Ceruloplasmin sequence alignments with ascorbate

The location of the copper with respect to the Greek key fold is interesting when compared to that of the cupredoxins. While the copper in the cupredoxins lies in the interior of the /8 barrel bound by three interior-facing residues of the carboxy-terminal loop in the )8 barrel, and by a histidine in an adjacent strand, the copper in SOD lies on the outside of its jS barrel, bound by one residue from the carhoxy-terminal loop and three from the adjacent strand (cf. Figs. 2c-5c with Fig. 8c.) A structural comparison of plastocyanin and SOD, coupled with sequence alignment of plastocyanin and ceruloplasmin (Ryden, 1988), showed that three of the SOD ligands correspond to putative copper ligands in ceruloplasmin. Why this is so will become more evident after the description of the ascorbate oxidase structure and its relationship to ceruloplasmin. 

A structurally based amino-acid sequence alignment strongly suggests a three-domain structure for laccase, closely related to ascorbate oxidase, and a six-domain structure for ceruloplasmin. These domains demonstrate homology with the small blue copper proteins. The relationship suggests that laccase, like ascorbate oxidase, has a mononuclear blue copper in domain 3 and a trinuclear copper between domain 1 and domain 3, and ceruloplasmin has mononuclear copper ions in domains 2, 4, and 6 and a trinuclear copper between domain 1 and domain 6. 

With the structure of ascorbate oxidase in hand, a new structurally based alignment of the sequences of ascorbate oxidase, laccase, and ceruloplasmin has been performed (Messerschmidt and Huber, 1990). In brief, while gene triplication for ceruloplasmin is still revelant, its sequence can be further subdivided into two domains per unit of triplicated sequence, or six domains in total. Each of these sequences bears some resemblance to each of the three domains of ascorbate oxidase, as does each of the two domains in laccase. The coppers of the trinuclear site of ceruloplasmin then are predicted to be bound between domains 1 and 6, with a type I site also lying in both domains 6 and 4 (see Huber, 1990). The relative orientation of each of these domains is not predicted by this alignment, but it turns out that the structure of nitrite reductase may shed some light on this (see Section V,C). 

---