Brucella abortus is a gram negative, facultative intracellular bacterial pathogen, capable of growth and replication within macrophages, and is the causative agent of bovine brucellosis. The progression of brucellosis within the host is determined by the interaction of Brucella with its host. Therefore, it was of interest to specifically examine several features of the Brucella-host interaction. The Brucella-macrophage interaction is central in the progression of brucellosis and, therefore, it was possible to study this interaction in vitro in the form of the abilities of Brucella strains to grow and replicate within macrophages. Furthermore, it was of interest to see if the in vitro model was capable of assessing the degree of attenuation of the Brucella. Various strains of B. abortus and B. melitensis were used to infect two murine macrophage-like cell lines to study their intracellular growth kinetics and to compare these kinetics with the growth characteristics observed in mice. It was determined that Brucella growth in one murine macrophage-like cell line (J774.A1) clearance pattern reflected the in vivo growth kinetics of the various Brucella tested. All strains tested in the macrophage model had a significant 1-4 log decrease in intracellular bacteria at 24 hours post infection. The decrease in intracellular numbers at 24 hours postinfection was due to the bactericidal activities of the macrophages as opposed to changes in the Brucella. This model was determined not a satisfactory means in vitro for assessing the degree of attenuation of Brucella mutants, as the model was not capable of predicting the reduced virulence of the B. melitensis RM1 rifampin resistant mutant. Vaccination with live B. abortus strain RB51 protects both mice and cattle against challenge with virulent strain 2308. As B. abortus is a facultative intracellular pathogen, the host's cell-mediated immunity is assumed to be important in the clearance of the bacteria. Therefore, selected antigens from strain RB51 could be used and tested in vitro for their ability to induce a cellular immune, specifically a T helper type 1 (T_H1) response, by splenocytes from mice vaccinated with strain RB51. An extract from strain RB51, designated S2, was found to stimulate the proliferation of splenic lymphocytes from strain RB51 vaccinated mice as well as the production of interleukin (IL)-2 and interferon (INF)-γ, but not IL-4. This cytokine profile is consistent with a cell-mediated T_H1 immune response. As the T_H1 response is assumed to be important in the clearance of Brucella by the host, mice were immunized with S2 extract in adjuvant. Upon virulent challenge with strain 2308, no protection was observed as compared with challenged control mice. The S2 extract contained nine proteins ranging in size from 10 kDa to 80 kDa. In order to determine which of the individual S2 proteins was responsible for the observed T_H1 activity, a means of separating the proteins into individual bands for testing was needed. Sufficient resolution of the nine proteins could not be accomplished by isoelectric focusing (Rotofor) or by fast performance liquid chromatography (FPLC). Therefore, SDS-polyacrylamide gel electrophoresis (SDS-PAGE) was used to separate S2 into individual proteins. Two proteins, 10.2 kDa and 11.5 kDa in size, were observed to stimulate the production of INF-γ by sensitized splenocytes. The N-terminal amino acid sequences of these two proteins were obtained and their putative DNA sequences deduced. Another antigen known to be a component of the S2 extract is the lipopolysaccharide (LPS). Polymyxin B is thought not to bind B. abortus LPS as it does not contain 3-myristic acid. However, it was important to evaluate the role, if any, of LPS contamination of the S2 extract in the in vitro responses observed with splenocytes from mice vaccinated with strain RB51. The S2 extract was treated with polymyxin B linked to agarose beads. The treated S2 extract contained less 2-keto-3- deoxyoctonate (KDO) which indirectly suggests a decrease in LPS. Furthermore, the proliferation of sensitized splenocytes induced by polymyxin B treated S2 extract was observed to decrease slightly, while the INF-γ levels were observed to increase following the treatment. Reactivity of a monoclonal antibody (Bru 48) specific for B. abortus rough LPS was observed by immunoblot analysis only with samples not treated with polymyxin B beads. This suggests that the amount of rough B. abortus LPS contained in the preparation is less than is detectable by the monoclonal antibody when the S2 extract was treated with polymyxin B beads. In contrast to published results, these data suggest that polymyxin B interacts with B. abortus LPS. The use of a live vector delivery system as a means of vaccination needs to be examined further in the case of brucellosis. In this manner, the expression of the S2 proteins may more accurately be assessed as to their role in eliciting a protective immune response. / Ph. D.