The nifV gene products from Azotobacter vinelandii and Klebsiella pneumoniae share a high level of primary sequence identity and are proposed to catalyze the synthesis of homocitrate. While searching for potential nif (nitrogen fixation) genes within the genomic region located downstream from the nifN-B gene of Clostridium pasteurianum, we observed two open reading frames (ORFs) whose deduced amino acid sequences exhibit nonoverlapping sequence identity to different portions of the nifV gene products from A. vinelandii and K. pneumoniae. Conserved regions were located between the C-terminal 195 amino acid residues of the first ORF and the C-terminal portion of the nifV gene product and between the entire sequence of the second ORF (269 amino acid residues) and the N-terminal portion of the nifV gene product. We therefore designated the first ORF nifVω and the second ORF nifVα. The deduced amino acid sequences of nifVω and nifVα were also found to have sequence similarity when compared with the primary sequence of the leuA gene product from Salmonella typhimurium, which encodes α-isopropylmalate synthase. Marker rescue experiments were performed by recombining nifVω and nifVα from C. pasteurianum, singly and in combination, into the genome of an A. vinelandii mutant strain which has an insertion and a deletion mutation located within its nifV gene. A NifV+ phenotype was obtained only when both the C. pasteurianum nifVω and nifVα genes were introduced into the chromosome of this mutant strain. These results suggest that the nifVω and nifVα genes encode separate domains, both of which are required for homocitrate synthesis in C. pasteurianum.