It is now accepted that Helicobacter pylori infection of the

Transcription

1 GASTROENTEROLOGY 1999;117: Gastric Ulcers in SCID Mice Induced by Helicobacter pylori Infection After Transplanting Lymphocytes From Patients With Gastric Lymphoma KENJI YOKOTA,* KEITA KOBAYASHI, YOSHIRO KAWAHARA, SHYUNJI HAYASHI, YOSHIKAZU HIRAI, MOTOWO MIZUNO, HIROYUKI OKADA, TSUNEATU AKAGI, TAKAO TSUJI, and KEIJI OGUMA* *Department of Bacteriology, Second Department of Pathology, and First Department of Internal Medicine, Okayama University Medical School, Okayama, Japan; and Department of Microbiology, Jichi Medical School, Tochigi, Japan Background & Aims: Several studies have indicated that host factors are important in Helicobacter pylori induced gastroduodenal diseases. We examined the pathological role of host immune responses in H. pylori infection by reconstituting components of the human immune system into severe combined immunodeficient (SCID) mice by transplantation of peripheral blood mononuclear cells (PBMCs) from H. pylori infected patients. Methods: PBMCs obtained from patients with mucosa-associated lymphoid tissue (MALT) lymphoma were injected intraperitoneally into SCID mice, designated MALToma-hu-SCID mice. One month after transplantation, H. pylori was administered orally to the mice. The mice were killed and examined for pathological changes and immunologic features. Results: Human lymphocytes were detected in hu-scid mice, and T- and B-cell functions were preserved for 1 month. Administration of H. pylori led to gastric ulcers with bleeding in the MALToma-hu- SCID mice. The gastric mucosa of control mice injected with Escherichia coli or transplanted with PBMCs from patients with peptic ulcers or gastritis or from healthy volunteers showed no pathological changes. Conclusions: Host immune responses against H. pylori appear to be involved in the development of gastric ulcers in patients who have MALT lymphoma. It is now accepted that Helicobacter pylori infection of the stomach can cause chronic active gastritis and duodenal and gastric ulcers and that infection may lead to a predisposition to gastric cancer. 1 3 The prevalence of H. pylori infection in adults has been estimated between 30% and 50%; however, gastroduodenal disease does not develop in all those infected. It is possible that an ulcerogenic strain causes peptic ulcer disease and that associated toxins injure gastric epithelial cells. 4 9 Host factors also seem to be important to the pathogenicity of H. pylori infection. 10 Large variations in disease severity were observed after infection of various inbred mouse strains with single isolates of H. felis 11 and H. pylori. 12 Therefore, it is important to investigate the role of host immunologic reactions against H. pylori. It has also been reported that gastric mucosa-associated lymphoid tissue (MALT) lymphoma is associated with H. pylori infection. 13,14 The regression of MALT lymphoma in response to antibiotic treatment suggests that the malignant lymphocytes require long-term stimulation by H. pylori to remain viable. 15,16 Also, gastric ulcers sometimes develop in patients with MALT lymphoma. We hypothesized that H. pylori specific lymphocytes exist in some patients with MALT lymphoma and that these cells are important to the development of gastroduodenal lesions. This study shows that gastric ulcers and erosions can be induced by immunologic reactions against H. pylori in the severe combined immunodeficient (SCID) mouse model. We show that the addition of peripheral blood mononuclear cells (PBMCs) from patients with MALT lymphoma is necessary for the development of ulcers after H. pylori infection in these mice. Materials and Methods Patients All patients with MALT lymphoma were found to be H. pylori positive by culture, serology, and histology. Diagnosis of low-grade MALT lymphoma was assessed according to the REAL classification based on histopathology. 