Idea Transcript
Laboratory Animal Science Copyright 1998 by the American Association for Laboratory Animal Science
Vol 48, No 3 June 1998
Experimentally Induced Infection with Helicobacter pylori in Squirrel Monkeys (Saimiri spp.): Clinical, Microbiological, and Histopathologic Findings Christian T. K.-H. Stadtländer,1,2* J. David Gangemi, 1,2 Fred J. Stutzenberger,2 John W. Lawson,2 Brian R. Lawson,2 Supriya S. Khanolkar,2 Kimberly E. Elliott-Raynor,2 Harold E. Farris, Jr.,3 Linda K. Fulton,3 Joseph E. Hill,4 F. Kay Huntington,5 Cynthia K. Lee,6 and Thomas P. Monath6 Helicobacter pylori is recognized as a pathogen in association with type-B gastritis, peptic ulcer disease, and gastric cancer (1). The motile bacterium prefers a microaerophilic environment and is a strong urease producer, allowing it to survive in the hostile environment of the stomach. Urease and motility have been described as virulence factors. Other factors that may also contribute to the development of disease include adhesions, superoxide dismutase, catalase, cytotoxin-associated protein (CagA), and vacuolating cytotoxin (VacA) (1). Helicobacter pylori can establish lifelong infections in humans and other primates (1, 2). Several animal models have been developed to study Helicobacter pathogenesis and to test vaccines and antibiotics for eradication of organisms from gastric mucosa. Mice, rats, ferrets, dogs, cats, pigs, and nonhuman primates have been used as natural or experimental infection models (1). Nonhuman primate species include rhesus monkeys (2), Japanese monkeys (3), pigtail macaques (4), cynomolgus monkeys (5), baboons (6), and chimpanzees (6). We recently reported that squirrel monkeys (Saimiri spp.) were susceptible to experimentally induced infection with H. pylori, and adapted the [13C]urea breath test to monitor the infection (7). We also used squirrel monkeys to examine the immunogenicity and safety of a recombinant H. pylori urease vaccine (8). We describe the clinical, histopathologic, and microbiological findings of infection with H. pylori in squirrel monkeys. Five female squirrel monkeys ( S. sciureus and S. boliviensis) weighing 650 to 750 g were obtained as wild catches from the University of South Alabama (Mobile, Ala.). Monkeys were pair or individually housed in 0.72 x 1.44 x 0.83-m cages under conditions of controlled temperature (258C), humidity (60 6 10%), lighting (12-h light/dark cycle), and air (12 to 14 complete air changes/h). Greenville Hospital System/Clemson University Biomedical Cooperative, 1 Department of Microbiology,2 and Research Services, 3 Clemson University, Clemson, South Carolina; Livestock Diagnostic Laboratory,4 Clemson University, Columbia, South Carolina; Gastroenterology Associates, P.A.,5 Professional Medical Center at St. Francis Hospital, Greenville, South Carolina; and OraVax, Inc.,6 Cambridge, Massachusetts *Address correspondence to Dr. C. T. K.-H. Stadtländer, School of Public Health, University of Alabama at Birmingham, 120 Ryals Building, 1665 University Blvd., Birmingham, AL 35294.
They were fed New World Monkey Chow (Harlan Teklad, Madison, Wis.), and water was available ad libitum. Routine care for the monkeys was provided by our Department of Research Services. Prior to inoculation with H. pylori, each monkey was examined and determined to be clinically healthy and free of Helicobacter spp., on the basis of results of culture of gastroscopy biopsy specimens and the urease test, and histopathologic findings. The H. pylori strain 12476, an isolate from a rhesus monkey (5), was used for inoculation experiments. Inocula were prepared by washing 72-h surface growth of H. pylori from Columbia blood agar plates with Brucella broth. Three S. sciureus (G4390, G657, G440) were inoculated with approximately 10 8 to 109 colony-forming units of H. pylori directly into the stomach via an adult/ pediatric feeding tube (no. 3642; Davol, Inc., Cranston, R.I.). Two S. boliviensis (B1024 and B1417) were initially retained in separate housing as controls and received only Brucella broth. The inoculation schedule was as follows: monkey G4390 was inoculated with H. pylori at 0, 2, 5, and 11 days and at 5 months; monkey G657 was inoculated at 0 and 11 days; and monkey G440 was inoculated at 0, 11, and 35 days and at 5 months. Monkeys B1024 and B1417 were inoculated with H. pylori at 5 months. Gastroscopies were performed by use of a 5.5-mm-diameter pediatric bronchoscope (Olympus America, Inc., Lake Success, N.Y.), with tracheal intubation for sufficient air ventilation. Gastric biopsy specimens (up to six/procedure) were collected from fundus and antrum at 1 month prior to inoculation, and 1, 2, 5, and 6 months after inoculation. Specimens were divided for culture and microscopic examination. In addition, brushings of the gastric mucosa were obtained for cytologic examination. Biopsy specimens for culture were ground in 2 ml of Brucella broth, and 0.3 ml of this suspension was inoculated onto Columbia blood agar. Agar plates were incubated for up to 15 days at 378C in an anaerobic jar (GasPak-100; Becton Dickinson, Cockeysville, Md.) with 10% carbon dioxide, using the Campy Pak Plus System (Becton Dickinson). Isolates were further investigated by urease testing (9) and analysis of proteins and antigens (10).
