Helicobacter pylori infection induces gastric cancer in Mongolian gerbils

GASTROENTEROLOGY 1998;115:642–648

Helicobacter pylori Infection Induces Gastric Cancer in Mongolian Gerbils TAKESHI WATANABE,* MAYUMI TADA,‡ HIROFUMI NAGAI,* SATOSHI SASAKI,* and MASAFUMI NAKAO‡ *Drug Safety Research Laboratories and ‡Pharmacology Laboratories, Takeda Chemical Industries, Ltd., Osaka, Japan

See editorial on page 780. Background & Aims: Although epidemiological studies have indicated that Helicobacter pylori infection plays a crucial role in gastric carcinogenesis in humans, there is no direct proof that H. pylori is actually associated with gastric carcinogenesis. The purpose of this study was to elucidate the relationship between H. pylori infection and gastric carcinogenesis using an animal model of long-term H. pylori infection. Methods: Mongolian gerbils were orally inoculated with H. pylori, and the sequential morphological changes in the stomach were examined for up to 62 weeks. Results: H. pylori was constantly detected in all infected animals throughout the study. At the 26th week, severe active chronic gastritis, ulcers, and intestinal metaplasia could be observed in infected animals. By the end of the study, adenocarcinoma had developed in the pyloric region of 37% of the infected animals. All tumors consisted of well-differentiated intestinal-type epithelium, and their development seemed to be closely related to intestinal metaplasia. Conclusions: We have successfully demonstrated that long-term infection with H. pylori induces adenocarcinoma in Mongolian gerbils. The observations are thus highly suggestive of the involvement of H. pylori infection in gastric carcinogenesis in humans.

lthough the overall incidence of gastric cancer has steadily declined over the past 50 years, it is still a major health problem and remains the second most common cancer in the world.1 Since the discovery of Helicobacter pylori in 1983,2 this microorganism has been implicated in gastric carcinogenesis on the basis of various epidemiological studies.3–7 A Working Group of the World Health Organization International Agency for Research on Cancer concluded in 1994 that H. pylori is a group 1 carcinogen in humans.8 However, the prevalence of H. pylori in patients with gastric carcinoma still remains considerably variable among the studies,9 and there has been no direct demonstration that H. pylori is

A

actually associated with gastric carcinogenesis. Gastric cancer is a multifactorial disease10; therefore, genetic factors and environmental factors other than H. pylori can complicate the elucidation of the association between H. pylori infection and gastric cancer. We believe that studies using animal models of H. pylori infection, especially those following the effects of long-term infection on the gastric mucosa, should be able to shed further light on this unresolved issue. The Mongolian gerbil (Meriones unguiculatus) is an appropriate animal in which various gastrointestinal diseases such as gastritis and ulcers that mimic human H. pylori infection can be studied.11,12 In our current research, we therefore examined the pathological consequences of long-term H. pylori infection using specific pathogen–free Mongolian gerbils with the aim of clarifying the association of H. pylori and gastric carcinogenesis.

Materials and Methods Five-week-old male specific pathogen–free Mongolian gerbils (MGS/Sea) were purchased from Seiwa Experimental Animals (Fukuoka, Japan) and maintained under standard laboratory conditions (room temperature, 23°C 6 2°C; relative humidity, 55% 6 5%; 12/12-hour light/dark cycle) with free access to a commercial rodent diet (CE-2; Clea Japan, Tokyo, Japan) and tap water. H. pylori TN2GF4 used in this study had been isolated from a patient with gastric ulcer and was motile and urease, catalase, and oxidase positive. The strain produced vacuolating cytotoxin and contained the cytotoxinassociated gene (cagA). H. pylori for the experimental inoculation was grown in brucella broth supplemented with 2.5% heat-inactivated fetal bovine serum in GasPak jars (BBL, Cockeysville, MD) containing CampyPak (BBL) at 37°C for about 24 hours. Sterilized glycerol was added to the cultures at a final concentration of 15%, and cultures were kept at 280°C until use. Fifty-five animals were inoculated with 1 mL of broth culture containing 107.54 colony-forming units (CFU) of H. pylori by intragastric gavage after fasting for 24 hours. Abbreviation used in this paper: AB-HID, alcian blue with high-iron diamine. r 1998 by the American Gastroenterological Association 0016-5085/98/$3.00

