key: cord-0007660-8kwcrh7x
authors: Spencer, S.A.; Broughton, E.S.; Hamid, S.; Young, D.B.
title: Immunoblot studies in the differential diagnosis of porcine brucellosis: an immunodominant 62 kDa protein is related to the mycobacterial 65 kDa heat shock protein (HSP-65)
date: 2002-11-13
journal: Vet Microbiol
DOI: 10.1016/0378-1135(94)90085-x
sha: 5acb4a64dd7a05fef7a6a3bed1609f2301d949d0
doc_id: 7660
cord_uid: 8kwcrh7x

Smooth Brucella spp. share certain lipopolysaccharide antigens with other bacteria, resulting in serological cross-reactions which can prevent the definitive diagnosis of brucellosis. To identify other antigens with serodiagnostic potential, immunoblot studies following sodium dodecyl sulphate - polyacrylamide gel electrophoresis were carried out. Sera from pigs experimentally infected with Brucella suis and naturally infected feral pigs, sera from pigs from a farm with a known history of Yersinia enterocolitica 0:9 infection, Brucella Complement Fixation Test (CFT) reactor pigs (aetiology unknown) and pigs from consistently Brucella CFT negative farms were examined. Although B. suis infected pigs recognized a total of nine B. melitensis antigens, individual pigs rarely recognized more than three antigens in the range. A 62 kDa antigen was recognized by the majority (73%) of the Brucella infected pigs, but only by 10 to 23% of pigs from the other groups. This antigen was shown to be the Brucella homologue of the ubiquitous 65 kDa heat shock protein (HSP-65) family by immunoblot studies with 14 monoclonal antibodies to the Mycobacterium leprae HSP-65. Only four of these monoclones (Y1.2, ML-30, D7C and IIIC8) identified the B. melitensis 62 kDa protein suggesting that unshared, potentially Brucella specific, regions exist. Sera from Y. enterocolitica 0:9 infected pigs, CFT reactor pigs (aetiology unknown), CFT negative pigs and hyperimmune pig serum raised to Y. enterocolitica 0:9 also recognized B. melitensis antigens, most notably a 17 kDa protein. This antigen appears to be a common cross-reactive protein.

Since 1988 there has been a marked increase in the number of pigs failing export tests for brucellosis in Great Britain. Wrathall et al. (1993) confirmed the Brucella free status of British pigs by showing that serological reactions observed in these tests were due not to Brucella infection but to infection with Yersinia enterocolitica serotype 0:9. The serological cross-reaction between Yersinia enterocolitica 0:9 and the Brucella A antigen is well known (Ahvonen et al., 1969; Corbel and Cullen, 1970; Mittal and Tizard, 1979a) and Corbel et al. (1984) showed that infection with Y. enterocolitica 0:9 can produce serological reactions to Brucella. Furthermore, studies with monoclonal antibodies have shown that the O-chain of the lipopolysaccharide (LPS) of these organisms is identical (Bundle et al., 1984; Palmer and Douglas, 1989) . This antigen has been characterized as a polymer of 4,6-dideoxy-4formamido-D-mannopyranosyl residues linked by a 1-2 bonds ( Bundle et al., 1989) . Since Brucella LPS forms the basis for the complement fixation (CFT) and serum agglutination tests (SAT) currently used for the diagnosis of brucellosis, pigs infected with Y. enterocolitica 0:9 will tend to fail these tests. To differentiate between these infections Corbel and Cullen (1970) and Mittal and Tizard (1979b) developed methods to quantify antibodies to the flagellar antigens of Y. enterocolitica 0:9, whilst Schoenerer et al. (1990) developed an enzyme linked immunoassay, using Yersinia plasmid encoded outer membrane proteins, to detect Yersinia infection. The former approach involves laborious absorption procedures, but most importantly neither test specifically detects Brucella infection nor excludes the possibility of dual infection. In addition, cross-reactions caused by other organisms (Corbel et al., 1984) may be wrongly attributed to brucellosis.

