Helicobacter mustelae-Associated Gastritis in Ferrets An Animal Model of Helicobacter Gastritis in Humans
JAMES G. FOX, PELAYO CORREA, NANCY S. TAYLOR, ADRIAN LEE, GLEN OTTO, JAMES C. MURPHY, and REBECCA ROSE Massachusetts Institute of Technology, Division of Comparative Medicine, Cambridge, Massachusetts; Louisiana State University Medical Center, Department of Pathology, New Orleans, Louisiana: University of New South Wales, School of Microbiology, Sydney, Australia
Gastric Helicobacter mustelae was present in 100% of 11 adult female ferrets (Mustela putorius furo). The high immunoglohulin G antibody levels to H. mustelae in all ferrets showed a significant immune response to the organism. Urease mapping of the ferret stomach indicated that the bacteria heavily colonized the proximal duodenum and antrum and, to a lesser extent, the corpus. The histological gastritis observed coincided with presence of H. mustelae. Superficial gastritis was noted in the oxyntic gastric mucosa, whereas in the distal antrum the chronic inflammatory response occupied the full thickness of the mucosa. In the proximal antrum and transitional mucosa, focal glandular atrophy and regeneration were observed. Seven control specific-pathogen-free ferrets were not colonized with the bacteria, did not have detectable levels of immunoglobulin G H. mustelae antibody, and did not have H. mustelaeassociated gastritis. The ferret lacks the polymorphonuclear-cell response seen in active chronic gastritis typically described with Helicobacter pylori gastritis in humans. However, the lesion in ferrets does closely resemble the diffuse antral gastritis seen in a subset of adults with H. pylori gastritis as well as children infected with H. pylori. Like H. pylori, H. mustefae adheres tightly to gastric mucosa. The ferret infected with H. mustelae, in addition to specific-pathogen-free uninfected control ferrets, will make longitudinal studies possible, enabling dissection of multiple host and environmental variables that influence the effect of H. mustelae colonization on progression and severity of gastroduodenal disease.
erious study of an infectious etiology in gastric disease followed when Marshall and Warren first described Campylobacter (now named Helicobacter) pylori and its association with gastritis in humans (1,Z). Helicobacter pylori is isolated from the antral and fundic gastric mucosa of infected individuals and is almost always associated with leukocytic infiltration (3-6). When infected individuals are treated and eradication of the organism occurs, there is corresponding elimination of gastric inflammation (4,5). Koch’s postulates also have been fulfilled in two human volunteers who ingested H. pyiori and subsequently developed gastritis with acute gastric symptoms (7,6). It has been proposed that after H. pylori induces an initial acute inflammatory response, the inflammation changes to an active chronic gastritis which may persist for years in asymptomatic patients. This observation is based on limited longitudinal studies in which ingestion (experimental or iatrogenic) of H. pylori caused acute gastritis which progressed in these patients to the active chronic gastritis typically associated with H. pylori infection (7,9,10). Because of the increasing evidence that H. pylori is a significant gastroduodenal pathogen, experimental animal models are needed to study the pathogenesis of the disease (11~2). Candidate animal models include gnotobiotic piglets and possibly nonhuman primates (13-15). Each of these has
Abbreviations used in this paper: JZLISA,enzyme-linked immunosorbent assay; HLO, Helicobocter-like organisnw; SG, super& cial gastritie; SPF, epeci5c pathogen free. 0 1999 by the American Gaatroenterological Aesodation 0016-5095/90/$3.00
disadvantages with regard to cost, availability, housing requirements, and ease of administering oral antimicrobials for treatment of H. pylori-associated gastritis and/or ulcers. Experimental oral challenge of H. pylori in rodents and rabbits does not establish gastric colonization or induce gastritis (15,16). The domestic ferret is a monogastric carnivore whose stomach has many anatomical and physiological characteristics similar to the human stomach [X7). The ferret is increasingly being used because of its commercial availability, relative low cost, ease of housing and husbandry, and tractable nature. We recently isolated and named another gastric Helicobacter, H. mustelae, with many of the same biochemical, molecular, and phenotypic characteristics as H. pylori, from the antrum and fundus of ferrets (18). In our experience, almost all adult ferrets are colonized with H. mustelae and have associated gastritis and/or ulcers (19,20). Since our original report, H. mustelae has been isolated from ferret stomachs in Canada, England, and Australia but not from ferrets from New Zealand (18,21,22). The purpose of this study is to describe the pathological presentation and immunologic response of H. mustelae-associated gastritis in ferrets and compare these lesions to those described in H. pyloriassociated gastritis in humans.