September 1998

HELICOBACTER PYLORI INDUCES GASTRIC CANCER

Thirty animals served as uninfected controls. Eight of the infected animals died between 40 weeks after inoculation and the end of the experiment. Data from these animals were excluded from analysis because of autolytic changes in the stomach. Six, 26, 39, and 52 weeks after inoculation, 5 infected animals were killed, and, in addition, all surviving animals, 27 infected and 30 uninfected animals, were killed 62 weeks after inoculation. At each time point, one half of the stomach from 5 infected animals was used for the quantitative determination of the bacterium. The other half of the stomach from these 5 animals and the entire stomach from the other animals, killed 62 weeks after inoculation, were examined for sequential morphological changes. After the stomach was opened along the greater curvature, the longitudinal half of the stomach was homogenized with physiological saline. An aliquot of dilutions was inoculated onto modified Skirrow’s agar and incubated at 37°C for 4 days under microaerobic conditions. The density of infection was estimated by counting the number of colonies per plate and expressed as log CFU per gastric wall. For histological examination, the stomach was stapled onto paper and fixed in 10% neutral buffered formalin. Fixed tissue was cut into longitudinal strips, about 3 mm wide, for histopathologic examination of the entire gastric mucosa. After processing for histology by routine methods, paraffin-embedded sections were cut and stained with H&E; for mucus clarification or for demonstration of neuroendocrine cells, when necessary, alcian blue (pH 2.5) with high-iron diamine (AB-HID) stain or Grimelius’ and Sevier–Mungar stains were applied. Histopathologic lesions of the glandular stomach were categorized as follows: (1) active chronic gastritis (characterized by severe inflammatory cell infiltration and multiple lymphoid follicle formation throughout the pyloric region and part of the fundic region); (2) ulcer; (3) regenerative hyperplasia; (4) invagination of glands into the submucosa; (5) hyperplastic polyp; (6) intestinal metaplasia (diagnosed by the presence of goblet cells that were positive for AB-HID staining); (7) adenocarcinoma (diagnosed by the presence of atypical glands that locally invaded the muscle layer and destroyed the original architec-

643

ture); and (8) carcinoid (diagnosed by neuroendocrine nests in excess of three glands in diameter). In addition, immunohistochemical staining to detect expression of p53 protein was performed. Formalin-fixed, paraffin-embedded sections from animals diagnosed with adenocarcinoma and from 5 uninfected animals killed 62 weeks after inoculation were pretreated by autoclaving (105°C, 10 minutes) in a target unmasking fluid (Monosan; Uden, The Netherlands). After routine inhibition of endogeneous peroxidase activity with 0.3% H2O2 in methanol, nonspecific proteins were blocked with normal goat serum. Specimens were incubated with a monoclonal antibody reactive to wild-type and mutant-type p53 protein (Dako Japan Corp., Kyoto, Japan). Immunolocalization was demonstrated using the Envision method (Dako Japan Corp.) with diaminobenzidine as the substrate. Only distinctly increased nuclear staining was regarded as positive, and when staining was observed in all or nearly all tumor cells, tumors were considered positive. Negative control tissues were prepared in exactly the same way as described above, except the primary antibody was omitted. All experimental protocols described were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals in Pharmaceutical Research Division, Takeda Chemical Industries, Ltd., and approved by the Ethical Committee for the animal experiments of our division.

