PATIENTS AND STUDY DESIGN

Table 1 shows the mean age, age range, and gender of the subjects examined in this study including five H pylori negative healthy volunteers, 47 H pylori positive patients with DU disease, and 35H pylori positive patients with gastric ulcer (GU) disease.

Endoscopic examination was used to diagnose the diseases and determine their stages. The ulcer stage was classified as active, healing, or scarring according to the classification of Sakita and colleagues9. Briefly, these stages are defined as follows: active, the surrounding mucosa is oedematously swollen and little regenerating epithelium is seen endoscopically; healing, the white coating becomes thin, and the regenerating epithelium extends into the ulcer; scarring, the regenerating epithelium completely covers the floor of the ulcer. Biopsy specimens were collected from the gastric antrum and the ulcer margin in cases of DU and GU disease respectively. In the case of GU disease, biopsy sampling from the antrum was not performed because it became difficult if the ulcer was located on the antral region. Infection with H pylori was confirmed by the rapid urease test on biopsy samples, and the number ofH pylori colonising the sample was determined by using Skirrow agar plates as previously reported.10 Some of the H pylori positive subjects including eight with DU (four active and four healing) and 10 with GU (six active and four healing) received a two week course of eradication therapy. The following regimen was used: 20 mg omeprazole/day, 1000 mg amoxicillin/day, and 400 mg clarithromycin/day. Just before and three months after this therapy, an endoscopic examination was performed to confirm that the ulcer was healed and also to collect biopsy samples for bacteriological examination. Eradication of infection by antibiotic therapy was defined as successful when both the rapid urease test of the biopsy sample and the bacterial culture of the sample using Skirrow plates were negative.

ANALYSIS OF APOPTOSIS

Apoptosis was investigated by cell cycle analysis using flow cytometry.11 Briefly, single gastric epithelial cells were fixed with 2% paraformaldehyde, permeabilised with 0.1% Nonidet P40, and treated with 0.05 mg/ml RNase A for 30 minutes at 37°C. After being washed, the cells were stained with propidium iodide and then analysed using a flow cytometer. The cells in a discrete subpopulation of signals under the G₀/G₁ cell cycle region (subdiploid cells) were designated as undergoing apoptosis. Single gastric epithelial cells were prepared from biopsy specimens by the method of Kobayashi and colleagues.12 In brief, the gastric tissue specimens were minced into pieces and stirred into Hanks balanced salt solution containing 0.1% (w/v) collagenase A (Boehringer Mannheim GmbH, Mannheim, Germany) for 60 minutes at room temperature. The tissue suspension was then filtered through a cotton gauze, and the filtrate was centrifuged at 750 g for five minutes to harvest single cells.

Examination of apoptosis in situ using the method of terminal uridine deoxynucleotide nick end labelling (TUNEL) histochemistry was performed as described by Gavrieli et al.13 Briefly, 4–6 μm sections were cut from paraffin embedded tissue blocks and mounted on glass slides. The sections were deparaffinised through xylene and alcohol, washed, and then digested with proteinase K 20 μg/ml for 15 minutes at room temperature, followed by washing and blocking of endogenous peroxidase with 2% hydrogen peroxide for five minutes.

After preincubation with terminal transferase reaction buffer containing 200 mM potassium cacodylate, 25 mM Tris/HCl, pH 6.6, and 0.25 mg/ml bovine serum albumin (Boehringer Mannheim) for 10 minutes, the sections were incubated at 37°C for 90 minutes with the same buffer containing, in addition, 1 mM cobalt chloride, terminal transferase (0.5 U/μl), and 0.4 μM digoxigenin-11-deoxyuridine triphosphate (digoxigenin-11-dUTP). The reaction was terminated by the addition of a solution containing 300 mM NaCl and 30 mM sodium citrate, and the slides were washed in water. Incorporated digoxigenin-11-dUTP was detected by incubation with peroxidase conjugated Fab fragments of anti-digoxigenin (Boehringer Mannheim) at a concentration of 1:300 in 100 mM Tris/HCl buffer, pH 7.5, containing 150 mM NaCl at room temperature for 30 minutes. The slides were then washed in water, counterstained with methyl green, and mounted.

