle and intraperitoneal physiological saline (control), oral omeprazole and intraperitoneal physiological saline (OPZ), oral omeprazole and intraperitoneal pirenzepine (OPZ 1 PZP), and oral vehicle and intraperitoneal pirenzepine (PZP) for 28 days. Levels of mRNA were quantified by slot-blot hybridization, as described in the text. Each data point represents the mean 6 SE of five rats. *P , 0.05; **P , 0.01, and ***P , 0.001 compared with the control group; †††P , 0.001 compared with the OPZ group, and §§§P , 0.001 compared with the OPZ 1 PZP group

β-actin mRNA (Fig. 3D), but there was an increase in somatostatin mRNA level (Fig. 3C) after pirenzepine treatment compared with the control levels. Furthermore, quantitation of mRNA levels of H1-K1-ATPase, histidine decarboxylase, and somatostatin by slot-blot hybridization and densitometry confirmed the results of Northern blot analysis (Fig. 4A, B, C). The administration of pirenzepine significantly increased somatostatin mRNA level to 1.9-fold of control in fundic mucosa. The morphological effects of pirenzepine administration are illustrated in the electron micrographs in Fig. 5 and in Tables 1 and 2.

Control animals receiving oral and intraperitoneal vehicles had 65% of parietal cells in the resting configuration and 4% classified as fully stimulated. With pirenzepine treatment, 77% of parietal cells were resting and none exhibited fully stimulated morphology. Pirenzepine administration thus clearly suppressed the parietal cell activity (Table 1). In the cytoplasm of ECL cells, pirenzepine treatment did not change the density of vesicles or granules, but it did significantly increase the diameters of both vesicles and granules, to 106.7% and 116.3% of control, respectively (Figs. 5, 6, and Table 2).

A. Tari et al.: Regulation of histamine in rat ECL cells

381

Table 1. Morphological analysis of rat parietal cells Morphological appearance, percentage of total Treatment Control Omeprazole 1 vehicle Omeprazole 1 pirenzepine Vehicle 1 pirenzepine

n

Resting

Partially stimulated

Stimulated

115 101 129 101

65 5 66 77

31 74 34 23

4 21* 0** 0**;***

* Significantly different from control (P , 0.001); ** significantly different from omeprazole 1 vehicle (P , 0.001); *** significantly different from control (P , 0.05) n, Number of cells

Table 2. Effects of omeprazole (OPZ) and pirenzepine on the ultrastructure of enterochromaffin-like cells Density (number/µm2) Vesicle Control OPZ 1 vehicle OPZ 1 pirenzepine Vehicle 1 pirenzepine

1.35 6 0.75 6 0.90 6 1.40 6

0.16 0.09** 0.14* 0.17

Diameter (nm)

Granule 0.45 6 0.07 0.18 6 0.03** 0.26 6 0.05* 0.45 6 0.05

Vesicle 400 6 600 6 509 6 427 6

7 5*** 8** 6*

Granule 153 6 4 207 6 8** 172 6 5** 178 6 4**

*P , 0.05; **P , 0.01; ***P , 0.001 vs control Values for results are expressed as means 6 SEM

Effect of omeprazole alone Omeprazole inhibits gastric H1-K1-ATPase by covalent binding to the cysteine-containing regions of the extracytoplasmic domain of the α-subunit, resulting in potent inhibition of gastric acid secretion. This potent inhibition was confirmed, as shown in Fig. 1A, which shows that the pH of the mucosa remained above 6 at 3 h after the last dose of omeprazole. Associated with this inhibition of gastric acid secretion, serum gastrin level increased markedly, by more than fourfold compared with that in vehicle-treated controls (Fig. 1B). Fundic histamine concentration was increased to 2.8fold of the control level (55.73 6 6.67 µg/g wet wt) by treatment with omeprazole (Fig. 2A). Omeprazole treatment significantly decreased the concentration of somatostatin-like immunoreactivity in the fundic mucosa, compared with the controls level and the level during treatment with pirenzepine alone (Fig. 2B). Northern blot analysis of fundic RNA is shown in Fig. 3. There was a reciprocal relationship in gene expression between fundic H1-K1-ATPase and somatostatin and between histidine decarboxylase and somatostatin; the mRNA levels of both H1-K1-ATPase and histidine decarboxylase were increased, while those of somatostatin were decreased in fundic tissues. No increase in β-actin mRNA was observed, as shown in Fig. 3D. Quantification of total mRNA by slot-blot hybridization confirmed the results of Northern analysis, as shown in Fig. 4. The administration of omeprazole thus