17 H&E-stained sections were prepared to confirm the presence of B-cell follicle centers and mantle zones surrounded by neoplastic infiltrates of centrocyte-like cells and the presence of lymphoepithelial lesions. Immunohistochemistry was also performed by using anti-immunoglobulin (Ig), Ig, UCHT-1, and Pan B antibodies (Dako, Glostrup, Denmark). Patients with gastric ulcers and gastritis were diagnosed by endoscopy and were Abbreviations used in this paper: ELISA, enzyme-linked immunosorbent assay; FCS, fetal calf serum; HGC, human gastric cell line; HSP, heat-shock protein; MALT, mucosa-associated lymphoid tissue; PBMS, peripheral blood mononuclear cell; PWM, pokeweed mitogen; SCID, severe combined immunodeficient by the American Gastroenterological Association /99/$10.00

2 894 YOKOTA ET AL. GASTROENTEROLOGY Vol. 117, No. 4 Table 1. Proliferation of PBMCs From Patients With MALT Lymphoma by H. pylori Sonicates and Effects of H. pylori Injection Into hu-scid Mice Group Origin of PBMCs (no. of injected cells/1 mouse) Proliferation a of PBMCs by H. pylori No. of injected bacteria Pathological future of stomach Bleeding b Ulcer or erosion c I. MALToma-hu-SCID 1 MALToma ( ) 3.20 H. pylori /6 6/6 2 MALToma ( ) 2.50 H. pylori /5 4/5 3 MALToma ( ) 2.01 H. pylori /5 4/5 4a MALToma ( ) 2.32 H. pylori /3 3/3 4b E. coli /2 0/2 II. Gastric-ulcer hu-scid 5 Gastric ulcer ( ) 0.95 H. pylori /4 0/4 6 Gastric ulcer ( ) 1.05 H. pylori /5 0/5 III. Gastritis-hu-SCID 8 Gastritis ( ) 1.20 H. pylori /4 0/4 9 Gastritis ( ) 0.97 H. pylori /5 0/5 IV. Healthy control hu-scid 8 Healthy ( ) 1.00 H. pylori /5 0/5 9 Healthy ( ) 0.98 H. pylori /5 0/5 10 Healthy ( ) 1.02 H. pylori /5 0/5 a Proliferation of the PBMCs is expressed as stimulation index (counts per minute [cpm] with H. pylori sonic extract divided by cpm without antigen). b Macroscopic findings of gastric mucosa. c Histopathologic findings. found to be H. pylori positive by culture, urease tests, and serology. Healthy volunteers not infected with H. pylori had only low anti H. pylori antibody titers detected in their sera. Construction of hu-scid mice. Peripheral blood samples (20 ml) were obtained from patients with MALT lymphoma, gastric ulcer, or gastritis and from healthy volunteers. Informed consent was obtained from all human subjects. PBMCs were isolated by Ficoll-Hypaque gradient centrifugation and suspended in RPMI 1640 medium (approximately cells/ml). Aliquots of the cell suspensions (0.2 ml) were injected intraperitoneally into groups of 6 7-week-old SCID mice (C.B.-17 scid/scid; Clea Japan Inc., Tokyo, Japan), resulting in 4 groups designated MALToma-hu-SCID (group I), gastric ulcer hu-scid (group II), gastritis-hu-scid (group III), and healthy-hu-scid mice (group IV). Mice were then kept in a clean isolator for 4 weeks to establish hu-scid mice. Oral inoculation and histopathology. H. pylori (ATCC 43504) was cultured on Brucella blood agar for 5 days and suspended in a viscous solution of Brucella broth containing 1% carboxymethyl cellulose. 18 The hu-scid mice established as described previously were inoculated with live bacteria ( or ) and killed after 24 hours. The stomach, intestine, liver, and spleen were removed and examined macroscopically. Tissues were then fixed with 4% buffered formalin, dehydrated, and embedded in paraffin. Serial sections were cut, mounted, and stained with H&E. Sampling techniques were standardized so that 2 longitudinal samples contained the forestomach, fundic mucosa, antrum, and duodenum. Erosive lesions, edema, and lymphoid follicles in the longitudinal extents of the stomach were assessed microscopically. Intensity of gastric inflammation was scored as follows: grade 0, no inflammatory changes in the gastric mucosa; grade 1, mild infiltration of inflammatory cells that were confined to the junction of the mucosa and submucosa; grade 2a, marked diffuse infiltration of inflammatory cells without lymph follicle