Results H. pylori was detected in all infected animals throughout the study for up to 62 weeks. The bacterial count reached 105.17 CFU/gastric wall at 6 weeks and then remained constant until the end of the experiment (Table 1). Sequential histopathologic changes are summarized in Table 1. By the 6th week, severe active chronic gastritis, characterized by dense neutrophil and mononuclear cell infiltration, and multiple lymphoid follicle formation in both the mucosa and submucosa could be observed in all infected animals (Figure 1). The gastritis

Table 1. Sequential Microbiological and Histopathologic Changes in the Glandular Stomach of Mongolian Gerbils Infected With H. pylori Duration after HP inoculation (wk) No. of animals examined Microbiology Bacterial count (log CFU/gastric wall) Histopathology Active chronic gastritis Ulcer Regenerative hyperplasia Invagination of glands Hyperplastic polyp Intestinal metaplasia Adenocarcinoma Carcinoid

6 5

26 5

39 5

52 5

62 27 a

5.17 6 0.41b

5.33 6 0.98

5.89 6 0.51

4.59 6 1.45

5.40 6 0.95

5 c (100) 0 (0) 5 (100) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)

5 (100) 5 (100) 5 (100) 5 (100) 0 (0) 3 (60) 0 (0) 0 (0)

5 (100) 5 (100) 5 (100) 5 (100) 0 (0) 4 (80) 0 (0) 0 (0)

5 (100) 2 (40) 5 (100) 5 (100) 3 (60) 5 (100) 0 (0) 0 (0)

27 (100) 16 (59) 27 (100) 27 (100) 4 (15) 23 (85) 10 (37) 3 (11)

HP, H. pylori. a Microbiological examination was performed on 5 animals. b Mean 6 SD. c Incidence with percentage in parentheses.

644

WATANABE ET AL.

Figure 1. (A ) Active chronic gastritis observed in the pyloric region of Mongolian gerbils 6 weeks after H. pylori inoculation. (B ) The gastritis is characterized by severe neutrophil and mononuclear cell infiltration in both the mucosa and submucosa. The normal mucosal architecture is almost completely lost and replaced by regenerative hyperplastic epithelium. The thickness of the mucosa increased remarkably compared with that of an uninfected animal (B) (H&E; original magnification 653).

originated in the pyloric antrum, extended to the fundic region, and was observed continuously throughout the study. The normal mucosal architecture was almost completely lost and was replaced by hyperplastic epithelium. The thickness of the mucosa increased remarkably, and pseudopyloric gland formation and invagination into the submucosa of hyperplastic glands were noted at the fundus-pylorus border. After the 26th week, ulcers, extending to the muscular layer, developed in the region of the fundus-pylorus border of all infected animals. Epithelial hyperplasia progressed, and intestinal metaplasia was observed in the mucosa of the pyloric region, especially in the area close to the ulcer. These metaplastic glands were composed of columnar cells containing little or no mucus and a moderate number of goblet cells producing sialomucins and occasionally sulfomucins (Figure 2). In addition, paneth cells were occasionally observed in some of the metaplastic glands. Cystic dilatation of the invaginated glands in the submucosa was a common finding during this period. These cystic submucosal glands were mainly lined by low, columnar, mucous cells with little cellular atypia. From the 39th to 52nd week, the structural irregularity of the hyperplastic glands was more prominent, and complex branching or cystic dilatation of these glands was frequently observed. Hyperplastic polyps, characterized by hyperplasia of the tall columnar surface epithelium and prominent edema, were also noted in the fundus. Despite such structural irregularities, little cellular atypia could be observed in the hyperplastic epithelium. The frequency of intestinal metaplasia increased during this period and could be observed not only in the mucosa but also in the

GASTROENTEROLOGY Vol. 115, No. 3

Figure 2. (A) Intestinal metaplasia observed near the ulcer in Mongolian gerbils 39 weeks after H. pylori inoculation. (B) The metaplastic glands are composed of columnar cells and a large number of goblet cells producing sialomucins (blue) and occasionally sulfomucins (gray) (H&E [A] and AB-HID [B]; original magnification 653 [A] and 1303 [B]).

