Epitope Mapping of the Brucella melitensis BP26 Immunogenic Protein: Usefulness for Diagnosis of Sheep Brucellosis

CLINICAL AND DIAGNOSTIC LABORATORY IMMUNOLOGY, July 2003, p. 647 651 Vol. 10, No. 4 1071-412X/03/$08.00 0 DOI: 10.1128/CDLI.10.4.647 651.2003 Copyright 2003, American Society for Microbiology. All Rights Reserved. Epitope Mapping of the Brucella melitensis BP26 Immunogenic Protein: Usefulness for Diagnosis of Sheep Brucellosis Patricia Seco-Mediavilla, 1 Jean-Michel Verger, 2 Maggy Grayon, 2 Axel Cloeckaert, 3 Clara M. Marín, 4 Michel S. Zygmunt, 2 Luis Fernández-Lago, 1 and Nieves Vizcaíno 1 * Departamento de Microbiología y Genética, Universidad de Salamanca, 37007 Salamanca, 1 and Unidad de Sanidad Animal, Servicio de Investigación Agroalimentaria, Diputación General de Aragón, 50080 Zaragoza, 4 Spain, and Station de Pathologie Infectieuse et Immunologie, 2 and Unité BioAgresseurs, Santé et Environnement, 3 Institut National de la Recherche Agronomique, 37380 Nouzilly, France Received 25 October 2002/Returned for modification 18 December 2002/Accepted 21 January 2003 Sequencing of bp26, the gene encoding the Brucella sp. immunogenic BP26 periplasmic protein, was performed in the reference strains of Brucella abortus, B. suis, and B. ovis. The three bp26 sequences were almost identical to that published for B. melitensis 16M bp26, and only minor nucleotide substitutions, without modifying the amino acid sequence, were observed between species. The bp26 genes of the seven B. abortus biovar reference strains and B. abortus S19 and RB51 vaccine strains were also sequenced. Again, only minor differences were found. Surprisingly, the bp26 nucleotide sequence for B. abortus S19 was almost identical to that found for B. melitensis 16M and differed from the sequence described previously by others (O. L. Rossetti, A. I. Arese, M. L. Boschiroli, and S. L. Cravero, J. Clin. Microbiol. 34:165-169, 1996) for the same B. abortus strain. The epitope mapping of BP26, performed by using a panel of monoclonal antibodies and recombinant DNA techniques, allowed the identification of an immunodominant region of the protein interesting for the diagnosis of B. melitensis and B. ovis infection in sheep. A recombinant fusion protein containing this region of BP26 reacted indeed, in Western blotting, as the entire recombinant BP26 against sera from B. melitensis- or B. ovis-infected sheep while it avoided false-positive reactions observed with sera from Brucella-free sheep when using the entire recombinant BP26. Thus, use of this recombinant fusion protein instead the entire recombinant BP26 could improve the specific serological diagnosis of B. melitensis or B. ovis infection in sheep. * Corresponding author. Mailing address: Departamento de Microbiología y Genética, Edificio Departamental, Universidad de Salamanca, Avda. Campo Charro s/n, 37007 Salamanca, Spain. Phone: 34-923-294532. Fax: 34-923-224876. E-mail: vizcaino@usal.es. Diagnosis of brucellosis is currently performed with serological techniques that mainly detect antibodies against lipopolysaccharide. However, antibodies against lipopolysaccharide are also induced in animals vaccinated with Brucella sp. attenuated strains. Therefore, an important goal in brucellosis research is the identification of protein antigens that induce an intense antibody response during infection and that are not essential for the induced protective immunity or for survival of the bacterium. Vaccination with a mutant of the vaccine strain lacking the gene coding for a protein of interest, in association with a serological test based on the purified protein, should allow the differentiation between vaccinated and infected animals. The most encouraging results until present have been obtained with the Brucella sp. BP26 protein, which has been simultaneously identified by three nonrelated research groups as an immunodominant antigen in infected cattle, sheep, goats, and humans (3, 6, 7, 8, 9). The most exhaustive studies about BP26 have evaluated its usefulness as a diagnostic antigen for sheep brucellosis that is caused by Brucella melitensis or B. ovis. A competitive enzyme-linked immunosorbent assay (ELISA) using BP26-specific monoclonal antibodies (MAbs) (9) and an indirect ELISA with the protein partially purified from Brucella spp. (8) provided good results in the differentiation between B. melitensis-infected sheep and B. melitensis Rev.1- vaccinated