303
Vol 48, No 3 Laboratory Animal Science June 1998
Table 1. Infection and gastritis in squirrel monkeys inoculated with H. pylori Infection with H. pyloria Urease LM TEM test
Gastritis scoresb Before After challenge challenge
Monkey no.
Clinical signs of disease
Biopsy culture
G4390
Restless, vomiting, diarrhea, decreased appetite
+[2]
+[2]
+[2]c
+[2]
1 [1]
2 [6]
G657
Restless, vomiting, diarrhea, decreased appetite
–
–
–
+[2]
2 [1]
2 [2]
G440
Restless, loose feces
+[1]
+[1;5]
+[5]d
–
2 [1]
3 [6]
Restless, loose feces
–
+[6]
–
NDe
2 [1]
2 [6]
Restless, loose feces
–
–
–
ND
1 [1]
2 [6]
B1024 B1417 a
+ Indicates detection of H. pylori; the number in brackets indicates the time in months after the first inoculation that H. pylori was detected; – indicates no detection of H. pylori b The gastritis score system of Lee et al. (15) was used: 1, minimal; 2, mild; 3, moderate; 4, severe; the number in brackets indicates the time in months after the first inoculation that gastric inflammation was observed. c Toluidine blue stain (14) d Silver stain (11) smooth gastric epithelial surfaces without bacteria (Fige Not done LM = light microscopy; TEM = transmission electron microscopy ure 1A). Histologic examination indicated that all monkeys
Three monkeys (G440, G657, B1024) were euthanized at 6, 11, and 12 months after inoculation, respectively. Exsanguination and euthanasia by barbiturate overdose were performed under deep surgical anesthesia with ketamine and xylazine. Tissue specimens, collected at necropsy, as well as gastric mucosa biopsy specimens were processed for microscopic evaluation. For light microscopy (LM), paraffin sections were stained either with hematoxylin and eosin or with silver, using the modified Steiner protocol (11) and Diff-Quik (12). Specimens were prepared for scanning electron microscopy as described (13), but were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde (prepared in 0.1 M cacodylic acid buffer), dried with Peldri II (Ted Pella, Inc., Redding, Calif.) or hexamethyldisilasane (HMDS; Polysciences, Inc., Warrington, Pa.), and coated with a 10- to 25-nm layer of gold, using a PS-2 Coating Unit (International Scientific Instruments/Polaron, Doylestown, Pa.). Transmission electron microscopy was performed as described (8). Semithin sections (0.5 mm) of durcupan-embedded specimens were stained with toluidine blue (14). The score system developed by Lee et al. (15) was used to determine severity of gastric inflammation in histologic sections: 1, minimal; 2, mild; 3, moderate; and 4, severe. Monkeys inoculated with H. pylori became restless, had decreased appetite and episodes of vomiting in the afternoon after inoculation, and developed loose feces or diarrhea (Table 1) that lasted up to 4 days after inoculation. Loose feces was not a good criterion for infection with H. pylori in squirrel monkeys because all monkeys had loose feces at some time during the course of the study. This clinical sign of disease may have been a stress-related phenomenon, as recently described in a larger group of 11 squirrel monkeys (8). Before inoculation, endoscopy of the stomach revealed no redness of gastric mucosa. Electron microscopy revealed
304
had minimal to mild (grade 1 to 2) gastritis (Table 1). Squirrel monkeys inoculated with H. pylori developed inflammation of variable severity. Mild generalized hyperemia of the gastric mucosa was observed endoscopically in monkey G440 2 months after the first inoculation, which exacerbated to moderately severe diffuse hyperemia of the fundus and antrum at 5 months. The gastric epithelial surface appeared eroded at 6 months, and spiral-shaped structures in clusters were observed by scanning electron microscopy in some areas of the gastric epithelium (Figure 1B). Histologic examination at necropsy indicated that monkey G440 developed lymphocytic gastritis (grade 3) with leukocytic infiltration of equal intensity in the body and antrum (Figure 2A). Silver-stained paraffin-embedded sections and Diff-Quik-stained