H. pylori was isolated from the antral biopsy specimens of patients with duodenal ulcer at Seoul National University Hospital. Biopsy specimens were placed in Brucella broth (Difco Laboratories, Detroit, MI, USA) immediately and homogenized with a tissue grinder. The homogenate was then inoculated on selective agar [GC medium base (Difco Laboratories). 0.024% yeast extract, 1% hemoglobin, 1% IsoVitaleX, 5 mg/L vancomycin, 1 mg/L mycostatin, 5% sheep blood]. The plates were incubated for 5–7 days at 37°C under microaerophilic conditions (5% O₂, 10% CO₂ and 85% N₂) in a CO₂ incubator (Napco 5410, Tualatin, Oregon, USA). Pure culture isolates were examined by Gram stain and biochemical assay such as urease test¹⁴⁾. Isolates were finally confirmed as members of the genus Helicobacter by a polymerase chain reaction (PCR) as described below. For a long-term storage, isolates were stored in Brucella broth containing 15% (v/v) glycerol and kept at −70°C.

DNA was extracted from the H. pylori isolates with proteinase K, sodium dodecyl sulfate and hexadecyltrimethyl ammonium bromide (Sigma, St. Louis, MO, USA). The cell lysate was extracted in sequential steps with equal volumes of phenol, phenol/chloroform/isoamylalcohol (25:24:1) and chloroform. DNA was then precipitated with isopropanol. The DNA pellet was washed with 70% ethanol, dried and resuspended in sterile Tris-EDTA buffer (pH 8.0).

Total RNA of H. pylori was extracted by the acid guanidinium thiocyanate-phenol-chloroform method¹⁵⁾ (4 M guanidinium thiocyanate, 25 mM sodium citrate, pH 7.0; 0.4% sarcosyl, 0.1 M 2-mercaptoethanol). The RNA was dissolved in diethyl pyrocarbonate-treated distilled water, and quantitated at absorbance of 260 nm. One μg of each RNA of H. pylori isolated from human and mouse was used for single-strand cDNA synthesis with oligo(dT)₁₅ primer (Promega, Madison, WI, USA) and Moloney murine leukemia virus reverse transcriptase (Gibco BRL, Gaithersburg, MD, USA) as described previously¹⁶⁾.

A number of PCRs were performed as described previously¹⁷⁾ to characterize the H. pylori isolates obtained from humans or mice. These were Helicobacter-specific PCR, ureA, cagA and vacA PCRs and random amplified polymorphic DNA (RAPD) PCR¹⁸⁾. Each PCR primer was designed on the basis of published sequences of H. pylori ^1,16,19,20) as shown in Table 1. The primer sequence used for RAPD PCR was based on the publication by Lee et al.¹²⁾ (5′-AACGCGCAAC-3′). Amplification of H. pylori genomic DNA sequences was carried out in a volume of 50 μl containing PCR buffer [50 mM KCl, 10 mM Tris-HCl (pH 8.3)], 1.5 mM MgCl₂, 200 M dNTP, 0.5 M primers, 2 U of Taq polymerase (Perkin-Elmer, Norwalk, CT, USA) and 100 ng of bacterial DNA or cDNA. Each reaction mixture was amplified for 33–39 cycles (shown in Table 2). In RAPD PCR, the MgCl₂ concentration was increased to 3 mmol/L and 20 pmol of a single primer was used. A sample omitting DNA or cDNA was included in every reaction as a negative control. Each PCR was performed with hot-start procedure^21,22) and the final extension was performed at 72°C for 10 min after the completion of the amplification cycles using a thermal cycler (Perkin Elmer). PCR products were separated in 2% NuSieve agarose gel (FMC Bioproducts, Rockland, ME, USA) and identified using ethidium bromide stains.

Twenty-two, specific pathogen-free, 6-week-old female BALB/c mice were divided into two groups (4 of control and 18 of experiment). Three of frozen cagA+/vacA+ H. pylori strains (strain # 99, # 232 and # 234) and one of fresh clinical cagA+/vacA+ H. pylori isolates (strain # 7) were used in this study. Each H. pylori strain was cultured under microaerophilic conditions and was harvested in the sterile phosphate buffered saline (PBS, pH 7.4) and mixed by equal density. Mice were infected with the H. pylori mixture as described previously²³⁾. Briefly, all animals were fed on a commercial diet and given water ad libitum. Mice were fasted overnight except for water. Mice were inoculated intragastrically with 10⁹ colony forming units (CFUs) of H. pylori mixture or PBS after treatment with 0.2 ml of 0.2 M NaHCO₃ intragastrically to neutralize gastric acidity. These procedures were repeated two or more times with a 2-day interval.

