Botulinum neurotoxins (BoNTs) produced by Clostridium botulinum are the causative agents of botulism and represents a family of seven structurally similar but antigenically different serotypes (A to G). The BoNTs are expressed in C. botulinum as a single polypeptide chain and then posttranslationally nicked, forming a di-chain polypeptide chain consisting of a 100-kDa heavy chain and a 50-kDa light chain held together by a disulfide bond. Topologically, the neurotoxins are composed of three domains, a binding domain (HC), a translocation domain (HN) and a catalytic domain. The BoNTs act preferentially on cholinergic nerve endings in both humans and animals and thus produce a flaccid paralysis that may result in death. In southern Africa, BoNT types C and D have been associated with botulism in cattle. To combat the disease, a bivalent vaccine consisting of formalin-inactivated type C and D holotoxins is currently available, and although it is efficacious, several concerns regarding its production has been raised, most notably its cost. The development of efficacious recombinant subunit vaccines may provide a means whereby many of the production problems may be eliminated or minimized. Consequently, the aim of this investigation was to produce a recombinant botulinum neurotoxin serotype D binding domain [BoNT/D(HC)] vaccine candidate for preventing BoNT/D intoxication. Towards this end, the gene fragment for the heavy chain (HC) of the BoNT produced by the C. botulinum type D vaccine strain D-50 was amplified, cloned in Escherichia coli and characterized by nucleotide sequence analyses. An alignment of the deduced amino acid sequence with that of characterized clostridial type C and D neurotoxins demonstrated that the heavy chains are composed of highly conserved domains interceded with tracts of amino acids exhibiting little overall relatedness, although considerable identity between the components ofa specific pair is apparent in certain of the regions. The deduced amino acid sequence exhibited 99, 66 and 73% identity with the reported amino acid sequences of BoNT/D-SA, BoNT/D and BoNT/C1, respectively. Attempts at expressing the native gene sequence for the HC from BoNT/D-50 in Brevibacillus brevis 47-5Q were unsuccessful. This may have been due to differences in codon bias between the heterologous gene and B. brevis. Consequently, a completely synthetic codonoptimized gene encoding the HC of BoNT/D-SA was constructed and expressed using a B. brevis 47-5Q mutant as expression host, obtained on mutagenesis with N-methyl-N’-nitro-Nnitrosoguanidine (NTG). Extracellular expression of the 48-kDa recombinant protein was verified by Western blot analyses with anti-BoNT/D antibodies. The recombinant BoNT/DSA(HC) protein was purified from the culture supernatant and used to vaccinate mice, after which their survival against challenge with active toxin was evaluated. Mice given two subcutaneous vaccinations were protected against intraperitoneal administration of 4 X 102 mouse lethal dosages (MLDs) of 16S BoNT/D-50 toxin. Antibody levels in mice surviving challenge were determined by enzyme-linked immunosorbent assays and confirmed that BoNT/D-SA(HC) was successful in evoking a protective immune response, whilst Western blot analyses indicated the presence of anti-16S BoNT/D-50 toxin antibodies in the serum. From these results it could be concluded that the recombinant BoNT/D-SA(HC) protein is an effective immunogen, able to protect against a high challenge dose of BoNT/D-50 neurotoxin.