formation, involving both the mucosa and submucosa; and grade 2b, localized inflammation with lymph follicle formation involving the mucosa and/or submucosa. Immunostaining was also performed by using CD45R/O (UCHL-1) as a human T-cell marker and CD20 as a human B-cell marker (Dako). Sections of gastric ulcer were also stained with anti-human IgA, IgG, and IgM (Dako). Proliferation of PBMCs from human and spleen cells from hu-scid mice with H. pylori antigens. PBMCs from patients and healthy volunteers and spleen cells ( in 0.1 ml of RPMI 1640 medium containing 10% fetal calf serum) from hu-scid mice were cultured in flat-bottomed 96-well microtiter plates for 72 hours in a 5% CO 2 incubator in the presence or absence of H. pylori antigens. [ 3 H]Thymidine (1 µci/well) was added for the last 16 hours of culture, and [ 3 H]thymidine uptake was measured by liquid scintillation in a beta counter. Sonicated cell extracts, urease, heat-shock protein (HSP), and other protein antigens were prepared as follows: whole H. pylori (ATCC 43504) and Escherichia coli (JC-2) cells (1 g each in 10 ml of ice-cold distilled water containing 1 mmol/l phenylmethylsulfonyl fluoride) were disrupted using an ultrasonic sonicator (Astrason, Farmingdale, NY). The sonicated cells were centrifuged at 20,000g for 20 minutes, and the supernatants were used as sonic extracts of H. pylori and E. coli. A sample of the H. pylori sonicated extract was heated at 100 C for 5 minutes in the presence of 2-mercaptoethanol. After sodium dodecyl sulfate polyacrylamide gel electrophoresis and staining with Coomassie brilliant blue, the bands with molecular weights of 66 kilodaltons (HSP), 60 kilodaltons (urease B subunit), 56 kilodaltons (function unknown), 30 kilodaltons (urease A subunit), and 25

3 October 1999 H. PYLORI INDUCED GASTRIC ULCERS BY IMMUNITY 895 kilodaltons (function unknown) were electroeluted by using Eutrap (Schleicher & Schuell, Dassel, Germany). The identities of the eluted proteins were checked by N-terminal sequencing, 19 and concentrations were measured by absorbance at optical density of 280 nm. Pokeweed mitogen (PWM; Sigma Japan, Tokyo, Japan) was used as control mitogen. Sonicated extracts, component antigens, and PWM were added to spleen cell cultures at final concentrations of 10, 5, and 5 µg/ml, respectively. Flow-cytometric analysis. Patient PBMCs and spleen cells from hu-scid mice were obtained and incubated for 1 hour on ice with CD3 fluorescein isothiocyanate (UCHT-1; Coulter Corp., Hialeah, FL) and B4-RD1 (B4; Coulter). Cells were washed, fixed with 2% formaldehyde in phosphatebuffered saline (PBS; ph 7.2), and analyzed by FACScan (Becton Dickinson Immunocytometry system, San Jose, CA). Human antibody to H. pylori and gastric cells in hu-scid mice. Human IgG antibodies against bacteria and HGC-27 cells (human gastric cancer cell line) were measured by enzyme-linked immunosorbent assay (ELISA). 20,21 Microtiter plates (96-well) were coated with sonicated H. pylori protein (4 µg/well) and blocked with 10% skim milk in PBS. Plates were reacted with 1:2 diluted mouse sera for 2 hours at room temperature. After washing, peroxidase-labeled anti-human IgG (Dako) and substrate (o-phenylenediamine 1 mg/ml in citrate buffer, ph 5.5) were successively reacted at room temperature. Optical densities were measured at 490 nm by an ELISA reader (Bio-Rad, Hercules, CA). To detect autoantibodies to human gastric cells, cell ELISA was performed using HGC-27 cells. HGC-27 cell monolayers were cultured in 96-well microtiter plates and washed, and cells were fixed with 2% formalin-pbs for 2 hours at room temperature. After washing and blocking with 10% skim milk-pbs, plates were reacted with sera as described previously. Results Proliferation of PBMCs From Patients and Pathology of hu-scid Mice After H. pylori Inoculation PBMCs were obtained from patients with MALT lymphoma, gastric ulcer, and gastritis and from healthy