Vol. 20, No. 1
INFECTION AND IMMUNITY, Apr. 1978, p. 6-11 0019-9567/78/0020-0006$02.00/0 Copyright © 1978 American Society for Microbiology
Printed in U.S.A.
Influence of the Antigenicity of Brucella Preparations on Modulation of the Immune Response to Sheep Erythrocytes Y. LE GARREC,l* L. TOUJAS,2 A. MARTIN,2 L. DAZORD,2 AND C. PILETI Service de Microbiologie, d'Immunologie et de Pathologie Generale, Ecole Nationale Vetrinaire, 94701 Maisons-Alfort, France,' and Centre Eugene Marquis and Unite INSERM 49, Hopital Pontchaillou, 35000
Rennes, France'
Received for publication 26 October 1977
The relationship between the expression of surface antigenicity of inactivated Brucella and the immunostimulant properties on the sheep erythrocyte response was studied in mice. The agglutinogenic Brucella abortus B19S preparation was compared to two non-agglutinogenic preparations (B19R and PB), using the plaque-forming and rosette-forming cell tests. When Brucella and antigen were injected together, only non-agglutinogenic preparations, even used at low doses, were able to increase the plaque-forming and rosette-forming cell responses measured 4 days after immunization. When the interval between the previous injection of non-agglutinogenic preparations and that of the antigen increased or when they were injected 24 h after the antigen, the modification of those two responses was no longer observed. After day 4, for the simultaneous injection of Brucella and antigen, B19S and PB preparations induced the best responses; this effect was dose dependent. Thus, in the present experimental situation, the modifications of the day 4 response appeared independent of those of the ensuing days.
Antigens. SRBC were obtained once a week from one sheep, collected in Alsever solution, and washed three times in saline before use. Bacterial preparations. The following preparations were used: (i) the fully antigenic vaccinal strain
The ability of inactivated Brucella abortus vaccine to increase non-specifically the titer of antibodies to Salmonella in the rabbit has been stressed by Ramon et al. (18). More recently, their role in affecting the development of bacterial or viral infections and of various tumors has been investigated (1, 10, 16, 17). However, earlier study of the modifications induced by inactivated B. abortus (strain B19S) on the response to sheep erythrocytes (SRBC) showed that the number of plaque-forming cells (PFC) per spleen on day 4 after immunization was, if anything, depressed by the simultaneous or previous injection of Brucella. Stimulation did occur, however, if the time between injection of Brucella and injection of SRBC was increased to 40 days (21). The finding that an injection of anti-Brucella serum given before injecting the B19S caused a partial reversal of the effect (20) led to the hypothesis that some antagonism might exist between the antigenicity of B. abortus and its immuno-potentiating activity. To test this hypothesis, we chose to compare the modulating effects of agglutinogenic and non-agglutinogenic strains of inactivated Brucella on the immune response to SRBC. MATERIALS AND METHODS
B19S of B. abortus, which provokes the formation of agglutinins; (ii) the B19R rough mutant, selected from B19S in the laboratory; and (iii) the PB preparation, corresponding to the Pilet-Bonneau vaccine, which is a non-agglutinogenic immune complex (15). Brucella organisms were grown on Albimi agar, harvested after 48 h, washed twice in phosphatebuffered saline, and then inactivated by heat for 1 h at 650C. The PB preparation was prepared by neutralizing agglutinogenic sites of inactivated B19S with a specific
immune serum prepared in cows. B19R and PB preparations do not induce agglutinins in mice (or rabbits). The bacterial preparations were injected intravenously at various intervals before, at the same time as, or 1 day after immunization. A dose of 500 gg (dry weight) was used except when stated otherwise. PFC. The number of plaque-forming cells (PFC) in the spleen was determined after intravenous injection of 2.5 x 108 SRBC. The Jerne and Nordin method (6) for direct PFC and the Dresser and Wortis method (4) for indirect PFC were used. For the latter, an antimouse immunoglobulin G serum produced in goats (Behring) was diluted 1/200. Animals. Female mice F2(C57Bl6xDBA2) (6 to 7 RFC. The number of rosette-forming cells (RFC) weeks old) from the Centre de Selection du Centre in the spleen was determined after the intravenous National de la Recherche Scientifique (Orleans) were injection of 2.5 x 108 SRBC. The McConnel et al. used for this study. technique (13) was used, in which cell mixtures are 6
VOL. 20, 1978
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
merely centrifuged in the cold without standing at
40C. The number of rosette-forming macrophages was determined by injecting colloidal carbon 0.5 h before excising spleens.
The resulting numbers of PFC and RFC were expressed per 106 spleen cells. The mean per group of six (PFC) or five (RFC) mice is given with the standard error (SE). The significance was calculated by Student's t test and accepted at P < 0.05. In both cases, an adjuvant index was used to express the ratios of the responses in treated and untreated mice: AI = PFC or RFC in treated mice/PFC or RFC in control mice. The PFC and RFC response to immunization with SRBC was tested in different experimental situations, injecting the Brucella preparations (i) before, (ii) at the same time as, and (iii) after the antigen.
