Search

To investigate the value of the BP26 protein in the serological diagnosis of ovine brucellosis caused by Brucella ovis, recombinant BP26 protein was produced in Echerichia coli and purified for use in an indirect enzyme-linked immunosorbent assay (I-ELISA). The majority of the recombinant protein was recovered from the supernatant of sonicated recombinant E. coli cells in a soluble form. This facilitated the purification of the recombinant BP26 protein which was achieved by using ion-exchange chromatography. After one step of purification, the purity of the recombinant BP26 protein was analyzed by using SDS-PAGE, Coomassie blue staining, and Western blot with a monoclonal antibody (MAb) directed against the BP26 protein. The degree of purity appeared satisfactory so that it could be directly used in I-ELISA. Although the recombinant BP26-ELISA appeared less useful than I-ELISA using the B. ovis hot saline (HS) extract as antigen, the high number of sera from B. ovis infected rams found positive (90%) in the recombinant BP26-I-ELISA indicated that the BP26 protein may be an additional suitable antigen for serological diagnosis of B. ovis infection in rams.

Smooth lipopolysaccharides (S-LPSs) from Brucella strains isolated from seals, dolphins, porpoises, an otter and a minke whale were characterized by ELISA using monoclonal antibodies (mAbs) directed against seven previously defined O-polysaccharide (O-PS) epitopes and by Western blot after SDS-PAGE. All strains studied were A-dominant as shown by specific polyclonal sera and this was also confirmed by the mAbs. However, binding patterns in ELISA of mAbs to the specific common (C) epitopes were rather heterogeneous, and for some strains, such as those isolated from striped dolphins, binding of these mAbs was much reduced or negative as had previously been shown for Brucella suis biovar 2 strains. Western blot after SDS-PAGE showed the typical A-dominant strain banding pattern for all marine mammal Brucella isolates, but the average S-LPS size was shorter in many of these compared to reference Brucella abortus strain 544. Thus, S-LPSs of the marine mammal isolates show heterogeneity with regard to their O-PS C epitope content and their average size.

Brucella abortus RB51 is a rough (R) stable vaccine strain used in cattle and is believed to be devoid of O-side chain. We analyzed by use of a panel of monoclonal antibodies (MAbs) directed against seven previously defined O-polysaccharide (O-PS) epitopes the O-chain expression in strain RB51. Two MAbs specific for the C/Y (A=M) and C (M>A) epitopes showed low bindings in ELISA to strain RB51. O-chain expression was further confirmed by Western blot after SDS-PAGE of strain RB51. In particular, the MAb of C (M>A) specificity, showing preferential binding to M-dominant smooth (S) Brucella strains, revealed in strain RB51 a typical smooth-lipopolysaccharide (S-LPS) pattern which resembled that of M-dominant S-LPS. Thus, the results clearly show that strain RB51 produces low levels of M-like O-antigen.

In the present study we completed the nucleotide sequence of a Brucella melitensis 16M DNA fragment deleted from B. abortus that accounts for 25,064 bp and show that the other Brucella spp. contain the entire 25-kb DNA fragment. Two short direct repeats of four nucleotides, detected in the B. melitensis 16M DNA flanking both sides of the fragment deleted from B. abortus, might have been involved in the deletion formation by a strand slippage mechanism during replication. In addition to omp31, coding for an immunogenic protein located in the Brucella outer membrane, 22 hypothetical genes were identified. Most of the proteins that would be encoded by these genes show significant homology with proteins involved in the biosynthesis of polysaccharides from other bacteria, suggesting that they might be involved in the synthesis of a Brucella polysaccharide that would be a heteropolymer synthesized by a Wzy-dependent pathway. This polysaccharide would not be synthesized in B. abortus and would be a polysaccharide not identified until present in the genus Brucella, since all of the known polysaccharides are synthesized in all smooth Brucella species. Discovery of a novel polysaccharide not synthesized in B. abortus might be interesting for a better understanding of the pathogenicity and host preference differences observed between the Brucella species. However, the possibility that the genes detected in the DNA fragment deleted in B. abortus no longer lead to the synthesis of a polysaccharide must not be excluded. They might be a remnant of the common ancestor of the alpha-2 subdivision of the class Proteobacteria, with some of its members synthesizing extracellular polysaccharides and, as Brucella spp., living in association with eukaryotic cells.

The Brucella melitensis sucB gene encoding the dihydrolipoamide succinyltransferase (E2o) enzyme (previously identified as an immunogenic protein in infected sheep) was cloned and sequenced. The amino acid sequence predicted from the cloned gene revealed 88.8 and 51.2% identity to the dihydrolipoamide succinyltransferase SucB protein from Brucella abortus and Escherichia coli, respectively. Sera from naturally infected sheep showed antibody reactivity against the recombinant SucB protein.

Previously a Brucella protein named CP28, BP26, or Omp28 has been identified as an immunodominant antigen in infected cattle, sheep, goats, and humans. In the present study we evaluated antibody responses of infected and B. melitensis Rev.1-vaccinated sheep to the BP26 protein using purified recombinant BP26 protein produced in Escherichia coli in an indirect enzyme-linked immunosorbent assay (I-ELISA). The specificity of the I-ELISA determined with sera from healthy sheep (n = 106) was 93%. The sensitivity of the I-ELISA assessed with sera from naturally infected and suspected sheep found positive in the current conventional diagnostic tests was as follows: 100% for bacteriologically and serologically positive sheep (n = 50), 88% for bacteriologically negative but serologically and delayed-type hypersensitivity-positive sheep (n = 50), and 84% for bacteriologically and serologically negative but delayed-type hypersensitivity-positive sheep (n = 19). However, the absorbance values observed did not reach those observed in an I-ELISA using purified O-polysaccharide (O-PS) as an antigen. In sheep experimentally infected with B. melitensis H38 the antibody response to BP26 was delayed and much weaker than that to O-PS. Nevertheless, the BP26 protein appears to be a good diagnostic antigen to be used in confirmatory tests and for serological differentiation between infected and B. melitensis Rev.1-vaccinated sheep. Weak antibody responses to BP26 in some of the latter sheep suggest that a B. melitensis Rev.1 bp26 gene deletion mutant should be constructed to ensure this differentiation.

Brucella melitensis is a facultative intracellular pathogen which is able to survive and replicate within phagocytic cells. Therefore, it has to adapt to a range of different hostile environments. In order to understand the mechanisms of intracellular survival employed by virulent B. melitensis 16M, an initial approach consisting of analysis of the differences in patterns of protein synthesis in response to heat, oxidative, and acid pH stresses by two-dimensional (2-D) polyacrylamide gel electrophoresis was used. Depending on the stress, this involved about 6.4 to 12% of the 676 protein spots detected in 2-D gel electrophoresis. On the basis of N-terminal sequence analysis and database searching, 19 proteins whose level of synthesis was up- or down-regulated by stress conditions were identified. Some of them were previously reported for Brucella, such as BvrR, DnaK, GroEL, and Cu-Zn superoxide dismutase (SOD). Eight other proteins closely matched proteins found in other bacteria: AapJ, alpha-ETF, ClpP, Fe and/or Mn SOD, malate dehydrogenase, IalB, 30S ribosomal protein S1, and pyruvate dehydrogenase E1 component beta subunit. Results indicated that B. melitensis could bring specific regulatory mechanisms into play in response to stress conditions. For example, the ribosome releasing factor in B. melitensis appeared to be a heat shock protein, whereas the ClpP protein, described as a heat shock protein for Escherichia coli, was strongly down-regulated in B. melitensis in response to heat stress. Some of the identified proteins and their potential specific regulation could be required for the adaptation of B. melitensis to environmental stresses encountered in phagocytic cells and possibly for bacterial virulence.