volunteers and then proliferated by H. pylori sonicates. The proliferations of PBMCs from patients with MALT lymphoma were significantly higher (P 0.016) than those of the other PBMCs (Table 1). PBMCs were injected intraperitoneally into SCID mice to establish 4 groups of hu-scid mice as described in Materials and Methods. Hu-SCID mice were orally inoculated with H. pylori or E. coli cells (Table 1). Gastric mucosal bleeding was macroscopically observed in MALToma-hu-SCID mice (group I), and erosions were commonly observed in histopathologic examination (Table 1 and Figure 1A). In some cases, tissue damage with edema extended to the muscular layers of the mucosa and appeared to be typical of ulcer lesions (Figure 1B). In contrast, the gastric mucosa of all Figure 1. Histology of gastric mucosa in MALToma-hu-SCID mice. (A and B) H&E staining; (C and D) immunohistochemistry. (A) Erosion (Er) with bleeding, originating from gastric body mucosa in MALToma-hu- SCID mice. (B) Ulcerative lesion (U) with edema (Ed) of fundic mucosa. (C) Ulcerative lesions stained with anti-human IgG antibody. (D) Normal mucosa in healthy-hu-scid mice stained with anti-human IgG antibody. Original magnifications: A, 100 ; B, 200 ; C and D, 400. hu-scid mice in groups II IV were macroscopically normal. Infiltration of inflammatory cells was not severe in any of the cases, and there were no significant differences between the groups; the intensity of inflammation was less than grade 1. Immunostaining of gastric tissue specimens with anti-human IgG revealed staining only in epithelial cells around the ulcer lesions in the MALToma-hu-SCID mice (Figure 1C and D), indicating that only these epithelial cells are reacted by human IgG. Detection of human lymphocytes in hu-scid mice. The presence of human lymphocytes in the gastric mucosa and spleens of SCID-hu mice was investigated by

4 896 YOKOTA ET AL. GASTROENTEROLOGY Vol. 117, No. 4 immunostaining with anti-cd45r/o and anti-cd20 antibodies. Although many human B and T cells were detected in the spleens, only a small number of T cells were detected in the gastric mucosa. Lymphocyte accumulation was sometimes observed in the gastric mucosa of the forestomach, although follicles with germinal centers were not detected (data not shown). Spleen cells from hu-scid mice were analyzed by 2-color flow cytometry. Human CD3 and CD19 lymphocytes were detected in spleens from hu-scid mice (Figure 2). Lymphocytes from all patients and healthy volunteers populated the hu-scid mice for at least 1 month after transplantation. Proliferation of mouse spleen cells with H. pylori antigens. The ability of the human lymphocytes detected in the murine spleen to react with H. pylori was confirmed by proliferation assays. Proliferation was expressed as a proliferation index (cpm with antigens divided by cpm without antigens). Uptake of [ 3 H]thymidine was significantly increased in spleen cells from MALToma-hu-SCID mice compared with those from healthy-hu-scid mice after stimulation with sonicated H. pylori extracts (MALToma, ; control, ; P 0.05), whereas stimulation with sonicated E. coli extract (MALToma, ; control, ) or PWM (MALToma, ; control, ) yielded no significant increase (Figure 3). Of the component-specific antigens tested, HSP (66-kilodalton) (MALToma, ; control, ), 60-kilodalton (urease B subunit) (MALToma, ; control, ), 56-kilodalton (MALToma, ; control, ), and 25-kilodalton (MALToma, ; control, ) antigens showed significant differences in stimula- Figure 2. Detection of human lymphocytes in hu-scid spleen cells. Spleen cells from MALToma-hu-SCID mice and PBMCs from patients with MALT lymphoma (group I-1) were reacted with anti CD3-fluorescein isothiocyanate (UCLH-1; human T-cell marker) and anti CD19-RD1 (B4; human B-cell marker) and analyzed by 2-color flow cytometry. Numbers in the lower right and upper left quadrants indicate percentages of CD3 and CD19 cells, respectively.

