Abstract
Severe acute respiratory syndrome (SARS) is a serious health threat and its early diagnosis is important for infection control and potential treatment of the disease. Diagnostic tools require rapid and accurate methods, of which a capture ELISA method may be useful. Toward this goal, we have prepared and characterized soluble full-length nucleocapsid proteins (N protein) from SARS and 229E human coronaviruses. N proteins form oligomers, mostly as dimers at low concentration. These two N proteins degrade rapidly upon storage and the major degraded N protein is the C-terminal fragment of amino acid (aa) 169-422. Taken together with other data, we suggest that N protein is a two-domain protein, with the N-terminal aa 50-150 as the RNA-binding domain and the C-terminal aa 169-422 as the dimerization domain. Polyclonal antibodies against the SARS N protein have been produced and the strong binding sites of the anti-nucleocapsid protein (NP) antibodies produced were mapped to aa 1-20, aa 150-170 and aa 390-410. These sites are generally consistent with those mapped by sera obtained from SARS patients. The SARS anti-NP antibody was able to clearly detect SARS virus grown in Vero E6 cells and did not cross-react with the NP from the human coronavirus 229E. We have predicted several antigenic sites (15-20 amino acids) of S, M and N proteins and produced antibodies against those peptides, some of which could be recognized by sera obtained from SARS patients. Antibodies against the NP peptides could detect the cognate N protein clearly. Further refinement of these antibodies, particularly large-scale production of monoclonal antibodies, could lead to the development of useful diagnostic kits for diseases associated with SARS and other human coronaviruses.