It is important, therefore, to identify antigens that demonstrate greater specificity for the definitive diagnosis of brucellosis. Previous immunoblot and immunoprecipitation studies with sera from B. ovis infected rams (Chin and Pang-Turner, 1990 ) and dogs infected with B. canis (Carmichael et al., 1989) have identified antigens in cytoplasmic extracts of Brucella which may be of diagnostic potential. In this study, sera from pigs infected with B. suis or Y. enterocolitica 0:9 and control animals were examined by immunoblot analysis after sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) of whole B. melitensis, to characterize the range of antigens recognized by the humoral response in porcine brucellosis. Lysozyme digests of digestion buffer insoluble material were also examined as Sowa et al. ( 1991 ) have shown the presence of additional group 2 and 3 outer membrane proteins (OMP) in such extracts. A rough strain of Brucella melitensis was used as immunoblot antigen to prevent smearing profiles due to lipopolysaccharide. The use of this antigen is discussed later.

Fifteen sera from pigs experimentally infected with B. suis, and from culture positive B. suis infected feral pigs were received from the Diagnostic Bacteriology Laboratory of the National Veterinary Services Laboratories (Ames, USA). Organs were cultured for B. suis as described by Payeur et al. (1990) .

Eighteen sera were examined from pigs from a farm with a known history of Y. enterocolitica 0:9 infection. Sera from reactor pigs exhibiting serological reactions to Brucella (aetiology unknown) were obtained from three geographically distinct farms. Four individual sera from one of these farms, with titres of < 8, 17, 40 and 106 international CFT units (ICFTU) were examined. Three pools were prepared from each of the remaining reactor farms by combining equal volumes of serum from at least eight pigs with similar CFT titres ( < 8, (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) .

The Brucella negative group comprised serum from five feral pigs negative by CFT and culture (received from the Diagnostic Bacteriology Laboratory, NVSL), and four British farms which were consistently CFI" negative. Two or three individual sera from each of three farms, and a pool comprising equal volumes of 100 sera from the fourth farm were tested.

Hyperimmune pig sera were prepared using dense suspensions of either heat killed (60°C for 1 h) B. melitensis (88/131) or formalin inactivated (0.5% v/v) Y. enterocolitica 0:9 (M90001, our own isolate) emulsified in Freunds incomplete adjuvant (Sigma). The antigens were administered twice, intramuscularly, at two week intervals and serum taken three weeks after the second injection.

The CFT was performed as described by Morgan et al. (1978) with reference to the 3rd British National Standard serum. The results are expressed as International CFT Units (ICFTU). The pass level for international trade is < 20 ICFTU per ml (Office International des Epizooties, 1986).

A rough strain of Brucella melitensis (88/131 ) from the Central Veterinary Laboratory (CVL) Brucella culture collection was grown on serum dextrose agar (SDA) at 37°C in air for five days. Pure growth was harvested as a dense suspension in electrophoresis digestion buffer consisting of stacking gel buffer ( see below) containing glycerol (20% v / v), 2-mercaptoethanol (2-ME, 2% v/v) and sodium dodecyl sulphate (SDS, 4% w/v). Suspensions were heated at 100°C for 10 rain, cooled and centrifuged at 11600 g for 10 min. Supernatants were stored at 4°C. The deposit was digested with lysozyme as described by Sowa et al. ( 1991 ) . Briefly, the pellet was washed twice in Lutkenhaus buffer ( 10 mM Tris/HC1, 5 mM EDTA and 1 mM 2-ME, pH 7.8) and resuspended in buffer containing egg white lysozyme (grade II, Sigma) at a ratio of one part enzyme to 10 parts detergent soluble protein. After incubation at 37°C for 17 h, two volumes of electrophoresis digestion buffer were added, the mixture was heated at 100°C for 10 min and the supernatant harvested as described above.

Electrophoresis was carried out by the method of Laemmli (1970) in a vertical slab gel system (Hoefer SE600) with 1.5 mm thick, 18 cm X 16 cm gels using a trough (blank) comb. The separation gel contained 10% or 15% acrylamide in 375 mM Tris/HCl buffer, pH 8.8, with a 5% stacking gel in 125 mM Tris/HC1 buffer, pH 6.8. The tank buffer comprised 250 mM Tris, 192 mM glycine, pH 8.3 containing 0.1% SDS (w/v). Separation was carried out at 10 mA per gel constant current for 16 h at 10°C. Rainbow molecular weight markers (Amersham) covering the range 200 to 14.2 kDa were used.