5 October 1999 H. PYLORI INDUCED GASTRIC ULCERS BY IMMUNITY 897 Figure 3. Proliferation assay of spleen cells from hu-scid mice. Spleen cells from MALToma-hu-SCID mice (n 21) from group I and healthy-hu-scid mice (n 9) from group IV were stimulated with sonicated extracts of E. coli and H. pylori and with 66-kilodalton (HSP), 60-kilodalton (urease B subunit), 56-kilodalton, and 25-kilodalton electroeluted antigens of H. pylori. Columns indicate mean SE. Data are expressed as stimulation indices (cpm with antigens/cpm without antigens). *P 0.05, statistically significant, Student t test. response to H. pylori antigen. We previously reported that antigen stimulation with HSP60 of H. pylori might enhance the development of gastric lesions in patients with MALT lymphoma. 21 The lymphocytes were proliferated with HSP60, urease-b fragment, and unknown antigens (56- and 25-kilodalton) of H. pylori. These results suggest that many lymphocytes that recognized H. pylori were present only in patients with MALT lymphoma and were essential for the development of gastric ulcers in MALToma-hu-SCID mice. Human antibodies were also clearly detected in hu- SCID mice sera (Figure 4A). Antibody levels against H. pylori were increased in ulcer-, gastritis-, and MALTomahu-SCID mice, whereas autoantibodies to human gastric cells were detected only in MALToma-hu-SCID mice by cell ELISA. Immunostaining of tissue specimens indicated that human IgG bound to epithelial cells around the ulcer, so that human IgG may recognize both human and mouse stomach epithelial cells. Some reports suggest that antibodies produced in response to H. pylori bind to cells within the gastric mucosa and contribute to gastric tory capability between MALToma- and healthy-hu- SCID mice. Only the 30-kilodalton urease antigen showed no significant difference between MALToma- ( ) and healthy-hu-scid ( ) mice. Human antibody to H. pylori and gastric cells in hu-scid mice. Human IgG antibodies against H. pylori in hu-scid mouse sera were measured by ELISA. Results are expressed as mean SE of optical density. Anti H. pylori IgG levels in MALToma-hu-SCID mice ( ) were significantly higher (P 0.01) than those in healthy-hu-scid mice ( ) (Figure 4A). The presence of human autoantibodies against gastric cells was tested by cell ELISA using the gastric cancer cell line HGC-27. Autoantibody levels ( ) of sera from MALToma-hu-SCID mice were significantly higher than those of sera from ulcer-hu- SCID mice ( ; P 0.05), gastritis-hu-scid mice ( ; P 0.05), and healthy-hu-scid mice ( ; P 0.01) (Figure 4B). Discussion Transplantation of lymphocytes from patients with MALT lymphoma induced gastric ulcerative lesions in SCID mice after gastric injection of H. pylori. The proliferation assay analysis of spleen cells from hu-scid mice (Figure 3) and patient PBMCs (Table 1) indicated that H. pylori reactive lymphocytes were present both in the hu-scid mice and patients with MALT lymphoma, whereas the lymphocytes from gastric ulcer or gastritis patients and healthy volunteers did not proliferate in Figure 4. Antibodies against (A) H. pylori sonic extract and (B) HGC-27 human gastric cells. Anti H. pylori sonicated extract and anti HGC-27 human IgG levels in group I MALToma-hu-SCID mice (n 21), group II ulcer hu-scid mice (n 9), group III gastritis hu-scid mice (n 9), and group IV healthy-hu-scid mice (n 15) are expressed as mean SE. Statistical significance among groups, Student t test: *P 0.05 and **P 0.01.

