Vol. 58, No. 8 INFECTION AND IMMUNITY, Aug. 1990, p. 2606-2612 0019-9567/90/082606-07$02.00/0 Copyright X 1990, American Society for Microbiology Helicobacter pylori Gastric Infection in Gnotobiotic Beagle Dogs M. JUDITH RADIN,1* KATHRYN A. EATON,1 STEVEN KRAKOWKA,1 DONNA R. MORGAN,2 ADRIAN LEE,3 GLEN OTTO,4 AND JAMES FOX4 The Ohio State University, Columbus, Ohio 432101; The Procter and Gamble Co., Miami Valley Laboratories, Cincinnati, Ohio 45239-87072; University of New South Wales, Kensington, Australia3; and Massachusetts Institute of Technology, Boston, Massachusetts 021394 Received 14 March 1990/Accepted 20 March 1990 Establishment of infection with Helicobacter pyloni and gastritis in nonhuman species is currently only successful in gnotobiotic piglets. This study was designed to determine whether H. pyloni will colonize the gastrointestinal tract of gnotobiotic dogs. Gnotobiotic beagle pups were derived by standard methods. Group A (five dogs) was orally challenged with 3 x 108 H. pylori at 7 days of age. Group B (two dogs) received only peptone water but was contact-exposed beginning on day 23 postinfection (p.i.). Necropsy was performed on dogs on day 30 p.i. H. pylori colonized the stomach of all dogs (groups A and B). Urease map analysis correlated with the microbiologic findings and indicated that the density of colonization was less than that observed in human tissue. Organisms were also recovered from the pharynx, esophagus, duodenum, and rectum of 1, 2, 2, and 1 dog, respectively. All group A and one group B dog developed serum immunoglobulin G specific for H. pylon by day 30 p.i. Gross lesions were restricted to the stomach and consisted of small (<1 mm) lymphoid follicles. Microscopically, there were focal to diffuse lymphoplasmacytic infiltrates with follicle formation and mild to moderate infiltration of neutrophils and eosinophils in the gastric lamina propria. With the Warthin-Starry silver stain, organisms were seen on the surface of the gastric epithelial cells, beneath the mucus layer. We conclude that H. pylori colonizes the stomachs of gnotobiotic dogs for at least 1 month and the lesions resemble those seen in humans. H. pylori is transmissible by contact from infected to noninfected dogs. Helicobacter pylori (formerly Campylobacter pylori) is a gram-negative microaerophilic bacterium that causes gastri- tis and is associated with nonulcer dyspepsia and gastroduo- denal ulcer in humans (6, 18). Oral challenge of volunteers results in histologic lesions of chronic active gastritis as well as symptoms of dyspepsia (19, 20). Treatment of H. pylori- associated gastritis with combinations of antimicrobial agents, bismuth, and H-2 antagonists in humans has met with limited success, and recurrence is frequent (10). Be- cause of this, the use of human volunteers to study this disease entity is not practical, and adequate models for study of the pathogenesis of this disease are needed. H. pylori will not colonize many of the usual laboratory animal species, including conventionally reared rats, mice, rabbits, guinea pigs, specific-pathogen-free pigs, colostrum- deprived piglets, and gnotobiotic rats and mice (9; Kra- kowka et al., Second Int. Symp. on C. pylori, in press). We and others have shown that the gnotobiotic neonatal piglet is susceptible to oral infection with H. pylori (15, 16). In this model, infection is limited to the stomach, and the lesions that develop are those of lymphoplasmacytic gastritis, re- sembling human infection (5). The gnotobiotic pig model, however, is limited because of the inability to maintain pigs in the gnotobiotic state for greater than 45 to 60 days because of size and nutritional constraints. In addition, study of ulcerogenesis may be confounded by the susceptibility of weanling pigs to the development of ulcers induced by diet and stress (3). The advantages of a dog model include the ability to maintain the animals in the gnotobiotic condition for years, the availability of well-established methods for studying immunologic and gastric physiologic responses, and the lack of propensity to develop spontaneous ulcers. * Corresponding author. The objective of this study was to determine whether gno- tobiotic dogs are susceptible to gastric infection by H. pylori. MATERIALS AND METHODS Animals. A litter of seven gnotobiotic beagle pups was derived from specific-pathogen-free bitches by standard methods (14). They were maintained in sterile Pentub isola- tion units and fed a diet of Esbilac (PatAg, Inc., Hampshire, Ill.). Initially, unchallenged control dogs were housed sepa- rately from inoculated dogs. Beginning on day 23 postinfec- tion (p.i.), controls were housed together with infected dogs to determine whether infection would spread via contact. Bacterial inoculum. A virulent strain of H. pylori, 26695, was used. This is a human isolate and is capable of coloniz- ing and producing gastritis in gnotobiotic piglets (7). Bacteria were grown in 250-ml Erlenmeyer flasks containing brucella broth (Difco Laboratories, Detroit, Mich.) supplemented with 10% fetal calf serum (BBL Microbiology Systems, Cockeysville, Md.). Flasks were incubated at 37°C in a 10% CO2 atmosphere on a rotary shaker at 150 rpm. Cultures were harvested by centrifugation 24 h after inoculation (logarithmic growth phase), washed, and suspended in pep- tone water. Organisms were enumerated with a hemacytom- eter and by standard plate count. An inoculum of 3 x 108 CFU in 2.0 ml of peptone water was prepared. Experimental design. At 7 days of age, five pups (group A) were orally challenged with 3 x 108 CFU of H. pylori in 2.0 ml of peptone water. Group B (two pups) served as controls and were given peptone water alone. Group B was initially housed separately from their infected littermates. On day 7 p.i., two pups from group A and one from group B were gavaged, and the stomach contents were cultured for reiso- lation of H. pylori. In order to ascertain whether contact infection was possible, the dogs in group B were subse- quently housed with their infected littermates beginning on day 23 p.i. Blood samples were collected prior to challenge 2606