RESULTS
Response to SRBC after a preliminary injection of Brucella preparations. The number of anti-SRBC PFC 4 days after antigen was measured in mice which received 500 ,tg of B19S, B19R, or PB preparations at intervals of 0 to 40 days before immunization (Table 1). With respect to the B19S preparation, a 50 to 70% inhibition was observed when its injection was separated from the SRBC by an interval of 5 to 30 days. No significant effect was seen when
7
the two were given simultaneously. The nonagglutinogenic preparations B19R and PB did not appear to affect the anti-SRBC response except on day 0. On the other hand, both the B19R and PB preparations appeared to enhance the RFC response when given either with or 5 days before the antigen (Table 2). The 19S vaccine did not affect the RFC response unless the data were plotted as RFC per spleen; a two- to fourfold increase then was observed on days 10 to 40. The expression of the results in number of PFC or RFC per 106 spleen cells is indeed affected by the number of spleen cells. As shown in Fig. 1, B19S induced the most prolonged increase in spleen cell number. Response to SRBC after the simultaneous injection of inactivated Brucella and SRBC. The mice received 500 ,tg of the Brucella preparations mixed with 2.4 x 108 SRBC intravenously. The maximum value for the direct PFC count was reached on day 4 in the control group (Table 3). This value was not significantly different for the group injected with B19S, but was surpassed in animals given B19R or PB. Subsequently, the number of direct PFC decreased quickly in controls, but remained elevated in all treated groups. The B19R and PB groups were both signifi-
TABLE 1. Modifications of the PFC response by injection of 500 Pg of different killed B. abortus preparations before SRBC immunization Interval (days) between B. ab~ortus and SRBC injections
No. of PFCa/106 spleen cells ± SE and adjuvant index after treatment with: PB B19R B19S
1,747 ± 120b (2.5) 456 ± 65 (0.7) 825 ± 99 (1.2) 769 ± 113 (1.1) 891 ± 105 (1.3) 769 ± 22 (1.1) 675 ± 56 Controls (no B. abortus) a Number of PFC measured 4 days after SRBC immunization. b Significantly different from the control groups. e ND, Not done. 224 ± 49 (0.7) 110 ± 26b (0.3) 163 ± 5b (0.5) 175 ± 9ob (0.5) 123 ± 63b (0.4) 246 ± 58 (0.8) 319 ± 95
0 5 10 20 30 40
1,851 ± 173b (2.6) 894 ± 134 (1.2) 856 ± 107 (1.2) 487 ± 38 (0.7) ND' 672 ± 66 (0.9) 719 ± 56
TABLE 2. Modifications of the RFC response by injection of 500 jg of killed B. abortus preparations before SRBC immunization No. of RFCa/10f spleen cells ± SE and adjuvant index after treatment with: Interval (days) between B. abortus and SRBC in-
sections
B19S
B19R
PB
20,516 ± 1,816b (1.6) 38,357 ± 4,432 (3.1) 13,022 ± 986 (1.0) 8,008 ± 301 (1.6) 10,229 ± 1,638b (2.0) 5,026 ± 519 (1.0) 10,652 ± 1,026 (0.9) 11,625 ± 676 (1.0) 15,452 ± 969 (1.4) 12,835 ± 2,621 (1.2) 7,172 ± 814 (0.6) 9,400 ± 962 (0.8) 12,556 ± 1,402 (1.3) 12,091 ± 1,016 (1.3) 12,638 ± 2,378 (1.3) 12,009 ± 1,136 (0.9) 16,265 ± 817 (1.2) 13,635 ± 1,390 (1.0) a Number of RFC measured 4 days after SRBC immunization. b Significantly different from the control group. 0 5 10 20 30 40
Nothing (controls)
12,536 ± 1,476 5,101 ± 581 11,095 ± 757 10,948 ± 1,456 9,295 ± 333 13,413 ± 694
8
INFECT. IMMUN.
LE GARREC ET AL.
cantly higher than the B19S group at days 4 and 6, but this was not maintained, except by the PB group, for the later days. The number of indirect PFC was measured on days 6, 9, 12, and 15 after immunization (Table 4). On day 9, when controls reached the maximum, the B19R and PB groups showed an increased response. On days 12 and 15, the difference between all treated animals and controls was still very significant, but whereas the B19R group was different from the B19S, the PB was not. The B19S group appeared to inhibit the day -0
o 250 z
IW 200.
i
150
Li
lu
6 response significantly. As mentioned above, only the non-agglutinogenic preparation modified the RFC response on day 4, but both the B19S and B19R preparations increased this response 7 and 11 days after injection with SRBC (Table 5). The effect of a minute dose (50 ,ug) of the Brucella preparations on the direct PFC response also was studied (Table 6). For the day 4 response, the B19R preparation was still stimulatory at 50 Ag, giving an adjuvant index of the same order as with 500 ,ug; the adjuvant index was larger with the 50-,ug dose than with the 500-/Ag dose in the B19S group. The prolongation effect, on the other hand, tended to increase with the dose. Response to SKBC when inactivated Brucella were injected after antigen. Non-agglutinogenic preparations injected 24 h after SRBC no longer affected the day 4 response (direct PFC), but an increase in the later response (day 7) was observed with both kinds of preparations
(Table 7).
100.
DISCUSSION The day 4 potentiation of the PFC and RFC 44 2'4 14 34 responses was only recorded with non-agglutinDAYS AFTER BRUCELLA INJECTION FIG. 1. Number of nucleated spleen cells at differ- ogenic Brucella preparations. It was maximum ent times after intravenous injection of 500 pg of when Brucella and SRBC were injected at the different killed B. abortus preparations. B19S (0); same time and decreased quickly when the time B19R (0); PB (A); controls (0). interval between the two injections increased. 50,
4
9
TABLE 3. Direct PFC response after simultaneous injection of 500 pg of different B. abortus preparations and SRBC No. of PFC/105 spleen cells + SE and adjuvant index in mice treated with:
Days after
immunzation
B19S
3 4 6 9
496 ± 46a (2.2) 578 ± 150 (0.8) 178 ± 14 (1.2)
B19R
521 ± 36a (2.4) 1,439 ± 606a, b (2.0) 247 ± 24a,b (1.6) 120 ± 33a (3.6) 92 ± 16a (2.8) 12 68 ± 6a (6.2) 44 ± 4a,b (4.0) 15 70 ± 7a (7.8) 39 ± 2.6a,b (4.3) aSignificantly different from the control group. b Significantly different from the B19S-treated group.