sheep. Cloning of the gene coding for BP26 (3, 6, 7) has allowed the construction of a mutant of B. abortus S19 vaccine strain unable to express bp26. This mutant vaccine strain protected mice from infection with pathogenic B. abortus 2308 to a level similar to that of the parental S19 strain (1). Although these results cannot be extrapolated to cattle, they indicate that the B. abortus S19 bp26 mutant vaccine strain might be used in combination with a BP26-based serological test for the differential diagnosis between infected and vaccinated animals. Similar results may be expected for the B. melitensis Rev.1 vaccine strain used against sheep and goat brucellosis. Additionally, the purification of the protein from recombinant Escherichia coli (14) has been possible. The use of recombinant BP26, instead of the protein extracted from Brucella spp., for diagnosis of brucellosis presents several advantages: (i) BP26 is available free of other Brucella sp. antigens that might interfere in the diagnostic test, (ii) the extraction of the protein is less time-consuming and high yields are obtained, and (iii) the manipulation of pathogenic Brucella spp. is avoided. Purified recombinant BP26 has revealed, by indirect ELISA, as a promising antigen for both confirmatory tests of infections caused by B. melitensis and B. ovis in sheep and differentiation between infected and B. melitensis Rev.1-vaccinated sheep (2, 14). However, some sera from Brucella-free sheep gave false-positive reactions (2) that might be reduced or even avoided if a selected immunodominant fragment of BP26, instead the entire 647

648 SECO-MEDIAVILLA ET AL. CLIN. DIAGN. LAB. IMMUNOL. TABLE 1. Brucella strains used in this study Species and strain a Biovar Host B. melitensis 16M (ATCC 23456; BCCN R1) protein, were used. In the present work, we describe the epitope mapping of BP26, by using a panel of specific MAbs and recombinant DNA techniques, and analyze the reactivity of the selected regions of BP26 against sera from Brucella-free sheep and B. melitensis- and B. ovis-infected sheep. MATERIALS AND METHODS Geographic origin 1 Goat United States B. abortus 544 (ATCC 23448; 1 Cattle England BCCN R4) 86/8/59 (ATCC 23449; 2 Cattle England BCCN R5) Tulya (ATCC 23450; 3 Human Uganda BCCN R6) 292 (ATCC 23451; 4 Cattle England BCCN R7) B3196 (ATCC 23452; 5 Cattle England BCCN R8) 870 (ATCC 23453; 6 Cattle Africa BCCN R9) C68 (ATCC 23455; 9 Cattle England BCCN R11) S19 (BCCN V1) b 1 Cattle United States RB51 (BCCN V5) b Cattle United States B. suis 1330 (ATCC 23444; BCCN R12) B. ovis 63/290 (ATCC 25840; BCCN R17) 1 Swine United States Sheep Africa a Abbreviations: ATCC, American Type Culture Collection; BCCN, Brucella Culture Collection, Nouzilly, France. b Vaccine strain. Bacterial strains and plasmids. Brucella strains (Table 1) were obtained from the Brucella Culture Collection maintained at the Institut National de la Recherche Agronomique, Nouzilly, France. Cultures were performed as described previously (13). Recombinant plasmids were propagated in E. coli JM109 (Promega, Madison, Wis.) and cultured by standard procedures in medium containing 50 g of ampicillin ml 1. Plasmid pcp2801, containing the B. melitensis 16M bp26 gene and DNA flanking both sides of the gene, was obtained as described previously (3). Recombinant plasmids bearing fragments of B. melitensis bp26 in pgem-7zf (Promega) were constructed during the epitope mapping of BP26 as described below. Immunological techniques. Culture supernatants of secreting hybridomas, produced as previously described (4), were used as source of MAbs specific for BP26. Sera from B. melitensis- and B. ovis-infected sheep were positive in conventional serology tests (Rose Bengal and complement fixation tests) and were obtained from naturally infected animals where the bacterium was isolated. They were provided by the Unidad de Sanidad Animal, Servicio de Investigación Agroalimentaria, Diputación General de Aragón, Zaragoza, Spain. Negative control sheep sera, bacteriologically and serologically negative by conventional tests, were obtained from the brucellosis-free flock of the Animal Production facilities, INRA PII unit, flock officially free of brucellosis for over 25 years. Colony and Western blotting with recombinant E. coli cultures induced with isopropyl-1-thio- -D-galactopyranoside (IPTG) were performed as previously described (13). DNA amplification and