INFECTION

AND

Vol. 20, No. 2

IMMUNITY, May 1978, p. 347-351

0019-9567/78/0020-0347$02.00/0

Printed in U.S.A.

Copyright © 1978 American Society for Microbiology

Cross-Protection of Mice Provided by Active and Passive Immunization Against Experimental Infections with Virulent Proteus rettgeri and Providencia Bacteria JOHN L. PENNER'* AND GORDON R. WHITELEY2 Department of Medical Microbiology' and Department of Microbiology and Parasitology,2 Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada M5G 1L5 Received for publication 14 October 1977

Immunization with Providencia and Proteus rettgeri Formalin-treated bacterial suspensions produced high levels of protection in mice against homologous and heterologous challenge. Mice were also cross-protected, but less effectively, by passive administration of rabbit type-specific antisera. The protective activity appeared to be due to an antigen common to strains of different 0-serotypes. It was not detectable in agglutination reactions, and preliminary results indicate that it is thermostable, not being inactivated in its antibody binding capacity at 121°C for 1 h.

Antibiotic-resistant Proteus rettgeri and Providencia stuartii have been shown in recent years to be the causes of serious nosocomial disease (7, 11, 15, 23-25). Numerous serologically different strains occur among these species and, in our laboratory, the number of 0-specificities had been noted to increase as isolates from different sources were examined. The use of immunotherapy for protection against these bacteria has not been investigated as has been the case for Pseudomonas aeruginosa (1-3, 5, 9, 10, 12). Previous studies on P. rettgeri and P. stuartii have shown considerable differences in virulence among strains (7, 23, 24), but relationships between virulence and the 0 antigens were not demonstrated. Since schemes for discriminating among strains of these proteeae have been described (20, 21), further studies on their virulence and the mechanisms of immunity to these bacteria were clearly indicated. The present communication is concerned with initial experiments designed to examine the production of immunity in mice through active and passive protection tests by using virulent, serologically unrelated strains of P. alcalifaciens, P. stuartii, and P. rettgeri. MATERIALS AND METHODS Selection of mouse-virulent bacterial strains. Eighty-four type strains of the P. rettgeri and 63 of the Providencia 0-serotyping schemes (20, 21) were screened for virulence by injecting two mice per strain intraperitoneally (i.p.) with 104 cells of each strain suspended in Trypticase soy broth (TSB) with 5% added hog gastric mucin. Death was caused in both mice within 48 h by only 18 of the 147 type strains. Six were selected from the 18 because they were unrelated

types previously shown to have no cross-reactions attributable to thermostable somatic (0) antigens. Characterization of strains. Characteristics of selected strains are shown in Table 1. Biotypes were classified according to reactions and schemes of classification previously reported (19, 21). Autoclaved cell suspensions of each strain agglutinated in homologous but not in heterologous typing antisera when tested by slide agglutination. Homologous reactions, but no cross-reactions, occurred at titers of 1:40 or greater when typing antisera and six mouse antisera prepared against the six strains were cross-titrated by the passive hemagglutination (PHA) technique in which supernatant from autoclaved cell suspensions was used to sensitize sheep erythrocytes. In cross-titrations of rabbit and mouse antisera by the microtiter method of agglutination of Formalin-treated bacterial suspensions, as described previously (18), homologous reactions were observed in all cases and relations were noted between two strains, P. stuartii 024 and 055. In PHA titrations of rabbit antisera after absorptions with homologous cell suspensions heated at 121°C for 1 h, neither homologous nor heterologous reactions were observed at titers of 1:40 or greater, but homologous reactions occurred for each antiserum and crossreactions occurred between P. stuartii 024 and 055 when titration was by agglutination of bacterial suspensions (ABS). Treatment of cells of P. stuartii 024 and 055 at 60°C for 1 h, but not treatment with 80% ethanol, abolished cross-reactions in subsequent ABS tests and, therefore, the relationship between these strains was attributed to a shared K antigen (13). Enterobacterial common antigen of Kunin (14). The six strains possessed enterobacterial common antigen in the non-immunogenic state. Each was agglutinated (by PHA) in Escherichia coli 014 antiserum (Table 1). Neither rabbit nor mouse antisera against the strains agglutinated (by PHA) antigen of

E. coli 014 at titers of 1:40 or greater. Determination of LDro. After two passages in 347

348

PENNER AND WHITELEY

INFECT. IMMUN.