were taken from the antrum, fundus, and the duodenal bulb. At 3, 4, and 5 months, biopsies were obtained through gastroscope from the antrum and body of the seven anesthetized, foster-reared, SPF ferrets raised in our animal facilities. Biopsies of stomach and duodenum obtained at necropsy were processed within 1 hour, minced using sterile scalpel blades, and inoculated onto blood agar plates supplemented with trimethoprim, vancomycin, and polymyxin B (Remel, Lenexa, KS]. The plates were incubated at 37% in microaerophilic conditions in vented jars containing N,, H,, and CO, (8O:lO:lO) for 3-7 days. Bacteria were identified as H. musteloe based on Gram’s stain, morphology of the organism, and strong oxidase and catalase positivity, sensitivity to nalidixic acid, and resistance to cephalothin. Using Christensen urea agar, investigators noted a positive urease reaction within 5 minutes (18).
Urease Mapping Test Five of the 11 adult ferret stomachs and proximal duodenum, (Figure 1) plus antral and fundus biopsies from seven SPF ferrets were assayed for urease production using the method of Hazel1 et al. (23).
Enzyme-Linked Immunosorbent Anti-H. mustelae Antibody
Antigen preparation. Antigen was prepared using whole cell extracts following previously
Materials and Methods Animals
Eleven adult (12-24 months old] female domestic ferrets [MusteJa putoriusfuro) were obtained from a commercial vendor who sells ferrets for biomedical research and to the pet market (Marshall Farms, North Rose, NY). Animals maintained at the commercial vendor are fed a pelleted ferret diet. While young kits are being nursed, the diet is supplemented with raw meat products. Upon arrival, animals were housed in our American Association for Accreditation of Laboratory Animal Care-accredited animal resource facility for a period of 1-2 months before they were euthanized. Adult ferrets were maintained as pairs in suspended stainless steel cages (13 x 24 x 17 in.). Ferrets were given food (Purina Cat Chow, Ralston Purina Co., St. Louis, MO) and water ad libitum. Four other adult ferrets, which received combined oral antibiotic and bismuth treatment for 4 weeks, have proven to be H. mustelae negative on gastric biopsy at 1,2, and 4 months after treatment. Seven H. mustelae-negative specific-pathogen-free (SPF) ferrets that were offspring of the H. mustelae-free jills served as controls (Otto G, Fox JG, Taylor NS. Antimicrob Agents Chemother, in press].
Microbiology Stomachs examined at necropsy were opened along the greater curvature using sterile instruments, and samples
duodenum Figure 1. Biopsy sites of m-ease mapping of ferret stomachs. A, upper lesser curvature; B, lower lesser curvahue; C,D, corpus; E,F, entrum.
354 FOX ET AL.
GASTROENTEROLOGY Vol. 99. No. 2
for H. pylori enzyme-linked immunosorbent-assay (ELISA) used in our laboratory (6). Three isolates of H. mustelae (including ATCC 43772) from gastric biopsies of ferrets were used. Antigen to Campylobacter jejuni was similarly prepared from two isolates of C. jejuni obtained from ferrets. Assay. The ELISA was carried out according to methods of Fox et al. (6). Titer of serum samples was expressed as the dilution of serum giving a reading equal to the mean plus 2 SD of the negative controls.
Antrum, fundus, and duodenum were collected at necropsy from each animal adjacent to sites of bacterial culture and fixed in 10% buffered neutral formalin. Gastric biopsies of fundus and antrum were taken at endoscopy in the seven SPF ferrets at 5 months of age. Tissues were processed by standard methods, embedded in paraffin, and five micron sections were stained with H&E, WarthinStarry, and alcin blue stains. Sections of antrum, body, and duodenum bulb were examined for histological changes and presence of H. mustelae.