Nothing (con-
PB
trols)
641 ± 56a (2.9) 1,646 ± 402a' b (2.3) 266 ± 14a, (1.8) 95 ± 13a (2.9) 59 ± 6" (5.4) 81 ± 6a (9.0)
220 ± 718 ± 150 ± 33 ± 11 ± 9±
28 205 23 3 1.3 1.4
TABLE 4. Indirect PFC response after simultaneous injection of 500 pg of different B. abortus preparations and SRBC No. of PFC/106 spleen cells ± SE and adjuvant index in mice treated with:
Days after
immunmzatiountion
B19S
B19S
B19RPBNothing (conB19R PB
230 ± 35" (0.4) 341 ± 70 (0.6) 1,270 ± 185 (1.0) 2,070 ± 170ab (1.7) 1,970 ± 165" (4.4) 1,650 ± 240" (3.7) 990 ± 119a (3.6) 749 ± 51a (2.7) a Significantly different from control group. 6 9 12 15
b
Significantly different from B19S-treated group.
trols)
1,059 ± 253a (1.9) 2,420 ± 257ab (2.0) 2,560 ± 355 (5.8) 1,290 ± 293a (4.6) b
559 ± 55 1,210 ± 135 443 ± 49 278 ± 36
9
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
VOL. 20, 1978
TABLE 5. RFC response after sunultaneous injections of 500 pg of different killed B. abortus preparations and SRBC Days after immunization
None (controls)
B19R
B19S
% of car%of car-
RFC bon RFC RDC bon RFCbboRFbnRC 14.0 14.5 12.8 16,200 ± 1,230" (1.3) 12,300 ± 2,360 4.4 9.0 4.8 6,800 ± 1,060 22,080 ± 3,880" (3.2) 6.0 4.1 6.1 32,250 ± 2,240" (2.3) 14,060 ± 3,130
RFCa
9,320 ± 1,090 (0.8) 20,650 ± 2,930" (3.0) 26,650 ± 6,350" (1.9) a Number of RFC per 106 spleen cells; adjuvant index in parentheses. b Percentage of RFC exhibiting colloidal carbon uptake. e Significantly different from the control groups. 4 7 11
TABLE 6. Direct PFC response after simultaneous injection of killed B. abortus preparations and SRBC and influence of the dose of B. abortuse Days after immunization 4
B19R
B19S 50g
None (con500
trols)
50
OSS~gtoS 0p 5p 602 ± 176 433 ± 126b (0.7) 1,238 ± 2046 (2.1) 816 ± 102 (1.4) 1,380 ± 337b (2.3) 73 ± 7.5 150 ± 16 (2.0) 113 ± 25b (1.5) 114 ± 12b,c (1.6) 176 ± 20b, (2.4) 6 24 ± 5 120 ± 23 (5.0) 83 ± 13 (3.5) 150 ± 16" (6.2) 103 ± 13" (4.3) 10 a Values refer to the number of PFC per 106 spleen cells ± SE and to the adjuvant index (in parentheses). b
Significantly different from control group. Difference between two neighboring groups significant.
TABLE 7. Direct PFC response when injecting different killed B. abortus preparations 24 h after SRBC immunization injection
B19R
B19S
Days after SRBC 50 tg
500,
g
50,ug
None 5
g
(controls)
1,288 ± 38 1,109 ± 79 (0.8) 1,405 ± 181 (1.1) 1,119 ± 102 (0.9) 1,172 ± 256 (0.9) 116 ± 14 329 ± 21b (2.8) 237 ± 24b (2.0) 280 ± 20b (2.4) 227 ± 18b (1.9) ± parentheses). (in index adjuvant the to refer to the number of PFC per 106 spleen cells SE and aValues b Significantly different from control groups. 4 7
would be in accordance with the fact that McArthur et al. (12) were able to transfer the depressive effect on the PFC response induced by the preliminary injection of B19S (at day 2) using T cells. The simultaneous injection of various preparations of Brucella and SRBC stressed differences between the agglutinogenic and non-agglutinogenic preparations. Indeed, a marked increase of the direct PFC and RFC response was observed on day 4 only for the two non-agglutinogenic preparations, even used at low doses. On day 4, the proportion of carbon-phagocyting RFC was increased. This could be due to cytophilic antibodies whose appearance is, according to Kapp and Benacerraf (8), transient. But the percentage of carbon-positive cells did not differ are substantially for B19S and non-agglutinogenic of RFC and role The nature pressor cells. not as clear as for PFC. The cold centrifugation preparations. During the ensuing response, both kinds of technique, used here, gives, according to Cone and Wilson (3), a large proportion of thymus- preparations provided stimulation, and B19S derived RFC, so a proportion of RFC could be and PB preparations induced the best results. specific T suppressor cells. This hypothesis The same relations were observed between the
The inoculation of the Brucella organisms provoked an augmentation of the number of spleen cells which was much more prolonged with B19S than with the other preparations. Recently, Nowotny et al. (14) were able to induce the production of colony-stimulating factor and to protect animals against irradiation damage using the polysaccharide fraction of the lipopolysaccharide (LPS) of Salmonella minnesota. So, the presence of a fully expressed S surface antigen, which is supposed to be a polysaccharide, seemed necessary for this long-term effect of Brucella. The lack of parallelism between the PFC (diminished) and RFC (unchanged) responses observed after B19S for 5to 30-day intervals has not yet received explanation. Such a discrepancy could be due to sup-
10
INFECT. IMMUN.
LE GARREC ET AL.
day 9 response and the following days for the indirect PFC response. So, the antagonism between the surface immunogenicity and the immuno-potentiating power seemed to act on the early phase of the response and did not preclude a further increase of the response; moreover, the ensuing response was favorably influenced by the presence of the S surface antigen, even if it was prevented from expressing its own antigenicity (PB preparation). For this last preparation, a special handling of such an antigen-antibody complex by immunocompetent cells could be responsible for the observed effect. Takahashi et al. (19) have demonstrated that B19S and SRBC do not cross-react, so the hypothesis that anti-B19S antibodies are simply masking the SRBC may be excluded. The active component common to S and R strains of B. abortus could be inhibited by antigenic competition due to the polysaccharide of the S antigen. The nature of the active component is not known. The studies of the adjuvant effect of LPS extracted from Enterobacteriaceae have not brought out this kind of early inhibition. Johnson (7) got an early increase of the antibody titer by the simultaneous injection of S-LPS and BGG. Leong and Rudbach (11), using the same antigen, did not observe any difference between S- and R-LPS: immediate stimulation appeared in both cases. Such an effect was obtained by Franzl and McMaster (5) after simultaneous injection of S-LPS and SRBC. The surface antigen of Brucella seems to present some peculiarities of extraction (M. S. Redfearn, Ph.D. thesis, University of Wisconsin, Madison, 1960; C. H. Lacave, Dr. Sci. thesis, Toulouse, 1969), and not much is known about correlations between chemical structure and biological activities. Whereas LPS of Enterobacteriaceae is strongly mitogenic by its lipid A (2), recently Kreutzer et al. (9) observed that LPS extracted from S and R Brucella in the aqueous or phenolic phase of the Westphal technique were not mitogenic. The observation that the day 4 increase of the PFC response was no longer observed with the B19R preparation injected 24 h after SRBC, whereas the day 7 response was augmented, again emphasized the absence of correlation between the modifications acting on those two
periods. The separation of well-defined fractions of B. abortus which can interact, such as LPS, its polysaccharide and lipid A moieties, and peptidoglycan, is needed for a better understanding of the observed modifications induced by inactivated Brucella on the immune response. ACKNOWLEDGMENTS This work 4008.
was
supported by INSERM
contract
no.