Mice were sacrificed 1, 2, 4 or 6 weeks after the last inoculation. The stomach of each mouse was bisected longitudinally. A specimen for rapid urease test (CLO test, Delta West Pty Ltd, Western Australia) was obtained from one half of the stomach and the remains of it were used for gastric scrapings for H. pylori cultures. They were fixed in buffered 10% formalin, using standard procedures, embedded in paraffin, sectioned at 4 ml, and stained with hemotoxylin & eosin for histology and Warthin-Starry silver to assess the level of bacterial colonization²⁴⁾. The glandular mucosae of the body, antrum and pylorus were examined histologically for a variety of inflammatory responses, epithelial changes and the presence of H. pylori.

ELISA was performed for the assessment of immune response in H. pylori-infected mice as described previously^25,26) with minor modification. Briefly, to obtain the antigen to coat the microtiter plate, colonies of H. pylori were sonicated for three 30 sec bursts with 30 sec resting periods in an MSE Soniprep 150, and ultracentrifuged at 100,000 × g for 60 min (Beckman TL-100, Palo Alto, CA, USA). The antigen was diluted in 0.1 M carbonate buffer (pH 9.6) to a final concentration of 10 μg/ml. Microtitre plates were coated with 100 μl/well of antigen solution and incubated overnight at 4°C, washed three times with PBS containing 0.05% Tween-20 and then blocked with 200 μl/well of 1% bovine serum albumin in PBS/Tween-20 at room temperature for 2 hours. After washing the plates, 100 μl of each serum sample diluted 1:75 in PBS were added to wells in duplicate and the plates were incubated at room temperature for 2 hours. The plates were then washed three times and incubated with 100 μl/well of a goat anti-mouse IgG peroxidase conjugate (Pierce, Rockford, IL, USA) and diluted 1:700 in 1% BSA-PBS at room temperature for 2 hours. After three washings, 100 μl/well of a substrate solution containing p-nitrophenyl phosphate in diethanolamine-MgCl₂ buffer was added to each well. The reaction was stopped with 50 μl/well of 2 N H₂SO₄, and read the absorbance read using Wellscan microplate reader at OD450 nm (Dynatech MR 700, Alexandria, VA, USA).

Specific pathogen-free BALB/c female mice were inoculated with H. pylori. Mice were sacrificed 1, 2, 4 and 6 weeks after the last inoculation. Gastric specimens from infected mice were processed for urease test and histologic evaluation. The positivity of CLO tests in gastric specimens obtained from mice sacrificed one and two weeks after the last inoculation was only 40% but 75% in those from 4 and 6 weeks. (Table 3). As presented the above, the positivity of CLO test was gradually increased from 4 weeks after treatment with H. pylori, which indicates that the colonized H. pylori proliferated slowly and was located in the stomach. The CLO test in the control group was negative during the experimental period.

In the first week after the last inoculation, no definite presence of bacteria and infiltration of inflammatory cells were shown in H. pylori-infected group (Fig. 1). However, the inflammatory infiltrates of lymphocytes, plasma cells and polymorphonuclear leukocytes in the antrum and corpus mucosa were gradually increased from 2 weeks after the last inoculation of H. pylori. At six weeks after the last inoculation, the gastric lesions were characterized by definite inflammatory infiltrates accompanied by the disruption of gastric architecture (Fig. 2, 3) and distinct H. pylori colonization (Fig. 4). Control mice showed no evidence of significant infiltration of inflammatory cells during the entire experimental period (Fig. 5).

Isolates from infected mice showed typical morphology and biochemical characteristics of H. pylori as the same as the original H. pylori strain (data not shown). Furthermore, the Helicobacter-specific PCR showed that the mouse and human isolates were identical to H. pylori (Fig. 6). The original human clinical isolates and the mouse isolates were also positive for cagA and vacA by PCR (Fig. 7). A comparison of the genomic DNA by RAPD PCR was performed for the mouse isolates and four original human clinical isolates which were used to infect the mouse. The mouse isolates gave an identical band pattern to only one (strain # 7) of the four human isolates (Fig. 8). These results suggest that strain # 7 may be the one that is able to colonize the mouse gastric mucosa among all the human isolates tested.

The experimentally infected mice showed serum antibody response to the colonizing strain that could be detected by an ELISA of mice sera from the first week of infection. Anti-H. pylori IgG antibody levels increased markedly from the second week after the last inoculation with H. pylori and reached the plateau after the fourth week. In contrast, the negligible serum antibody response was shown in the control group during the entire experiment period (Fig. 9).