We constructed infectious but replication-deficient Semliki Forest virus (SFV) particles carrying recombinant RNA encoding Brucella abortus Cu,Zn superoxide dismutase (SOD). The recombinant SFV particles (SFV-SOD particles) were then evaluated for their ability to induce a T-cell immune response and to protect BALB/c mice against a challenge with B. abortus 2308. Intraperitoneal injection of mice with recombinant SFV-SOD particles did not lead to the induction of SOD-specific antibodies, at least until week 6 after immunization (the end of the experiment). In vitro stimulation of splenocytes from the vaccinated mice with either recombinant Cu,Zn SOD (rSOD) or crude Brucella protein resulted in a T-cell proliferative response and the induction of gamma interferon secretion but not interleukin-4. In addition, the splenocytes exhibited significant levels of cytotoxic T-lymphocyte activity against Brucella-infected cells. The SFV-SOD particles, but not the control virus particles, induced a significant level of protection in BALB/c mice against challenge with B. abortus virulent strain 2308. These findings indicated that an SFV-based vector carrying the SOD gene has potential for use as a vaccine to induce resistance against B. abortus infections.Brucella abortus is gram-negative, facultative intracellular bacterial pathogen that causes brucellosis in humans and cattle. In the infected host B. abortus multiplies within the phagosomes of reticuloendothelial cells, avoiding the killing effect of the macrophage cells by inhibiting phagosome-lysosome fusion (22). Like resistance to other facultative intracellular bacterial pathogens, resistance to B. abortus depends on acquired cellmediated immunity (CMI) (51). In this respect, the development of the Th1 subset of CD4 ϩ lymphocytes that secrete gamma interferon (IFN-␥), a crucial cytokine that up-regulates the macrophage anti-Brucella activity, and the development of CD8 ϩ T lymphocytes that are able to lyse Brucella-infected cells (33,34) are the two main components of the protective response. Live, attenuated vaccines, such as B. abortus strains S19 and RB51, that can stimulate a strong CMI response are usually very effective against brucellosis and have been used to control brucellosis in domestic animals. However, these vaccine preparations are far from ideal, since they have disadvantages; e.g., they are considered virulent or unsafe for human use, and they induce abortion in pregnant cattle (43). In the search for methods that can provide immunity against brucellosis, DNA-based vaccines have been shown to be effective for the delivery of antigenic proteins (3, 22) and to generate a strong cellular response (28). In previous reports, we showed that a DNA vaccine encoding the Cu,Zn superoxide dismutase (SOD) protein of B. abortus is able to induce cytotoxic Tlymphocyte (CTL) activity, high levels of IFN-␥ production, and a significant level of protection against challenge with B.abortus virulent strain 2308 in BALB/c mice (30, 37). However, repeated doses and high c...