Three-Dimensional Model of a Human Interferon-α Consensus Sequence

Publisher Website
Google Scholar
Abstract
A computer-built, three-dimensional, atomic-level model for human interferon-α (IFN-α) was constructed. This model was prepared using the primary amino acid sequence of consensus IFN-α (IFN-αCon1) and the α-carbon Cartesian coordinates of murine IFN-β as a homolog guide to the model building. In agreement with an earlier report from this laboratory, the two domains 29–35 and 123–140 are in close spatial proximity in this model, and may constitute a receptor recognition domain, whereas the region bounded by residues 78–95 is somewhat removed from this region on the molecule and may constitute an alternative active site. Extrapolating from the model, we propose that, of the stretch 123–140, the residues that are exposed are 123, 125, 126, 128–130, and 132–139; and of the stretch 29–35, all are accessible. Additionally, we propose that there may be sufficient complexity in the Type 1 IFN receptor to account for the differential sensitivites between IFN-αs and IFN-β that may be associated with residue differences in the region 78–95, specifically at residues 84, 86, and 87. This model conforms with experimental data that identify specific amino acid residues in human IFN-α that either do, or do not, affect the active conformation and biological activities of the molecule.