72-
The authors thank F. B. Barrat, Cl. M. Gavinet, D. C. Le Rhun, D. J. Thibault, and M. Kessler for excellent technical assistance.
LITERATURE CITED 1. Berger, F. M., G. M. Fukui, B. J. Ludwig, and J. P. Rosselet. 1970. The increase of non specific resistance to bacterial infections by extracts obtained from Brucella abortus. International Symposium on Brucellosis, Tunis 1968. Symp. Ser. Immunobiol. Stand. 12:296-280. 2. Chiller, J. M., B. J. Skidmore, D. C. Morrison, and W. Weigle. 1973. Relationships of the structure of bacterial lipopolysaccharide to its function in mitogenesis and adjuvanticity. Proc. Natl. Acad. Sci. U.S.A. 70:2129. 3. Cone, R. E., and J. D. Wilson. 1972. Adjuvant action of Poly AU on T and B rosette-forming cells in SRBC immunized mice. Int. Arch. Allergy Appl. Immunol. 43:123-130. 4. Dresser, D. W., and H. H. Wortis. 1965. Use of an antiglobulin serum to detect cells producing antibody with low haemolytic efficiency. Nature (London) 208:858-859. 5. Franzl, R. E., and P. D. McMaster. 1968. The primary immune response in mice. I. The enhancement and suppression of hemolysin production by a bacterial endotoxin. J. Exp. Med. 127:1087-1107. 6. Jerne, N. K., and A. A. Nordin. 1963. Plaque formation in agar by single antibody producing cell population. Science 140:405-407. 7. Johnson, A. G. 1964. Adjuvant action of bacterial endotoxins on the primary antibody response, p. 252-262. In M. Landy and W. Braun (ed.), Bacterial endotoxins. Rutgers University Press, New Brunswick, N.J. 8. Kapp, J., and B. Benacerraf. 1972. The rosette cell response of several mouse strains to immunization with both sheep and pigeon erythrocytes; magnitude of errors caused by cytophilic antibody. Eur. J. Immunol. 2:467-472. 9. Kreutzer, D. L., J. W. Scheffer, L. R. Draper, and D. C. Robertson. 1977. Mitogenic activity of cell wall components from smooth and rough strains of Brucella abortus. Infect. Immun. 15:842-845. 10. Le Garrec, Y., D. Sabolovic, L. Toujas, J. Guelfi, and C. Pilet. 1974. Activity of inactivated Brucella on murine tumors: prophylactic effect and combination with specific stimulation. Biomedicine 21:40-43. 11. Leong, D. L., and J. A. Rudbach. 1971. Antigenic competition between an endotoxic adjuvant and a protein antigen. Infect. Immun. 3:308-317. 12. McArthur, W. P., G. W. Siskind, and G. J. Thorbecke. 1974. Studies on antigenic competition using a cell transfer system. Cell. Immunol. 13:66-75. 13. McConnel, I., A. Munro, B. W. Guerner, and R. R. A. Coombs. 1969. Studies on actively allergised cells. I. The cyto dynamics and morphology of rosettes forming lymph node cells in mice and inhibition of rosettes formation with antibody to mouse immunoglobulins. Int. Arch. Allergy Appl. Immunol. 35:209-227. 14. Nowotny, A., U. G. Belhing, and H. L. Ghang. 1975. Relation of structure to function in bacterial endotoxins. VIII. Biological activities in a polysaccharide rich fraction. J. Immunol. 115:199-203. 15. Pilet, C., and M. Bonneau. 1970. Sur un nouveau vaccin anti-brucellique non agglutinogene: le vaccin P. B. Rec. Med. Vet. 146:25-36. 16. Pilet, C., Y. Le Garrec, L. Toujas, D. Sabolovic, J. C. Monteil, F. Rothier, U. Mishra, B. Ghebrehiwet, and J. Guelfi. 1974. Non specific stimulation by inactivated or ultrasonicated Brucella abortus. Recent Results Cancer Res. 47:294-301. 17. Pilet, C., and D. Sabolovic. 1970. Brucella abortus et immunotherapie active non specifique de la tumeur d'Ehrlich. Bull. Assoc. Vet. Microbiol. 7:43-57.
VOL. 20. 1978
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
18. Ramon, G., R. Richou, J. P. Thiery, and C. Gerbeaux. 1949. De influence comparee de diverses substances stimulantes sur l'accroissement de l'immunite. C. R. Acad. Sci. (Paris) 229:278-280. 19. Takahashi, T., J. J. Mond, E. A. Carswell, and G. J. Thorbecke. 1971. The importance of theta and Ig bearing cells in the immune response to various antigens. J. Immunol. 107:1520. 20. Toujas, L., L. Dazord, and J. Guelfi. 1973. Augmenta-
11
tion des proprietes immunostimulantes de Brucella abortus B19S inactivee par action conjugu6e d'un antiserum specifique. C. R. Acad. Sci. (Paris) 276D:433-435. 21. Toujas, L., D. Sabolovic, L. Dazord, Y. Le Garrec, J. P. Toujas, J. Guelfi, and C. Pilet. 1972. The mechanism of immunostimulation induced by inactivated Brucella abortus. Rev. Eur. Etud. Clin. Biol. 17:267-273.