Outer membrane proteins (OMPs) and rough lipopolysaccharide (R-LPS), the main surface antigens of Brucella ovis, display surface-exposed epitopes. Mixtures of monoclonal antibodies (mAbs) to both antigens were previously shown to protect mice against a B. ovis challenge. To further identify the antigens involved, seven mAbs against Brucella OMPs (Omp10, Omp16, Omp19, Omp25, Omp31, Omp2b and Omp1) and three to R-LPS were tested for protection either individually or in combinations. Significant reduction in spleen infection in challenged mice, relative to controls, was used as the protection criteri. Controls included nonimmunized mice and mice given an irrelevant, anti-O-polysaccharide (OPS), mAb. For comparison, a group received a mouse serum containing antibodies to both OMPs and R-LPS; this serum was prepared by immunization with a B. ovis hot-saline extract which, as described previously, induces protective immunity in mice and rams. Significant protection was observed with both mAbs to OMPs and R-LPS. mAbs to Omp16, Omp19 and Omp31 afforded the highest protection and prevented the development of splenomegaly. The protective effect of mAb to Omp31 was not interfered with by nonprotective mAbs in different mixtures. The data presented confirm the protective role of antibodies to OMPs and R-LPS against B. ovis, and identify several OMPs, especially Omp31, which are promising candidates for a subunit vaccine against ram epididymitis.

Brucellae have been reported to be phylogenetically related to bacteria of the family Rhizobiaceae. In the present study, we used a panel of monoclonal antibodies (MAbs) to Brucella outer membrane proteins (OMPs) to determine the presence of common OMP epitopes in some representative bacteria of this family, i.e., Ochrobactrum anthropi, Phyllobacterium rubiacearum, Rhizobium leguminosarum, and Agrobacterium tumefaciens, and also in bacteria reported to serologically cross-react with brucella, i.e., Yersinia enterocolitica O:9, Escherichia coli O:157, and Salmonella urbana. In particular, most MAbs to the Brucella outer membrane lipoproteins Omp10, Omp16, and Omp19 cross-reacted with O. anthropi and P. rubiacearum, which are actually the closest relatives of brucellae. Some of them also cross-reacted, but to a lower extent, with R. leguminosarum and A. tumefaciens. The putative Omp16 and Omp19 homologs in these bacteria showed the same apparent molecular masses as their Brucella counterparts. None of the antilipoprotein MAbs cross-reacted with Y. enterocolitica O:9, E. coli O:157, or S. urbana.

A Brucella melitensis 16M DNA fragment of 17,119 bp, which contains a large region deleted in B. abortus strains and DNA flanking one side of the deletion, has been characterized. In addition to the previously identified omp31 gene, 14 hypothetical genes have been identified in the B. melitensis fragment, most of them showing homology to genes involved in the synthesis of a polysaccharide. Considering that 10 of the 15 genes are missing in B. abortus and that all the polysaccharides described in the Brucella genus (lipopolysaccharide, native hapten, and polysaccharide B) have been detected in all the species, it seems likely that the genes described here might be part of a cluster for the synthesis of a novel Brucella polysaccharide. Several polysaccharides have been identified as important virulence factors, and the discovery of a novel polysaccharide in the brucellae which is probably not synthesized in B. abortus might be interesting for a better understanding of the pathogenicity and host preference differences observed between the Brucella species. However, the possibility that the genes described in this paper no longer encode the synthesis of a polysaccharide cannot be excluded. Brucellae belong to the alpha-2 subdivision of the class Proteobacteria, which includes other microorganisms living in association with eucaryotic cells, some of them synthesizing extracellular polysaccharides involved in the interaction with the host cell. The genes described in this paper might be a remnant of the common ancestor of the alpha-2 subdivision of the class Proteobacteria, and the brucellae might have lost such extracellular polysaccharide during evolution if it was not necessary for survival or for establishment of the infectious process. Nevertheless, further studies are necessary to identify the entire DNA fragment missing in B. abortus strains and to elucidate the mechanism responsible for such deletion, since only 9,948 bp of the deletion was present in the sequenced B. melitensis DNA fragment.

Smooth Brucella strains are classified into three serotypes, i.e., A+M-, A-M+, and A+M+, according to slide agglutination with A and M monospecific polyclonal sera. The epitopes involved have been located on the O-polysaccharide (O-PS) moiety of the smooth lipopolysaccharide (S-LPS), which represents the most exposed antigenic structure on the surface of Brucella spp. By use of monoclonal antibodies (MAbs) a number of epitope specificities on the O-PS have been reported: A, M, and epitopes shared by both A and M dominant strains, which have been named common (C) epitopes. The latter have been further subdivided, according to relative MAb binding in enzyme-linked immunosorbent assays (ELISA) to A- and M-dominant Brucella strains and to cross-reacting Yersinia enterocolitica O:9, into five epitopic specificities: C (M>A), C (M=A), C/Y (M>A), C/Y (M=A), and C/Y (A>M). In the present study, we studied the occurrence of these epitopes at the surface of representatives of all Brucella species and biovars including the live vaccine strains by analyzing the levels of MAb binding to whole Brucella cells in ELISA and flow cytometry assays. In ELISA, the level of MAb binding correlated well with the previously defined epitope specificity and the serotype defined by polyclonal sera for each Brucella species, biovar, or strain. However, MAbs to the C (M=A) and C (M>A) epitopes showed insignificant binding to B. suis biovar 2 strains and bound at lower titers to B. suis biovar 3 and B. neotomae than to the other Brucella strains. Some of the flow cytometry results were contradictory to those obtained by ELISA. In fact, it appeared by flow cytometry that all O-PS epitopes, including the A and M epitopes, are shared to different degrees by Brucella spp. which nevertheless show a high degree of O-PS heterogeneity according to MAb binding intensities. The subdivision of MAb specificities and Brucella serotypes was therefore less evident by flow cytometry than by ELISA. Whereas in ELISA the MAb specific for the A epitope showed insignificant binding to Y. enterocolitica O:9, this MAb bound strongly to Y. enterocolitica O:9 in flow cytometry. One of the two MAbs specific to the C (M=A) epitope also bound at a low but significant level to B. suis biovar 2 strains. However, as in ELISA the MAb specific for the C (M>A) epitope did not bind at all to B. suis biovar 2 strains in flow cytometry. Flow cytometry provided new information regarding specificity of the MAbs and may further explain some aspects of the capacity of passive protection of some MAbs against smooth Brucella infection in mice. As shown in the present study the occurrence of Brucella strains apparently completely devoid of one specific C O-PS epitope (e.g., B. suis biovar 2 devoid of the C [M>A] epitope) offers the possibility of obtaining vaccine strains devoid of a diagnostic O-PS epitope, which could further help to resolve the problem of discriminating infected from vaccinated animals that remains a major goal in brucellosis research.