Following electrophoresis, gels were equilibrated in transfer buffer (25 mM Tris, 110 mM glycine buffer, pH 8.6 and 20% v/v methanol) for 1 h at room temperature (RT). Separated antigens were transferred to Immobilon-P (Millipore) at 100 mA per gel constant current for 22 h, then 200 mA per gel for 1 h (LKB Transphor 2005). The membranes were stored dry between filter paper at RT. For use membranes were cut into 1 cm wide strips, wetted in neat methanol and blocked for 1 h at 37°C in 3 % (w / v) skimmed milk in phosphate buffered saline-Tween (PBS-T, PBS pH 7.2 with 0.02% v/v Tween 20). After washing the strips in PB S-T (3 X 5 min), serum (diluted 1 / 100 in PBS-T with 1% skimmed milk) was added and incubated for 1 h at 37°C. The strips were then washed (3 X 5 min) and incubated at 37°C with horseradish peroxidase (HRP) conjugated rabbit anti-swine IgG (Nordic, 1/1000 in PBS-T with 1% w/v skimmed milk). After washing (6 x 5 min), 4chloro-1-napthol substrate was added (stock solution of 30 g/1 in methanol, diluted 1/100 in 50 mM Tris/HC1 pH 7.6, and 100 /xl 30 vol. H202 added just prior to use). After incubation at 37°C for 10 rain the reaction was stopped by washing with distilled water. For monoclonal antibodies the primary antibody was incubated at RT for 90 min or overnight, and the secondary antibody was HRP conjugated goat anti-mouse IgG (Bio-Rad or Sigma, diluted 1 / 1000).

Stock solutions (40 u/ml) of pronase E (protease XXV, Sigma) and proteinase K (protease XXVIII, Sigma) were prepared in cold (4°C) PBS pH 7.6 and stored at 4°C. Just prior to use an aliquot of pronase E stock solution was diluted 1 / I0 in PBS pH 7.6 at 37°C and 5 ml added to each strip of blotted antigen. A similar volume of warmed buffer was added to control strips. After incubation at 37°C for 2 h with shaking, all strips were washed (3 X 5 rain) in buffer at 37°C. This procedure was repeated with a 1 / 10 dilution of proteinase K. Control strips received 5 ml of warmed buffer. After the final wash the strips were immunostained as described above.

Mouse IgG monoclonal antibodies Y1.2, IIC8, IIIC8, IIIE9, CI.1, D7C, D5H, F67-2, F67-19, F67-13, F47-10, ML-30, IVD8 and IIH9, raised to the HSP-65 of Mycobacterium leprae, were gifts from Dr T.P. Gillis (LSU School of Veterinary Medicine, LA, USA) (IIC8, IIIC8, IIIE9, IVD8, IIH9), Dr A.H.J. Kolk (Royal Tropical Institute, Amsterdam, The Netherlands) (F67-2, F67-19, F67-13, F47-10), and Dr T.M. Shinnick (Centres for Disease Control, Atlanta, USA) (ML-30, originally cloned by Prof. J. Ivanyi, MRC Tuberculosis and Related Infections Unit, London, UK). They were received as tissue culture supernatants or ascites and were used at 1 / 100 or 1/2000 respectively.

Preliminary immunoblot studies using whole cell extracts of smooth B. suis showed smearing which obscured the banding profile (data not shown). The use of rough B. melitensis overcame this smearing to give clearer profiles.