TABLE 1. Characteristics of six bacterial strains virulent for mice Titers of homologous and heterologous reactions"

Species

P. alcalifaciens P. stuartii

01ati

geanti-

Bio-

b

PHA technique'

ABS technique'

ters of

antise-

E. ~~~~~abitrum Rabbit Mouse (ab)bt Rabbit Mouse Rabbit (ab) coli 014

gen type

018 024

PHA ti-

10,240 1,280

2,560 640

LDro

1,280 160 640 2,560 4.0 x 104 1,280 10,240 10,240 5,120 2.5 x 104 (055; 160) (055; 2,560) (055; 80) P. stuartii 5 320 640 320 1,280 1.9 X 104 055 2,560 320 (024; 20) (024; 20) 1,280 160 640 5,120 1.5 x 104 P. rettgeri 05 1 2,560 5,120 P. rettgeri 041 3 80 40 80 2,560 4.4 x 104 640 320 2 P. rettgeri 052 80 80 80 5,120 2.0 x i0' 5,120 10,240 0 antigens classified according to schemes for Providencia and P. rettgeri as previously reported (20, 21). Biotypes of Providencia and P. rettgeri as previously described (19, 21). 'Titers are expressed as reciprocals of highest dilution of antiserum showing agglutination. Heterologous titers are shown with corresponding reactants in parentheses. d No heterologous reactions were observed at the initial antiserum dilution of 1:40. 'Initial dilution of antisera was 1:10; negative reactions of this dilution are not shown. fRabbit antisera were absorbed (ab) with homologous cell suspensions heated for 1 h at 121°C. -, Negative reactions of antiserum dilution of 1:40. 2 5

-

mice, the 50% lethal dose (LD50) (Table 1) was determined for each strain by the method of Reed and Muench (22). Active protection tests. Swiss White mice (Connaught Laboratories, Willowdale, Ontario, Canada) were injected i.p. with Formalin-treated cell suspensions prepared as described previously (18) in five doses at 5-day intervals increasing from 2.5 x 105 to 7.5 x 106 cells over the 3-week immunization period. Seven days later, mice were challenged i.p. with 100 LD5o of organisms suspended in TSB with 5% added hog gastric mucin. The mice were observed for 1 week. Deaths among control mice or unsuccessfully protected mice usually occurred within 48 h. Passive protection tests. The 50% effective dose (ED50) values of antisera were based on survival 1 week postchallenge of 50% of mice challenged with 100 LD50 of homologous bacteria by the method of Reed and Muench (22). Mice were immunized passively by i.p. injection of 0.5 ml of rabbit antiserum previously diluted in sterile saline to provide a protective dose equivalent to 10 EDs0. Four hours later a 100 LD5o challenge dose was administered i.p. Normal rabbit serum was administered i.p. to control mice. Protection was recorded if mice were healthy 1 week later. Deaths usually occurred within 48 h. Antisera. Typing antisera were produced in rabbits against Formalin-treated bacterial suspensions by procedures reported previously (18, 21). Mouse antisera were prepared in Swiss White mice through i.p. injection of Formalin-treated cells in five doses, increasing from 2.5 x 105 to 7.5 x 106 cells and harvested by displac"ement of the eye from the retro-orbital plexus. Antisera from five mice were pooled. Challenge doses. From an overnight growth (at 37°C) of bacteria in 5 ml of TSB, 0.01 ml was transferred to 10 ml of TSB and incubated at 37°C for 4 h. Growth from this culture was diluted in TSB with 5% hog gastric mucin to produce cell concentrations as determined in previous experiments to be equivalent to 100 LD50. Viable cell counts of the suspensions were confirmed by dilution plate counts by the method of Miles and Misra (17).