An indirect enzyme-linked immunosorbent assay (ELISA) was developed for the serological diagnosis of bovine anaplasmosis with purified recombinant major surface protein 5 (MSP5) of Anaplasma marginale produced in Escherichia coli. Serum antibody responses against MSP5 were detected in calves experimentally infected with A. marginale as early as 21 days postinfection and reached maximum titers at 28 days postinfection. The MSP5 ELISA performed with serum samples taken from field cattle from different regions of Venezuela showed a seroprevalence of 47%, which seems to be in accordance with the reported epidemiological status of bovine anaplasmosis in Venezuela. Positive results obtained in the MSP5 ELISA were further confirmed by immunoblotting, with the recombinant MSP5 as the antigen. Thus, these results confirmed the importance of MSP5 as a suitable antigen for the serological diagnosis of bovine anaplasmosis.

A panel of 45 Brucella ovis serologically positive sera were tested in immunoblots against B. ovis outer membrane proteins Omp31 and Omp25, purified by preparative SDS-gel electrophoresis. Forty-three sera reacted with Omp31, while only 11 reacted with Omp25, suggesting that Omp31 is identical to the previously reported immuno-dominant 29-kDa protein. Attempts to purify Omp31 on a larger scale by using procedures such as ion exchange-, reversed phase-, affinity- and gel filtration chromatography suggested that the outer membrane proteins were aggregated with rough lipopolysaccharide. Only denaturing SDS-gel filtration chromatography was able to separate proteins of about 29 kDa from rough lipopolysaccharide but did not separate Omp31 from Omp25 in B. ovis preparations. When used in an enzyme-linked immunosorbent assay, this 29-kDa protein preparation was less sensitive and less specific than the routinely used heat-extracted B. ovis antigen. A readily available recombinant E. coli, expressing the gene for Omp31 from Brucella melitensis 16 M, was used to extract and enrich recombinant Omp31 by a temperature-dependent Triton X-114-based technique. When this material was used in immunoblots with the 45 sera from B. ovis-infected sheep and with 10 monoclonal antibodies, raised against B. ovis Omp31, major differences in the antibody reactivity between the recombinant B. melitensis Omp31 and the B. ovis Omp31 were found. Such differences were unexpected because of the known structural and immunological relatedness of outer membrane proteins from various Brucella species. These results indicated that the antibody-response in B. ovis naturally-infected sheep against the immuno-dominant Omp31 was directed against epitopes which were only accessible when the protein was aggregated with rough lipopolysaccharides, or which were formed after aggregation but were not present in the recombinant protein.

The antibody response specific to the 25-kDa major outer-membrane protein (Omp25) of Brucella melitensis expressed in Escherichia coli was assessed in BALB/c mice. Groups of mice were immunised and boosted either with sonicated E. coli carrying plasmid pAC2533-E. coli (pAC2533)-expressing the gene coding for Omp25 (omp25 gene) of B. melitensis, or with E. coli carrying plasmid pUC19-E. coli (pUC19). One control group received saline. The evolution of antibody responses was investigated by indirect ELISA with whole rough (R) B. melitensis H38 cells as antigen. Serum antibody titres of mice immunised with E. coli (pAC2533) were appreciably higher than those of mice immunised with E. coli (pUC19). The specificity to Omp25 of murine antibodies induced by E. coli (pAC2533) was demonstrated by SDS-PAGE and immunoblotting of five B. melitensis strains. Binding of antibody in E. coli (pAC2533) immune sera to the surface of B. melitensis strains differing in their smooth lipopolysaccharide (S-LPS) expression was also studied by whole-cell ELISA and by flow cytometry. Antibody reactivity to R and smooth-rough (S-R) was much stronger than that to smooth (S) B. melitensis strains, indicating a much better accessibility of Omp25 to antibody on strains lacking or expressing less O-polysaccharide on their surface. The antibodies to Omp25 were predominantly of IgG2a isotype. The capacity of E. coli (pAC2533) to induce protective immune responses against four challenge strains of B. melitensis was further evaluated in mice. Significant reductions in splenic infections, in comparison with mice immunised with E. coli (pUC19) and unimmunised (saline injection) mice, were observed in R B. melitensis B115, S-R B. melitensis EP and S B. melitensis H38 infected mice. Protection against S B. melitensis 16M was not significant. The data from the present study, together with previous results, suggest that humoral immunity against probably conformational, well-exposed epitopes of the Omp25 could contribute to protective mechanisms against B. melitensis infection in mice.

A bacterioferritin (BFR) deletion mutant of Brucella melitensis 16M was generated by gene replacement. The deletion was complemented with a broad-host-range vector carrying the wild-type bfr gene, pBBR-bfr. The survival and growth of the mutant, B. melitensis PAD 2-78, were similar to those of its parental strain in human monocyte-derived macrophages (MDM). These results suggest that BFR is not essential for the intracellular survival of B. melitensis in human MDM.

We have previously produced and characterized four monoclonal antibodies to the Brucella DnaK protein which were derived from mice infected with B. melitensis or immunized with the B. melitensis cell wall fraction. By use of a recombinant DNA technique, we have localized a linear epitope, recognized by two of these monoclonal antibodies (V78/07B01/G11 and V78/09D04/D08), in the last 21 amino acids of the C-terminal region of the Brucella DnaK protein. The C-terminal region has been reported to be the most variable region among DnaK proteins. The two other monoclonal antibodies (A53/09G03/D02 and A53/01C10/A10) failed to react with the recombinant clones and might recognize discontinuous epitopes of the Brucella DnaK protein. The four monoclonal antibodies reacted with all recognized Brucella species and biovars in immunoblotting after SDS-PAGE. Monoclonal antibodies V78/07B01/G11 and V78/09D04/D08 did not react with reported cross-reacting bacteria nor with bacteria of the alpha-2 subdivision of the class Proteobacteria for which a close genetic relationship with Brucella spp. has been reported. However, monoclonal antibodies A53/09G03/D02 and A53/01C10/A10 reacted with Phyllobacterium rubiacearum and/or Ochrobactrum anthropi, both bacteria of the alpha-2 subdivision of the class Proteobacteria. The Brucella genus DnaK specific epitopes could be of importance for diagnostic purposes.

Brucellergene is a commercial allergen prepared from Brucella melitensis strain B115 and containing at least 20 cytoplasmic proteins. These proteins were separated by SDS-PAGE. The unstained gel was divided into 18 fractions and proteins were eluted from the gel fractions. The capacity of the separated proteins to elicit delayed-type hypersensitivity (DTH) in infected guinea-pigs or to induce the production of interferon-gamma (IFN-gamma) by blood cells from infected cattle was evaluated. The biological activity of the corresponding protein fractions blotted on to nitrocellulose was measured in a lymphocyte blastogenesis assay. Among the 18 fractions tested, two-spanning the mol. wt ranges 17-22 (fraction 8) and 35-42-kDa (fraction 17)-showed the maximum biological activity in the three tests. These fractions contain two antigens, the Brucella bacterioferritin (BFR) and P39 proteins. Both proteins are good candidates for the detection of cellular immunity to Brucella.