6 898 YOKOTA ET AL. GASTROENTEROLOGY Vol. 117, No. 4 inflammation. 22,23 Both anti-idiotypic antibodies produced against MALToma cells and Ig secreted from lymphoma cells have been shown to react with normal tissue components including follicular dendritic cells, 24 and autoantibodies were frequently detected in patients with H. pylori infected atrophic gastritis. 25 We also reported antibodies to human gastric epithelial cells and HSP60 in H. pylori positive MALT lymphoma. 26 Therefore, autoantibody to HS60 may have an important role in mediating tissue damage in this model. Some recent reports have suggested that T-helper 1 phenotype cells are closely associated with the pathogenesis of H. pylori infection In our MALToma-hu-SCID mice model, inflammatory cell infiltration was modest even though the mice had pathological damage. Immunohistopathology revealed only a few CD45 T cells in the gastric mucosa of hu-scid mice. Serum levels of tumor necrosis factor (a T-helper 1 cytokine) did not differ greatly between MALToma- and healthy-hu-scid mice (data not shown). To clarify the phenotype of the cells responsible for the severe gastric lesions, we constructed hu-scid mice transplanted with only T or B cells separated by nylon wool columns. The lymphocytes transplanted survived for less than 1 month, indicating that T or B cells alone cannot survive. The lymphocytes may need reciprocal stimulation with cytokines produced by the other lymphocytes to allow their proliferation. Therefore, the T-cell phenotype responsible for epithelial cell damage in this mouse model remains unclear. ATCC produces both vaca and caga (actually many proteins coded by genes in the cag region). The existence of vaca and caga genes was confirmed by polymerase chain reaction using specific primers for these genes, and the ability of vaca was checked by bioassay. VacA and caga proteins are considered to cause epithelial cell damage during long-term H. pylori infection. 32 We were unable to assess the role of these proteins because we do not have a mutant of this strain, lacking only the ability to produce vaca and/or caga. Because gastric cell damage in MALToma-hu-SCID originated within 1 day after H. pylori inoculation, we think that these proteins may have only a minor role in causing cell damage. This study suggests that immune reactions against H. pylori and gastric cells are important to the development of mucosal cell damage. Cytotoxic T cells and/or antibodies reacting with both H. pylori and gastric epithelial cells may cause tissue damage. We expect that the hu-scid mouse model will continue to be useful in clarifying the relationship between H. pylori infection, host immunity, and the disease. Detailed in vivo and in vitro experiments are under way in an attempt to clarify these mechanisms. References 1. Marshall BJ, Armstrong JA, McGechie DB, Glancy RJ. Attempt to fulfill Koch s postulates for pyloric Campylobacter. Med J Aust 1985;142: Alper J. Ulcers as an infectious disease. Science 1993;260: The Eurogast Study Group. An international association between Helicobacter pylor infection and gastric cancer. Lancet 1993;341: Covacci A, Censini S, Bugnoli M, Petracca R, Burroni D, Macchia G, Massone A, Papini E, Xiang Z, Figura N, Rappuoli R. Molecular characterization of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci USA 1993;90: Censini S, Lange C, Xiang Z, Crabtree JE, Ghiara P, Borodovsky