simultaneously caused a significant increase in both H1-K1-ATPase mRNA, to 2.4-fold of control level, and histidine decarboxylase mRNA, to 1.8-fold of control level, and a significant decrease in somatostatin mRNA, to 0.8-fold of control level. Morphologically, omeprazole-treated fundic mucosa contained only 5% resting parietal cells and 21% fully stimulated parietal cells (Table 1). Omeprazole markedly reduced the numbers of vesicles and granules in the cytoplasm of ECL cells, to 55.6% and 40% of those of the control, respectively, and significantly increased the diameter of both vesicles and granules, to 150% and 135% of those of the control, respectively (Figs. 5, 6, and Table 2).

Effect of combination treatment with omeprazole and pirenzepine The results of combination treatment with omeprazole plus pirenzepine on intragastric pH and serum gastrin are shown in Fig. 1. The effect on acid secretion was not significantly different from that of treatment with omeprazole alone. Serum gastrin was suppressed to 48% of the level caused by treatment with omeprazole alone, but was still 3.2-fold the control level and 2.7fold the level caused by pirenzepine treatment alone. The increase in histamine concentration caused by omeprazole was partially abolished by treatment with pirenzepine (Fig. 2A). Histamine concentration was

382

A

C

A. Tari et al.: Regulation of histamine in rat ECL cells

B

D

Fig. 5A–D. Electron micrographs showing enterochromaffin-like (ECL) cells of A a control rat, B an omeprazole (OPZ) treated rat, C an omeprazole 1 pirenzepine (OPZ 1 PZP) treated rat, and D a pirenzepine (PZP) treated rat. The numbers of both secretory vesicles and granules were markedly reduced in the rats treated with omeprazole and omeprazole 1 pirenzepine, while the diameters of both secretory vesicles and granules were markedly increased in the rats treated with omeprazole and omeprazole 1 pirenzepine. Bar, 1µm

Fig. 6A,B. Histograms illustrating the size distribution of A secretory vesicles and B granules in ECL cells after 4 weeks treatment with oral vehicle and intraperitoneal physiological saline (control; open circles), oral omeprazole and intraperitoneal physiological saline (OPZ; diamonds), oral omeprazole and intraperitoneal pirenzepine (OPZ 1 PZP; triangles), and oral vehicle and intraperitoneal pirenzepine (PZP; Squares). The histograms are based on the analysis of 306 vesicles and 92 granules in 13 ECL cells from OPZ-treated rats, 499 vesicles and 136 granules in 15 ECL cells from OPZ 1 PZP-treated rats, 469 vesicles and 207 granules in 18 ECL cells from PZP-treated rats, and 455 vesicles and 161 granules in 14 ECL cells from control rats

A. Tari et al.: Regulation of histamine in rat ECL cells

decreased by 18.7% (45.3 6 8.07 µg/g wet wt) by the addition of pirenzepine treatment, but it was still significantly higher than that of control, at 229.4% of control level. The concentration of somatostatin-like immunoreactivity in the fundic mucosa achieved with combination treatment was between those achieved by omeprazole treatment alone and pirenzepine treatment alone (Fig. 2B). Northern blot analysis of the RNA isolated from fundic tissue is shown in Fig. 3. There were decreases in the mRNA levels of both H1-K1-ATPase and histidine decarboxylase with combination treatment, compared with the levels obtained by omeprazole alone (Fig. 3A,B). The somatostatin mRNA level achieved was between those obtained by omeprazole alone and pirenzepine alone, as shown in Fig. 3C. No change was found in the β-actin mRNA level, as shown in Fig. 3D. Analysis of total mRNA confirmed the results of the slot-blot analysis, as shown in Fig. 4. Combination treatment with omeprazole plus pirenzepine significantly decreased the mRNA levels of H1-K1-ATPase and histidine decarboxylase, to 59% and 71% of those obtained with omeprazole alone, respectively, and significantly increased the somatostatin mRNA level, to 1.7-fold of the level obtained with omeprazole alone. No change in β-actin mRNA level was observed. Omeprazole and pirenzepine administration significantly affected parietal cell morphology; 66% were classified as resting, while only 5% were classified as resting with omeprazole treatment only. No cell was classified as stimulated in morphology with the combination treatment, while 21% were stimulated with omeprazole alone. Pirenzepine thus significantly inhibited the morphological change of parietal cells from resting to stimulated state induced by omeprazole (Table 1). In the cytoplasm of ECL cells, the densities of both the vesicles and granules achieved by treatment with omeprazole and pirenzepine were between those achieved by omeprazole alone and pirenzepine treatment alone. The diameter of vesicles obtained by combination treatment with omeprazole plus pirenzepine was also between those obtained by omeprazole alone and pirenzepine treatment alone, and was significantly decreased to 84.8% of that achieved with omeprazole alone. The diameter of granules obtained by combination treatment was significantly larger than that of control, and almost the same as that achieved with pirenzepine alone (Figs. 5, 6, and Table 2).