Immunization procedure. Two positive and two neg ative control serum samples were run on each plate. Positive control sera were obtained from adult ferrets hyperimmunized with three isolates of H. mustelae (ATCC no. 43722 and two other strains] and one isolate of C. jejuni obtained from a diarrheic ferret. Bacterin for immunization was prepared by harvesting growth from H. mustelae plates at 3-4 days and C. jejuni plates at 48 hours. Organisms were grown on blood agar plates (Remel) at 37°C under microaerophilic conditions. The OD, of the suspensions was adjusted to approximately 0.7. After heating at 62°C for 45 minutes, the cultures were mixed such that 3 mL of each H. mustelae culture and 2 mL of C. jejuni were mixed together. The bacterin was cultured to determine sterility prior to immunization Vaccination consisted of mixing bacterin 1:lwith Freund’s incomplete adjuvant and administering 0.5 cc SC at four sites along the back of each animal. This procedure was repeated after 2 weeks. Animals were test bled before immunization and at intervals after immunization to determine antibody response. Negative control sera were obtained from two caesarian-derived gnotobiotic ferret kits that had no exposure to Helicobacter species, kindly supplied by Dr. 1. Bell of the University of Wisconsin (24). Adsorption/inhibition. To assess the specificity of the C. jejuni and H. mustelae ELISAs, individual serum samples at a dilution of 1:50 were incubated at 37’C with 30 bugof antigen preparation from the appropriate organism for 60 minutes. Following incubation, the samples were tested in the ELISA.
Pathology Pathological classification of a histological diagnosis of stomachs for each of 11 ferrets and gastric biopsies of the seven SPF ferrets was made by independent interpretation of the H&E slides: this following criteria used in human pathological classification of inflammation of the stomach: superficial gastritis, diffuse antral gastritis, and multifocal chronic atrophic gastritis (25). Superficial gastritis refers to inflammatory infiltrates [either polymorphonuclear or mononuclear] confined to the superficial layers of the mucosa. Diffuse antral gastritis refers to infiltrates of lymphocytes and plasma cells occupying the full thickness of the antral mucosa; this pathological entity is most frequently a part of the duodenal ulcer syndrome. Chronic atrophic gastritis refers to multifocal glandular atrophy and inflammation observed in antrum and body.
For transmission electron microscopy (TEM), intact biopsies of the antrum and fundus were fixed and embedded using previously described techniques (26). Thin sections (50-70 pm) were examined with a Hitachi H-7000 transmission electron microscope (Hitachi Ltd., Tokyo, Japan).
Results Microbiology All 11 ferrets had H. mustelae isolated from the antrum and proximal duodenum. Gastric biopsies taken from the SPF ferrets biopsied at 3, 4, and 5 months
of age were negative
for H. mustelae.
All 11 animals had antibody titers to H. mustelae ranging from 1:157 to 1:2048 as measured in the ELISA [See Table 1). This ELISA was specific for H. mustelae since 71%-87% of activity in sera as measured by the ELISA was adsorbed with H. mustelae antigen; however, only O%-26% of the H. mustelae activity was absorbed from serum by C. jejuni antigen. 70%Using C. jejuni antigen, activity was reduced 97% when adsorbed with sera from ferrets immu-
nized with C. jejuni. Also, adsorption of these two sera with H. mustelae antigen did not reduce C. jejuni activity in the ELISA. Antibody to H. mustelae was not detected in the two gnotobiotic ferret kits or in the seven SPF &month-old ferrets. Pathology At necropsy, in three of the 11 ferrets, 5-mm to l-cm ulcers were present at the pyloroduodenal junction. No other lesions were observed grossly in the stomach or duodenum. In our histological assessment of the pylorus and proximal duodenum, Brunner’s glands (submucosal glands] began at the pyloric sphincter and extended into the duodenum for approximately 1 cm. Mucosa over the Brunner’s glands was foveolar gastric epithe-
GASTRITIS IN FERRETS
Table 1. Enzyme-Linked lmmunosorbent Assay Immunoglobulin G Antibody Titers With and Without Absorption Using Whole Cell-Soluble Extracts of H. mustelae and C. jejuni in Ferrets Colonized With H. mustelae C. jejuni ELISA
H. mustelae ELISA
% Inhibition when absorbed with:
% Inhibition when absorbed with: H. mustelae
1:512 1:580 1:157 1:294 1:478 1:1885 1:548 1:2048 1:416 1:338 1:724 >1:2000 >1:2000
71 78 78 79 82 85 76 74 87 75 81 50 54
7 6 25 12 0 26 0 20 0 0 6 0 0
Sample no. I
2 3 4 5 6 7 8 9 10 11 8594b 1482b
+ + + + + + + + + + + + ND
H. mustelae antigen 0
27 0 25 21 0 0 0 37 18 27 20 ND
C. jejuni antigen 84 78 74 88 84 95 93 74 95 92 97 95 ND
“OD,,, > the mean + 2 SD of the negative control.