M, Rappuoli R, Covacci A. Cag, a pathogenicity island of Helicobacter pylori, encodes type I specific and disease-associated virulence factors. Proc Natl Acad Sci USA 1996;93: Xiang Z, Censini S, Bayeli PF, Telford JL, Figura N, Rappuoli R, Covacci A. Analysis of expression of CagA and VacA virulence factors in 43 strains of Helicobacter pylori reveals that clinical isolates can be divided into two major types and that CagA is not necessary for expression of the vacuolating cytotoxin. Infect Immun 1995;63: Telford JL, Ghiara P, Dell Orco M, Comanducci M, Burroni D, Bugnoli M, Tecce MF, Censini S, Covacci A, Xiang Z, Paponi E, Montecucco C, Parente L, Rappuoli R. Gene structure of the Helicobacter pylori cytotoxin and evidence of its key role in gastric disease. J Exp Med 1994;179: Ghiara P, Marchetti M, Blaser MJ, Tummuru MK, Cover TL, Segal ED, Tompkins LS, Rappuoli R. Role of the Helicobacter pylori virulence factors vacuolating cytotoxin, caga, and urease in a mouse model of disease. Infect Immun 1995;63: Molinari M, Salio M, Galli C, Norais N, Rappuoli R, Lanzavecchia A, Montecucco C. Selective inhibition of Ii-dependent antigen presentation by Helicobacter pylori toxin VacA. J Exp Med 1998;187: Ernst PB, Crowe SE, Reyes VE. How does Helicobacter pylori cause mucosal damage? The inflammatory response. Gastroenterology 1997;113:S35 S Mohammadi M, Redline R, Nedrud J, Czinn S. Role of the host in pathogenesis of Helicobacter-associated gastritis: H. felis infection of inbred and congenic mouse strains. Infect Immun 1996;64: Sakagami T, Dixon M, O Rourke J, Howlett R, Alderuccio F, Vella J, Shimoyama T, Lee A. Atrophic gastric changes in both Helicobacter felis and Helicobacter pylor infected mice are host dependent and separate from antral gastritis. Gut 1996;39: Wotherspoon AC, Ortiz-Hidalgo C, Falzon MR, Isaacson PG. Helicobacter pylori associated gastritis and primary B-cell gastric lymphoma. Lancet 1991;338: Hussell T, Isaacson PG, Crabtree JE, Spencer J. The response of cells from low-grade B-cell gastric lymphomas of mucosaassociated lymphoid tissue to Helicobacter pylori. Lancet 1993; 342: Horstmann M, Erttmann R, Winkler K. Relapse of MALT lymphoma associated with Helicobacter pylori after antibiotic treatment. Lancet 1994;343: Bayerdorffer E, Neubauer A, Rudorph B, Thiede C, Lehn N, Eidt S, Stolte M. Regression of primary gastric lymphoma of mucosaassociated lymphoid tissue type after cure of Helicobacter pylori infection. MALT Lymphoma Study Group. Lancet 1995;345: Harris NL, Jaffe ES, Stein H, Banks PM, Chan JK, Cleary ML, Delsol G, De Wolf-Peeters C, Falini B, Gatter KC. A revised European-American classification of lymphoid neoplasms: a pro-

7 October 1999 H. PYLORI INDUCED GASTRIC ULCERS BY IMMUNITY 899 posal from the international lymphoma study group. Blood 1994; 84: Yokota K, Kurebayashi Y, Takayama Y, Hayashi S, Isogai H, Isogai E, Imai K, Yabana T, Yachi A, Oguma K. Colonization of Helicobacter pylori in the gastric mucosa of Mongolian gerbils. Microbiol Immunol 1991;35: Yokota K, Hirai Y, Haque M, Hayashi S, Isogai H, Sugiyama T, Nagamachi Y, Tsukada N, Fujii N, Oguma K. Heat shock protein produced by Helicobacter pylori. Microbiol Immunol 1994;38: Akagi T, Kimoto T. Human cell line (HGC-27) derived from the metastatic lymph node of gastric cancer. Acta Med Okayama 1976;30: Kobayashi K, Yokota K, Yoshino T, Kawahara Y, Dey A, Hirai Y, Oguma K, Akagi T. Detection of Helicobacter pylori associated antigen and heat shock protein 60 on follicular dendritic cells in the germinal centers of low grade B cell lymphoma of gastric mucosa associated lymphoid tissue (MALT). Clin Pathol 1998;51: Negrini R, Savio A, Poiesi C, Appelmelk BJ, Buffoli F, Paterlini A, Cesari P, Graffeo M, Vaira D, Franzin G. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis. Gastroenterology 1996;111: Faller G, Steininger H, Kranzlein J, Maul H, Kerkau T, Hensen J, Hahn EG, Kirchner T. Antigastric autoantibodies in Helicobacter pylori infection: implications of histological and clinical parameters of gastritis. Gut 1997;41: Greiner A, Marx A, Heesemann J, Leebmann J, Schmausser B, Muller-Hermelink HK. Idiotype identity in a MALT-type lymphoma and B cells in Helicobacter pylori associated chronic gastritis. Lab Invest 1994;70: Negrini R, Savio A, Poiesi C, Appelmelk BJ, Buffoli F, Paterlini A, Cesari P, Graffeo M, Vaira D, Franzin G. Antigenic mimicry between Helicobacter pylori and gastric mucosa in the pathogenesis of body atrophic gastritis. Gastroenterology 1996;111: Kawahara Y, Yokota K, Kobayashi K, Yunoki N, Yoshinaga F, Okada H, Mizuno M, Tsuji T, Hirai Y, Oguma K. Antibodies to human gastric epithelial cells and heat shock protein in Helicobacter pylori positive mucosa-associated lymphoid tissue lymphoma Gut 1999;45: Fan XJ, Chua A, Shahi CN, McDevitt J, Keeling PW, Kelleher D. Gastric T responses to Helicobacter pylor in patients with H pylori colonisation. Gut 1994;35: Mohammadi M, Czinn S, Redline R, Nedrud J. Helicobacterspecific cell-mediated immune responses display a predominant Th1 phenotype and promote a delayed-type hypersensitivity in the stomachs of mice. J Immunol 1996;156: D Elios MM, Manghetti M, De Carli M, Costa F, Baldari CT, Telford JL, Romagnani S, Del Prete G. T helper 1 effector cell specific for Helicobacter pylori in the gastric antrum of patients with peptic ulcer disease. J Immunol 1997;158: Haeberle HA, Kubin M, Bamford KB, Garofalo R, Graham DY, El-Zaatari F, Karttunen R, Crowe SE, Reyes VE, Ernst PB. Differential stimulation of interleukin-12 (IL-12) and IL-10 by live and killed Helicobacter pylori in vitro and association of IL-12 production with gamma interferon producing T cells in the human gastric mucosa. Infect Immun 1997;65: Bamford KB, Fan X, Crowe SE, Leary JF, Gourley WK, Luthra GK, Brooks EG, Graham DY, Reyes VE, Ernst PB. Lymphocytes in the human gastric mucosa during Helicobacter pylori have a T helper cell 1 phenotype. Gastroenterology 1998;114: Marchetti M, Arico B, Burroni D, Figura N, Rappuoli R, Ghiara P. Development of a mouse model of Helicobacter pylori infection that mimics human disease. Science 1995;267: Received May 5, Accepted June 15, Address requests for reprints to: Keiji Oguma, M.D., Department of Bacteriology, Okayama University Medical School, 2-5-1, Shikatacho, Okayama , Japan. kuma@med.okayamau.ac.jp; fax: (81) Supported in part by grant no from the Ministry of Education, Science, Sports, and Culture, Japan, and by Mitsui Life Social Welfare Foundation. Portions of this work were presented at the Xth International Workshop on Gastroduodenal Pathology and Helicobacter pylori and have been published in abstract form in (Gut 1997;41[suppl]): A ). The authors thank Dr. William R. Brown (University of Colorado School of Medicine) for assistance in the preparation of the manuscript.