submucosal cystic glands. These metaplastic glands appeared more atypical than the surrounding nonmetaplastic, hyperplastic glands (Figure 3). Budding of glands, enlarged and vesicular nuclei with prominent nucleoli, and stratification of nuclei were frequently observed. By the end of the study, adenocarcinoma had developed in 10 of the remaining 27 (37%) infected animals. Grossly, almost all tumors were located in the pyloric region and detected as crater-like areas with irregular elevated ridges. Gastric wall thickening was frequently observed in these regions. The histomorphological features of these tumors were quite similar (Figure 4). The tumors arose in the areas of intestinal metaplasia in deep mucosa or submucosa and grew downward. They penetrated into the muscle layer and occasionally spread into the serosa. Metastasis or vascular invasion, however, was not ob-

Figure 3. (A ) Intestinal metaplasia observed in the pyloric region of Mongolian gerbils 52 weeks after H. pylori inoculation. (B ) The metaplastic glands show complex branching and budding, and enlarged and vesicular nuclei with prominent nucleoli and stratification of nuclei are frequently recognized (H&E; original magnifications 503 [A ] and 1303 [B ]).

September 1998

HELICOBACTER PYLORI INDUCES GASTRIC CANCER

645

Figure 4. (A ) An adenocarcinoma observed in the pyloric region of a Mongolian gerbil 62 weeks after H. pylori inoculation. (B ) The tumor has arisen in the area of intestinal metaplasia in the deep mucosa, and the direction of growth is downward. Penetration into the lamina muscularis and destruction of the original architecture can be observed. (C ) Spreading into the serosa of the neoplastic glands is also noted, and these glands show nuclear pleomorphism. (D ) The neoplastic glands are composed of intestinal-type epithelium, which contains goblet cells producing predominantly sialomucins (blue). Cystic dilatation of the neoplastic glands with an accumulation of mucus and extracellular mucus pools are frequently seen in the submucosa. (E ) Immunohistochemically, almost all tumor cells express p53 protein (H&E; original magnifications 303 [A ], 653 [B ], 1303 [C ]; AB-HID, 1303 [D ]; p53, 1303 [E ]).

646

WATANABE ET AL.

served in any case. The tumors retained a well preserved glandular structure and consisted of well-differentiated columnar intestinal-type epithelium, which contained a moderate to large number of goblet cells producing predominantly sialomucins. Irregular budding or branching of glands and nuclear pleomorphism were frequently observed. Copious mucous production was a characteristic feature, and cystic dilatation of neoplastic glands with an accumulation of mucus and extracellular mucus pools were commonly seen in the submucosa. The stroma of these tumors showed various degrees of inflammation and fibrosis, and osseous metaplastic foci were occasionally noted. Immunohistochemically, intranuclear overexpression of p53 protein was observed in 4 of the 10 (40%) adenocarcinomas. In these cases, p53 protein–expressing cells were densely distributed in almost all tumor cells. The relationship between p53 protein expression and tumor histology, including the invasive pattern, was not clear. There was no increase in immunoreactivity in the uninfected animals, and negative controls stained as expected. In addition to the adenocarcinoma, carcinoids derived from fundic mucosa were also detected in 3 infected animals. Tumor cells were moderately pleomorphic, with pale or eosinophilic cytoplasm and argyrophilic, and were arranged in solid sheets or small nests in lamina propria. One of three carcinoids spread into the submucosa. No other hyperplastic changes in neuroendocrine cells were observed in any infected animal. In the uninfected animals, no significant changes had been observed by the end of the study.

Discussion We successfully showed that long-term infection with H. pylori induces gastric adenocarcinoma in Mongolian gerbils. The tumor incidence was 37% (10 of 27 animals) 62 weeks after H. pylori inoculation. Although H. pylori infection has been categorized as a risk factor for gastric cancer based on the results of various epidemiological studies,3–7 no direct evidence of this relationship has been presented to date. In animal studies, Fox et al.13 reported that Helicobacter mustelae, a relative of H. pylori, is associated with gastric carcinogenesis in naturally infected ferrets. To the best of our knowledge, our study is the first showing a direct relationship between H. pylori infection and gastric carcinogenesis. All tumors induced in the H. pylori–infected Mongolian gerbils showed similar histological features. They were located in the pyloric region and consisted of well-differentiated intestinal-type epithelium, corresponding to intestinal-type adenocarcinoma in humans. In comparing the histology of the cancer associated with