cytologic brush specimens collected from monkey G440 at 6 months after the first inoculation contained bacteria with morphologic characteristics of Helicobacter spp. (Figure 2B). The increase in gastric inflammation in monkey G440 correlated with H. pylori infection. It was, however, more difficult to evaluate the establishment of infection in the remaining four monkeys: monkeys G4390, G657, and B1024 had detectable organisms and mild (grade 2) inflammation in the fundus and antrum of the stomach, and monkey B1417 had mild inflammation (grade 2) without detectable microorganisms. Gross lesions were not observed in monkeys G440, G657, and B1024 at necropsy. Monkey B1024 had, at necropsy, diffuse infiltration of lymphocytes, plasma cells, neutrophils, and eosinophils in the gastric mucosa and in the mucosal submucosal interface. In addition, a nodular-like aggregate of lymphocytes was evident in the mucosa of monkey B1024. Helicobacter spp. were not apparent in stained thin sections from monkeys G657, B1024, and B1417. Examination of tissue specimens collected from monkey G657 at 2 months after inoculation revealed some gastric epithelial cells with peripherally oriented condensed nuclear chromatin (Figure 3A). Other cells (sometimes of
Note
Figure 1. Scanning electron micrographs of gastric tissue specimens collected from squirrel monkey G440. Prior to inoculation, mucosa was smooth without epithelial surface damage (A). Six months after the first inoculation with H. pylori, there were rough surfaces with erosions (B). Filamentous or spiral-shaped structures, some of which resemble H. pylori, were detected on the epithelial surface and in intercellular spaces (B, arrowheads). Bars = 50 mm.
the same specimen) did not have these alterations and appeared uninfected. Tissue specimens collected from monkey G657 at necropsy revealed mild inflammation at the mucosal submucosal interface in the gastric body and antrum. The inflammation was characterized by infiltration of lymphocytes and plasma cells. A recent study (8) with 11 squirrel monkeys indicated that low-grade gastritis is common in wild-caught monkeys kept in captivity. It is desirable to further investigate whether this was due to the effect of past bacterial infection. At present, however, it seems that H. pylori does not naturally colonize in squirrel monkeys. Thus, squirrel monkeys serve as a good source of immune-naive animals for studying the immunogenicity and safety of H. pylori vaccine candidates in nonhuman primates (8), but the inconsistent experimentally induced infection of these monkeys with H. pylori precludes their use in vaccine protection studies.
Bacteria resembling H. pylori were recovered from monkey G440 1 month after the first inoculation and from monkey G4390 2 months after the first inoculation. The isolate from monkey G4390 could be subcultured, and pure cultures were obtained for analysis by electrophoresis and immunoblotting. For one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (1-D-SDS-PAGE), bacterial samples (1 mg/ml in SDS) were heated at 1008C for 2 min (whole cell lysate under reduced conditions), and proteins were separated on 10% resolving gels and stained with Coomassie brilliant blue G-250 stain (Serva Feinbiochemica, Heidelberg, Germany). For Western blotting, SDS-PAGE-separated proteins (unstained) were transferred to nitrocellulose sheets and incubated with an antirecombinant H. pylori urease mouse serum diluted 1:500 in blocking buffer. To obtain this serum, five BALB/c mice (Charles River,
305
Vol 48, No 3 Laboratory Animal Science June 1998
Figure 2. Photomicrographs of sections of gastric mucosa and mucus from monkey G440 at 6 months after the first inoculation with H. pylori. Sections stained with hematoxylin and eosin revealed lymphocytes in the lamina propria (A). Sections stained with silver, using the modified Steiner protocol, revealed spiral-shaped organisms resembling H. pylori in the mucus layer of the gastric mucosa (B, arrowheads). Bars = 10 mm.