INFECTION AND IMMUNITY, Apr. 1978, p. 6-11 0019-9567/78/0020-0006$02.00/0 Copyright © 1978 American Society for Microbiology
Printed in U.S.A.
Influence of the Antigenicity of Brucella Preparations on Modulation of the Immune Response to Sheep Erythrocytes Y. LE GARREC,l* L. TOUJAS,2 A. MARTIN,2 L. DAZORD,2 AND C. PILETI Service de Microbiologie, d'Immunologie et de Pathologie Generale, Ecole Nationale Vetrinaire, 94701 Maisons-Alfort, France,' and Centre Eugene Marquis and Unite INSERM 49, Hopital Pontchaillou, 35000
Rennes, France'
Received for publication 26 October 1977
The relationship between the expression of surface antigenicity of inactivated Brucella and the immunostimulant properties on the sheep erythrocyte response was studied in mice. The agglutinogenic Brucella abortus B19S preparation was compared to two non-agglutinogenic preparations (B19R and PB), using the plaque-forming and rosette-forming cell tests. When Brucella and antigen were injected together, only non-agglutinogenic preparations, even used at low doses, were able to increase the plaque-forming and rosette-forming cell responses measured 4 days after immunization. When the interval between the previous injection of non-agglutinogenic preparations and that of the antigen increased or when they were injected 24 h after the antigen, the modification of those two responses was no longer observed. After day 4, for the simultaneous injection of Brucella and antigen, B19S and PB preparations induced the best responses; this effect was dose dependent. Thus, in the present experimental situation, the modifications of the day 4 response appeared independent of those of the ensuing days.
Antigens. SRBC were obtained once a week from one sheep, collected in Alsever solution, and washed three times in saline before use. Bacterial preparations. The following preparations were used: (i) the fully antigenic vaccinal strain
The ability of inactivated Brucella abortus vaccine to increase non-specifically the titer of antibodies to Salmonella in the rabbit has been stressed by Ramon et al. (18). More recently, their role in affecting the development of bacterial or viral infections and of various tumors has been investigated (1, 10, 16, 17). However, earlier study of the modifications induced by inactivated B. abortus (strain B19S) on the response to sheep erythrocytes (SRBC) showed that the number of plaque-forming cells (PFC) per spleen on day 4 after immunization was, if anything, depressed by the simultaneous or previous injection of Brucella. Stimulation did occur, however, if the time between injection of Brucella and injection of SRBC was increased to 40 days (21). The finding that an injection of anti-Brucella serum given before injecting the B19S caused a partial reversal of the effect (20) led to the hypothesis that some antagonism might exist between the antigenicity of B. abortus and its immuno-potentiating activity. To test this hypothesis, we chose to compare the modulating effects of agglutinogenic and non-agglutinogenic strains of inactivated Brucella on the immune response to SRBC. MATERIALS AND METHODS
B19S of B. abortus, which provokes the formation of agglutinins; (ii) the B19R rough mutant, selected from B19S in the laboratory; and (iii) the PB preparation, corresponding to the Pilet-Bonneau vaccine, which is a non-agglutinogenic immune complex (15). Brucella organisms were grown on Albimi agar, harvested after 48 h, washed twice in phosphatebuffered saline, and then inactivated by heat for 1 h at 650C. The PB preparation was prepared by neutralizing agglutinogenic sites of inactivated B19S with a specific
immune serum prepared in cows. B19R and PB preparations do not induce agglutinins in mice (or rabbits). The bacterial preparations were injected intravenously at various intervals before, at the same time as, or 1 day after immunization. A dose of 500 gg (dry weight) was used except when stated otherwise. PFC. The number of plaque-forming cells (PFC) in the spleen was determined after intravenous injection of 2.5 x 108 SRBC. The Jerne and Nordin method (6) for direct PFC and the Dresser and Wortis method (4) for indirect PFC were used. For the latter, an antimouse immunoglobulin G serum produced in goats (Behring) was diluted 1/200. Animals. Female mice F2(C57Bl6xDBA2) (6 to 7 RFC. The number of rosette-forming cells (RFC) weeks old) from the Centre de Selection du Centre in the spleen was determined after the intravenous National de la Recherche Scientifique (Orleans) were injection of 2.5 x 108 SRBC. The McConnel et al. used for this study. technique (13) was used, in which cell mixtures are 6
VOL. 20, 1978
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
merely centrifuged in the cold without standing at
40C. The number of rosette-forming macrophages was determined by injecting colloidal carbon 0.5 h before excising spleens.
The resulting numbers of PFC and RFC were expressed per 106 spleen cells. The mean per group of six (PFC) or five (RFC) mice is given with the standard error (SE). The significance was calculated by Student's t test and accepted at P < 0.05. In both cases, an adjuvant index was used to express the ratios of the responses in treated and untreated mice: AI = PFC or RFC in treated mice/PFC or RFC in control mice. The PFC and RFC response to immunization with SRBC was tested in different experimental situations, injecting the Brucella preparations (i) before, (ii) at the same time as, and (iii) after the antigen.