sequencing. PCR was performed, as described previously (12), with the Expand long-template PCR system (Roche Diagnostics GmbH, Mannheim, Germany) according to the instructions of the manufacturer. The bp26 gene was amplified from the Brucella strains by using primers CP-141 (5 -GCGCAGATATTCAGTTGC-3 ) and CP-1317 (5 -GTGACATTTGCCGA TACG-3 ), selected according to the published sequenced of B. melitensis 16M (3). The PCR products were electrophoresed through an agarose gel, purified from the gel with the Geneclean II kit (Bio 101, La Jolla, Calif.) and sequenced, with primers CP-141 and CP-1317, by primer-directed dideoxy method with an ABI Prism 377 DNA sequencer (Perkin-Elmer, Foster City, Calif.). Primers ERY1 (5 -TTGGCGGCAAGTCCGTCG-3 ) and ERY2 (5 -CAGA AGCGAGACGAAACG-3 ) for amplification of the ery locus from B. abortus S19 were selected according to the published sequence (10). The same primers were used for sequencing. DNA inserts of the recombinant plasmids obtained during the BP26 epitope mapping were sequenced with the universal and reverse puc19 primers, using plasmid DNA obtained with the Wizard Plus SV Miniprep system (Promega) as specified by the manufacturer. Epitope mapping. Recombinant plasmid pcp2801, containing B. melitensis 16M bp26 and adjacent DNA to both sides of the gene, was digested with ApaI to release the insert that was then purified from an agarose gel after electrophoresis by using the Geneclean II kit. DNase I (Roche Diagnostics GmbH) digestion of the insert, purification of DNA fragments ranging from 100 to 500 bp, their end repairing with T4 DNA polymerase and ligation into the SmaI site of pgem-7zf were performed as described previously (13). E. coli JM109 was transformed with the ligation mixture and plated on Luria-Bertani agar containing ampicillin, IPTG, and 5-bromo-4-chloro-3-indolyl- -D-galactopyranoside (X- Gal). Bacterial colonies were transferred to new plates and screened by colony blotting with the BP26-specific MAbs. The colonies reacting with one or more MAbs were selected, their plasmid DNA was extracted, and the insert DNA was sequenced and converted into amino acids. The shortest region on BP26 common to all the plasmids giving reactivity with an individual MAb delimits its specific epitope. DNA and protein analysis. Multiple DNA and amino acid alignments were performed with CLUSTAL W (11) (http://www2.ebi.ac.uk/clustalw/). Nucleotide sequence accession numbers. The bp26 nucleotide sequences of the Brucella strains have been deposited in GenBank/EMBL/DDBJ databases under accession numbers AY166759 to AY166769. RESULTS AND DISCUSSION Sequencing of bp26 in Brucella species and biovars. The Brucella sp. BP26 protein has been identified by three independent laboratories as an interesting diagnostic antigen for brucellosis (3, 6, 7, 8, 9). Nucleotide sequencing has revealed that the bp26 DNA sequence of B. abortus S19 differs significantly from that of B. melitensis 16M (3, 6, 7). DNA polymorphism in genes encoding other Brucella sp. immunogenic proteins has been shown to lead to antigenic differences in the proteins (5, 13). Differences in antigenicity may be a drawback for a protein intended to be a diagnostic antigen for infections caused by several species of the same genus. The B. ovis Omp31 differs in only 7 amino acids from the B. melitensis Omp31, but these minor differences highly modify the antigenic properties of the protein, as measured by reactivity with specific MAbs and sera from infected sheep (13). Similarly, a deletion of 36 bp in the B. ovis omp25 gene, compared to omp25 of the other Brucella species, resulted in an antigenic shift in the encoded protein (5). Differences between the published bp26 sequences from B. melitensis and B. abortus S19 involve two deletions in B. melitensis 16M of 24 and 6 bp, respectively, and several nucleotide substitutions (3, 6, 7). If these differences modify the antigenicity of the protein, the choice of the BP26 protein used for diagnosis might be of great importance. Accordingly, to determine the degree of conservation of bp26 in the genus Brucella, sequencing of the gene was performed in the reference strains of the four Brucella species with more relevance in veterinary and human health: B. melitensis, B. abortus, B. suis, and B. ovis. In the case of B. abortus, the