RESULTS Protection of mice against homologous and heterologous lethal challenges of P. alcalifaciens, P. stuartii, and P. rettgeri bacteria by active immunization. One P. alcalifaciens strain (018), one P. stuartii strain (055), and two P. rettgeri strains (05 and 052) were selected for use in experiments to determine the degree of protection provided in mice through active immunization. Each mouse was immunized and challenged by the procedures described above. It was found that all mice challenged with homologous cell suspensions survived and, of the mice challenged with heterologous suspensions, all but three survived. All mice that received saline without the immunogen died (Table 2). Similar results were obtained in a separate experiment with P. stuartii 024, P. stuartii 055, and P. rettgeri 041 strains. All mice survived except one that was immunized with P. stuartii 055 and challenged with P. rettgeri 041. The high degree of cross-protection observed in these experiments was unexpected because the strains had been selected on the basis of their biochemical and serological unrelatedness. However, it was conceivable that the strains were of one immunotype since they were initially selected on the basis of their virulence in mice. It was the cross-protective activity among these otherwise unrelated strains that prompted further investigation by passive protection experiments. Passive protection of mice with immune rabbit antisera against homologous and heterologous challenges of P. alcalifaciens, P. stuartii, and P. rettgeriL The results of i.p. administration to mice of 10 ED5o of immune

IMMUNIZATION WITH P. RETTGERI AND PROVIDENCIA

VOL. 20, 1978

TABLE 2. Protection of mice by active immunization against homologous and heterologous lethal challenges of P. alcalifaciens, P. stuartii, and P. rettgeri

TABLE 3. Passive protection of mice against homologous and heterologous lethal challenges of P. alcalifaciens, P. stuartii, and P. rettgeri with immune rabbit antisera Survivors/no. of mice challenged

Survivors/no. of mice challenged Immunogen'

P. alcalifaciens

P. alcalifaciens 018

p. stuarti

055b v

05

5/5

5/5

4/5 4/5

P. rettgeri

Rabbit antiseraa

052b

018

P. stuartii 055 5/5 5/5 5/5 5/5 P. rettgeri 05 5/5 5/5 5/5 4/5 P. rettgeri 052 5/5 5/5 5/5 5/5 Saline (control) 0/5 0/5 0/5 0/5 a were Formalin-treated bacterial susImmunogens pensions. b Challenge doses of 100 LD50 were administered i.p. 5 days after last immunizing injection.

rabbit antisera 4 h before challenge with the homologous or heterologous strains used in the preceding experiment are shown in Table 3. Mice were effectively protected against homologous challenge but were considerably less wellprotected against heterologous challenge. No clear pattern of reciprocal cross-protection was evident, and patterns of reciprocal absence of protection were not noted except in the cases of P. rettgeri 052 and P. stuartii 055. However, all heterologous antisera failed to show or showed only minor protective activity against challenge with P. stuartii 055 bacteria. Since 24 of the 60 mice challenged with heterologous bacteria did survive, it was evident that a degree of crossprotective activity was present in each of the rabbit antisera. This passive cross-protection was studied further in three experiments similar in design to the one above but with antisera that had been absorbed with homologous bacterial cell suspensions. When antisera were absorbed with suspensions heated at 600C for 1 h, passive protection experiments showed neither homologous nor heterologous protection. Of the 80 mice that received antiserum, only 1 survived. In the next experiment the antisera were absorbed with homologous cell suspensions that had been subjected to flowing steam (95 to 9800) for 1 h. Again, only 1 of the 80 mice survived in the passive protection experiments. When antisera were absorbed with cell suspensions that had been autoclaved (1210C) for 1 h, the passive protection experiments showed, as described above, that the absorptions had virtually abolished cross-protective activity (Table 4). Furthermore, homologous protective activity was also abolished as in two preceding experiments except that, in this experiment, the autoclaved cells failed to remove homologous

349

p alcalifa- P. stuarti ciens 018 055.

P. rettgeri 05b 052b

P. alealifaciens 5/5 1/5 3/5 1/5 018 P. stuartii 055 4/5 5/5 5/5 0/5 P. rettgeri 05 0/5 0/5 5/5 3/5 P. rettgeri 052 4/5 0/5 3/5 5/5 NRSC (control) 0/5 0/5 0/5 0/5 a Rabbit immune antisera were prepared against the strains listed. Doses were equivalent to 10 EDso. b Challenge doses of 100 LD5o were administered i.p. 4 h after passive immunization. c NRS, Normal rabbit serum. TABLE 4. Effect of absorptions on protective activity of antisera in passive protection experiments Survivors/no. of mice challenged Rabbit antisera

P. alcalifaciens

P. alcalifa- P. stuarti ciens 018b 055b

1/5

0/5

P.

