Você está na página 1de 5

Journal of Medical Microbiology (2005), 54, 527531 DOI 10.1099/jmm.0.

45936-0

Reduced interleukin-18 secretion in Brucella


abortus 2308-infected murine peritoneal
macrophages and in spleen cells obtained from
B. abortus 2308-infected mice
Luis Fernandez-Lago,1 Antonio Orduna2 and Nieves Vizcano1
1
Correspondence Departamento de Microbiologa y Genetica, Edificio Departamental, Universidad de Salamanca,
Luis Fernandez-Lago Plaza Doctores de la Reina s/n, 37007 Salamanca, Spain
lrlago@usal.es 2
Departamento de Microbiologa, Facultad de Medicina, Universidad de Valladolid, Valladolid, Spain

Th1 immune responses in which gamma interferon (IFN-) production predominates are associated
with protective immunity against intracellular bacteria. Following infection, interleukin-18 (IL-18) may
contribute, in association with IL-12, to optimal IFN- production. In this study, the secretion of IL-18
following intracellular infection with virulent Brucella abortus 2308 in CD-1 cultured peritoneal
macrophages and splenocyte cultures was investigated. The production of IL-18 was reduced in
both CD-1 mouse peritoneal macrophages infected with B. abortus 2308 and splenocyte cultures
obtained from B. abortus 2308-infected mice at 3, 6 and 10 days post-infection (p.i.). In contrast,
splenocyte cultures obtained from B. abortus 2308-infected mice at 3 days p.i. secreted significant
amounts of IFN-. Stimulation of these cells with recombinant IL-18 (rIL-18) and/or rIL-12 did not
significantly increase IFN- secretion at the splenocyte level. These data suggest that once the
Received 21 October 2004 infection has been established, B. abortus 2308 selectively limits IL-18 secretion without affecting
Accepted 4 February 2005 endogenous IFN- production.

INTRODUCTION synthesized as a biologically inactive 24 kDa precursor


protein. Cleavage to an 18 kDa active mature protein,
Brucella abortus is a facultative intracellular bacterium and is
mediated by the interleukin-1 (IL-1)-converting enzyme,
one of the causative agents of brucellosis in animals and
also called caspase-1, is essential for IL-18 to become
humans (Fernandez-Lago et al., 1999). Professional phago-
biologically active (Ghayur et al., 1997). Although IL-18
cytes are the first target of B. abortus invasion and the bacteria
alone does not induce significant IFN- production, it
are able to survive and multiply within these cells (Pizarro-
synergistically enhances IL-12-stimulated IFN- production
Cerda et al., 1998). Resistance to B. abortus depends on the
(Yoshimoto et al., 1998) and promotes cell-mediated im-
effective generation of a T-cell-mediated response (Araya et
munity (Takeda et al., 1998). Besides its IFN- inducing
al., 1989) as well as on an effective activation of macrophages
effect, IL-18 has a direct proinflammatory effect on T and
by gamma interferon (IFN-)-producing CD4 T-lympho-
natural killer cells, enhancing proliferation and cytotoxicity
cytes (Zhan et al., 1993). IFN- is a key mediator in
and stimulating the production of cytokines such as tumour
conferring protection against Brucella spp. infections both
necrosis factor alpha (TNF-), IL-1, IL-6 and IL-8 (Netea et
in vivo (Zhan & Cheers, 1995) and in vitro (Jiang & Baldwin,
al., 2000).
1993), and an elevated IFN- production, mainly in the early
phase of the infection, could be crucial in their control
Previous studies have demonstrated the role of IL-18 in the
(Pasquali et al., 2002). In consequence, cytokines secreted in
hosts response to infection. Endogenously synthesized IL-18
this initial response, which can affect IFN- secretion, can
has been demonstrated to be required in mice for an adequate
strongly affect the outcome of infection.
host defence against mycobacterial infections (Sugawara et al.,
Interleukin-18 (IL-18) is a proinflammatory cytokine pro- 1999), disseminated candidiasis (Stuyt et al., 2004), pneumo-
duced by several cell types, including activated macrophages, coccal pneumonia (Lauw et al., 2002), and Pseudomonas
dermal keratinocytes, osteoblasts, adrenal cortex cells and aeruginosa (Schultz et al., 2003), Yersinia enterocolitica (Bohn
intestinal epithelial cells (Gerdes et al., 2002), that is et al., 1998), Salmonella typhimurium (Mastroeni et al., 1999),
Leishmania major (Ohkusu et al., 2000) and Listeria mono-
Abbreviations: HK, heat-killed; i.p., intraperitoneal; p.i., post-infection; cytogenes (Neighbors et al., 2001) infections. Recently, it
S-LPS, smooth lipopolysaccharide. has been shown that only a combined administration of