VOL. 10, 2003 B. MELITENSIS BP26 EPITOPE MAPPING FOR DIAGNOSIS 649 FIG. 2. Reactivity in Western blotting of BP26-specific MAbs against recombinant E. coli/pcp2801. Positions of protein molecular mass markers are shown on the left. MAbs: V78/09B12/B02 (lane 1), V78/11A07/G06 (lane 2), V78/06E10/H10 (lane 3), V78/04D01/A10 (lane 4), V78/10F02/E10 (lane 5), V78/10A07/H09 (lane 6), V78/ 03G01/F01 (lane 7), V78/02D06/C08 (lane 8), V78/05B10/F07 (lane 9), V78/02E08/F03 (lane 10), V78/05G03/H03 (lane 11), V78/09C08/A01 (lane 12), V78/05G06/C01 (lane 13), V78/04G07/H05 (lane 14), V78/ 07G06/A09 (lane 15), V78/10A03/A02 (lane 16), V78/06D11/G06 (lane 17), V78/04D06/B04 (lane 18). FIG. 1. Alignment of the BP26 amino acid sequences from Brucella spp. Amino acid differences for each strain in comparison to the published B. melitensis 16M BP26 are highlighted in black. Abbreviations: Ba1, B. abortus 544 (biovar 1); Ba3, B. abortus Tulya (biovar 3); S19, B. abortus S19 vaccine strain; RB51, B. abortus RB51 vaccine strain. The BP26 amino acid sequence for the five remaining B. abortus biovar reference strains and for B. suis and B. ovis reference strains was identical to B. abortus 544 and B. melitensis 16M BP26. reference strains of the seven biovars and two vaccine strains, S19 and RB51, were also included in the study. The same nucleotide sequence was found for bp26 in the reference strains of B. melitensis and B. abortus, and only one nucleotide substitution in B. suis and another one in B. ovis were detected (data not shown). However, at the BP26 amino acid level no differences were found between the four species reference strains. As the bp26 sequence for B. abortus S19 published by Rossetti et al. in 1996 significantly differs from that found for the B. abortus reference strain, bp26 sequencing of the reference strains of the seven B. abortus biovars and S19 and RB51 strains was also performed. Surprisingly, an almost identical bp26 nucleotide sequence was found for all the strains, including B. abortus S19. The same single nucleotide substitution that leads to a amino acid difference was found in S19 and RB51 B. abortus vaccine strains (Fig. 1). This nucleotide substitution creates a ClaI-specific recognition site (data not shown) that might be a DNA molecular marker for the B. abortus S19 and RB51 vaccine strains. However, other Brucella strains should be analyzed to verify this point. Another single nucleotide substitution was found in B. abortus bv. 3 reference strain, also leading to an amino acid difference (Fig. 1). PCR amplification and sequencing of the ery locus was performed with the B. abortus S19 vaccine strain. The nucleotide sequence was as expected (data not shown), with the specific deletion described for the ery locus in this strain (10), which confirms that the strain used to sequence bp26 corresponds well to B. abortus S19 and not to another Brucella strain. Differences found in the published B. abortus S19 bp26 nucleotide sequence (7) when compared to the bp26 sequence determined in the present work for all the B. abortus biovars reference strains and B. abortus S19 are difficult to explain. Only a mistake in sequencing or a specific mutation might explain it. According to the results presented, BP26 is highly conserved in the genus Brucella, and the recombinant protein from one Brucella species might be used for the serological diagnosis of infections caused by all the Brucella species, provided that the animal host is able to induce an immune response against this antigen. BP26 epitope mapping. Antigenic characterization of BP26 was performed by using a panel of 18 BP26-specific MAbs and recombinant DNA techniques. Reactivity of the MAbs with recombinant BP26 synthesized in E. coli/pcp2801 was tested in Western blotting. As expected, the MAbs reacted with two protein bands (Fig. 2), with a molecular mass close to 31 kda; these MAbs probably correspond to the preprotein with the signal peptide and the mature protein (3, 6, 7). However, some MAbs reacted better with the mature protein than with the preprotein, suggesting that their reactivity is more dependent on the conformation of the protein. To delimit the specific epitope of each MAb, fragments of the B. melitensis 16M BP26 were synthesized in recombinant E. coli as fusion proteins with LacZ of pgem-7zf and their reactivity with all the MAbs was tested. Twelve recombinant E. FIG. 3. Fragments of BP26, synthesized as fusion proteins with LacZ in E. coli, reacting in colony blotting with the BP26-specific MAb V78/10A07/H09.