RESULTS
Response to SRBC after a preliminary injection of Brucella preparations. The number of anti-SRBC PFC 4 days after antigen was measured in mice which received 500 ,tg of B19S, B19R, or PB preparations at intervals of 0 to 40 days before immunization (Table 1). With respect to the B19S preparation, a 50 to 70% inhibition was observed when its injection was separated from the SRBC by an interval of 5 to 30 days. No significant effect was seen when
7
the two were given simultaneously. The nonagglutinogenic preparations B19R and PB did not appear to affect the anti-SRBC response except on day 0. On the other hand, both the B19R and PB preparations appeared to enhance the RFC response when given either with or 5 days before the antigen (Table 2). The 19S vaccine did not affect the RFC response unless the data were plotted as RFC per spleen; a two- to fourfold increase then was observed on days 10 to 40. The expression of the results in number of PFC or RFC per 106 spleen cells is indeed affected by the number of spleen cells. As shown in Fig. 1, B19S induced the most prolonged increase in spleen cell number. Response to SRBC after the simultaneous injection of inactivated Brucella and SRBC. The mice received 500 ,tg of the Brucella preparations mixed with 2.4 x 108 SRBC intravenously. The maximum value for the direct PFC count was reached on day 4 in the control group (Table 3). This value was not significantly different for the group injected with B19S, but was surpassed in animals given B19R or PB. Subsequently, the number of direct PFC decreased quickly in controls, but remained elevated in all treated groups. The B19R and PB groups were both signifi-
TABLE 1. Modifications of the PFC response by injection of 500 Pg of different killed B. abortus preparations before SRBC immunization Interval (days) between B. ab~ortus and SRBC injections
No. of PFCa/106 spleen cells ± SE and adjuvant index after treatment with: PB B19R B19S
1,747 ± 120b (2.5) 456 ± 65 (0.7) 825 ± 99 (1.2) 769 ± 113 (1.1) 891 ± 105 (1.3) 769 ± 22 (1.1) 675 ± 56 Controls (no B. abortus) a Number of PFC measured 4 days after SRBC immunization. b Significantly different from the control groups. e ND, Not done. 224 ± 49 (0.7) 110 ± 26b (0.3) 163 ± 5b (0.5) 175 ± 9ob (0.5) 123 ± 63b (0.4) 246 ± 58 (0.8) 319 ± 95
0 5 10 20 30 40
1,851 ± 173b (2.6) 894 ± 134 (1.2) 856 ± 107 (1.2) 487 ± 38 (0.7) ND' 672 ± 66 (0.9) 719 ± 56
TABLE 2. Modifications of the RFC response by injection of 500 jg of killed B. abortus preparations before SRBC immunization No. of RFCa/10f spleen cells ± SE and adjuvant index after treatment with: Interval (days) between B. abortus and SRBC in-
sections
B19S
B19R
PB
20,516 ± 1,816b (1.6) 38,357 ± 4,432 (3.1) 13,022 ± 986 (1.0) 8,008 ± 301 (1.6) 10,229 ± 1,638b (2.0) 5,026 ± 519 (1.0) 10,652 ± 1,026 (0.9) 11,625 ± 676 (1.0) 15,452 ± 969 (1.4) 12,835 ± 2,621 (1.2) 7,172 ± 814 (0.6) 9,400 ± 962 (0.8) 12,556 ± 1,402 (1.3) 12,091 ± 1,016 (1.3) 12,638 ± 2,378 (1.3) 12,009 ± 1,136 (0.9) 16,265 ± 817 (1.2) 13,635 ± 1,390 (1.0) a Number of RFC measured 4 days after SRBC immunization. b Significantly different from the control group. 0 5 10 20 30 40
Nothing (controls)
12,536 ± 1,476 5,101 ± 581 11,095 ± 757 10,948 ± 1,456 9,295 ± 333 13,413 ± 694
8
INFECT. IMMUN.
LE GARREC ET AL.
cantly higher than the B19S group at days 4 and 6, but this was not maintained, except by the PB group, for the later days. The number of indirect PFC was measured on days 6, 9, 12, and 15 after immunization (Table 4). On day 9, when controls reached the maximum, the B19R and PB groups showed an increased response. On days 12 and 15, the difference between all treated animals and controls was still very significant, but whereas the B19R group was different from the B19S, the PB was not. The B19S group appeared to inhibit the day -0
o 250 z
IW 200.
i
150
Li
lu
6 response significantly. As mentioned above, only the non-agglutinogenic preparation modified the RFC response on day 4, but both the B19S and B19R preparations increased this response 7 and 11 days after injection with SRBC (Table 5). The effect of a minute dose (50 ,ug) of the Brucella preparations on the direct PFC response also was studied (Table 6). For the day 4 response, the B19R preparation was still stimulatory at 50 Ag, giving an adjuvant index of the same order as with 500 ,ug; the adjuvant index was larger with the 50-,ug dose than with the 500-/Ag dose in the B19S group. The prolongation effect, on the other hand, tended to increase with the dose. Response to SKBC when inactivated Brucella were injected after antigen. Non-agglutinogenic preparations injected 24 h after SRBC no longer affected the day 4 response (direct PFC), but an increase in the later response (day 7) was observed with both kinds of preparations
(Table 7).
100.
DISCUSSION The day 4 potentiation of the PFC and RFC 44 2'4 14 34 responses was only recorded with non-agglutinDAYS AFTER BRUCELLA INJECTION FIG. 1. Number of nucleated spleen cells at differ- ogenic Brucella preparations. It was maximum ent times after intravenous injection of 500 pg of when Brucella and SRBC were injected at the different killed B. abortus preparations. B19S (0); same time and decreased quickly when the time B19R (0); PB (A); controls (0). interval between the two injections increased. 50,
4
9
TABLE 3. Direct PFC response after simultaneous injection of 500 pg of different B. abortus preparations and SRBC No. of PFC/105 spleen cells + SE and adjuvant index in mice treated with:
Days after
immunzation
B19S
3 4 6 9
496 ± 46a (2.2) 578 ± 150 (0.8) 178 ± 14 (1.2)
B19R
521 ± 36a (2.4) 1,439 ± 606a, b (2.0) 247 ± 24a,b (1.6) 120 ± 33a (3.6) 92 ± 16a (2.8) 12 68 ± 6a (6.2) 44 ± 4a,b (4.0) 15 70 ± 7a (7.8) 39 ± 2.6a,b (4.3) aSignificantly different from the control group. b Significantly different from the B19S-treated group.