rettgeri

05b

052b

0/5

0/5

018

P. stuartii 055 0/5 5/5 0/5 0/5 P. rettgeri 05 0/5 0/5 0/5 0/5 P. rettgeri 052 0/5 0/5 0/5 0/5 NRSC (control) 0/5 0/5 0/5 0/5 a Rabbit immune antisera were prepared against the strains listed and absorbed with homologous bacterial suspensions heated at 121°C for 1 h. Doses were equivalent to 10 ED5o. b Challenge doses of 100 LDs were administered i.p. 4 h after passive administration of antiserum. NRS, Normal rabbit serum.

protective activity of the 055 antiserum. Evidently, cross-protective activity of the unabsorbed immune rabbit antisera was mediated by antibody removable by absorptions with homologous suspensions of heat-treated bacteria. This indicated that the antibody binding property of the bacterial factor associated with protective activity was thermostable to at least 1210C for 1 h. In addition, a factor borne by the P. stuartii 055 strain and apparently associated with homologous protective activity appeared to be stable in its antibody binding capacity at 95 to 980C, but labile at 121°C. DISCUSSION The mechanisms of immunity to P. rettgeri and Providencia bacteria are not well known. Schemes for discriminating among strains of

350

PENNER AND WHITELEY

INFECTr. IMMUN. I

these groups on the basis of 0 antigens have 4. Barber, C., and E. Eylan. 1976. Cross-protection induced I in mice by immunizations with proteins of related sperecently been described (20, 21), and the design cies. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. of the experiments to investigate type-specific Hyg. Abt. Orig. Reihe A 234:46-52. and cross-protective immunity was based on 5. Bass, J. A., and J. C. McCoy. 1971. Passive immunization against experimental Pseudomonas infection: corthese schemes. Type-specific protection was obrelation of protection to Verder and Evans "O" seroserved but, contrary to expectations, high levels types. Infect. Immun. 3:51-58. i.lI of cross-protective activity were also observed 6. Braude, A. I., E. J. Ziegler, H. Douglas, and J. A. among a set of unrelated strains in both active McCutchan. 1977. Protective properties of antisera to R core, p. 253-256. In D. Schlessinger (ed.), Microbioland passive protection experiments. Protection ogy-1977. American Society for Microbiology, Washprovided through active immunization, presumington, D.C. ably due to both humoral and cell-mediated 7. Curreri, P. W., H. M. Bruck, R. B. Lindberg, A. D. v,I immunity, was greater than that provided Mason, and B. A. Pruitt. 1973. Providencia stuartii sepsis: a new challenge in the treatment of thermal through passive administration of type-specific injury. Ann. Surg. 117:133-138. antiserum, although higher doses would likely 8. Edwards, L. D., A. Cross, S. Levin, and W. Landau. have given higher levels of protection. 1974. Outbreak of a nosocomial infection with a strain Since the bacterial strains were unrelated in of Proteus rettgeri resistant to many antimicrobials. Am. J. Clin. Pathol. 61:41-46. antigenic components detectable by basic imM. W., H. B. Devlin, and F. J. Gnabasik. 1969. I,lj munological techniques, the familiar gram-neg- 9. Fisher, New immunotype schema for Pseudomonas aerugiative 0 and K antigens could be excluded as nosa based on protective antigens. J. Bacteriol. mediators of the observed cross-protection. Ev98:835-836. idence was obtained to show that the strains 10. Harvath, L., and B. R. Andersen. 1976. Evaluation of type-specific and non-type-specific pseudomonas vacpossessed enterobacterial common antigen, but cine for treatment of pseudomonas sepsis during grananti-enterobacterial common antigen antibodies ulocytopenia. Infect. Immun. 13:1139-1143. in not involved because were cross-protection 11. Janis, B., R. C. Evans, and P. D. Hoeprich. 1968. Providence bacillus bacteremia and septicopyemia. Am. antibody of this specificity was not detected in J. Med. 45:943-947. either the rabbit antisera used in passive protecR. J., E. A. Roe, E. J. L. Lowbury, J. J. Miler, tion or in mouse antisera raised against the 12. Jones, and J. F. Spilsbury. 1976. A new pseudomonas vacrole in A by strains. significant cross-protection cine: preliminary trial on human volunteers. J. Hyg. 76:429-439. nonspecific antibody arising from mitogenic acof L antigens tivity of endotoxin on B-cells was discounted 13. Kauffmann, F. 1974. On the significance for the serology, immunology and pathogenicity of because of the high levels of observed protection Escherichia species. Zentralbl. Bakteriol. Parasitenkd. 11 and because cross-protective activity was readily Infektionskr. Hyg. Abt. Orig. Reihe A 229:178-189. removable by specific absorption. Evidently, the 14. Kunin, C. M., M. V. Beard, and N. E. Halmagyi. 1962. Evidence for a common hapten associated with endostrains were related through a protective comtoxin fractions of E. coli and other enterobacteriaceae. mon antigen thermostable to 1210C and not Proc. Soc. Exp. Biol. Med. 111:160-166. detectable in agglutination reactions. Previous 15. Lindsey, J. O., W. T. Martin, A. C. Sonnenwirth, and reports have associated the core polysaccharide J. V. Bennett. 1976. An outbreak of nosocomial Proteus rettgeri urinary tract infection. Am. J. Epidemiol. of Enterobacteriaceae (6, 16) and thermostable 103:261-269. protein (1, 4) with cross-protective activity. 16. McCabe, W. R. 1972. Immunization with R mutants of S. It is expected that further use of the particular minnesota. I. Protection against challenge with heterstrains described in this study will serve toward ologous Gram-negative bacilli. J. Immunol. 108: defining the antigens involved in protective ac- 17. 601-610. Miles, A. A., and S. S. Misra. 1938. The estimation of tivity among P. rettgeri and Providencia. the bactericidal power of the blood. J. Hyg. 38:732-749. il