650 SECO-MEDIAVILLA ET AL. CLIN. DIAGN. LAB. IMMUNOL. FIG. 4. BP26-specific MAbs reacting in colony blotting with the BP26 fragments synthesized as fusion proteins with LacZ in E. coli/pcp28112 and E. coli/pcp28124. coli colonies reacted in colony blotting with MAb A78/10A07/ H09 (Fig. 3), the MAb giving the lowest reactivity in Western blotting with the entire BP26 protein (Fig. 2, lane 6). Sequencing of their plasmid inserts and translation into amino acids revealed that the shortest BP26 region common to the 12 plasmids was the C-terminal region between amino acids 220 and 250 which is synthesized in E. coli/pcp28126 (Fig. 3). The fusion protein of pcp28126, of about 7.1 kda, is constituted by the N-terminal end of LacZ and the C-terminal end of BP26. In contrast to the high number of colonies reacting with MAb A78/10A07/H09, only two recombinant E. coli colonies (E. coli/pcp28112 and E. coli/pcp28124) reacted with five MAbs: V78/04D01/A10, V78/10F02/E10, V78/05B10/F07, V78/06D11/G06 and V78/04G07/H05 (Fig. 4). pcp28112 synthesizes the region of BP26 between amino acids 1 and 191 (including the signal peptide). When analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blotting (data not shown), the molecular mass of the protein seems to correspond to a protein constituted by the N terminus of BP26, with the signal peptide cleaved, fused to the C terminus of LacZ. According to the nucleotide sequence and Western blotting results, pcp28124 synthesizes a fusion protein with the N and C terminus of LacZ and the BP26 region of 98 amino acids from amino acid 55 to 152 between both ends. Only E. coli/pcp28112 reacted with MAbs V78/09B12/ B02, V78/02D06/C08 and V78/04D06/B04 (Fig. 4). The remaining nine MAbs gave no reactivity with any recombinant E. coli colony. The low number of recombinant E. coli colonies showing reactivity with most of the BP26-specific MAbs seems to indicate that their reactivity is highly dependent on the conformation of the protein, in spite of their reactivity in Western blotting with E. coli/pcp2801. Their specific epitopes would be presented in a different way in the entire BP26, synthesized in E. coli/pcp2801, and in the fragments of BP26 synthesized as fusion proteins in recombinant E. coli. Another possibility might be the eventual toxicity of the BP26 fragments in E. coli, as BP26 has been shown to be toxic for E. coli (3). This toxic effect has also been observed for the fusion proteins of pcp28112 and pcp28124. The optical density (OD) at 600 nm of IPTG-induced cultures of E. coli/pcp2801, /pcp28112, or /pcp28124 was shown to be reduced at the same level, during the4hofinduction, when compared to E. coli/pgem-7zf and E. coli/pcp28126, synthesizing the C terminus of BP26 (data not shown). Synthetic peptides might help to establish a better characterization of the BP26 epitope specific for each MAb. Reactivity of sheep sera. Sera from Brucella-free sheep and B. melitensis or B. ovis-infected sheep were tested in Western blotting for reactivity with recombinant E. coli bearing plasmids pcp2801, pcp28112, pcp28124, or pcp28126. The Brucella-free sheep group of sera (negative sera) included six sera giving high OD values, in indirect ELISA, against purified recombinant BP26 and three sera giving low OD values (2). Sera from Brucella sp.-infected sheep did not react with the fusion protein of pcp28126 that contains the C terminus of FIG. 5. Reactivity in Western blotting of sera from sheep naturally infected by B. melitensis or from Brucella-free sheep with E. coli/pcp2801 synthesizing the entire recombinant BP26 (A) or E. coli/pcp28124 synthesizing amino acids 55 to 152 of BP26 (B). Reactivity with the BP26-specific MAb V78/04D01/A10 is shown in lanes 1. The same lane number corresponds to the same serum in both panels. O.D., optical density provided by sera in indirect ELISA with purified recombinant BP26 (2).

