Abstract
Available sequence information (a total of 650 residues out of about 1050) of human ceruloplasmin, a blue Cu-containing oxidase, was examined for internal homologies and relationships to other blue proteins. The peptide chain has an evident 3-fold repeat of about 340 residues, and weak evidence for a 6-fold repeat of 170 residues exists. When another method was used to compare the longer sequence with the sequences of small blue proteins, azurins and plastocyanins, a 109-residue-long sequence at the COOH terminus of ceruloplasmin was homologous to the plastocyanins. The alignment obtained, was used to construct on a graphic display, a 3-dimensional model of this part of ceruloplasmin by using the coordinates for poplar plastocyanin. Deletions and insertions could be accommodated in turns and kinks in the essentially 8-stranded pleated sheet molecule wherein each of the hydrophobic core residues was conserved or conservatively replaced. Eight of the 12 histidine side chains were clustered at or close to the binding site for the blue (type 1) Cu. On the assumption that these are Cu ligands, a model for the active site of ceruloplasmin containing 4 Cu ions could be constructed in a manner consistent with known spectroscopic and kinetic data. In particular, 2 of the Cu are close enough (3 .ANG.) to form a binuclear center. The positions of the 2 additional Cu (the fifth and the sixth) in ceruloplasmin are suggested on the basis of the internal homologies.