45936 & 2005 SGM Printed in Great Britain 527


L. Fernandez-Lago, A. Orduna and N. Vizcano

recombinant IL-18 and IL-12 induces protection against B. phagocytosed bacteria were washed off. Following this, 1 ml RPMI
abortus 2308 infection in mice, and this effect is possibly due supplemented as described above and containing 50 g gentamicin
to an increase in IFN- production in the early phases of ml 1 was added to the macrophages to kill any remaining extracellular
bacteria. The cells were then cultured in the same medium with
infection (Pasquali et al., 2001, 2002). However, at present
gentamicin (50 g ml 1 ) at 37 8C under a 5 % (v/v) CO2 atmosphere
the dynamics of endogenous IL-18 production in response to in air for 48 h, and culture supernatants were obtained from each well,
a B. abortus infection have not been thoroughly investigated. filtered through 0.45 m pore-size filters, and kept at 70 8C until used
for cytokine analysis. Supernatants obtained at the same time from
The aim of the present study was to investigate the capacity of
uninfected macrophage cultures were used as negative controls. Unin-
B. abortus 2308 to stimulate the secretion of IL-18 by cultured fected macrophage cultures (five wells per sample) were also incubated,
CD-1 peritoneal macrophages and splenocyte cultures, and in the same medium with gentamicin (50 g ml 1 ), in the presence of
to evaluate the role of this cytokine in IFN- production in either purified S-LPS (20 g ml 1 ) or a 1/10 dilution of culture
response to B. abortus 2308 infection. supernatant obtained from B. abortus 2308, and S-LPS from E. coli
O157 (1 g ml 1 ; Sigma). Macrophage supernatants were also obtained
at 48 h from each well, filtered, and kept at 70 8C until assayed.
METHODS
Splenocyte cultures. Infected CD-1 mice were killed on days 3, 6 and
Mice. Female outbred CD-1 mice (Charles River, Barcelona, Spain) 10 p.i. The spleens were removed aseptically, and individual cell
were purchased at 8 weeks of age and kept for 1 week before use. The suspensions were prepared by gentle teasing through cell strainers
animals were housed in microisolator cages in Horsefall units, and were (Becton Dickinson). Erythrocytes were lysed by adding lysis buffer
cared for in accordance with standard guidelines. (0.155 M NH4 Cl, 10 mM KHCO3 , 10 mM EDTA, pH 7.4) to the
pellet for 3 min. The cells were then washed three times in RPMI,
Bacteria and infection of animals. Virulent B. abortus 2308, used as 20 mM HEPES supplemented with 2 mM L-glutamine, 10 5 M -
the challenge strain, was cultured on tryptic soy agar (TSA; BBL) mercaptoethanol and 10 % (v/v) FCS. Spleen cells (106 ) were cultured in
enriched with 0.3 % (w/v) yeast extract (YE; Difco) for 72 h at 37 8C 24-well plates (0.5 ml per well) and stimulated with 0.5 ml heat-killed
under a 5 % (v/v) CO2 atmosphere in air. To infect the mice, the bacteria (HK) B. abortus 2308 at 107 cells ml 1 . Cultures were incubated at 37 8C
were suspended in sterile PBS (pH 7.4) and adjusted turbidimetrically to under a 5 % (v/v) CO2 atmosphere in air. Supernatants were harvested
the desired concentration. The exact dose was established retrospec- at 48 h, filtered through a 0.45 m pore-size filter, and frozen at 70 8C
tively by viable cell counting. Mice were infected by the intraperitoneal for use in cytokine assays. Controls including cultures of splenocytes
(i.p.) route with 0.2 ml PBS containing 5 3 105 c.f.u. of the B. abortus obtained from uninfected mice or from infected mice stimulated with
2308 virulent strain. Control groups received 0.2 ml PBS. At the the mitogen concanavalin A (5 g per well; Sigma) were also established.
indicated time in each experiment, five mice per group were killed by In some experiments, splenocytes obtained at 3 days p.i. from mice