GASTROENTEROLOGY Vol. 115, No. 3

H. pylori infection in humans and Mongolian gerbils, the topographic location was similar.6,14–17 The histological subtype of cancer is difficult to assess accurately because of the existing controversy about the difference between intestinal and diffuse types of gastric cancer in their association with H. pylori infection in humans.5–7,14–17 The reason for these conflicting observations is considered to be mainly the different methods used for the detection of H. pylori.15 In studies based on histological detection of H. pylori, the prevalence has generally been higher in the intestinal rather than in the diffuse cancer type.7,18,19 This difference has been less distinct or nonexistent in most serological studies,5,6,14,17 although there are some exceptions.15,16 However, in focusing on early gastric cancer to eliminate any possible complicating effects of tumor progression on colonization with H. pylori, the prevalence of H. pylori infection is significantly higher in intestinal-type gastric cancer regardless of the method used to detect H. pylori infection.16,20,21 In addition, several reports have shown that H. pylori infection and intestinal-type gastric carcinoma share some similar epidemiological characteristics.15 The pathogenesis of gastric cancer is certainly multifactorial, and genetic host and/or environmental factors, other than H. pylori, may complicate the resolution of this issue in humans. It seems to be important, therefore, that tumors induced in the H. pylori–infected Mongolian gerbils, in which other environmental factors could be minimized, were all intestinal-type adenocarcinoma. Epidemiological surveys,22,23 histopathologic examination,24 and biochemical25 and genetic26 analyses have shown that intestinal metaplasia, especially the sulfomucin-secreting type, may be a possible precancerous lesion leading to intestinal-type adenocarcinoma in humans.27 In the H. pylori–infected Mongolian gerbils, intestinal metaplasia occurred in the mucosa or submucosal invaginated glands in the pyloric region before the cancer developed. Although these metaplastic glands produced sialomucins predominantly, the metaplastic epithelium was more dysplastic than the surrounding nonmetaplastic hyperplastic epithelium, and the tumors arose in association with metaplastic epithelium. Furthermore, all tumors found in the present study consisted solely of intestinal-type epithelial cells. All these results indicate a close relationship between intestinal metaplasia and gastric carcinogenesis, and it can be concluded that intestinal metaplasia is a precancerous lesion leading to gastric cancer in this model. Intestinal metaplasia has been widely assumed to be linked to chronic gastritis with mucosal atrophy.28 Although the pathogenesis of intestinal metaplasia has not been clarified as yet, Scott et al.29 suggested that a

September 1998

hyperproliferative state in the inflamed gastric mucosa may promote the progression from normal mucosa to metaplastic epithelium. Moreover, Recavarren-Arce et al.30 reported that an altered gastric microflora, subsequent to hypochlorhydria due to atrophy, increases N-nitroso compounds in the gastric lumen, and these substances may induce intestinal metaplasia.30 In our study, so-called mucosal atrophy, characterized by a decrease in mucosal height and gland density, was not evident; however, the original glands, including oxyntic glands, were almost completely lost and were replaced by hyperplastic epithelium. This mucosal state would cause the generation of metaplastic epithelium. Metaplastic epithelium was frequently noted around the ulcers; therefore, ulcer formation might have contributed to intestinal metaplasia in this model, possibly by enhancement of the mucosal proliferative activity in its regenerative process. No significant changes were noted in uninfected animals in this study. Also, severe gastritis in H. pylori–infected Mongolian gerbils can be reversed by clearance of the organism.12 This indicates that H. pylori infection alone caused severe inflammatory changes including intestinal metaplasia, and may also induce gastric cancer. Several animal models for H. pylori produce severe gastritis, including gnotobiotic piglets, monkeys, and mice.31–33 However, induction of ulceration or intestinal metaplasia has never been reported in these models, although higher levels of colonization were noted in some of them.33 Possibly host factors such as differences in mast cell subtype or cytokine production are responsible for the susceptibility of Mongolian gerbils to H. pylori.12 Although the pathogenesis of the tumor is unknown, there are several possible mechanisms by which H. pylori infection may play a part in the mutagenesis of the gastric epithelium. One explanation is increased epithelial turnover in the inflamed mucosa. Enhanced cell replication increases the frequency of mutation by errors in gene replication, by conversion of endogenous or exogenous DNA adducts to mutation, or by diminishing the time for the DNA repair processes.34 Cahill et al.35 have stated that gastric mucosal cell proliferation is significantly higher in patients with H. pylori–positive gastritis than in those with H. pylori–negative gastritis. Recently, a novel cell proliferation gene ( pag) has been shown to share approximately 50% homology with a 26-kilodalton antigen of H. pylori.36 In the Mongolian gerbil model, prominent regenerative hyperplasia was a characteristic feature and constantly observed throughout the study. This was suggestive of persistently and highly enhanced cellular proliferation in the gastric mucosa in this model. Therefore, H. pylori itself and/or an indirect hyperplastic