The omp-31 gene, encoding a major outer-membrane protein in Brucella melitensis, was PCR-amplified from Brucella strains representing all species and known biovars by using primers selected according to the B. melitensis 16M omp-31 published sequence. Amplification of omp-31 was achieved from DNA of all Brucella species with the exception of Brucella abortus, the only Brucella species where expression of omp-31 was not detected by reactivity with an mAb specific for an epitope located in Omp-31. Southern blot hybridization of plasmid probes, bearing inserts (4.4-17 kb) containing B. melitensis 16M omp-31 and adjacent DNA of different sizes, with HindIII-digested total DNA showed that a large fragment, comprising the entire omp-31 gene and flanking DNA, was actually absent in B. abortus strains. The size of this DNA fragment has been determined to be about 10 kb. Southern blot hybridization with the different plasmid probes identified species-specific markers for B. abortus and B. melitensis. At the biovar level, a specific marker for B. melitensis bv. 1 was also identified. Additionally, PCR-RFLP studies of omp-31 revealed specific markers for Brucella ovis, Brucella canis and Brucella suis bv. 2. Using a combination of omp-31 PCR-RFLP patterns and Southern blot hybridization profiles Brucella species were differentiated with the sole exception of Brucella neotomae which was not differentiated from B. suis bv. 1, 3, 4 and 5. Results presented in this paper demonstrate the potential of omp-31 for differentiating the brucellae and show that B. abortus lacks a large DNA fragment of about 10 kb containing omp-31 and flanking DNA. In such a large deletion, other genes in addition to omp-31 are probably involved. Sequencing of this DNA fragment will help to identify the missing genes in B. abortus which could possibly be involved in the differences of pathogenicity and host preference seen in Brucella species.

In a previous report, proteins from Brucella melitensis were characterized by two-dimensional polyacrylamide gel electrophoresis (2-D PAGE) and N-terminal microsequencing. In the present report, we have extended this study to the second etiologic agent in ovine brucellosis, B. ovis, responsible for ram epididymitis and infertility. The combination of 2-D gel electrophoresis and protein microsequencing facilitated the location and identification of the major proteins of B. ovis on the 2-D pattern. These proteins comprised cytoplasmic, periplasmic, and some membrane proteins except the major outer membrane proteins. By comparing 2-D gel profiles of B. ovis with that of B. melitensis described previously, a few proteins with different expression levels were readily identified. Serum from a ram naturally infected with B. ovis was used in immunoblotting studies to identify immunogenic proteins recognized during the course of infection. This serum showed antibody reactivity against approximately 82 protein spots. Twenty-one of these proteins were identified either by use of monoclonal antibodies or by N-terminal microsequencing. Several proteins previously described in earlier Brucella works were identified: the 89 kDa outer membrane protein, DnaK, GroEL, BP26, and Cu-Zn superoxide dismutase. Eight proteins had amino acid sequences homologous to those of various proteins from other bacteria found in protein databases; NikA, dihydrolipoamide succinyltransferase, a hypothetical 31 kDa protein, malate dehydrogenase, succinyl-CoA synthetase alpha subunit, an amino acid ABC type transporter, Leu/Ile/Val-binding protein precursor, and ClpP. The remaining eight proteins had N-terminal sequences lacking similarity to existing databases entries. Thus, the 2-D PAGE analysis provided a convenient first approach in the characterization of immunogenic proteins.

Two-dimensional (2-D) gel electrophoresis was used to map Brucella melitensis proteins. The 2-D proteins map of B. melitensis B115 revealed 595 silver-stained protein spots separated by both isoelectric point and molecular mass. Twenty-five proteins were identified either by immunoblotting using monoclonal antibodies (MAbs) or by N-terminal microsequencing. The protein spots identified by MAbs were the 89 kDa outer membrane protein, DnaK, bacterioferritin, CP24, and BP26. Some spots were identified by N-terminal microsequencing as proteins whose sequences had been reported previously from Brucella, such as three heat-shock proteins, namely DnaK, GroEL and GroES; bacterioferritin; Cu-Zn superoxide dismutase; and the 50S ribosomal protein L7/L12. Other proteins had amino acid sequences homologous with those of various proteins from other bacteria found in protein databases: ClpP; the 10K-S protein; the ORFU phosphoprotein; succinyl-CoA synthetase alpha sub-unit; an inorganic pyrophosphatase; the Fe and/or Mn superoxide dismutase; the nucleoside diphosphate kinase, an amino acid ABC type transporter, and an electron transfer flavoprotein small subunit. Seven proteins were identified with N-terminal sequences not yet reported in databases. The 2-D map established in this study will be the basis for comparative studies of protein expression in Brucella.

Competitive enzyme-linked immunosorbent assay (C-ELISA) was performed using 15 monoclonal antibodies (MAbs), specific for Brucella BP26 (previously also called CP28), a periplasmic protein antigen, to investigate antibody responses in naturally and B. melitensis H38 experimentally infected and B. melitensis Rev.1 vaccinated sheep. The antigen preparation consisted of cytosoluble protein extract (CPE) of B. melitensis B115. By combining the C-ELISA results of several MAbs, a high percentage of naturally infected animals were detected which showed different status in the current conventional diagnostic tests. Indeed, 90% of sheep which were positive in the conventional bacteriological and serological tests were positive in C-ELISA. 72% of the bacteriologically negative but serologically and delayed type hypersensitivity positive sheep were also positive in the C-ELISA. Moreover, 79% of the bacteriologically and serologically negative sheep but delayed type hypersensitivity positive were also detected by C-ELISA. Thus, these results confirmed the importance of BP26 as a frequently recognized target of the humoral immune response of infected sheep. The 8 B. melitensis H38 experimentally infected sheep showed various degrees of antibody responses at the 90th day after infection, which was delayed in comparison to that against O-polysaccharide (O-PS). Of the 15 MAbs tested, only one MAb was weakly inhibited (20 to 35% inhibition) by 56% of negative control sera. Furthermore, no antibody response against BP26 was detected in B. melitensis Rev.1 vaccinated sheep. Results of the C-ELISA with the 15 MAbs showed individual variability of the antibody responses against BP26. Thus, it is suggested that several epitopes of BP26 are of interest for diagnosis of B. melitensis infection in sheep.

The gene coding for a Brucella melitensis cytosoluble protein (CP24) that is immunogenic in infected sheep and a major component of brucellin INRA was cloned and sequenced. As in Brucella cells, CP24 was located in the cytoplasm of recombinant Escherichia coli. The amino acid sequence predicted from the cloned gene revealed 48 and 46% identity with the ribosome releasing factor, a protein factor required for release of the 70S ribosome from the mRNA, of E. coli and Haemophilus influenzae Rd, respectively. Sera from naturally infected sheep and sheep experimentally infected with B. melitensis H38 showed antibody reactivity against recombinant CP24.