Sera from the 15 B. suis infected pigs identified a total of nine antigens in the range 85 to 11 kDa, but no one antigen was recognized by all the sera ( Table 1 ). The 62 kDa antigen was the most frequently recognized with 11 of 15 sera (73%) showing weak ( _ ) to strong ( + + + ) immunostaining, strongest reactions generally accompanying increased CFF titre. Recognition of this antigen tended to agree with the CFT pass/fail threshold for international trade; ie sera with a CFT titre of > 20 ICFTU generally recognized the 62 kDa antigen. However, one serum (pig 91-2) with a high CFT titre (106 ICFFU) failed to recognize this antigen whilst another (pig 117-6) with a CFT titre of 17 ICFFU (ie acceptable for international trade) showed weak recognition. The immunodominance of the 62 kDa antigen for B. suis infected pigs is illustrated in Fig. 1 , lanes c, d, e and f. A 17 kDa antigen was the next most frequently identified. This antigen was recognized by 7 of the 15 sera (47%). Antigens of 85, 55, 40, 37, 28, 19 and 11 kDa were recognized infrequently and each of these antigens was never recognized by more than three sera. Similarly individual B. suis sera rarely recognized more than three antigens and showed little consistency in the profile exhibited.

Fourteen of the l 8 sera from pigs infected with Yersinia enterocolitica 0:9 also recognized B. melitensis antigens. Most notably a 17 kDa antigen was identified by 13 of the 18 sera (72%) ( Table 2) , three showing strong ( + + ) staining. Each of the 62, 28 and 19 kDa antigens were recognized strongly ( + +, + + + ) by two or more sera, whilst antigens of 85, 52, 40, 37, 21, 12.5, 12 and 11 kDa were stained less intensely and were rarely recognized by more than two sera. Yersinia enterocolitica 0:9 infected pigs showed no correlation between CFT titre and recognition of the 62 kDa antigen, since a CFT negative serum (pig 80) and an international trade fail (27 ICFTU) serum (pig 3) both recognized this antigen strongly ( + + + ). A range of cross-reactions shown by sera from pigs infected with Y. enterocolitica 0:9 is shown in Fig. 1 , lanes i, j, k and 1.

A summary of the results of the CFT reactor (aetiology unknown) and CFT negative groups are given in Table 3 together with the results from B. suis and Y. enterocolitica 0:9 infected pigs for comparison. Examples of these immunoblot profiles are shown in Fig. 1 qCFTU International complement fixation test units * % of sera recognizing band lanes m, n, and o (reactor) and g, h, p, q and r (negative). Complement fixation test reactor and negative groups identified fewer antigens than the B. suis or Y. enterocolitica 0:9 sera. The 17 kDa antigen was most consistently recognized by sera from the reactor (9 of 10, 90%) ( Fig. 1 lanes m, n and o) and negative (6 of 14, 64%) groups (Fig. 1, lane h) . The 52 and 28 kDa antigens were the next most frequently recognized, whilst the 62, 55, 37, 32, 21, 19 and 11 kDa antigens were recognized less frequently. In contrast to the results with B. suis sera, only one serum ( 105 ICFTU) from the reactor group ( ___ immunostaining) and three sera from the negative group ( + to + + + immunostaining) detected the 62 kDa antigen (data not shown). 

Pig hyperimmune serum raised to B. melitensis recognized a broad range of B. melitensis antigens well, most notably those of 85, 62, 55, 28, 21, 19, 17, 12.5 and 11 kDa ( Fig. 1  lane b) . Serum raised to Y. enterocolitica 0:9 recognized the 17 kDa Brucella antigen strongly and demonstrated weak recognition of a 58 kDa antigen ( Fig. 1 lane a) .

The nature of the immunodominant 62 kDa and cross-reactive 17 kDa B. melitensis antigens was evaluated by digestion with pronase E and proteinase K. The ability of pig sera to recognize these antigens was abrogated by protease digestion ( Fig. 2) , demonstrating that both antigens are proteins.

The size, proteinaceous nature, immunodominance and cross-reactivity of the 62 kDa antigen suggested that this was the Brucella homologue of the Mycobacterium leprae HSP-65. The presence of epitopes common to the immunodominant 62 kDa B. melitensis protein and the HSP-65 of M. leprae was demonstrated by reactivity with the anti-M, leprae monoclonals Y1.2, IIIC8, D7C and ML-30 (Fig. 3, lanes a, c, f and 1 respectively) . The remaining 10 monoclones failed to recognize the B. melitensis protein, demonstrating the presence of both common and unshared regions on the Brucella and Mycobacterial proteins. Sowa et al. ( 1991 ) identified additional group 2 and 3 Brucella OMP's by digesting the SDS-insoluble fraction of B. abortus with lysozyme. In the present study the two additional group 2 proteins of 35 and 37 kDa, and one further group 3 protein of 27 kDa recovered by lysozyme treatment were not identified by immunoblotting with sera from B. suis infected pigs (data not shown). These proteins therefore do not appear to be immunologically significant and were not studied further. 