cervical dislocation. infected with 5 3 105 c.f.u. B. abortus 2308 were cultured, once
stimulated with 107 HK B. abortus 2308 cells, in the presence of
Reagents. B. abortus 2308 smooth lipopolysaccharide (S-LPS) and recombinant IL-18 (10 ng ml 1 ) and/or rIL-12 (10 ng ml 1 ). Super-
culture supernatant extracts were obtained as described previously natants were also removed after 48 h, filtered, and tested for IFN-
(Vizcano et al., 1991; Joubier-Maurin et al., 2001). The S-LPS con- production.
tained, as a percentage of dry weight, 1.5 % protein and 1.1 % 2-keto-3-
deoxyoctulosonic acid. S-LPS from Escherichia coli O157 was obtained Cytokine ELISA. Cytokine contents in macrophage and splenocyte
from Sigma. Recombinant mouse IL-18 and IL-12 were obtained from supernatants were determined by ELISA, using commercial kits for
Pharmingen. IL-18 (biologically active and non-active forms) and IFN- (OptEIA;
Pharmingen). Protocols were as recommended by the manufacturer.
Isolation and infection of CD-1 mouse peritoneal macrophages. Known concentrations of each cytokine were used to generate standard
Macrophages were isolated as described previously (Fernandez-Lago et curves in each assay and used as references to calculate cytokine
al., 1999), with some modifications. Briefly, CD-1 mice were inoculated concentrations. Samples were assayed in duplicate and values are
i.p. with 0.3 ml incomplete Freunds adjuvant (Sigma). Three days later, expressed as means  SD. The lowest limit of detection for IL-18 and
the mice were killed and peritoneal cells were harvested by washing with IFN- was 30 pg ml 1 .
cold RPMI (Gibco) with 10 % (v/v) fetal calf serum (FCS; Gibco). The
peritoneal cells thus collected were washed twice by centrifugation at Statistical methods. All experiments in this study were performed at
1200 g for 5 min, and resuspended in RPMI, 20 mM HEPES (Life least twice. Experimental values are given as means  SD. The data
Technologies) supplemented with 2 mM L-glutamine (Life Tech- obtained were analysed by one-way analysis of variance, comparing the
nologies), 10 5 M -mercaptoethanol and 10 % (v/v) FCS. The cell mean values with the Fishers protected LSD (least significant differ-
suspensions were adjusted to 106 cells ml 1 , seeded on 24-well plates ences, P , 0.05).
(1 ml per well), and macrophages were allowed to adhere for 90 min at
37 8C under a 5 % (v/v) CO2 atmosphere. Non-adhering cells were
washed off and adhering cells were washed three times with 1 ml of the RESULTS AND DISCUSSION
same medium. The number and viability of the macrophages obtained
by this procedure were determined by morphological examination and IL-18 secretion is reduced in B. abortus
trypan blue exclusion (Fernandez-Lago et al., 1999). For macrophage
infection, bacteria from overnight cultures were opsonized at 37 8C for
2308-infected macrophages
1 h in PBS containing a subagglutinating dilution (1 : 200) of the To investigate the effect of B. abortus 2308 infection on
monoclonal antibody BmE10-5, which reacts specifically with the macrophage-derived IL-18 secretion, CD-1 mouse elicited-
S-LPS of Brucella spp. (Vizcano et al., 1991). Opsonized B. abortus
2308 cells were then centrifuged, washed and diluted in RPMI
peritoneal macrophage cultures were used. As shown in Fig.
supplemented as described above. For infection, macrophage cells were 1, 24 h culture supernatants of uninfected macrophages
incubated with 1 ml of the opsonized bacterial suspension at a m.o.i. constitutively secreted IL-18. A significant (P , 0.05) in-
of 30 : 1 (bacteria : macrophage) for 1 h. After incubation, non- crease in the secretion of this cytokine was observed when

528 Journal of Medical Microbiology 54


Reduced IL-18 secretion by B. abortus

100 abortus 2308 to induce the secretion of IL-18 mediated by


Interleukin-18 (pg ml21)