Monoclonal antibodies (MAbs) were produced to Brucella melitensis cytosoluble proteins (CP) with apparent molecular masses of 12, 24 and 28 kDa (CP12, CP24 and CP28) which were previously shown by immunoblotting to differentiate antibody responses of infected sheep from those of B. melitensis strain Rev. 1 vaccinated sheep. These MAbs were derived from mice infected with virulent smooth (S) B. melitensis strain H38. Most MAbs obtained were directed to CP28, which indicated (as was shown in infected sheep) that this protein was also highly immunogenic in mice. A large number of MAbs that showed reactivity to CP in ELISA but did not show reactivity in immunoblotting of CP were also obtained and might recognise conformational epitopes of these proteins. MAbs were used to localise CP12, CP24 and CP28. None of the MAbs reacted with whole B. melitensis cells in ELISA but showed reactivity with sonicated bacteria in ELISA, which indicated an internal localisation of these proteins. Among several B. melitensis B115 subcellular fractions tested, the anti-CP12 and anti-CP28 MAbs reacted essentially with the CP extract (CPE) in both ELISA and immunoblotting, whereas the anti-CP24 MAbs reacted with both CPE and cell envelope fraction (CEF)-although with lower intensity to the latter fraction. The internal localisation of these proteins was confirmed by immuno-electron microscopy of thin-sectioned B. melitensis B115 or B. melitensis 16M cells. Immunogold labelling was mainly observed in the cytoplasm and, consequently, CP12, CP24 and CP28 are probably cytoplasmic proteins. Immunoblotting of whole cell lysates with the MAbs also showed the presence of these proteins in all Brucella species and biovars, including the vaccine strains B. melitensis Rev. 1 and B. abortus B19. The use of these MAbs should help further study of antibody responses in sheep and other hosts and may be of considerable value for developing new diagnostic tests for ovine brucellosis.

The gene coding for the major outer membrane protein (OMP) of 31 to 34 kDa, now designated Omp31, of Brucella melitensis 16M was cloned and sequenced. A B. melitensis 16M genomic library was constructed in lambda GEM-12 XhoI half-site arms, and recombinant phages expressing omp31 were identified by using the anti-Omp31 monoclonal antibody (MAb) A59/10F09/G10. Subcloning of insert DNA from a positive phage into pGEM-7Zf allowed the selection of a plasmid bearing a 4.4-kb EcoRI fragment that seemed to contain the entire omp31 gene under control of its own promoter. omp31 was localized within a region of the EcoRI insert of approximately 1.1 kb. Sequencing of this region revealed an open reading frame of 720 bp encoding a protein of 240 amino acids and a predicted molecular mass of 25,307 Da. Cleavage of the first 19 amino acids, showing typical features of signal peptides for protein export, leaves a mature protein of 221 amino acids with a predicted molecular mass of 23,412 Da. The predicted amino acid sequence of B. melitensis 16M Omp31 showed 35.2% identity with the RopB OMP of Rhizobium leguminosarum bv. viciae 248 and 34.3% identity with Omp25 of B. abortus 544. As in Brucella spp., Omp31 was located in the outer membrane of recombinant Escherichia coli, but its reported peptidoglycan association in Brucella cells was not detected in E. coli. The ability of Omp31 to form oligomers resistant to sodium dodecyl sulfate denaturation at low temperatures, a characteristic described for several bacterial porins, was observed in both B. melitensis and recombinant E. coli. The epitope recognized by the anti-Omp31 MAb A59/10F09/G10, for which a protective activity has been suggested, has been delimited to a region of 36 amino acids of Omp31 covering the most hydrophilic part of the protein. The availability of recombinant Omp31 and the identification of the antigenic determinant recognized by MAb A59/10F09/G10 will allow the evaluation of their potential protective activity and their potential for the development of subcellular vaccines against brucellosis.

We have previously identified a Brucella melitensis 28 kDa cytosoluble protein (CP28) which was highly immunogenic in infected sheep and which in addition made possible the serological differentiation between infected and B. melitensis Rev. 1 vaccinated sheep. Monoclonal antibodies against CP28 were used to screen a B. melitensis 16M genomic library and to clone the corresponding gene. DNA sequencing of the gene encoding CP28 of B. melitensis 16M revealed that it was nearly identical to that of the recently published bp26 gene of Brucella abortus vaccine strain S19 coding for a periplasmic protein. The differences between the B. melitensis 16M gene and that of B. abortus S19 consisted of single nucleotide substitutions, one or two codon deletions, one codon addition, and most importantly a 21-bp deletion. The corresponding region of B. abortus S19 contains two 10-bp direct repeats which could have been involved in the genesis of the deletion. Expression of the B. melitensis 16M bp26 gene in Escherichia coli studied by the use of the monoclonal antibodies showed the same characteristics as reported for the B. abortus S19 bp26 gene, i.e. the presence of a higher molecular mass preprotein and a lower molecular mass band which probably corresponds to the mature protein exported to the periplasm. Immunoblotting performed with sera from either naturally infected or B. melitensis H38 experimentally infected sheep confirmed the importance of the B. melitensis CP28/BP26 protein as diagnostic antigen.

The nucleotide sequences encoding the major 25-kDa outer membrane protein (OMP) (omp25 genes) of Brucella ovis 63/290, Brucella melitensis 16M, Brucella suis 1330, Brucella canis RM6/66, and Brucella neotomae 5K33 (all reference strains) were determined and compared with that of Brucella abortus 544 (P. de Wergifosse, P. Lintermans, J. N. Limet, and A. Cloeckaert, J. Bacteriol. 177:1911-1914, 1995). The major difference found was between the omp25 gene of B. ovis and those of the other Brucella species; the B. ovis gene had a 36-bp deletion located at the 3' end of the gene. The corresponding regions of other Brucella species contain two 8-bp direct repeats and two 4-bp inverted repeats, which could have been involved in the genesis of the deletion. The mechanism responsible for the genesis of the deletion appears to be related to the "slipped mispairing" mechanism described in the literature. Expression of the 25-kDa outer membrane protein (Omp25) in Brucella spp. or expression from the cloned omp25 gene in Escherichia coli cells was studied with a panel of anti-Omp25 monoclonal antibodies (MAbs). As shown by enzyme-linked immunosorbent assay (ELISA) and immunoelectron microscopy, Omp25 was exported to the outer membrane in E. coli expressing either the truncated omp25 gene of B. ovis or the entire omp25 genes of the other Brucella species. Size and antigenic shifts due to the 36-bp deletion were demonstrated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunoblotting and by the differences in binding patterns in ELISA of the anti-Omp25 MAbs at the cell surface of E. coli cells harboring the appropriate gene and of cells of B. ovis and other Brucella species. In particular, MAbs directed against discontinuous epitopes of the entire Omp25 showed the absence of, or a significant reduction in, antibody reactivity with the B. ovis truncated Omp25. The results indicated that, as defined by the MAbs, exported Omp25 probably presents similar topologies in the outer membranes of E. coli and Brucella spp. and that the short deletion found in the omp25 gene of B. ovis has important consequences for the expression of surface B-cell epitopes which should be considered for the development of vaccines against B. ovis infection.

The problem of differentiating sheep infected with Brucella melitensis from those vaccinated or exposed to cross-reaching organisms has not been resolved by conventional serological tests or through the use of the smooth lipopolysaccharide in primary binding assays. We therefore analyzed sera from ewes experimentally infected with B. melitensis H38, from ewes naturally infected with B. melitensis, and from B. melitensis Rev.1-vaccinated ewes by enzyme-linked immunosorbent assay with three antigenic fractions: O polysaccharide, a cytosoluble protein extract (CPE) from the rough strain B. melitensis B115, and a partially purified cytosoluble protein of 28 kDa (CP28) from the CPE. Immunoglobulin G anti-O polysaccharide and anti-CPE responses were detected in all groups of animals tested (Rev.1 vaccinated and B. melitensis infected). However, false-positive reactions with CPE occurred with sera from Brucella-free ewes. The use of partially purified CP28 abolished these false-positive reactions. Furthermore, no immunoglobulin G antibodies against CP28 were detected in sera from vaccinated ewes, whereas 80% (8 of 10) of ewes experimentally infected with B. melitensis H38 and 89% (25 of 28) of naturally infected ewes showed various degrees of anti-CP28 reactivity (absorbance values of between 0.5 and 2.5). The results obtained with CP28 showed the potential usefulness of this antigen to permit the detection of B. melitensis-infected ewes and their differentiation from B. melitensis Rev.1-vaccinated ones.