HELICOBACTER PYLORI INDUCES GASTRIC CANCER

647

response to cellular damage or inflammatory products stimulate the proliferation of gastric epithelial cells and may cause an increased opportunity for mutation.37 An additional explanation is the effect of the formation of nitric oxide, a putative endogenous mutagen in the presence of oxide. It has been reported that a soluble protein from H. pylori enhances the expression of the NO synthase gene38 and NO production in macrophage cell lines.39 In addition, high concentrations of NO induce wild-type p53 protein accumulation,40 and the NOrelated deamination of DNA has been reported to cause GC-AT transitions, which are frequently found in p53 mutations in gastric cancer.39 The overexpression of p53 protein in gastric cancer observed in the present model, may therefore suggest the participation of NO in the gastric carcinogenesis in H. pylori–infected Mongolian gerbils. We have successfully shown that long-term H. pylori infection induces gastric cancer in Mongolian gerbils. All of the induced tumors were located in the pyloric region and consisted of well-differentiated intestinal-type epithelium. The development of these tumors seemed to be closely related to intestinal metaplasia formed in the chronically inflamed mucosa. Although careful interpretation is required in extrapolating the data from such an animal model to humans, our present findings are highly suggestive of the involvement of H. pylori infection in gastric carcinogenesis in humans.

References 1. Correa P, Haenszel W. Epidemiology of gastric cancer. In: Correa P, Haenszel W, eds. Epidemiology of cancer of the digestive tract. Hague: Martinus-Nijhoff, 1982:58–84. 2. Warren JR, Marshall BJ. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 1983;1:1273– 1275. 3. The Eurogast Study Group. An international association between Helicobacter pylori infection and gastric cancer. Lancet 1993;341: 1359–1362. 4. Nomura A, Stemmermann GN, Chyou P-H, Kato I, Perez-Perez GI, Blaser MJ. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. N Engl J Med 1991;325: 1132–1136. 5. Parsonnet J, Friedman GD, Vandersteen DP, Chang Y, Vogelman JH, Orentreich N, Sibley RK. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med 1991;325:1127–1131. 6. Talley NJ, Zinsmeister AR, Weaver A, DiMagno EP, Carpenter HA, Perez-Perez GI, Blaser MJ. Gastric adenocarcinoma and Helicobacter pylori infection. J Natl Cancer Inst 1991;83:1734–1739. 7. Parsonnet J, Vandersteen D, Goates J, Sibley RK, Pritikin J, Chang Y. Helicobacter pylori infection in intestinal- and diffuse-type gastric adenocarcinomas. J Natl Cancer Inst 1991;83:640–643. 8. International Agency for Research on Cancer. IARC Monographs on the evaluation of carcinogenic risks to humans, Volume 61, Schistosomes, liver flukes and Helicobacter pylori. Lyon, France: 1994. 9. Sipponen P, Kosunen TU, Valle J, Riihela¨ M, Seppa¨la¨ K. Helicobacter pylori infection and chronic gastritis in gastric cancer. J Clin Pathol 1992;45:319–323.