* CD-1 mouse elicited-peritoneal macrophages. As shown in


80
Fig. 1, neither B. abortus 2308 culture supernatants nor B.
60 abortus 2308 S-LPS had a significant effect on IL-18 secretion.
Similar results were obtained upon using different concen-
40 trations of both preparations (data not shown). These
20
findings suggest, as happens with wild-type Salmonella
,30 pg ml21 dublin infections (Elhofy & Bost, 1999), that this decrease
0 in IL-18 synthesis in mouse macrophages is possibly depen-
Uninfected LPS-Ec Ba-infected LPS-Ba Extract-Ba
dent on the intracellular replication of B. abortus 2308,
although the mechanism responsible for this remains to be
Fig. 1. B. abortus 2308-infected peritoneal macrophages secrete elucidated.
reduced amounts of IL-18. Cultured peritoneal macrophages were
uninfected, infected with virulent B. abortus 2308 (Ba-infected), or IFN- production in B. abortus 2308-infected
treated with S-LPS from E. coli (LPS-Ec; 1 g ml 1 ), S-LPS from B. splenocyte cultures is not dependent on the
abortus 2308 (LPS-Ba; 20 g ml 1 ) or a B. abortus 2308 culture presence of IL-18
extract (Extract-Ba; 1 : 10 dilution). After 2 days of culture, super-
natants were harvested and assayed for IL-18 by ELISA. Results are In subsequent experiments we analysed the effect that the
expressed as means  SD (five wells per point). *, Statistically differ- reduction in IL-18 secretion, observed in cultured macro-
ent from uninfected macrophage cultures (P , 0.05). phages as a result of B. abortus 2308 infection, had on IFN-
secretion in B. abortus 2308-infected mice at the splenocyte
level. As shown in Table 1, a 48 h splenocyte culture super-
cultured peritoneal macrophages were exposed to S-LPS natant obtained from uninfected CD-1 mice constitutively
from E. coli O157 (1 g ml 1 ) (Fig. 1). In contrast, when expressed IL-18. A significant decrease (P , 0.05) in the
infected with opsonized B. abortus 2308 for 90 min, followed secretion of this cytokine was observed in 48 h splenocyte
by removal of extracellular bacteria, supernatants of these cultures obtained from B. abortus 2308-infected mice on days
macrophage cultures showed no IL-18 activity (, 30 pg 3, 6 and 10 p.i., and stimulated in vitro with 107 HK B. abortus
ml 1 ; Fig. 1). These results demonstrate that B. abortus 2308 2308 cells. In contrast, a statistically significant (P , 0.05)
infection induces a reduction of IL-18 secretion in mice at the increase in IFN- secretion by the splenocyte cultures was
macrophage level. It has previously been demonstrated that observed on days 3, 6 and 10 p.i. (Table 1). These results show
Brucella culture supernatants contain a protein factor(s) that that the limited secretion of IL-18, as a result of B. abortus
is able to inhibit the secretion in LPS-activated human 2308 infection, does not significantly affect IFN- produc-
macrophages of other macrophage-derived cytokines, such tion, and therefore that IL-18 would not be involved in B.
as TNF- (Joubier-Maurin et al., 2001). Accordingly, we also abortus-induced IFN- secretion at the splenocyte level. To
analysed the capacity of both the culture supernatants (1 : 10 investigate this further, splenocytes were obtained from B.
dilution) and the S-LPS (20 g ml 1 ) obtained from B. abortus 2308-infected mice at 3 days p.i. and, once stimulated

Table 1. Production of IL-18 and IFN- in stimulated spleen cells from B. abortus 2308-
infected mice
CD-1 mice were i.p.-infected with 5 3 105 c.f.u. B. abortus 2308. Mice, five per group, were killed
at 3, 6 and 10 days after infection. Data are mean values of five animals  SD. An asterisk indicates
that the results are statistically significant (P , 0.05) as compared to those for the uninfected
control group.