The major 25-kDa outer membrane protein (Omp25) of Brucella abortus was purified and antigenically characterized by use of monoclonal antibodies (mAbs). Purification was achieved from the sodium dodecyl sulphate-insoluble (SDS-I) cell wall (CW) fraction of vaccine strain B. abortus B19 which was shown by use of mAbs to contain the two major outer membrane proteins of 25 and 36 kDa linked to peptidoglycan, smooth lipopolysaccharide (S-LPS), and rough LPS (R-LPS). Purity of Omp25 was checked with a number of mAbs directed to the different components of the SDS-I fraction. In ELISA, five anti-Omp25 mAbs, which showed significant binding to B. abortus whole cells and which are probably directed to conformational epitopes well-exposed on the bacterial surface, reacted poorly or not at all with the purified Omp25. Addition of R-LPS to purified Omp25 restored the binding capacity of these mAbs, which suggested that R-LPS may play an important role in reconstitution and exposure of conformational epitopes of Omp25. Immunoelectron microscopy showed that Omp25 was inserted into the R-LPS vesicles. Four of these anti-Omp25 mAbs probably recognize the same or closely located epitopes on Omp25, since one of the mAbs conjugated to peroxidase was inhibited in its binding in ELISA by the three others. Other anti-Omp25 mAbs showed strong binding to purified denatured Omp25 and their binding capacity was not affected by the addition of R-LPS to the purified Omp25. Thus, these results confirmed, as defined by the mAbs, the presence of both sequential and at least one conformational epitope on Omp25.

The Brucella melitensis dnaK gene was amplified by the polymerase chain reaction using primers chosen according to the published sequence of B. ovis and cloned in multiple copy plasmids enabling expression under the control of the Plac promoter. Monoclonal antibodies (mAb) obtained by immunizing mice with B. melitensis B115 cell wall (CW) fraction or by infecting mice with virulent B. melitensis strain H38 and recognizing a 73-kDa band in immunoblotting of the B. melitensis CW fraction reacted with the cloned dnaK gene product and were thus shown to be specific for the heat shock protein DnaK. The anti-Dnak protein mAbs did not react with Escherichia coli control cells or cell lysates and could therefore be specific to Brucella DnaK protein epitopes. These mAbs were further used to study overproduction of the DnaK protein. B. melitensis DnaK overproduction in E. coli resulted in a defect in cell septation and formation of cell filaments. Immunogold labelling with the mAbs and electron microscopy localized the DnaK protein inside as well as outside the E. coli cells, probably resulting from lysis due to toxicity of the overproduced DnaK protein. These results indicated that overproduction of the B. melitensis DnaK protein in E. coli had similar physiological consequences as that of E. coli overproduced in E. coli. The DnaK protein localization in B. melitensis cells was essentially cytoplasmic, as shown by immunoelectron microscopy. Heat shock treatment of these cells resulted in increased binding of mAbs and labelling in the cytoplasm. However, in subcellular fractions the DnaK protein was predominantly found in the cell envelope fraction of B. melitensis, which could perhaps be due to interaction of the DnaK protein with membrane proteins.

Brucella ovis, a naturally virulent rough Brucella species, is the aetiological agent of ram epididymitis. The identification of protective antigens is necessary to obtain a safe, specific subcellular vaccine. Monoclonal antibodies (MAbs) directed at both brucella outer-membrane proteins (OMPs) and rough lipopolysaccharide (R-LPS) in a mouse protection test were used to identify potential targets for humoral immunity. Mixtures of MAbs directed at the 16.5-, 25-27-, 31-34- and 36-38-kDa OMPs conferred significant protection 7 days after challenge with reference strain B. ovis 63/290 compared with controls receiving either saline or an anti-brucella O-polysaccharide MAb. Furthermore, an anti-R-LPS MAb tested alone conferred protection at a level comparable with that obtained with the mixture of anti-OMP MAbs. The combination of protective OMP MAbs with the anti-R-LPS MAb was also strongly protective. One combination of OMP MAbs, which bound intensely to B. ovis in vitro, was ineffective. These results indicate that B. ovis OMPs and R-LPS are targets for protective antibodies and that they can be regarded as candidates for ram epididymitis subcellular vaccines.

Seven surface-exposed outer membrane proteins (OMPs) in Brucella supp. have been previously described (A. Cloeckaert, P. de Wergifosse, G. Dubray, and J. N. Limet, Infect. Immun. 58:3980-3987, 1990). OMPs were shown to be more accessible to monoclonal antibodies (MAbs) on rough (R) Brucella melitensis and B. abortus strains than to MAbs on their smooth (S) counterparts. In this work, we have extended this study to representatives of the main Brucella species, using MAbs specific for OMPs and S and R lipopolysaccharides (S-LPS and R-LPS). Enzyme-linked immunosorbent assay (ELISA), flow cytometry, and immunoelectron microscopy showed important differences between strains in the binding of OMP- and R-LPS-specific MAbs which were in part related to the particular expression of S-LPS, irrespective of the species. Results indicated that both the amount and the length of O polysaccharide on S-LPS greatly influenced the accessibility of OMP and R-LPS epitopes to MAbs. S-R B. melitensis EP and S B. suis 40, for instance, which express O-polysaccharide chains in small amounts and with short mean length, respectively, bound a greater number of OMP- and R-LPS-specific MAbs than the other S Brucella strains. The major 31- to 34-kDa OMP was the most exposed OMP on S strains of B. melitensis and B. suis. In most cases, flow cytometry results agreed with those of ELISA and supplied additional data, such as the homogeneity or heterogeneity of OMP expression at the strain level. However, there were some discordances between flow cytometry and ELISA results concerning the surface exposure of the 25- to 27-kDa and 31- to 34-kDa OMPs on S strains and that of minor OMPs in vaccine strain B. melitensis Rev.1. Immunoelectron microscopy confirmed the poor accessibility of OMPs to MAbs on the surface of S Brucella strains. The naturally R pathogenic species B. ovis and B. canis bound the majority of OMP-specific MAbs as well as the R-LPS-specific MAbs. Therefore, the conserved OMP and R-LPS epitopes could play a role as targets of protective antibody-mediated immunity in infections caused by naturally R B. ovis and B. canis.