bFerrets immunized with H. mustelae and C. jeiuni. There were two germ-free mustelae or C. jejuni.
lium. The distal duodenum displayed Brunner’s glands covered by duodenal epithelium with absorptive and goblet cells. The stomachs of the 11 ferrets had a similar histological appearance. The histopathologic lesions consisted of the following basic elements: mucus depletion, gland loss and regeneration, and leukocytic infiltrate. The characteristics of each element varied according to topography. In the oxyntic mucosa (corpus-fundus),
Figure 2. Oxyntic mucosa with mudn deqhtion of superilcial foveolar cell8 and monoimdear lenbqtfc inllltrate prima&y in the superiidal lamha propria, with focal in6ltrates at the be of the mucosa. Super5dal gastritis(HBE).
and seven SPF ferrets with no detectable
antibody to H.
mucus depletion and leukocytic infiltrate were the only elements present and were specifically localized in the superficial part of the mucosa. The leukocytic infiltrate was almost exclusively localized in the lamina propria and not in the epithelial layer: it was predominantly mononuclear (lymphocytes and plasma cells) with only occasional eosinophilic and neutrophilic polymorphonuclear cells (Figure 2). This is the classical picture of the so-called superficial gastritis
356 FOX ET AL.
Figure 3. Helicobacter organisms in the gastric pits of the corpus and adjacent to the foveolar cells (Warthindtarry stain).
(SC). The Warthin-Starry stains showed abundant Helicobacter organisms at the surface and in the lumen of the foveola (pits]; the organisms were located in the topography of the inflammation (Figure 3). In the antrum, the lesions were different in the proximal and the distal portions. In the proximal
GASTROENTEROLOGY Vol. 99, No. 2
portion and in the transitional zone (antrum-corpus junction), all the elements described earlier were present. The mucus depletion was evident in the foveolar cells. The leukocytic infiltrate occupied the full thickness of the mucosa; it was almost totally mononuclear and interstitial with only occasional presence of polymorphonuclear leucocytes in the epithelium or the lumen (so called “activity”) (Figure 4). Gland necrosis (atrophy) was focal and affected clusters of glands from top to bottom, leaving neighboring glands unaffected. Active glandular regeneration was evident in these areas (Figure 5). The foci of atrophy and regeneration were not abundant and rather far apart from each other. Warthin-Starry stain demonstrated very large colonies of H. musteiae at the surface, in the pits, and occasionally penetrating deep into the glandular lumens in groups of neighboring glands (Figure 6). The distal antrum, including the prepyloric, pyloric, and proximal duodenal mucosa lined by foveolar cells, displayed diffuse interstitial mononuclear infiltrate and mucus depletion but no epithelial necrosis or regeneration (Figure 7). Warthin-Starry stains demonstrated abundant H. mustelae on the surface, in the pits, and in the superficial portion of the gland lumens. The electron microscopic evaluation clearly depicts H. mustelae closely adhering or partially penetrating the epithelial cells (Figure 8). In many sections, the microvilli appear to be surrounding the bacteria. Biopsies of gastric antrum and fundus in seven SPF 5-month-old ferrets did not have evidence of gastritis,
Figure 4. Superficial portion of the proximal antral mucosa showing superficial erosion, mucus depletion, mononuclear interstitial Mltrate, and occasional polymorphonuclear leukocytes in the glandular lumina and the surface (arrows) (H&E].