Time after infection that Spleen wt Cytokine level (pg ml21 )


splenocytes were obtained (mg)
(days) IL-18 IFN-

Uninfected control 100  17 125  10 , 30


3 164  20* 60  8* 137  20*
6 204  15* 35  3* 120  15*
10 404  47* , 30 130  12*

Splenocytes were cultured in RPMI medium and stimulated with 107 HK B. abortus 2308 cells as
described in Methods. Supernatants were collected at 48 h for cytokine analysis.
Culture supernatants were tested for IL-18 and IFN- by ELISA. The level of sensitivity for both
cytokines was 30 pg ml 1 .

http://jmm.sgmjournals.org 529
L. Fernandez-Lago, A. Orduna and N. Vizcano

Table 2. Effect of IL-18 and IL-12 on IFN- production by Elhofy, A. & Bost, K. (1999). Limited interleukin-18 response in
stimulated spleen cells from B. abortus 2308-infected mice Salmonella-infected murine macrophages and in Salmonella-infected
mice. Infect Immun 67, 50215026.
CD-1 mice were i.p.-infected with 5 3 105 c.f.u. B. abortus 2308. Mice,
Fernandez-Lago, L., Rodrguez-Tarazona, E. & Vizcano, N. (1999).
five per group, were killed at 3 days after infection. Data are mean values Differential secretion of interleukin-12 (IL-12) subunits and hetero-
of five animals  SD. dimeric IL-12p70 protein by CD-1 mice and murine macrophages in
response to intracellular infection by Brucella abortus. J Med Microbiol
Treatment* IFN- level (pg ml21 ) 48, 10651073.
Gerdes, N., Sukhova, G. K., Libby, P., Reynolds, R. S., Young, J. L. &
None 130  26 Schonbeck, U. (2002). Expression of interleukin (IL)-18 and functional
IL-18 145  60 IL-18 receptor on human vascular endothelial cells, smooth muscle
IL-12 142  33 cells, and macrophages: implications for atherogenesis. J Exp Med 195,
IL-18 + IL-12 145  50 245257.
Concanavalin A 550  125 Ghayur, T., Banerjee, S., Hugunin, M. & 11 other authors (1997).
Caspase-1 processes IFN-gamma-inducing factor and regulates LPS-
*Cytokines (10 ng ml 1 ) were added to splenocyte cultures. Concana- induced IFN-gamma production. Nature 386, 619626.
valin A was used at 5 g ml 1 . Jiang, X. & Baldwin, C. L. (1993). Effects of cytokines on intracellular
Splenocytes were cultured in RPMI medium and stimulated with 107 growth of Brucella abortus. Infect Immun 61, 124134.
HK B. abortus 2308 cells as described in Methods. Supernatants were Joubier-Maurin, V., Boigegrain, R. A., Cloeckaert, A. & 8 other authors
collected at 48 h and tested for IFN- by ELISA. The level of sensitivity (2001). Major outer membrane protein Omp25 of Brucella suis is
was 30 pg ml 1 . involved in inhibition of tumor necrosis factor alpha production during
infection of human macrophages. Infect Immun 69, 48234830.
Lauw, F. N., Branger, J., Florquin, S., Speelman, P., van Deventer,
with 107 HK B. abortus cells, were treated with rIL-12 (10 ng S. J. H., Akira, S. & van der Poll, T. (2002). IL-18 improves the early
antimicrobial host response to pneumococcal pneumonia. J Immunol
ml 1 ) and/or rIL-18 (10 ng ml 1 ) or concanavalin A (5 g
168, 372378.
ml 1 ). The levels of IFN- were then measured in the culture
supernatants 48 h after stimulation. As shown in Table 2, the Mastroeni, P. S., Clare, S., Khan, S., Harrison, J. A., Hormaeche, C. E.,
Okamura, H., Kurimoto, M. & Dougan, G. (1999). Interleukin 18
addition to splenocyte cultures of rIL-12, rIL-18 or both contributes to host resistance and gamma interferon production in
cytokines simultaneously did not significantly affect IFN- mice infected with virulent Salmonella typhimurium. Infect Immun 67,
secretion as compared to untreated cells. A significant 478483.
(P , 0.05) increase in IFN- production was only seen in Neighbors, M., Xu, X., Barrat, F. J. & 7 other authors (2001). A critical
the concanavalin A-treated splenocyte cultures (Table 2). It role for interleukin 18 in primary and memory effector responses to
may therefore be concluded that once the infection has been Listeria monocytogenes that extend beyond its effects on interferon
established, the exogenous addition of IL-12, IL-18 or both gamma production. J Exp Med 194, 343354.
cytokines simultaneously does not contribute to IFN- Netea, M. G., Kullberg, B. J., Verschueren, I. & van der Meer, J. W.
production in response to B. abortus 2308 infection. Thus (2000). Interleukin-18 induces production of proinflammatory cyto-
it is possible that both cytokines are active, inducing a kines in mice: no intermediate role for the cytokines of the tumor
protective endogenous IFN- production only when phar- necrosis factor family and interleukin-1. J Immunol 30, 30573060.
macologically effective levels of them in a biologically active Ohkusu, K., Yoshimoto, T., Takeda, K., Ogura, T., Kashiwamura, S.-I.,
form are present in the first stages of infection (Pasquali et al., Iwakura, Y., Akira, S., Okamura, H. & Nakanishi, K. (2000). Potentiality
2002). of interleukin-18 as a useful reagent for treatment and prevention of
Leishmania major infection. Infect Immun 68, 24492456.
Pasquali, P., Adone, R., Gasbarre, L. C., Pistola, C. & Ciuchini, F. (2001).
Mouse cytokine profiles associated with Brucella abortus RB51 vaccina-
ACKNOWLEDGEMENTS tion or B. abortus 2308 infection. Infect Immun 69, 65416544.
This work was financed by project G03/204 Red para la Investigacion de Pasquali, P., Adone, R., Gasbarre, L. C., Pistoia, C. & Ciuchini, F. (2002).
la Brucelosis, Instituto de Salud Carlos III, Ministerio de Sanidad y Effect of exogenous interleukin-18 (IL-18) and IL-12 in the course of
Consumo, Spain. Brucella abortus 2308 infection in mice. Clin Diag Lab Immunol 9,
491492.
Pizarro-Cerda, J., Moreno, E., Sanguedolce, V., Mege, J. L. & Gorvel,
J. P. (1998). Virulent Brucella abortus prevents lysosome fusion and is
REFERENCES distributed within autophagosome-like compartments. Infect Immun
Araya, L. N., Elzer, P. H., Rowe, G. E., Enright, F. M. & Winter, A. J. (1989). 66, 23872392.
Temporal development of protective cell-mediated and humoral
Schultz, M. J., Knapp, S., Florquin, S., Pater, J., Takeda, K., Akira, S. &
immunity in BALB/c mice infected with Brucella abortus. J Immunol
van der Poll, T. (2003). Interleukin-18 impairs the pulmonary host
143, 33303337.
response to Pseudomonas aeruginosa. Infect Immun 71, 16301634.
Bohn, E., Sing, A., Zumbihl, R., Brielfeldt, C., Okamura, H., Kurimoto, Stuyt, R. J., Netea, M. G., van Krieken, J. H., van der Meer, J. W. &
M., Heesemann, J. & Autenrieth, I. B. (1998). IL-18 (IFN--inducing Kullberg, B. J. (2004). Recombinant interleukin-18 protects against
factor) regulates early cytokine production in, and promotes resolution disseminated Candida albicans infection in mice. J Infect Dis 189, 1524
of, bacterial infection in mice. J Immunol 160, 299307. 1527.