Wilmington, Md.) were intraperitoneally injected with 0.15 ml of antigen mixture. At priming, this mixture contained 20 mg of recombinant H. pylori urease (OraVax, Inc., Cambridge, Mass.) emulsified in 0.05 ml of phosphate-buffered saline (PBS) and 0.1 ml of Freund’s complete adjuvant (Sigma Chemical Co, St. Louis, Mo.). A second and third injection were administered after 21 and 35 days: 10 mg of antigen in 0.05 ml of PBS and 0.1 ml of Freund’s incomplete adjuvant (Sigma Chemical). Four booster injections were given on days 49, 50, 51, and 52: 20 mg of antigen in 0.15 ml of PBS. Mice were exsanguinated by cardiac puncture under deep surgical anesthesia (4% isoflurane in 2 L of oxygen). Serum was obtained by centrifugation of blood at 8,160 X g and was stored at -208C until used in immunoblot analysis. A biotin-avidin peroxidase complex was used in the Western blot detection system to visualize antigens. Electrophoretic analysis indicated that the protein profile of the isolate from monkey G4390 was not different from that of the challenge strain, and the two urease subunits were identical in electrophoretic mobility to those expressed by the challenge strain 12476 (Figure 4). Positive urease reactions were observed on several occasions when suspensions of gastric mucosal biopsy specimens were inoculated into test broth (Table 1). However, we were unable to propagate bacterial isolates for further analysis. Microorganisms were observed between organelles of gastric epithelial cells on two occasions in monkey G657.
306
One bacterium appeared to be attached to a mitochondrion (Figure 3B), and the cytoplasm surrounding the bacterium was loose. The affected host cell was identified as a chief cell due to the extensively developed rough endoplasmic reticulum. It is uncertain whether this location can be considered as intracellular, or the cytoplasmic membrane of the host cell was damaged and organisms entered the cell. However, recent reports (16, 17) indicate that H. pylori may penetrate cells, and these features may contribute to the pathogenic action of the organism. Diagnostic procedures to detect infection with H. pylori should take this into consideration. During the course of this study, we successfully adapted the [13C]urea breath test to these squirrel monkeys for the detection of urease-producing bacteria (7). The breath test was sufficiently sensitive to determine conversion of the experimentally infected squirrel monkeys B1024 and B1417 from breath test negative to positive within 3 to 5 days after inoculation. Additionally, we documented that re-inoculation of monkey G440 with H. pylori, when superimposed on an existing infection (which remained breath test positive after 4 months), resulted in a three-fold increase in atom percent excess [ 13C], which gradually decreased over the ensuing 2 weeks. Thus, results of the breath test confirmed (7) that monkeys G440, B1024, and B1417 were infected with H. pylori. In conclusion, results of this study indicated that squirrel monkeys can be infected with H. pylori, but infection is
Note
Figure 3. Transmission electron micrographs of biopsy specimens from the gastric mucosa of squirrel monkey G657 collected 2 months after the first inoculation with H. pylori. Some cells have peripheral orientation of the nuclear chromatin (A). In some sections, organisms were detected between cell organelles (B, arrowheads). mi = mitochondria; N = nucleus; P = parietal cell; tvs = tubulovesicular system; and rER = rough endoplasmic reticulum. Bars = (A), 2.5 mm; (B), 0.5 mm.
307
Vol 48, No 3 Laboratory Animal Science June 1998
Figure 4. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblot analysis of the challenge strain (H. pylori) and isolate recovered from monkey G4390 at 2 months after the first inoculation. Helicobacter mustelae was used as a control. Proteins were reacted on Western blots with an anti-recombinant (H. pylori) urease mouse serum. Relative molecular masses are indicated to the left in kilodaltons (kDa). Protein and antigen patterns of the H. pylori challenge strain and isolate are nearly identical (showing major bands at approximately 66 and 30 kDa), but different from H. mustelae.
transient, and only one of five monkeys developed moderate lymphocytic gastritis. Several diagnostic tools were necessary to confirm infection with H. pylori, including gastroscopy with biopsy, microbiologic examination, urease testing, histologic examination, electron microscopy, and [13C]urea breath test (7). In addition, preexisting minimal to mild gastritis complicated evaluation of gastritis induced by infection with H. pylori. On the basis of these observations, the squirrel monkey is not as useful for the study of experimentally induced gastritis as was expected, but is useful for evaluation of immune responses against H. pylorivaccine candidates as well as the conduct of vaccine safety trials, as recently documented (8). It is likely that this animal model will have further applications in H. pylori research because of the increased interest in development of newer, more effective drugs and vaccines for treatment and prevention of H. pylori infection.