H. MLJSTELAE GASTRITIS
and H. mustelae was not observed stains.
Gastric Biopsy Urease The positive urease reactions on five ferrets are indicative of H. mustelae colonization, and correlate with positive cultures of H. mustelae in the five cases from the antrum and proximal duodenum (Table 2). The antrum and proximal duodenum were heavily colonized as indicated by the rapid urease production; at least one of the four antral biopsies from each ferret was positive on the urease assay at 2-4 hours, whereas the proximal duodenum was positive in 3:s ferrets by 6 hours (Table 2). The corpus was colonized in much lower numbers; positive reactions were not recorded from corpus biopsies (except one site in one ferret) until 6 hours. This assay correlated well with H. mustelae visualized by Warthin-Starry stain, i.e., sparse colonization and focal superficial gastritis in the corpus, and heavy H. mustelae colonization with diffuse antral gastritis extending into the proximal duodenum. None of the gastric biopsy samples from the seven SPF ferrets biopsied at 3, 4, and 5 months had positive urease tests.
Figure 5. Proximal and mucosa with diffuseinterstitialmononuclear infiltrate occupying the full thickness of the mucosa. Mucus depletion and glandular regeneration are present (H&E).
Figure 6. Gland of the proximal antrum containing large numbers of Heiicobucter organisms (Warthin-Starry stain j
It is reasonable to assume that if H. mustelae is a true pathogen, an appreciable immune response would be noted in infected ferrets. Our results indicate that the correlation between prevalence of H.
FOX ET AL.
GASTROENTEROLOGY Vol. 99. No. 2
Figure 7. Distal antral mucosa showing diffuse interstitial mononuclear infiltrate occupying the full thickuess of the mucosa. Diffuse antral gastritis (H&E).
mustelae gastric infection and corresponding elevation of immunoglobulin G (IgG) antibodies to H. mustelae is very high as is the case with H. pylori infection in man. In comparison with H. pylori infection in man and gnotobiotic piglets, elevated IgG antibodies to H. mustelae does not result in eradication of the organism. Our ELISA uses a whole cell antigen extract preparation and provides an accurate discrimination between H. pylori positive and negative sera (6,271. The two germ-free and seven SPF young ferrets in our study, which were not colonized with H. musteiae, had neither histological gastritis nor detectable IgG antibody to H. mustelae. In our previous studies, we ascertained that H. mustelae colonization of the ferret stomach occurs at a very young age, probably at weaning [i.e., 5-6 weeks of age], and is apparently a persistent infection with nearly 100% of adult ferrets sampled being infected (20). Therefore, it is not surprising that all 11 adult ferrets in this study were heavily colonized with H. mustelae. The urease mapping done in the stomach of the ferrets indicates heavier colonization in the antrum than the fundus; interestingly, the organism also colonizes in large numbers in the proximal duodenum which, from our evaluation, is lined with foveolar epithelium. Comparative studies on gastrointestinal mucosubstances indicate that mammalian species differ in their distribution of antral tissue. Unlike most rodents, nonhuman primates, and man, the hamster has antral tissue extending beyond the normal duodenal antral junction; this appears analogous to what we have observed in the ferret (28).