The smearing profiles in immunoblots with smooth B. suis antigen were believed to be due to antibodies binding both specifically and non-specifically to smooth lipopolysaccharide (S-LPS) present in these extracts (Gamazo et al., 1989; Belzer et al., 1991; . This problem was resolved by using a rough strain of B. melitensis, a strategy previously used by others . This was adopted in view of the high DNA relatedness of all Brucella species (Verger et al., 1985 ) ; the antigenic similarity of proteins of the different species ofBrucella (Diaz et al., 1967; Berman et al., 1975; Verstreate et al., 1982; Santos et al., 1984; Holman et al., 1985; and Gamazo et al., 1989) ; and improved sensitivity of B. melitensis outer membrane proteins in the detection of human infection by all species of Brucella (Hunter et al., 1986) .

Pigs infected with B. suis generally recognized B. melitensis antigens poorly. In similar studies B. ovis infected sheep produced strong antibody responses to the 63, 55, 41-42, 38-40.5, 36-37, 32-33.5, 25-27, 19-20.5 and 15-18 kDa antigens (Chin and Pang-Turner, 1990) . Although a similar range of antigens was identified by B. suis infected pigs, individual sera rarely identified more than three antigens and, in contrast to the results obtained with infected sheep, there was little consistency in the profiles obtained. This relatively weak serological response in B. suis infected pigs parallels the results obtained in traditional serological tests where the detection of 80-90% of infected pigs is considered optimal (Deyoe, 1987) . For diagnostic use in individual animals a single protein antigen should detect brucellosis, specifically, in a high proportion of pigs. However, to differentiate brucellosis from infection with known cross-reacting organisms (Corbel, 1985) in countries with no B. suis infection a test of lower sensitivity but high specificity, ie a herd test, may be acceptable.

The antigen recognized most frequently by B. suis infected pigs (the 62 kDa protein) was identified by 73% of these animals (Table 1 ). This antigen is distinct from the outer membrane proteins: group 1 (88-94 kDa), group 2 (3840 kDa), group 3 (25-30 kDa) , and the 15 kDa or 8 kDa proteins identified by Dubray and Bezard (1980) and Gomez-Miguel and Moriyon (1986) . Previous studies have shown that serum from B. ovis infected sheep (Chin and Pang-Turner, 1990 ) and B. canis infected dogs (Carmichael et al., 1989) also recognize a significantly immunogenic antigen in the range 63-68 kDa. The experiment reported here with monoclonals to the M. leprae HSP-65 demonstrates that the 62 kDa protein is the Brucella homologue of the mycobacterial antigen. This family of proteins is immunodominant in infections due to Mycobacteria (Thole and Van der Zee, 1990) and Legionella (Sampson et al., 1986; Plikaytis et al., 1987) and the results reported here confirm its immunodominance in porcine brucellosis. In addition, this group of proteins is known to be cross-reactive, demonstrating sequence homology and serological cross-reactions with a wide range of prokaryotic and eukaryotic organisms (Hoiby, 1975; Shinnick et al., 1988; Thole et al., 1988; Waldinger et al., 1988; Sampson et al., 1990) . It is therefore not surprising that the immunodominant 62 kDa protein was also recognized by some sera from pigs infected with Y. enterocolitica 0:9, CFT reactor and CFT negative groups of pigs. Indeed, two Yersinia sera and one serum from the CFT negative group identified the protein strongly. This implies that antibodies are raised to conserved regions of this antigen during infection with other unidentified organisms, and confirms the cross-reactive nature of the HSP-62 in Brucella.

Other cross-reactions were also seen, most notably to antigens of 52, 37, 28, 19 and 17 kDa. The most extensive of these were to the 17 kDa Brucella protein. This antigen was frequently recognized by sera from all groups tested and was identified by hyperimmune pig sera raised to Y. enterocolitica 0:9 (Fig. 1 ) . The range of cross-reactions observed indicates that the use of undefined antigen preparations in diagnostic assays will not provide adequate specificity for the diagnosis of brucellosis.