648

WATANABE ET AL.

10. Correa P. Human gastric carcinogenesis: a multistep and multifactorial process. Cancer Res 1992;52:6735–6740. 11. Hirayama F, Takagi S, Kusuhara H, Iwao E, Yokoyama Y, Ikeda Y. Induction of gastric ulcer and intestinal metaplasia in Mongolian gerbils infected with Helicobacter pylori. J Gastroenterol 1996;31: 755–757. 12. Matsumoto S, Washizuka Y, Matsumoto Y, Tawara S, Ikeda F, Yokota Y, Karita M. Induction of ulceration and severe gastritis in Mongolian gerbil by Helicobacter pylori infection. J Med Microbiol 1997;46:391–397. 13. Fox JG, Dangler CA, Sager W, Borkowski R, Gliatto JM. Helicobacter mustelae–associated gastric adenocarcinoma in ferrets (Mustela putorius furo). Vet Pathol 1997;34:225–229. 14. Hansson L-E, Engstrand L, Nyre´n O, Lindgren A. Prevalence of Helicobacter pylori infection in subtypes of gastric cancer. Gastroenterology 1995;109:885–888. 15. Martin-De-Argila C, Boixeda D, Redondo C, Alvarez I, Gisbert JP, Garcia Plaza A, Canton R. Relation between histologic subtypes and location of gastric cancer and Helicobacter pylori. Scand J Gastroenterol 1997;32:303–307. 16. Wu M-S, Chen S-Y, Shun C-T, Lee W-J, Wang H-P, Wang T-H, Chen C-J, Lin J-T. Increased prevalence of Helicobacter pylori infection among patients affected with intestinal-type gastric cancer at non-cardiac locations. J Gastroenterol Hepatol 1997;12:425– 428. 17. Hansson L-E, Engstrand L, Nyre´n O, Evans DJ, Lindgren A, Bergstro¨m R, Andersson B, Athlin L, Bendtsen O, Tracz P. Helicobacter pylori infection: Independent risk indicator of gastric adenocarcinoma. Gastroenterology 1993;105:1098–1103. 18. Buruk F, Berberoglu U, Pak I, Aksaz E, Celen O. Gastric cancer and Helicobacter pylori infection. Br J Surg 1993;80:378–379. 19. Loffeld RJLF, Willems I, Flendrig JA, Arends JW. Helicobacter pylori and gastric carcinoma. Histopathology 1990;17:537–541. 20. Endo S, Ohkusa T, Saito Y, Fujiki K, Okayasu I, Sato C. Detection of Helicobacter pylori infection in early stage gastric cancer: a comparison between intestinal- and diffuse-type gastric adenocarcinomas. Cancer 1995;75:2203–2208. 21. Tatsuta M, Iishi H, Okuda S, Taniguchi H, Yokota Y. The association of Helicobacter pylori with differentiated-type early gastric cancer. Cancer 1993;72:1841–1845. 22. Correa P, Cuello C, Duque E. Carcinoma and intestinal metaplasia of the stomach in Colombian migrants. J Natl Cancer Inst 1970;44:297–306. 23. Correa P, Cuello C, Duque E, Burbano LC, Garcia FT, Bolanos O, Brown C, Haenszel W. Gastric cancer in Colombia. III. Natural history of precursor lesions. J Natl Cancer Inst 1976;57:1027– 1035. 24. Nakamura K, Sugano H, Takagi K. Carcinoma of the stomach in incipient phase: its histogenesis and histological appearances. Gann 1968;59:251–258. 25. Kawachi T, Kurisu M, Numanyu N, Sasajima K, Sano T, Sugimura T. Precancerous changes in the stomach. Cancer Res 1976;36: 2673–2677. 26. Tahara E, Kuniyasu H, Yasui W, Yokozaki H. Gene alterations in intestinal metaplasia and gastric cancer. Eur J Gastroenterol Hepatol 1994;6(suppl 1):S97–S101. 27. Silva S, Filipe MI, Pinho A. Variants of intestinal metaplasia in the

GASTROENTEROLOGY Vol. 115, No. 3

28.