Acknowledgements The animal care and use described in this report were approved by the Institutional Animal Care and Use Committee at Clemson University and conducted in compliance with the Animal Welfare Act and the Public Health Service Policy. We thank Gary E. Zurenko (Upjohn Company, Kalamazoo, Mich.) for providing us with a culture of H. pylori strain 12476. We also thank Debbie L. Garrett (Gastroenterology Associates, P.A., Greenville, S.C.) for excellent technical assistance in the animal work. We express our appreciation to the St. Francis Hospital for donation of the endoscopic equipment and to the Electron Microscopy Facility at Clemson
308
University for the opportunity to use the electron microscopic equipment. This investigation was funded by OraVax, Inc. (Cambridge, Mass.).
References 1. Cover, T. L., and M. J. Blaser. 1995. Helicobacter pylori: a bacterial cause of gastritis, peptic ulcer disease, and gastric cancer. ASM News 61:21–26. 2. Baskerville, A., and D. G. Newell. 1988. Naturally occurring chronic gastritis and C. pylori infection in the rhesus monkey: a potential model for gastritis in man. Gut 29:465–472. 3. Shuto, R., T. Fujioka, T. Kubota, et al. 1993. Experimental gastritis induced by Helicobacter pylori in Japanese monkeys. Infect. Immun. 61:933–939. 4. Bronsdon, M. A., and F. D. Schoenknecht. 1988. Campylobacter pylori isolated from the stomach of the monkey Macaca nemestrina. J. Clin. Microbiol. 26:1725–1728. 5. Euler, A. R., G. E. Zurenko, J. B. Moe, et al. 1990. Evaluation of two monkey species (Macaca mulatta and Macaca fascicularis) as possible models for human Helicobacter pylori disease. J. Clin. Microbiol. 28:2285–2290. 6. Hazell, S. L., J. W. Eichberg, D. R. Lee, et al. 1992. Selection of the chimpanzee over the baboon as a model for Helicobacter pylori infection. Gastroenterology 103:848–854. 7. Stadtländer, C. T. K.-H., and F. J. Stutzenberger. 1995. Adaptation of the [13C]urea breath test as a non-invasive method for detection of Helicobacter pylori infection in squirrel monkeys (Saimiri spp.). Lab. Anim. Sci. 45:239–243. 8. Stadtländer, C. T. K.-H., J. D. Gangemi, S. S. Khanolkar, et al. 1996. Immunogenicity and safety of recombinant Helicobacter pylori urease in a nonhuman primate. Dig. Dis. Sci. 41:1853–1862.
Note
9. Power, D. A., and P. J. McCuen. 1988. Manual of BBL® products and laboratory procedures, p. 281. 10. Stadtländer, C. T. K.-H., C. Zuhua, H. L. Watson, et al. 1991. Protein and antigen heterogeneity among strains of Mycoplasma fermentans. Infect. Immun. 59:3319–3322. 11. Garvey, W., A. Fathi, and F. Bigelow. 1985. Modified Steiner for the demonstration of spirochetes. J. Histotechnol. 8:15–17. 12. Skipper, R., and D. B. DeStephano. 1989. A rapid stain for Campylobacter pylori in gastrointestinal tissue sections using Diff-Quik®. J. Histotechnol. 12:303–304. 13. Stadtländer, C. T. K.-H., H. L. Watson, J. W. Simecka, et al. 1993. Cytopathogenicity of Mycoplasma fermentans (including strain incognitus). Clin. Infect. Dis. 17(Suppl. 1):S289–S301.
14. Kramer, H., and G. M. Windrum. 1955. The metachromatic staining reaction. J. Histochem. Cytochem. 3:227–237. 15. Lee, A., J. G. Fox, G. Otto, et al. 1990. A small animal model of human Helicobacter pylori active chronic gastritis. Gastroenterology 99:1315–1323. 16. Noah, L. A., T. M. Rolf, and G. N. Tytgat. 1994. Electron microscopic study of association between Helicobacter pylori and gastric and duodenal mucosa. J. Clin. Pathol. 47:699–704. 17. El-Shoura, S. M. 1995. Helicobacter pylori: I. Ultrastructural sequences of adherence, attachment, and penetration into the gastric mucosa. Ultrastruct. Pathol. 19:323–333.
309