In England, ferret stomachs have been examined for the presence of gastric Helicobacter-like organisms (HLO) (19). Although specific location of gastric sampling was not included, HLO were isolated from all of 17 gastric samples; morphologically and biochemically, the bacteria fulfilled the criteria for H. mustelae (18). Histologically, bacteria were identified in 1~7 antral specimens in antral gastric pits, but none were seen in the body or the duodenum. Chronic gastritis was not observed; 7:14 colonized antral specimens did have occasional debris within gastric pits and accompanying intraepithelial polymorphonuclear cell infiltrate. These changes were not seen in noncolonized areas of the antrum nor in the body. Unfortunately, the authors did not state the age of the ferrets, but instead referred to them as mature animals. The mild lesions noted may be indicative of sparse colonization of H. mustelae in younger animals. Others examining H. mustelae-colonized stomachs from mature male ferrets noted that 1 of 14 had small antral erosions which histologically consisted of surface epithelial loss with vascular congestion and reepithelization at the lesion margin. Gastric HLO were abundant adjacent to the lesions [29). All 14 ferrets had chronic gastritis deep in the antral mucosa, which sometimes extended into the submucosa. Gastric HLO were seen in the gastric pits, but did not extend to the deep mucosa (29). The moderate to severe lesions noted in ferrets included in this study perhaps reflect the age of the animals and a longer period of persistent infection with H. mustelae. In addition, constant exposure to the organism, which is hyperendemic within the confined
GASTRITIS IN FERRETS
0. Large overell view of Helicobacter mwtelae in geshk epithelium of antrum from ferret no. 5 (original magniflcatfon ~13,730). Bar = 10 pm.
A. Small photograph of cross section of the bacterium with cross section of villi surrounding it [original magnification x 26,600). B. Small photograph showing organism lying next to villi (original magnification x33.000). C. Small photograph showing the depth the bacteria penetrate [original magnification x 26.600).
mustelae. In the oxyntic mucosa, both are limited to the superficial portion [superficial gastritis). In the distal antrum, the inflammation occupies the full thickening of the mucosa, the so-called diffuse antral
environment of our ferrets maintained at the commercial facility, may elicit a more pronounced pathology. The histopathological changes that were observed closely coincide in topography with the presence of H. Table 2. Urease Mapping of Ferret Stomachs for H. mustelae Animal No. 1 4 5 7 8
Stomach sites biopsied” Hoursb 2 6 2 6 2 6 4 16 2 6
+ + _
+ + _
+ _ -
+ + + +
+ + + _
+ + + + _
+ + +
+ + +
+, positive urease reaction; -, negative urease reaction. “See Figure 1 for stomach sites biopsied. bTime elapsed from the point of biopsy placement into urea broth.
+ + +
360 FOX ET AL.
gastritis described in humans. H. mustelae are seen at the surface, the pits, and the superficial portion of the glands. In the proximal antrum and the transitional mucosa, the element of focal glandular atrophy and regeneration is added to the lesions described in the distal antrum. This coincides with deep focal HLO colonization of groups of antral glands. The pathology of Helicobacter-associated gastritis in the ferret model has many similarities with the human disease and contributes considerably to the interpretation of chronic gastritis in humans. The lesions observed in the distal antrum and the oxyntic mucosa bear a close resemblance to the diffuse antral gastritis observed in humans which, like the ferret model, is usually accompanied by SG of the corpus. This is the clinicopathological entity which underlies the duodenal ulcer syndrome (1~5). The changes observed in the proximal antrum and the transitional zone appear to represent early stages of the multifocal atrophic gastritis of humans, the entity which underlies the gastric ulcer and gastric carcinoma syndromes (6,25). Another interesting feature that may be associated with H. mustelae infection in the ferret is gastric and duodenal ulcers. Gastric ulcers and, to a lesser extent duodenal ulcers, are a common necropsy finding in both pet and research ferrets in the United States; gastric ulcers also have been observed in ferrets from In the present England and New Zealand (17,30,31). study, H. mustelae was seen in association with antral ulcers. These ulcers were also accompanied by occasional microabscesses in