Distinction between Brucella melitensis infected and vaccinated sheep is needed to fully achieve ovine brucellosis eradication in several countries. For this purpose, we probed immunoblots of cytosoluble protein extract (CPE) of the rough (R) B. melitensis strain B115 with sera of Brucella-free, naturally infected, B. melitensis H38 experimentally infected and B. melitensis Rev.1 vaccinated sheep to identify immunogenic Brucella cytosoluble proteins which may lead to the development of more useful diagnostic tests and which may eventually differentiate vaccinated sheep from infected sheep. Brucella-free sheep sera showed IgM antibody reactivity to protein bands between 19 and 92 kDa. The use of conjugate specific for sheep IgG avoided non specific reactivity to all these bands except to the 39 and 50 kDa bands which were still detected in some Brucella-free sheep sera. In sera of B. melitensis H38 experimentally infected sheep, a specific IgG antibody response was observed against proteins of molecular masses of 19, 24, 28, 32, 39, 50 and 54 kDa. These proteins were also variably detected by IgG of sera from naturally infected sheep. Other proteins of 10, 12, 23, 36, 38, 42, 46, 68, 80 and 92 kDa were also detected by the latter sera. In sera from B. melitensis Rev.1 vaccinated sheep, an IgG antibody response was only observed against proteins with molecular masses of 39 and 50 kDa. These results suggest that the 19, 24, 28, 32 and 54 kDa proteins (the first recognized after experimental infection and that provoked a consistent humoral response during natural infection) could be interesting to develop serological tests for differentiating B. melitensis infection from B. melitensis Rev.1 vaccination.

The 40 N-terminal amino acids of the 20 kDa antigen A2 from Brucella melitensis were sequenced and showed important similarities with 4 bacterioferrins. A monoclonal antibody raised against this antigen cross-reacted with Escherichia coli bacterioferritin. Hybridization of two sets of degenerate primers with B. melitensis HindIII-digested genomic DNA identified a 3.8 kb fragment. This fragment was shown to contain a bacterioferritin gene (bfr) encoding a 161-amino acid protein. The sequence of the Brucella bacterioferritin is 69% similar to that of E. coli, and many of the ferroxidase centre and haem-ligation residues are conserved.

Changes in Brucella cell envelope protein profiles were investigated with batch cultures of B. melitensis strain 16M in a 2-litre fermenter. Analysis of expression of outer membrane proteins (OMP) (apparent molecular masses of 10, 16.5, 19, 25-27, 31-34, 36-38 and 89 kDa) and heat-shock protein DnaK (73 kDa) was performed with monoclonal antibodies (mAb) and immunoblotting techniques. Synthesis of the 89-kDa OMP and the heat-shock protein DnaK was invariant during B. melitensis growth. Expression of the 10-, 19- and 36-38-kDa minor OMPs was never detected. Variations in profiles of some OMPs, i.e. 25-27-kDa and 31-34-kDa major proteins and 16.5-kDa minor protein, occurred during growth stages, principally at the end of the exponential growth phase. These variations consisted of shifts in apparent molecular masses for the 25-27-kDa and 31-34-kDa OMPs and of peptidoglycan association for the 16.5-kDa OMP. Therefore, whereas the strong association of major OMPs with peptidoglycan was confirmed, results suggested that the 16.5-kDa minor OMP is also a peptidoglycan-associated protein.

Cell envelope fraction (CEF) of Brucella melitensis B115 was used to investigate antibody responses of B. melitensis naturally and strain H38 experimentally infected sheep by immunoblotting, indirect enzyme-linked immunosorbent assay (ELISA) (I-ELISA), and competitive ELISA (C-ELISA) with monoclonal antibodies (MAbs). MAbs used were directed to outer membrane proteins with molecular masses of 10, 16.5, 19, 25 to 27, 31 to 34, 36 to 38, and 89 kDa; to the heat shock protein DnaK, to O-polysaccharide (O-PS) common (C) and M epitopes; and to rough lipopolysaccharide (R-LPS) epitopes. In immunoblotting, all infected sheep sera tested recognized a large number of protein bands, including the above-cited proteins and other proteins for which MAbs have not been defined. The antibody response pattern was different from one animal to another, even within the experimentally infected sheep which were infected under the same experimental conditions. A number of protein bands were recognized by the sheep sera prior to experimental infection and by other uninfected sheep sera. The antibody reactivity to these antigens and others might explain the nonspecific antibody reactivity of sera in I-ELISA with CEF. C-ELISA confirmed also the individual variability of the antibody responses of infected sheep to protein antigens. Antibody responses to O-PS C and M epitopes were detected in all experimentally infected sheep and in half of the naturally infected sheep, but these responses were also heterogeneous in relation to their intensities. Antibody responses to R-LPS epitopes detected by use of C-ELISA with the anti-R-LPS MAbs were low or negative in most of the infected animals. Despite antibody response heterogeneity for CEF antigens, immunoblot and C-ELISA results indicated that, among the CEF antigens, the O-PS epitopes (C and M epitopes) and epitopes of the major 25- to 27- and 31- to 34-kDa outer membrane proteins seem to be the most promising for detecting B. melitensis infection in sheep.

Sera from Brucella infected and B. melitensis Rev1 vaccinated sheep were analysed by immunoblotting using the cell envelope fraction (CEF) of B. melitensis B115. The CEF of B. melitensis B115 was analysed using a bank of monoclonal antibodies. The fraction consisted mainly of S-LPS like molecules, R-LPS and outer membrane proteins (OMPs) of molecular masses of 10, 16.5, 19, 25-27, 31-34, 36-38, 73 and 89 kDa. Immunoblot analysis indicates that the antibody response in infected sheep was mainly directed against the major OMPs of 25-27, 31-34, 36-38 kDa, against 55 to 62, 70-73 and 89 to 94 kDa proteins associated with the CEF and, against S-LPS like molecules. Some infected sheep reacted with antigens of molecular mass lower than 20 kDa. Sera from vaccinated sheep reacted only with OMPs of 36-38, 60, 70-73 and 89 kDa. The major 25-27 and 31-34 kDa OMPs and proteins below 20 kDa were only detected by the sera of infected sheep. These differences may be due to the persistence of the field infection also reflected by the fact that antibody response against O-polysaccharide (O-PS), as measured by enzyme-linked immunosorbent assay (ELISA), was more intense in infected sheep than in vaccinated ones. These results also indicate that these OMPs could be useful to differentiate B. melitensis infection from B. melitensis Rev.1 vaccination in sheep.

Twenty-two monoclonal antibodies (mAbs) specific for smooth lipopolysaccharide (S-LPS) were generated by fusion of spleen cells from mice infected with the rough Brucella melitensis strain B115 with the NSO myeloma. According to reactivity in enzyme-linked immunosorbent assay (ELISA) with O-polysaccharide (O-PS) and absence of reactivity with rough lipopolysaccharide (R-LPS), it was postulated that these mAbs recognized epitopes present on the O-PS. Most of the mAbs reacted equally well in ELISA and immunoblotting with S-LPS types of Brucella A and M dominant strains and were designated as specific for common (C) epitopes. Three mAbs were highly specific for M dominant S-LPS. All these mAbs, in contrast to a mAb specific for the A epitope, showed little or no cross-reactivity with Yersinia enterocolitica O:9 S-LPS. S-LPS of B. melitensis B115 was extracted and analysed by immunoblotting and ELISA with mAbs specific for A, M and C epitopes. Reactivity of the mAbs with this S-LPS was compared to reactivity with S-LPS of A and M dominant smooth Brucella strains. The results suggest that S-LPS of B. melitensis B115 bears mainly C epitopes and a few M epitopes. The very weak reactivity of this S-LPS with the mAb specific for the A epitope and the fact that the mAbs specific for C and M epitopes showed little or no cross-reactivity with Y. enterocolitica O:9 S-LPS suggest that O-PS from this rough strain could be used to distinguish Y. enterocolitica O:9 infection from Brucella infection.(ABSTRACT TRUNCATED AT 250 WORDS)

An internal protein was purified from cell extracts of Brucella melitensis B115 by a combination of preparative isoelectric focusing and high-performance size exclusion chromatography. The protein has an apparent molecular mass of 230 kDa as determined by size exclusion chromatography. The protein was resolved to a single band of 20 kDa after sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The native protein had an isoelectric point of 4.9. The N-terminal sequence of the 20-kDa protein was determined. The 20-kDa protein has been identified as antigen A-2 with a previously described anti-antigen A-2 serum (B. Stemshorn, K. Nielsen, and B. Samagh, Can. J. Comp. Med. 45:77-81, 1981). Antigen A-2 reacted with sera from infected sheep in immunoblotting and may be useful in developing diagnostic tests for brucellosis.