In view of the known cross-reactive and immunodominant nature of the HSP-65 family of antigens it is surprising that antibodies to the 62 kDa antigen were not detected in hyperimmune pig serum raised to Y. enterocolitica 0:9. Yamaguchi et al. (1990) identified an HSP-65 homologue in Yersinia and demonstrated cross-reactivity with a wide range of organisms, although these workers did not examine Brucella. The absence of antibody to the 62 kDa Brucella protein could be due to growth of the Yersinia inocutum under controlled non-stress conditions. This explanation seems unlikely as HSP' s are produced constitutively and account for approximately 1% of the total protein in E. coli grown at 37°C (Thole and Van der Zee, 1990) . Alternatively, the epitopes recognized by the humoral response to the HSP-65 family during infection may be more specific than the results with cross-reactive monoclonals imply. This may be due to suppression of the response to broadly crossreacting epitopes (Cooper et al., 1984) . Furthermore, the observation that a greater proportion of Brucella sera recognized the 62 kDa antigen provides some evidence for greater specificity.

The reaction of four of the 14 M. leprae HSP-65 monoclones with the Brucella homologue confirms the existence of cross-reactive epitopes on the 62 kDa protein. However, the failure of 10 monoclones to react demonstrates that many epitopes are not shared and suggests that Brucella specific epitopes are present. Recent studies would appear to support this since the deduced amino acid sequence of the HSP-65 homologue of B. abortus (Roop et al., 1992) shows regions of variability when compared to other published HSP-65 homologue sequences. Indeed, studies with the mycobacterial HSP-65 have shown an M. leprae specific epitope in the most variable region (amino acids 420-480) (Anderson et al., 1988) , and work with the HSP-60 of Y. enterocolitica has demonstrated an epitope specific for this organism (Yamaguchi et al., 1990) . In addition studies with Legionella (Steinmetz et al., 1991 ) have demonstrated a genus specific epitope.

The observed cross-reactions and the presence of an HSP-65 homologue in Y. enterocolitica (Yamaguchi et al., 1990) precludes the use of the entire Brucella HSP-62 in a differential diagnostic role. To develop a specific test for brucellosis based on this immunodominant antigen the epitopes which confer specificity must be defined. Specificity may be introduced by the use of specific monoclonals in a competition assay. Indeed, Levis et al. (1986) developed a highly specific enzyme-linked immunoassay for lepromatous and borderline lepromatous leprosy patients based on competition with the monoclonal IVD8 for HSP-65. Unfortunately this test detected only 32% of these patients. The use of monoclonals to more immunodominant specific regions, or a panel of specific monoclonals could improve this sensitivity. Therefore, although the Brucella HSP-65 homologue carries crossreacting epitopes, its immunodominance and the potential existence of both genus and species specific epitopes show potential as a candidate for a differential serological test for brucellosis in pigs and other animals. It is essential that fragments of recombinant Brucella HSP-62 or synthetic peptides are evaluated to identify the B-cell epitopes recognized specifically by Brucella infected pigs.

Marked cross-agglutination between Brucella and a sub-type of Yersinia enterocolitica

Exact definition of species -specific and cross-reactive epitopes of the 65-kilodalton protein of Mycobacterium leprae using synthetic peptides

Differentiation by Western blotting of immune responses of cattle vaccinated with Brucella abortus strain 19 or infected experimentally or naturally with virulent Brucella abortus

Characterization of Brucella abortus soluble antigen employed in immunoassay

Characterization of antibody-reactive epitopes on the 65-kilodalton protein of Mycobacterium leprae

Serological confirmation ofBrucella abortus and Yersinia enterocolitica 0:9 0-antigen by monoclonal antibodies

Definition ofBrucella A and M epitopes by monoclonal typing reagents and synthetic oligosaccharides

Characterization of Brucella canis protein antigens and polypeptide antibody responses of infected dogs

Profiles of serological reactivity against cytosoluble antigens of Brucella ovis in experimentally infected rams