Nothing (con-
PB
trols)
641 ± 56a (2.9) 1,646 ± 402a' b (2.3) 266 ± 14a, (1.8) 95 ± 13a (2.9) 59 ± 6" (5.4) 81 ± 6a (9.0)
220 ± 718 ± 150 ± 33 ± 11 ± 9±
28 205 23 3 1.3 1.4
TABLE 4. Indirect PFC response after simultaneous injection of 500 pg of different B. abortus preparations and SRBC No. of PFC/106 spleen cells ± SE and adjuvant index in mice treated with:
Days after
immunmzatiountion
B19S
B19S
B19RPBNothing (conB19R PB
230 ± 35" (0.4) 341 ± 70 (0.6) 1,270 ± 185 (1.0) 2,070 ± 170ab (1.7) 1,970 ± 165" (4.4) 1,650 ± 240" (3.7) 990 ± 119a (3.6) 749 ± 51a (2.7) a Significantly different from control group. 6 9 12 15
b
Significantly different from B19S-treated group.
trols)
1,059 ± 253a (1.9) 2,420 ± 257ab (2.0) 2,560 ± 355 (5.8) 1,290 ± 293a (4.6) b
559 ± 55 1,210 ± 135 443 ± 49 278 ± 36
9
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
VOL. 20, 1978
TABLE 5. RFC response after sunultaneous injections of 500 pg of different killed B. abortus preparations and SRBC Days after immunization
None (controls)
B19R
B19S
% of car%of car-
RFC bon RFC RDC bon RFCbboRFbnRC 14.0 14.5 12.8 16,200 ± 1,230" (1.3) 12,300 ± 2,360 4.4 9.0 4.8 6,800 ± 1,060 22,080 ± 3,880" (3.2) 6.0 4.1 6.1 32,250 ± 2,240" (2.3) 14,060 ± 3,130
RFCa
9,320 ± 1,090 (0.8) 20,650 ± 2,930" (3.0) 26,650 ± 6,350" (1.9) a Number of RFC per 106 spleen cells; adjuvant index in parentheses. b Percentage of RFC exhibiting colloidal carbon uptake. e Significantly different from the control groups. 4 7 11
TABLE 6. Direct PFC response after simultaneous injection of killed B. abortus preparations and SRBC and influence of the dose of B. abortuse Days after immunization 4
B19R
B19S 50g
None (con500
trols)
50
OSS~gtoS 0p 5p 602 ± 176 433 ± 126b (0.7) 1,238 ± 2046 (2.1) 816 ± 102 (1.4) 1,380 ± 337b (2.3) 73 ± 7.5 150 ± 16 (2.0) 113 ± 25b (1.5) 114 ± 12b,c (1.6) 176 ± 20b, (2.4) 6 24 ± 5 120 ± 23 (5.0) 83 ± 13 (3.5) 150 ± 16" (6.2) 103 ± 13" (4.3) 10 a Values refer to the number of PFC per 106 spleen cells ± SE and to the adjuvant index (in parentheses). b
Significantly different from control group. Difference between two neighboring groups significant.
TABLE 7. Direct PFC response when injecting different killed B. abortus preparations 24 h after SRBC immunization injection
B19R
B19S
Days after SRBC 50 tg
500,
g
50,ug
None 5
g
(controls)
1,288 ± 38 1,109 ± 79 (0.8) 1,405 ± 181 (1.1) 1,119 ± 102 (0.9) 1,172 ± 256 (0.9) 116 ± 14 329 ± 21b (2.8) 237 ± 24b (2.0) 280 ± 20b (2.4) 227 ± 18b (1.9) ± parentheses). (in index adjuvant the to refer to the number of PFC per 106 spleen cells SE and aValues b Significantly different from control groups. 4 7
would be in accordance with the fact that McArthur et al. (12) were able to transfer the depressive effect on the PFC response induced by the preliminary injection of B19S (at day 2) using T cells. The simultaneous injection of various preparations of Brucella and SRBC stressed differences between the agglutinogenic and non-agglutinogenic preparations. Indeed, a marked increase of the direct PFC and RFC response was observed on day 4 only for the two non-agglutinogenic preparations, even used at low doses. On day 4, the proportion of carbon-phagocyting RFC was increased. This could be due to cytophilic antibodies whose appearance is, according to Kapp and Benacerraf (8), transient. But the percentage of carbon-positive cells did not differ are substantially for B19S and non-agglutinogenic of RFC and role The nature pressor cells. not as clear as for PFC. The cold centrifugation preparations. During the ensuing response, both kinds of technique, used here, gives, according to Cone and Wilson (3), a large proportion of thymus- preparations provided stimulation, and B19S derived RFC, so a proportion of RFC could be and PB preparations induced the best results. specific T suppressor cells. This hypothesis The same relations were observed between the
The inoculation of the Brucella organisms provoked an augmentation of the number of spleen cells which was much more prolonged with B19S than with the other preparations. Recently, Nowotny et al. (14) were able to induce the production of colony-stimulating factor and to protect animals against irradiation damage using the polysaccharide fraction of the lipopolysaccharide (LPS) of Salmonella minnesota. So, the presence of a fully expressed S surface antigen, which is supposed to be a polysaccharide, seemed necessary for this long-term effect of Brucella. The lack of parallelism between the PFC (diminished) and RFC (unchanged) responses observed after B19S for 5to 30-day intervals has not yet received explanation. Such a discrepancy could be due to sup-
10
INFECT. IMMUN.
LE GARREC ET AL.
day 9 response and the following days for the indirect PFC response. So, the antagonism between the surface immunogenicity and the immuno-potentiating power seemed to act on the early phase of the response and did not preclude a further increase of the response; moreover, the ensuing response was favorably influenced by the presence of the S surface antigen, even if it was prevented from expressing its own antigenicity (PB preparation). For this last preparation, a special handling of such an antigen-antibody complex by immunocompetent cells could be responsible for the observed effect. Takahashi et al. (19) have demonstrated that B19S and SRBC do not cross-react, so the hypothesis that anti-B19S antibodies are simply masking the SRBC may be excluded. The active component common to S and R strains of B. abortus could be inhibited by antigenic competition due to the polysaccharide of the S antigen. The nature of the active component is not known. The studies of the adjuvant effect of LPS extracted from Enterobacteriaceae have not brought out this kind of early inhibition. Johnson (7) got an early increase of the antibody titer by the simultaneous injection of S-LPS and BGG. Leong and Rudbach (11), using the same antigen, did not observe any difference between S- and R-LPS: immediate stimulation appeared in both cases. Such an effect was obtained by Franzl and McMaster (5) after simultaneous injection of S-LPS and SRBC. The surface antigen of Brucella seems to present some peculiarities of extraction (M. S. Redfearn, Ph.D. thesis, University of Wisconsin, Madison, 1960; C. H. Lacave, Dr. Sci. thesis, Toulouse, 1969), and not much is known about correlations between chemical structure and biological activities. Whereas LPS of Enterobacteriaceae is strongly mitogenic by its lipid A (2), recently Kreutzer et al. (9) observed that LPS extracted from S and R Brucella in the aqueous or phenolic phase of the Westphal technique were not mitogenic. The observation that the day 4 increase of the PFC response was no longer observed with the B19R preparation injected 24 h after SRBC, whereas the day 7 response was augmented, again emphasized the absence of correlation between the modifications acting on those two
periods. The separation of well-defined fractions of B. abortus which can interact, such as LPS, its polysaccharide and lipid A moieties, and peptidoglycan, is needed for a better understanding of the observed modifications induced by inactivated Brucella on the immune response. ACKNOWLEDGMENTS This work 4008.
was
supported by INSERM
contract
no.