VOL. 10, 2003 B. MELITENSIS BP26 EPITOPE MAPPING FOR DIAGNOSIS 651 from Brucella-free sheep, for the serological diagnosis by indirect ELISA of sheep brucellosis caused by B. melitensis or B. ovis. This region of BP26 might be either obtained as a synthetic peptide or purified from recombinant E. coli/pcp28124 as fusion protein with LacZ. ACKNOWLEDGMENTS We thank Manuel Sánchez Hernández for his valuable help with DNA sequence determination. This work and Nieves Vizcaíno were financed by project FAIR5- CT97-3360 from the European Union. FIG. 6. Reactivity in Western blotting of sera from rams naturally infected by B. ovis with E. coli/pcp2801 synthesizing the entire recombinant BP26 (A) or E. coli/pcp28124 synthesizing amino acids 55 to 152 of BP26 (B). Reactivity with the BP26-specific MAb V78/04D01/ A10 is shown in lane 1 in panels A and B. The same lane number on panels A and B corresponds to the same serum. BP26 (data not shown). Therefore, this region of BP26 would not be useful for the serological diagnosis of sheep brucellosis. Sera belonging to sheep naturally infected by B. melitensis (seven sera giving high OD values in indirect ELISA against purified recombinant BP26 and six sera giving low OD values [2]) reacted with the two protein bands of recombinant E. coli/pcp2801 recognized by MAb V78/04D01/A10, bands corresponding to the BP26 preprotein and the entire mature BP26 (3, 6, 7) (Fig. 5A, lanes 2 to 14). The three negative sera showing low OD values in indirect ELISA against recombinant BP26 did not react with the entire BP26 of E. coli/pcp2801 (Fig. 5A, lanes 21 to 23). However, the entire BP26 protein was detected with the six negative sera giving high OD values in indirect ELISA (Fig. 5A, lanes 15 to 20), showing that some nonspecific binding of antibodies to BP26 may occur with sera of Brucella-free sheep. Similar results were obtained with the fusion protein of E. coli/pcp28112 (data not shown). However, with E. coli/ pcp28124 no binding to the fusion protein was detected with any of the nine negative sera (Fig. 5B, lanes 15 to 23), while all the B. melitensis-infected sheep sera (Fig. 5B, lanes 2 to 14) reacted with the same protein band recognized by MAb V78/ 04D01/A10 (Fig. 5B, lane 1), corresponding to the fusion protein of pcp28124 that contains amino acids 55 to 152 of BP26. Some other protein bands, with a molecular mass not corresponding to Brucella BP26 or fusion protein, were detected in some strips (i.e., Fig. 5B, lanes 7, 8, 9, 11, and 17) and presumably correspond to E. coli proteins reacting with serum antibodies developed in response to common exposure of animals to this bacterium. Additionally, the six sera from B. ovis-infected rams that were tested reacted with the fusion protein of E. coli/pcp28124 (Fig. 6B) as they also did with the entire Brucella BP26 synthesized in E. coli/pcp2801 (Fig. 6A). Accordingly, the region of BP26 between amino acids 55 and 152 might provide better specificity results than the entire recombinant BP26, avoiding false-positive reactions with sera REFERENCES 1. Boschiroli, M. L., S. L. Cravero, A. I. Arese, E. Campos, and O. L. Rossetti. 1997. Protection against infection in mice vaccinated with a Brucella abortus mutant. Infect. Immun. 65:798 800. 2. Cloeckaert, A., S. Baucheron, N. Vizcaíno, and M. S. Zygmunt. 