Antibody response to Brucella outer membrane proteins in bovine brucellosis: Immunoblot analysis and competitive enzyme-linked immunosorbent assay using monoclonal antibodies

IgA and IgM monoclonal antibodies specific for a common epitope shared by the Brucella A and M smooth lipopolysaccharide

Antibody response to myoglobins: effect of host species

Recent advances in the study of Brucella antigens and their serological cross-reactions

Differentiation of the serological response to Yersinia enterocolitica serotype IX and Brucella abortus in cattle

Observations on serological cross-reactions between smooth Brucella species and organisms of other genera

Brucellosis

Antigenic relationship of Brucella ovis and Brucella melitensis

Isolation of three Brucella abortus cell-wall antigens protective in murine experimental brucellosis

Comparative analysis of proteins extracted by hot saline or released spontaneously into outer membrane blebs from field strains of BruceUa ovis and Brucella melitensis

Demonstration of a peptidoglycan-linked lipoprotein and characterization of its trypsin fragment in the outer membrane of Brucella spp

Cross-reactions between Pseudomonas aeruginosa and thirty-six other bacterial species

Development of monoclonal antibodies to Brucella cell surface antigens

Enzyme-linked immmunosorbent assay with major outer membrane proteins of Brucella melitensis to measure immune response to Brucella species

Cleavage of structural proteins during the assembly of the head of bacteriophage T4

Serodiagnosis of leprosy: relationships between antibodies to Mycobacterium leprae phenolic glycolipid 1 and protein antigens

Experimental studies on the serological relationships between Yersinia enterocolitica and Brucella abortus

A simple technique to differentiate between animals infected with Yersinia enterocolitica IX and those infected with Brucella abortus

Brucellosis Diagnosis -Standard Laboratory Techniques. 2nd edn. London. Ministry of Agriculture, Fisheries and Food

International Animal Health Code, 5th edn

Analysis of Brucella lipopolysaccharide with specific and cross-reacting monoclonal antibodies

Comparison of five serologic tests and culture for brucellosis in swine experimentally infected with Brucella suis Biovar 1

Purified 60-kilodalton Legionella protein antigen with Legionella-specific and nonspecific epitopes

Molecular cloning and nucleotide sequence analysis of the gene encoding the immunoreactive Brucella abortus Hsp60 protein

Immunologic response of patients with legionellosis against major protein-containing antigens of Legionella pneumophila serogroup 1 as shown by immunoblot analysis

Nucleotide sequence ofhtpB, the Legionellapneumophila gene encoding the 58 kDa common antigen, formerly designated the 60 kDa common antigen

Outer membrane proteins from rough strains of four Brucella species

Differentiation of serological responses to Yersinia enterocolitica serotype 0:9 and Brucella species by immunoblot or enzyme-linked immunosorbent assay using whole Bacteria and Yersinia outer membrane proteins

The Mycobacterium tuberculosis 65-Kilodalton antigen is a heat shock protein which corresponds to common antigen and to the Escherichia coli GroEL protein

SDS-insoluble and peptidoglycan-bound proteins in the outer membrane-peptidoglycan complex of Brucella abortus

Genus-specific epitope on the 60-kilodalton Legionella heat shock protein recognized by a monoclonal antibody

The 65 kD antigen: molecular studies on a ubiquitous antigen

Antigenic relatedness of a strongly immunogenic 65 kDa mycobacterial protein antigen with a similarly sized ubiquitous bacterial common antigen

Brucella, a monospecific genus as shown by deoxyribonucleic acid hybridization

Outer membrane proteins ofBrucella abortus: Isolation and characterization

Comparison of sodium dodecyl sulphate-polyacrylamide gel electrophoresis profiles and antigenic relatedness among outer membrane proteins of 49 Brucella abortus strains

Amino-acid sequence homology of a polymorphic cellular protein from human lymphocytes and the chaperonins from Escherichia coli (groEL) and chloroplasts (Rubisco-binding protein)

Serological reactions to Brucella in British pigs

Yersinia enterocolitica immunodominant 60 kDa antigen, common to a broad range of bacteria, is a heat shock protein

This study was funded by the Ministry of Agriculture, Fisheries and Food, and received financial support from the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases (TDR) by the provision of ML-30.