29.

30.

31.

32.

33.

34.

35.

36.

37. 38.

39.

40.

evolution of chronic atrophic gastritis and gastric ulcer. A follow up study. Gut 1990;31:1097–1104. Correa P, Haenszel W, Cuello C, Zavala D, Fontham E, Zarama G, Tannenbaum S, Collazos T, Ruiz B. Gastric precancerous process in a high risk population: cross-sectional studies. Cancer Res 1990;50:4731–4736. Scott N, Lansdown M, Diament R, Rathbone B, Murday V, Wyatt JI, McMahon M, Dixon MF, Quirke P. Helicobacter gastritis and intestinal metaplasia in a gastric cancer family (letter). Lancet 1990;1:728. Recavarren-Arce S, Leo´n-Baru´a R, Cok J, Berendson R, Gilman RH, Ramı´rez-Ramos A, Rodrı´guez C, Spira WM. Helicobacter pylori and progressive gastric pathology that predisposes to gastric cancer. Scand J Gastroenterol 1991;26(suppl 181):51–57. Krakowka S, Morgan DR, Kraft WG, Leunk RD. Establishment of gastric Campylobacter pylori infection in the neonatal gnotobiotic piglet. Infect Immun 1987;55:2789–2796. Shuto R, Fujioka T, Kubota T, Nasu M. Experimental gastritis induced by Helicobacter pylori in Japanese monkeys. Infect Immun 1993;61:933–939. Lee A, O’Rourke J, Corazon De Ungria M, Robertson B, Daskalopoulos G, Dixon MF. A standardized mouse model of Helicobacter pylori infection: introducing the Sydney strain. Gastroenterology 1997;112:1386–1397. Butterworth BE, Goldsworthy TL. The role of cell proliferation in multistage carcinogenesis. Proc Soc Exp Biol Med 1991;198:683– 687. Cahill RJ, Sant S, Beattie S, Hamilton H, O’Morain C. Helicobacter pylori and increased epithelial cell proliferation: a risk factor for cancer. Eur J Gastroenterol Hepatol 1994;6:1123–1127. Prospe´ri M-T, Ferbus D, Karczinski I, Goubin G. A human cDNA corresponding to a gene overexpressed during cell proliferation encodes a product sharing homology with amoebic and bacterial proteins. J Biol Chem 1993;268:11050–11056. Goldstone AR, Quirke P, Dixon MF. Helicobacter pylori infection and gastric cancer. J Pathol 1996;179:129–137. Chi DS, Shen D, Li C, Ha T, Ferguson D, Thomas E. TNF-a and nitric oxide (NO) synthase gene expressions in cultured macrophages enhanced by soluble proteins from Helicobacter pylori (abstr). FASEB J 1994;8:A253. Tsuji S, Kawano S, Tsujii M, Takei Y, Tanaka M, Sawaoka H, Nagano K, Fusamoto H, Kamada T. Helicobacter pylori extract stimulates inflammatory nitric oxide production. Cancer Lett 1996;108:195–200. Ambs S, Hussain SP, Harris CC. Interactive effects of nitric oxide and the p53 tumor suppressor gene in carcinogenesis and tumor progression. FASEB J 1997;11:443–448.

Received October 28, 1997. Accepted May 5, 1998. Address requests for reprints to: Takeshi Watanabe, M.D., Drug Safety Research Laboratories, Takeda Chemical Industries, Ltd., 17-85, Jusohonmachi 2-chome, Yodogawa-ku, Osaka 532-8686, Japan. Fax (81) 6-300-6918. The authors thank Jeffrey A. Hogan, Leslie I. Brezak, Toshimasa Kyutoku, Yasuharu Yamamoto, Naomi Inui, and Yuri Ozaki for their helpful assistance.