The gene encoding a protective protein antigen of the gram-positive bacterium Erysipelothrix rhusiopathiae, an important veterinary pathogen responsible for erysipelas in swine and a variety of diseases in animals, was cloned and sequenced. The gene encodes a polypeptide of 597 amino acids plus a putative signal sequence of 29 amino acids, resulting in a mature protein with a molecular mass of 69,017 Da. Sequence analysis of the gene product revealed a C-terminal region composed of nine tandem repeats of 20 amino acids and a total sequence that is nearly identical to that of the 64-kDa cell surface protein (SpaA) of the bacterium. Because of this similarity, the protein was designated SpaA.1. In this study, we examined whether the SpaA.1 protein could induce protective antibodies and whether we could identify the region involved in protective immunity. Both the mature SpaA.1 protein and its C-terminal repeat region, but not the N-terminal segment, were expressed in Escherichia coli and purified as a histidine-tagged fusion recombinant protein. Rabbit antiserum raised against the mature SpaA.1 protein passively protected mice from lethal challenge with a virulent homologous strain, Fujisawa-SmR, suggesting that protection is mediated by humoral antibodies. To determine which domain of the SpaA.1 protein is responsible for the observed protection, mice were actively immunized with either the mature SpaA. 1 protein or the C-terminal repeat region and then challenged with Fujisawa-SmR. The result showed that mice immunized with the mature SpaA.1 protein, but not the C-terminal repeat region, were protected, suggesting that the protection-eliciting epitope(s) is located within the N-terminal two-thirds of the SpaA.1 molecule. This was confirmed by passive immunization experiments in which the protective activity of rabbit antiserum, raised against mature SpaA. 1 protein, was not abolished by absorption with the purified recombinant C-terminal repeat region. In addition, antibodies specific for the C-terminal repeat region were unable to protect mice from lethal challenge. These results show that the N-terminal two-thirds of the SpaA.1 molecule may constitute a good vaccine candidate against erysipelas.