2001. Use of recombinant BP26 protein in serological diagnosis of Brucella melitensis infection in sheep. Clin. Diagn. Lab. Immunol. 8:772 775. 3. Cloeckaert, A., H. Salih-Alj Debbarh, N. Vizcaíno, E. Saman, G. Dubray, and M. S. Zygmunt. 1996. Cloning, nucleotide sequence, and expression of the Brucella melitensis bp26 gene coding for a protein immunogenic in infected sheep. FEMS Microbiol. Lett. 140:139 144. 4. Cloeckaert, A., H. Salih-Alj Debbarh, M. S. Zygmunt, and G. Dubray. 1996. Production and characterisation of monoclonal antibodies to Brucella melitensis cytosoluble proteins that are able to differentiate antibody responses of infected sheep from Rev. 1 vaccinated sheep. J. Med. Microbiol. 45:206 213. 5. Cloeckaert, A., J. M. Verger, M. Grayon, M. S. Zygmunt, and O. Grépinet. 1996. Nucleotide sequence and expression of the gene encoding the major 25-kilodalton outer membrane protein of Brucella ovis: evidence for antigenic shift, compared with other Brucella species, due to a deletion in the gene. Infect. Immun. 64:2047 2055. 6. Lindler, L. E., T. L. Hadfield, B. D. Tall, N. J. Snellings, F. A. Rubin, L. L. Van de Verg, D. Hoover, and R. L. Warren. 1996. Cloning of a Brucella melitensis group 3 antigen gene encoding Omp28, a protein recognized by the humoral immune response during human brucellosis. Infect. Immun. 64:2490 2499. 7. Rossetti, O. L., A. I. Arese, M. L. Boschiroli, and S. L. Cravero. 1996. Cloning of Brucella abortus gene and characterization of expressed 26-kilodalton periplasmic protein: potential use for diagnosis. J. Clin. Microbiol. 34:165 169. 8. Salih-Alj Debbarh, H., A. Cloeckaert, G. Bézard, G. Dubray, and M. S. Zygmunt. 1996. Enzyme-linked immunosorbent assay with partially purified cytosoluble 28-kilodalton protein for serological differentiation between Brucella melitensis-infected and B. melitensis Rev.1-vaccinated sheep. Clin. Diagn. Lab. Immunol. 3:305 308. 9. Salih-Alj Debbarh, H., M. S. Zygmunt, G. Dubray, and A. Cloeckaert. 1996. Competitive enzyme-linked immunosorbent assay using monoclonal antibodies to the Brucella melitensis BP26 protein to evaluate antibody responses in infected and B. melitensis Rev.1 vaccinated sheep. Vet. Microbiol. 53:325 337. 10. Sangari, F. J., J. M. García-Lobo, and J. Agüero. 1994. The Brucella abortus vaccine strain carries a deletion in the erythritol catabolic genes. FEMS Microbiol. Lett. 121:337 342. 11. Thomson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignments through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673 4680. 12. Vizcaíno, N., A. Cloeckaert, M. S. Zygmunt, and L. Fernández-Lago. 1999. Molecular characterization of a Brucella species large DNA fragment deleted in Brucella abortus strains: evidence for a locus involved in the synthesis of a polysaccharide. Infect. Immun. 67:2700 2712. 13. Vizcaíno, N., R. Kittelberger, A. Cloeckaert, C. M. Marín, and L. Fernández- Lago. 2001. Minor nucleotide substitutions in the omp31 gene of Brucella ovis result in antigenic differences in the major outer membrane protein that it encodes compared to those of the other Brucella species. Infect. Immun. 69:7020 7028. 14. Zygmunt, M. S., S. Baucheron, N. Vizcaíno, R. A. Bowden, and A. Cloeckaert. 2002. Single-step purification and evaluation of recombinant BP26 protein for serological diagnosis of Brucella ovis infection in rams. Vet. Microbiol. 87:213 220.