ACKNOWLEDGMENT This research was supported by Medical Research Council grant (Canada) no. MA-5648. LITERATURE CITED 1. Abe, C., H. Shinoya, Y. Hirao, K. Okada, and J. Y. Homma. 1975. Common protective antigens (OEP) of Pseudomonas aeruginosa. Jpn. J. Exp. Med. 45: 355-359. 2. Alexander, W. J., and M. W. Fisher. 1974. Immunization against Pseudomonas in infection after thermal injury. J. Infect. Dis. 130:S152-S158. 3. Alms, T. H., and J. A. Bass. 1967. Immunization against Pseudomonas aeruginosa. I. Induction of protection by an alcohol-precipitated fraction from the slime layer. J. Infect. Dis. 117:249-256.

18. Penner, J. L., N. A. Hinton, and J. Hennessy. 1974. Serotyping of Proteus rettgeri on the basis of 0 antigens. Can. J. Microbiol. 20:777-789. 19. Penner, J. L., N. A. Hinton, and J. Hennessy. 1975. Biotypes of Proteus rettgeri. J. Clin. Microbiol. 1:136-142. 20. Penner, J. L., N. A. Hinton, and J. N. Hennessy. 1976. Evaluation of a Proteus rettgeri 0-serotyping system for epidemiological investigation. J. Clin. Microbiol. 3:385-389. 21. Penner, J. L., N. A. Hinton, J. N. Hennessy, and G. R. Whiteley. 1976. Reconstitution of the somatic (O-) antigenic scheme for Providencia and preparation of 0typing antisera. J. Infect. Dis. 133:283-292. 22. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty per cent endpoints. Am. J. Hyg. 27:493-497. 23. Solberg, C. D., and J. M. Matsen. 1971. Infections with

VOL. 20, 1978

IMMUNIZATION WITH P. RETTGERI AND PROVIDENCIA

Providence bacilli. Am. J. Med. 50:241-246. 24. Traub, W. H., M. E. Craddock, E. A. Raymond, M. Fox, and C. E. McCaU. 1971. Characterization of an unusual strain of Proteus rettgeri associated with an outbreak of nosocomial urinary-tract infection. Appl.

351

Microbiol. 22:278-283. 25. Wenzel, R. P., K. J. Hunting, C. A. Osterman, and M. A. Sande. 1976. Providencia stuartii, a hospital pathogen: potential factors for its emergence and transmission. Am. J. Epidemiol. 104:170-180.

Cross-protection of mice provided by active and passive immunization against experimental infections with virulent Proteus rettgeri and Providencia bacteria.

INFECTION AND Vol. 20, No. 2 IMMUNITY, May 1978, p. 347-351 0019-9567/78/0020-0347$02.00/0 Printed in U.S.A. Copyright © 1978 American Society f...
705KB Sizes 0 Downloads 0 Views