530 Journal of Medical Microbiology 54


Reduced IL-18 secretion by B. abortus

Sugawara, I., Yamada, H., Kaneko, H., Mizuno, S., Takeda, K. & Akira, Yoshimoto, T., Nagai, N., Ohkusu, K., Ueda, H., Okamura, H. &
S. (1999). Role of interleukin-18 (IL-18) in mycobacterial infection in Nakanishi, K. (1998). IL-12 up-regulates IL-18 receptor expression on
IL-18-gene-disrupted mice. Infect Immun 67, 25852589. T cells, Th1 cells, and B cells: synergism with IL-18 for IFN-
Takeda, K., Tsutsui, H., Yoshimoto, T., Adachi, O., Yoshida, N., production. J Immunol 161, 34003407.
Kishimoto, T., Okamura, H., Nakanishi, K. & Akira, S. (1998). Defective
Zhan, Y. & Cheers, C. (1995). Endogenous interleukin-12 is involved in
NK cell activity and Th1 response in IL-18-deficient mice. Immunity 8,
resistance to Brucella abortus infection. Infect Immun 63, 13871390.
383390.
Vizcano, N., Chordi, A. & Fernandez-Lago, L. (1991). Characterization Zhan, Y., Yang, J. & Cheers, C. (1993). Cytokine response of T-cell
of smooth Brucella lipopolysaccharides and polysaccharides by mono- subsets from Brucella abortus-infected mice to soluble Brucella proteins.
clonal antibodies. Res Microbiol 142, 971978. Infect Immun 61, 28412847.

http://jmm.sgmjournals.org 531