the glandular portion of the antrum. Both of these pathological features are seen in man but have not been reported in the neonatal gnotobiotic piglet infected with H. pylori (15). In general, therefore, H. mustelae-associated gastritis in the ferret is an approximate replica of many aspects of human H. pylori gastric infection, i.e., diffuse antral gastritis and SG in the corpus. This lesion is typical of what is described in the gnotobiotic piglet model (15). What the ferret, the rhesus monkey, and piglet models do not depict is a severe degree of the active component of the gastritis. The activity of the syndrome is defined as polymorphonuclear infiltrate of the epithelial cells extending to the lumen of the gland and the mucosal surface (25). Many authors consider active chronic gastritis a hallmark of H. pylori-associated infection. However, a subset of adult patients with H. pylori infection, as well as children with elevated IgG H. pylori antibodies, have follicular gastritis comparable to that seen in ferrets and gnotobiotic piglets (32,331. As in H. pylori infection in humans, H. mustelae only appears to colonize foveolar epithehum and tightly adheres to epithelial cells (34). Further longitudinal studies of the naturally occurring disease in the ferret, coupled with the availability of
Vol. 99, No. 2
ferrets not colonized with H. mustelae to serve as controls, should prove to be an extremely useful model in studying the pathophysiology of H. pylori gastroduodenal disease in humans. References I. Marshall BJ, Warren JR. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet i984;i:i3ii-i3i5. 2. Goodwin CS, Armstrong JA, Chilvers T. Peters M, Collins MD, Sly L, McConnell W, Harper WES. Transfer of Campylobacter pylori to Helicobacter gen. nov. as Helicobacter pylori comb. nov. and Helicobacter mustelae comb. nov.. respectively. Int J Syst Bacterial 1969;39:397-465. 3. Marshall BJ, McGechie OB, Rogers PA, Glancy RJ. Pyloric Campylobacter infection and gastroduodenal disease. Med J Australia 1985;142:436-439. 4. Rauws EAJ, Langenberg W, Houthoff HJ, Zanen HC, Tytgat GNJ. Campylobacter pyloridis associated chronic active antral gastritis. Gastroenterology 1988;94:33-40. 5. Blaser MJ. Gastric Campylobacter-like organisms, gastritis and peptic ulcer disease. Gastroenterology 1987;93:371-383. 6. Fox JG, Correa P, Taylor NS, Zavala D, Fontham E, Janney F, Rodriguez E, Hunter F, Diavolitsis S. Campylobacter pylori associated gastritis and immune response in a population at increased risk of gastric carcinoma. Am J Gastroenteroll969;89: 775-781. 7. Morris A, Nicholson G. Ingestion of Campylobacter pyloridis causes gastritis and raised fasting gastric pH. Am J Gastroenterol 1987;82:192-199. a. Marshall BJ, Armstrong JA, McGechie DB, Glancy R. Attempts to fulfill Koch’s postulates for pyloric Campylobacter. Med J Australia 1965;142:436-439. 9. Frommer DJ, Carrick J, Lee A, Hazel1 SL. Acute presentation of Campylobacter pylorigastritis. Am J Gastroentero11986;83:11681171. 10. Graham DY, Klein PD. Campylobacter pyloridis gastritis: the past, the present, and speculations about the future. Am J Gastroenteroll987;82:283-286. 11. Graham DY. Campylobacter pylori and peptic ulcer disease. Gastroenterology 1969;96:615-625. 12.Marshall BJ, Goodwin CS, Warren JR, Murray R, Blincow ED, Blackbourn SJ, Phillips M, Waters TE, Sanderson CR. Prospective double-blind trail of duodenal ulcer relapse after eradication of Campylobacter pylori. Lancet 1988;2:1437-1442. 13.Baskerville A, Newell DB. Naturally occurring chronic gastritis and C. pylori infection in the rhesus monkey: a potential model for gastritis in man. Gut 1966;29:465-472. 14.Brondson MA, Schoenknecht FD. C. pylori isolated from the stomach of the monkey, Macaca nemistrina. J Clin Microbial i988;26:1725-1728. 15.Krakowka S, Morgan DR, Kraff WG, Leunk RD. Establishment of gastric Campylobacter pylori infection in the neonatal gnotobiotic piglet. Infect Immun 1987;55:2789-2796. 16.Fox JG, Lee A. Gastric Campylobacter-like organisms: their role in gastric disease in laboratory animals. Lab Anim Sci 1989;39: 543-553. 17.Fox JG. Bacterial mycoplasmal diseases. In: Fox JG, ed. Biology and diseases of the ferret. Philadelphia: Lea and Febiger, 1988:202. ia. Fox JG, Chilvers T, Goodwin CS, Taylor NS, Edmonds P, Sly LI, Brenner DJ. Campylobacter mustelae, a new species resulting from the elevation of Campylobacter pylori subsp. mustelae to species status. Int J System Bacterial 1969;39:301-303. 19.Fox JG. Edrise BM, Cabot E, Beaucage C, Murphy J, Prostak KS.