Discussion Rat ECL cells are stimulated by gastrin and acetylcholine through gastrin/cholecystokinin (CCK)-B receptors and muscarinic receptors, respectively, and inhibited by

383

somatostatin through somatostatin type 2 receptors.24 The expression of genes encoding histidine decarboxylase, chromogranin A, and vesicular monoamine transporter type 2 is physiologically controlled.24 Endogenous gastrin regulates histidine decarboxylase gene expression in rodent gastric corpus, and does so within the physiological range of circulating gastrin concentrations. It is therefore possible to address the question whether the well documented gastrin-mediated stimulation of histidine decarboxylase activity and histamine production can be explained largely in terms of increases in ECL cell histidine decarboxylase mRNA.21 Products of translation of chromogranin A mRNA and vesicular monoamine transporter type 2 mRNA are inserted into endoplasmic reticulum and progress via the Golgi complex to secretory granules. Cytoplasmic enzyme histidine decarboxylase generates histamine, which is sequestered into granules in exchange for protons.24 The histamine released from ECL cells then reacts with H2 receptors of parietal cells, increasing H1K1-ATPase α-subunit transcription and playing an important role in acid secretion in the fundic mucosa. In this study, omeprazole treatment markedly increased serum gastrin level in rats. The demonstrable responses of ECL cells to gastrin were increased concentrations of histamine and increased mRNA levels of both histidine decarboxylase and H1-K1-ATPase in the fundic mucosa. In ECL cells, the numbers of cytoplasmic vesicles and granules were greatly reduced 3 h after the last administration of 4 weeks of OPZ treatment, compatible with exocytosis of the vesicles and granules. However, the sizes of cytoplasmic vesicles and granules were significantly increased. In particular, the histogram illustrating the size distribution of cytoplasmic vesicles in the ECL cells of OPZ-treated rats shifted markedly rightward, and the increase in vesicle size was larger than that in cytoplasmic granule size. The vesicles are numerous, of various sizes, and often quite large; they contain a small, eccentrically located dense core. By comparison, granules are few in number and small, and have a dense core that is separated from the enclosing membrane by a thin electron-lucent halo. However, examination of serially sectioned ECL cells has revealed that the vesicles almost invariably contain an electron-dense core, suggesting that granules and vesicles are different manifestations of the same organelle.25 Andersson et al.4 speculated that granules develop into vesicles as a result of progressive accumulation of preformed histamine from the cytoplasm, and that they grow in size during this process, possibly through osmotic forces generated, at least in part, by the resulting high local concentration of histamine. Depletion of ECL cell histamine by α-fluoromethylhistidine, a powerful and irreversible inhibitor of histidine decarboxylase, is associated with loss of vesicles, supporting