72-
The authors thank F. B. Barrat, Cl. M. Gavinet, D. C. Le Rhun, D. J. Thibault, and M. Kessler for excellent technical assistance.
LITERATURE CITED 1. Berger, F. M., G. M. Fukui, B. J. Ludwig, and J. P. Rosselet. 1970. The increase of non specific resistance to bacterial infections by extracts obtained from Brucella abortus. International Symposium on Brucellosis, Tunis 1968. Symp. Ser. Immunobiol. Stand. 12:296-280. 2. Chiller, J. M., B. J. Skidmore, D. C. Morrison, and W. Weigle. 1973. Relationships of the structure of bacterial lipopolysaccharide to its function in mitogenesis and adjuvanticity. Proc. Natl. Acad. Sci. U.S.A. 70:2129. 3. Cone, R. E., and J. D. Wilson. 1972. Adjuvant action of Poly AU on T and B rosette-forming cells in SRBC immunized mice. Int. Arch. Allergy Appl. Immunol. 43:123-130. 4. Dresser, D. W., and H. H. Wortis. 1965. Use of an antiglobulin serum to detect cells producing antibody with low haemolytic efficiency. Nature (London) 208:858-859. 5. Franzl, R. E., and P. D. McMaster. 1968. The primary immune response in mice. I. The enhancement and suppression of hemolysin production by a bacterial endotoxin. J. Exp. Med. 127:1087-1107. 6. Jerne, N. K., and A. A. Nordin. 1963. Plaque formation in agar by single antibody producing cell population. Science 140:405-407. 7. Johnson, A. G. 1964. Adjuvant action of bacterial endotoxins on the primary antibody response, p. 252-262. In M. Landy and W. Braun (ed.), Bacterial endotoxins. Rutgers University Press, New Brunswick, N.J. 8. Kapp, J., and B. Benacerraf. 1972. The rosette cell response of several mouse strains to immunization with both sheep and pigeon erythrocytes; magnitude of errors caused by cytophilic antibody. Eur. J. Immunol. 2:467-472. 9. Kreutzer, D. L., J. W. Scheffer, L. R. Draper, and D. C. Robertson. 1977. Mitogenic activity of cell wall components from smooth and rough strains of Brucella abortus. Infect. Immun. 15:842-845. 10. Le Garrec, Y., D. Sabolovic, L. Toujas, J. Guelfi, and C. Pilet. 1974. Activity of inactivated Brucella on murine tumors: prophylactic effect and combination with specific stimulation. Biomedicine 21:40-43. 11. Leong, D. L., and J. A. Rudbach. 1971. Antigenic competition between an endotoxic adjuvant and a protein antigen. Infect. Immun. 3:308-317. 12. McArthur, W. P., G. W. Siskind, and G. J. Thorbecke. 1974. Studies on antigenic competition using a cell transfer system. Cell. Immunol. 13:66-75. 13. McConnel, I., A. Munro, B. W. Guerner, and R. R. A. Coombs. 1969. Studies on actively allergised cells. I. The cyto dynamics and morphology of rosettes forming lymph node cells in mice and inhibition of rosettes formation with antibody to mouse immunoglobulins. Int. Arch. Allergy Appl. Immunol. 35:209-227. 14. Nowotny, A., U. G. Belhing, and H. L. Ghang. 1975. Relation of structure to function in bacterial endotoxins. VIII. Biological activities in a polysaccharide rich fraction. J. Immunol. 115:199-203. 15. Pilet, C., and M. Bonneau. 1970. Sur un nouveau vaccin anti-brucellique non agglutinogene: le vaccin P. B. Rec. Med. Vet. 146:25-36. 16. Pilet, C., Y. Le Garrec, L. Toujas, D. Sabolovic, J. C. Monteil, F. Rothier, U. Mishra, B. Ghebrehiwet, and J. Guelfi. 1974. Non specific stimulation by inactivated or ultrasonicated Brucella abortus. Recent Results Cancer Res. 47:294-301. 17. Pilet, C., and D. Sabolovic. 1970. Brucella abortus et immunotherapie active non specifique de la tumeur d'Ehrlich. Bull. Assoc. Vet. Microbiol. 7:43-57.
VOL. 20. 1978
KILLED BRUCELLA ABORTUS AND ANTI-SRBC RESPONSE
18. Ramon, G., R. Richou, J. P. Thiery, and C. Gerbeaux. 1949. De influence comparee de diverses substances stimulantes sur l'accroissement de l'immunite. C. R. Acad. Sci. (Paris) 229:278-280. 19. Takahashi, T., J. J. Mond, E. A. Carswell, and G. J. Thorbecke. 1971. The importance of theta and Ig bearing cells in the immune response to various antigens. J. Immunol. 107:1520. 20. Toujas, L., L. Dazord, and J. Guelfi. 1973. Augmenta-
11
tion des proprietes immunostimulantes de Brucella abortus B19S inactivee par action conjugu6e d'un antiserum specifique. C. R. Acad. Sci. (Paris) 276D:433-435. 21. Toujas, L., D. Sabolovic, L. Dazord, Y. Le Garrec, J. P. Toujas, J. Guelfi, and C. Pilet. 1972. The mechanism of immunostimulation induced by inactivated Brucella abortus. Rev. Eur. Etud. Clin. Biol. 17:267-273.