Campylobacter-like organisms isolated from gastric mucosa of ferrets. Am J Vet Res 1986;47:238-239. 20.Fox JG, Cabot EB. Taylor NS, Laraway R. Gastric colonization of Campylobacter pylori subsp. mustelae in the ferret. Infect Immun 1988;58:2994-2996. 21.Morris A. Thomasen L, Tasman-Jones C, Nicholson G. Heap M. Failure to detect gastric Campylobacter-like organisms in a group of ferrets in New Zealand. N Z Med J 1988;101:275. 22.Tompkins DS, Wyatt JI, Rathbone BJ, West AP. The characterization and pathological significance of gastric Campylobacterlike organisms in the ferret: a model for chronic gastritis? Epidemiol Infect 1988;101:289-278. 23.Hazel1 SL, Brody TJ, Gal A, Lee A. Campylobacter pyloridis gastritis I: detection of urease as a marker of bacterial colonization and gastritis. Am J Gastroenterol1987;82:292-296. 24.Bell JA, Manning DD. Derivation of gnotobiotic ferrets: perinatal diet and hand-rearing requirements, Lab Anim Sci [in press]. 25.Correa P. Chronic gastritis. A clinico-pathological classification. Am J Gastroenterol1988;83:504-509. 26.Lee A, Hazel1 SL. O’Rourke J, Kouprach S. Isolation of a spiral-shaped bacterium from the cat stomach. Infect Immun 1988;56:2843-2850. 27.Perez-Perez GI, Dworkin BM, Chodes JE, Blaser MJ. CampyJobatter pylori antibodies in humans. Ann Intern Med 1988;109: 11-17. 28.Sheahan DG, Jervis HR. Comparative histochemistry of gastrointestinal mucosubstances. Am J Anat 1977;146:103-132. 29.Jeffries L, Buckley DE, Blower PR, Plumb JN. Comparative sensitivities to antimicrobial agents of C. pylori and the gastric
GASTRITIS IN FERRETS
Campylobacter-like organism from the ferret [letter). J Clin Path01 1987;40:1265-1268. 30.Andrews PL, Illman 0, Mellersh A. Some observations of anatomical abnormalities and disease states in a population of 35Oferrets(MusteJaputoriusfuro). Z Versuchstierkd. 1979;21:346353. 31.Anonymous. Diseases of the Fitch. Surveillance. 1984;11:1-28. 32.Czinn SJ, Dahms BB, Jacobs GH, Kaplan B, Rothstein FC. Campylobacter-like organisms in association with symptomatic gastritis in children. J Pediatr 1986;109:80-83. 33.Czinn SJ, Carr H, Sheffler L, Aponoff S. Serum IgG antibody to the outer membrance proteins of Campylobacter pylori in children with gastroduodenal disease. J Infect Dis 1989;159:586589. 34.Chen XG, Correa P. Offerhaus J. Rodriguez E, Janney F. Hoffmann E, Fox J, Hunter F, Diavolitsis S. Ultrastructure of the gastric mucosa harboring Campylobacter-like organisms. Am J Clin Path01 19888657%582.
Received November 27,1989. Accepted February 13,199O. Address requests for reprints to: James G. Fox, D.V.M., 37 Vassar Street (45-104). Division of Comparative Medicine, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139. This research was supported in part by Public Health Service Grants ROl-A125631 from the National Institute of Allergy and Infectious Diseases, POl-CA-28731 from the National Cancer Institute, and RR01046-14 and T32-RR07036-02 from the Division of Research Resources; and by a grant from Proctor and Gamble and the National Health and Medical Research Council of Australia.