384

the hypothesis that histamine is predominantly contained in vesicles and that lack of histamine prevents young and “immature” granules from developing into “mature” vesicles. The suggestion that histamine is predominantly stored in vesicles is also supported by the observation that ECL cells in the human stomach, which have a predominance of dense-core granules over clear vesicles, display relatively poor and inconsistent histamine immunoreactivity.26 In omeprazole-treated rats, the marked increase in histidine decarboxylase mRNA in fundic mucosa appeared to strongly drive the synthetic pathways of histamine in ECL cells, to develop young and “immature” granules into “mature” vesicles, and to markedly shift rightward the size distribution of cytoplasmic vesicles in the ECL cells. Muscarinic cholinergic agents, such as carbachol, inhibit both baseline and stimulated release of somatostatin from gastric D cells. Both atropine and pirenzepine reversed these effects, in a dose-dependent fashion. Pirenzepine caused a progressive parallel rightward shift in the dose-response curves for somatostatin inhibition and gastrin stimulation by carbachol, suggesting competitive inhibition.13 These findings indicate that high-affinity muscarinic receptors (M1) on D cells govern gastrin and somatostatin release from the stomach. The levels of both intragastric pH and serum gastrin induced by the muscarinic receptor (M1) antagonist pirenzepine were significantly lower than those induced by omeprazole. Compared with the control group, the level of intragastric pH was significantly higher with pirenzepine treatment, but serum gastrin level was not so altered. Pirenzepine increased the levels of somatostatin expression in both antral and fundic tissues. A reciprocal relationship exists between gastrin and somatostatin secretion in the baseline state, as well as during stimulation or inhibition of either peptide.14 In the rat, D cells extend dendrite-like cytoplasmic processes that are in close apposition with G cells.27 These findings show that pirenzepine inhibits cholinergic agonists at M1 receptors and increases somatostatin transcription in rat mucosal D cells. The participation of a muscarinic inhibitory mechanism in the regulation of somatostatin14 has also been reported by Schubert,28 in accordance with findings obtained for isolated rat, mouse, and porcine stomachs, in which muscarinic agonists and activation of intramural cholinergic neurons inhibit somatostatin secretion. Hence, binding of somatostatin to the putative somatostatin receptors on G cells inhibits the synthesis of gastrin in antrum and suppresses the elevation of the serum gastrin level induced by intragastric pH elevation. When pirenzepine was administered in addition to omeprazole to block activation of the muscarinic receptor-dependent pathway, the effects of omeprazole on serum gastrin concentration and gastrin and soma-

A. Tari et al.: Regulation of histamine in rat ECL cells

tostatin mRNA levels were partially blocked, despite sustained elevation of intragastric pH. These findings suggest that the effect of omeprazole on gastrin mRNA synthesis in rat antral G cells is reduced by pirenzepine via a mediation in somatostatin synthesis through muscarinic receptors on antral D cells.14 In fundic mucosa, the increase in the level of somatostatin by pirenzepine treatment directly suppressed the synthesis and secretion of histamine in ECL cells through somatostatin type 2 receptors. Thus, the addition of pirenzepine to omeprazole suppresses the synthesis and secretion of histamine in ECL cells, directly through somatostatin receptor type 2 (SSTR2), and indirectly through gastrin/CCK-B receptors, by reducing the increase in serum gastrin concentration. Pirenzepine may also decrease the expression of genes encoding chromogranin A and vesicular monoamine transporter type 2. Therefore, ultrastructurally, pirenzepine inhibits the development of “immature” granules into “mature” vesicles.

References 1. Walsh JH. Role of gastrin as a trophic hormone. Digestion 1990;47 (Suppl 1):11–16. 2. Watson SA, Durrant LG, Morris DL. Growth-promoting action of gastrin on human colonic and gastric tumor cells cultured in vitro. Br J Surg 1988;75:342–5. 3. Ekman L, Hanssone E, Havu N, Carlsson E, Lundberg C. Toxicological studies on omeprazole. Scand J Gastroenterol 1985;20 (Suppl 108):53–69. 4. Andersson K, Chen D, Hakanson R, Mattsson H, Sundler F. Enterochromaffin-like cells in the rat stomach: effect of αfluoromethylhistidine-evoked histamine depletion. A chemical, histochemical and electron microscopic study. Cell Tissue Res 1992;270:7–13. 5. Hakanson R, Sundler F. Do histamine-storing cells in the gastric mucosa mediate the acid-stimulating action of gastrin? In: Uvnas B, editor. Histamine and histamine antagonists. Handbook of experimental Pharmacology, vol. 97. Berlin Heidelberg New York Tokyo: Springer Verlag; 1990. p. 325–46. 6. Kahlson G, Rosengren E, Svahn D, Thunberg R. Mobilization and formation of histamine in the gastric mucosa as related to acid secretion. J Physiol 1964;174:400–16. 7. Sandvik AK, Waldum HL, Kleveland PM, Schulze-Sognen B. Gastrin produces an immediate and dose-dependent histamine release preceding acid secretion in the totally isolated, vascularly perfused rat stomach. Scand J Gastroenterol 1987;22: 803–8. 8. Prinz C, Kajimura M, Scott DR, Mercier F, Helander HF, Sachs G. Histamine secretion from rat enterochromaffin-like cells. Gastroenterology 1993;105:449–61. 9. Tari A, Yamamoto G, Sumii K, Sumii M, Takehara Y, Haruma K, et al. Role of histamine(2) receptor in increased expression of rat gastric H1/K1-ATPase α-subunit induced by omeprazole. Am J Physiol 1993;265:G752–8. 10. Karasawa H, Tani N, Miwa T. The effect of omeprazole on ultrastructure changes in gastric parietal cells. Gastroenterol Jpn 1988; 23:1–8. 11. Karam SM, Forte JG. Inhibiting gastric H1-K1-ATPase activity by omeprazole promotes degradation and production of parietal cells. Am J Physiol 1994;266:G745–58.