A monoclonal antibody (3D6) was produced which reacted only with Brucella sonicated cell extracts that had been lysozyme-treated after sonication. The monoclonal antibody (mAb) reacted with the three major outer-membrane proteins (OMPs) of B. melitensis B115 in Western blots. A large number of reactive bands ranging from 12 to 43 kDa were present in lysozyme-treated Escherichia coli and Yersinia enterocolitica sonicated cell extracts. In a latex agglutination inhibition immunoassay, mAb 3D6 showed better reactivity with purified peptidoglycan (PG) of B. melitensis B115 than with that of Escherichia coli. This mAb was also used in immunogold electron microscopy with whole Brucella cells and sections. No binding was observed on whole cells and immunogold labelling in sections was observed close to the outer membrane, in the periplasmic space and in the cytoplasm. These findings indicate that mAb 3D6 is specific for PG subunits. Immunoblot analysis of B. melitensis B115 rough sonicated cell extracts after SDS-PAGE, with or without lysozyme treatment, was performed using mAbs specific for Brucella OMPs of molecular masses of 10, 16.5, 19, 25-27, 31-34, 36-38 and 89 kDa, for PG and for rough lipopolysaccharide (R-LPS) and smooth lipopolysaccharide (S-LPS). mAbs specific for the 25-27, 31-34 and 36-38 kDa OMPs reacted with three to six bands. All of them except the band of lowest molecular mass reacted with the PG-specific mAb and not with R-LPS- and S-LPS-specific mAbs.(ABSTRACT TRUNCATED AT 250 WORDS)

Spleen cells from mice infected with the rough Brucella melitensis strain B115 were fused with NSO myeloma cells. Hybridoma supernatants were screened in ELISA with cell walls (CW), sonicated cell extracts (CE) and rough lipopolysaccharide (R-LPS) of B. melitensis strain B115 and whole B. melitensis B115 cells. Surprisingly, 22 monoclonal antibodies (mAbs) reacting in ELISA with both CW and CE but not with R-LPS and bacterial cells were shown by immunoblot analysis and ELISA to react with smooth lipopolysaccharide (S-LPS). These mAbs also reacted in ELISA with O polysaccharides (OPS) from the smooth Brucella abortus strain 99 and the smooth B. melitensis strain 16M and thus recognize epitopes present on the O-chain. Proteinase K LPS preparations from B. melitensis B115 analysed by immunoblotting with one mAb (12G12) recognizing S-LPS of both A and M specificity displayed the typical S-LPS high-molecular-mass ladder pattern but no S-LPS was detected in the phenol/water/chloroform/light petroleum LPS preparation of the same strain. mAb 12G12, specific for S-LPS, and a mAb (A68/03F03/D05) specific for R-LPS were used to localize the O-chain and R-LPS expressed in B. melitensis strain B115 by immunoelectron microscopy. Immunogold labelling was observed at the surface of B. melitensis B115 cells with the anti-R-LPS mAb but not with the anti-S-LPS mAb. In ultrathin sections, immunogold labelling with the S-LPS specific mAb was observed in the cytoplasm and in the periphery of the cytoplasm, probably at the cytoplasmic membrane.(ABSTRACT TRUNCATED AT 250 WORDS)

The protein X of Steptococcus agalactiae is a surface antigen included in the typing scheme of group B streptococci (GBS). We have developed a monoclonal antibody to the protein X and used it to purify this antigen by affinity chromatography. Electrophoresis in polyacrylamide, and immunoblotting using the monoclonal antibody or a rabbit antiserum raised with the affinity purified protein X, revealed a major band in the region of 200 kDa and a smaller one at 100 kDa. The isolated protein X will make possible investigations of its potential role in virulence and protection.

Analysis of the immune response towards the facultative intracellular bacterium, Brucella melitensis, was studied by immunoblotting after either isoelectric focusing (IEF) or sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). A cytoplasmic extract (CPE) of Brucella melitensis was used as antigen to analyse the response in 17 sera from naturally infected goats. CPE analysed by IEF exhibited 25 proteins within the pH range of 4.35 to 6. Immunoblotting revealed most of the stained bands around pH 4.5-5.4. CPE analysed by SDS-PAGE showed more than 20 silver stained proteins in the molecular range of 16-18 kDa to 70 kDa but immunoblotting revealed only 1 to 6 bands according to the sera tested. Because proteins are preserved in their native state with IEF, in contrast to SDS-PAGE treatment, this technique may be best suited for analysis of the overall response to natural infection.

Purification of the Brucella polysaccharide referred to as native hapten (NH) and extracted from cells by the autoclaving procedure, was accomplished by ultrafiltration, followed by repetitive gel filtration using high-performance liquid chromatography on a "TSK-G2000-SW" column. The purified NH was analysed by SDS-PAGE, gas-liquid chromatography mass spectroscopy, and 13C and 1H NMR spectroscopy. NH from B. abortus B19 (NH-A) was shown to have a structure identical to that of A polysaccharide from B. abortus 1119-3, a linear homopolymer of alpha-1,2-linked-4,6-dideoxy-4-formamido-D-mannopyrannosyl residues. The structure of the NH from B. melitensis 16M (NH-M) was identified as a linear homopolysaccharide of the same sugar but composed of a pentasaccharide repeating unit in which four alpha 1,2-linked-4,6-dideoxy-4-formamido-D-mannopyrannosyl residues are linked alpha-1,3 to the last monosaccharide of the sequence. This structure is similar to that determined for the Brucella M polysaccharide from B. melitensis 16M. The discovery in highly purified NH preparations of covalently bound monosaccharides characteristic of lipopolysaccharide inner core regions e.g., quinovosamine, mannose and 3-deoxy-D-manno-octulosonate (KDO), indicates that this polysaccharide is derived from lipopolysaccharides (LPS) by hydrolytic conditions fortuitously generated during the extraction protocol. The antigenically important polysaccharides of Brucella are now established to be either A or M antigens. Polysaccharide B is a cyclic glucan with no structural or serological relationship to A or M polysaccharides, its apparent activity in diagnostic tests of infected cattle results from O polysaccharide contamination. This artefact, previously referred to as NH, results from LPS hydrolysis under the extraction conditions used to prepare polysaccharide B.

An effective method was developed for the preparation of Brucella abortus native hapten by use of an ultrafiltration system. The membranes PM30 and YM5 were used to separate the lipopolysaccharide and the native hapten. The yield of the native hapten was higher than that obtained by a more complex procedure reported previously. This method is economical for the large-scale preparation of B. abortus native hapten. The effects of ultrafiltration were evaluated by a quick and sensitive high-performance liquid chromatographic technique.