FLOW CYTOMETRY

To analyse the surface cell marker, single gastric epithelial cells were washed once with Hanks balanced salt solution and then stained with anti-epidermal growth factor (EGF) receptor mouse monoclonal antibody (clone 6F1; Immunotech, Marseille, France) according to the manufacturer’s instructions. Bound antibodies were disclosed with fluorescein isothiocyanate conjugated anti-mouse IgG antibody (Tago Inc, Burlingame, California, USA). The machine was gated to include single cells but to exclude any cell debris and clumps of cells according to forward and side scatter patterns.

CHARACTERISATION OF H PYLORI STRAINS

H pylori was isolated from the gastric biopsy materials. The wet weight of the biopsy material was measured before the procedure in order to determine bacterial number. The bacteria were grown on Skirrow plates, so that the number of colony forming units in the biopsy material (number/g wet tissue) could be counted, and then inoculated into a Brucella broth containing 5% fetal calf serum in a microaerophilic atmosphere at 37°C to prepare a fresh sample of bacteria as previously described.10 To determine the urease activity of H pylori, freshly prepared H pylori was suspended at 1 × 10⁹ colony forming units/ml in phosphate buffered saline. Then 0.1 ml of this bacterial solution was mixed with 0.9 ml of a urease substrate solution (50 mM urea in 100 mM acetate buffer), incubated at 37°C for 20 minutes, and then applied to an Ammonia-Test-Wako Kit (Wako, Osaka, Japan) to measure the ammonia concentration according to the manufacturer’s instructions. The amount of protein was measured using a Bio-Rad Protein Assay Kit (Bio-Rad Laboratories, Hercules, California, USA). In a preliminary experiment, the amount of protein in a bacterial solution correlated closely with the colony forming units contained in the solution so that the urease activity per protein can also be considered as the urease activity per bacterial body. To determine the vacuolating cytotoxin activity, 1 × 10⁷ colony forming units of H pyloriwere inoculated into 10 ml Brucella broth containing 5% fetal calf serum and then incubated for seven days as described above. The culture was then diluted serially in Dulbecco’s modified Eagle’s medium containing 10% fetal calf serum and 10 mM ammonium chloride before being added to RK13 cells.13 ,14 The maximum dilution at which more than 80% of RK13 cells become vacuolated was regarded as positive and this value was recorded.

HISTOLOGICAL ANALYSIS

Sections from biopsy samples were stained with haematoxylin and eosin. Multiple high power fields were examined by one experienced pathologist blinded to the patient’s clinical diagnosis. The histological features of inflammation and epithelial damage were graded from 1 to 3 for each biopsy specimen. The scored features included lymphocyte infiltration, neutrophil infiltration, intestinal metaplasia, and epithelial erosion. Lymphocyte infiltration was graded as follows: 0, lymphocytes present in normal numbers; 1, a minor and equivocal increase in the lymphocyte number; 2, a moderate increase in the lymphocyte number; 3, a maximal or submaximal increase in the lymphocyte number in most of the section. Neutrophil infiltration was graded as follows: 0, neutrophils difficult to find; 1, a minor increase in neutrophils but they were not in groups and did not invade the gland necks; 2, the neutrophils definitely increased, forming groups or invading the gland necks; 3, groups of neutrophils were visible in nearly every field. Intestinal metaplasia and epithelial erosion were graded as follows: 0, not found; 1, found.

STATISTICAL ANALYSIS

Statistical analysis was by Student’st test, the pairedt test, and the scattergram plot. Histological scores were non-parametric and were therefore analysed by the Mann-Whitney U test. Numerical data in the text represent the mean (SEM) except where indicated otherwise.