A. Tari et al.: Regulation of histamine in rat ECL cells 12. Negulescu PA, Reenstra WW, Machen TE. Intracellular Ca requirements for stimulus-secretion coupling in parietal cells. Am J Physiol 1989;256:C241–51. 13. Sue R, Toomy ML, Todisco A, Soll AH, Yamada T. Pirenzepinesensitive muscarinic receptors regulate gastric somatostatin and gastrin. Am J Physiol 1985;248:G184–7. 14. Tari A, Hamada M, Kamiyasu T, Fukino Y, Sumii M, Sumii K, et al. Effects of pirenzepine on omeprazole-induced gastrin gene expression in rat antral tissues. Dig Dis Sci 1996;41:1150–6. 15. Tari A, Hamada M, Kamiyasu T, Fukino Y, Sumii M, Haruma K, et al. Effects of pirenzepine on omeprazole-induced hypergastrinemia and acid suppression in peptic ulcer patients. J Gastroenterol 1996;31:167–70. 16. Sumii K, Sumioka M, Yoshihara M, Tari A, Haruma K, Kajiyama G. Somatostatin in gastric juice in normal subjects and patients with duodenal ulcer. Dig Dis Sci 1992;37:1020–4. 17. Yamatodani A, Fukuda H, Wada H, Iwaeda T, Watanabe T. High-performance liquid chromatographic determination of plasma and brain histamine without previous purification of biological samples: cation-exchange chromatography coupled with post-column derivatization fluorometry. J Chromatogr 1985;344: 115–23. 18. Chomczynski P, Sacchi N. Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156–9. 19. Melton DA, Krieg PA, Rebagliati MR, Maniatis T, Zinn K, Green MR. Efficient in vitro synthesis of biologically active RNA and RNA hybridization probes from plasmids containing a bacteriophage SP6 promoter. Nucleic Acids Res 1984;12:7035– 56.

385 20. Shull GE, Lingrel JB. Molecular cloning of the rat stomach (H11K1)-ATPase. J Biol Chem 1986;261:16788–91. 21. Dimaline R, Sandvik AK. Regulation of histidine decarboxylase mRNA abundance by endogenous gastrin in the rat. In: Walsh JH, editor. Gastrin. New York: Raven; 1993. p. 273–84. 22. Stachura J, Konturek SJ, Cieszkowsky M, Debros W, Zakrewska J, Konturek J. Comparison of the effect of omeprazole—a substituted benzimidazole—and ranitidine—a potent H2 receptor antagonist—on histamine-induced gastric acid secretion and the ultrastructure of canine parietal cells. Hepatogastroenterology 1983;30:205–10. 23. Chen D, Monstein H-J, Nylander A-G, Zhao C-M, Sundler F, Hakanson R. Acute responses of rat stomach enterochromaffinlike cells to gastrin: secretory activation and adaptation. Gastroenterology 1994;107:18–27. 24. Dockray GJ, Varro A, Dimaline R. Gastric endocrine cells: gene expression, processing, and targeting of active products. Physiol Rev 1996;76:767–98. 25. Hakanson R, Chen D, Andersson K, Monstein H-J, Zhao CM, Ryberg B, et al. The biology and physiology of the ECL cells. Yale J Med Biol 1994;67:123–34. 26. Delwaide J, Belaiche J, Courtoy R, Vivario M, Louis E, Gast P, et al. Ultrastructural demonstration of histamine in human enterochromaffin-like cell granules. Gut 1991;32:834. 27. Saffouri B, Weir GC, Bitar KN, Makhlouf GM. Gastrin and somatostatin secretion by perfused rat stomach: functional linkage of antral peptides. Am J Physiol 1980;238:G495–501. 28. Schubert ML. Neural and paracrine regulation of gastrin secretion. In: Walsh JH, editor. Gastrin. New York: Raven; 1993. p. 129–37.