Structural domains in venom proteins: evidence that metalloproteinases and nonenzymatic platelet aggregation inhibitors (disintegrins) from snake venoms are derived by proteolysis from a common precursor.

A comparison of the structures of a precursor of trigramin (a disintegrin), metalloproteinases, disintegrins and related proteins, suggests the existence of common precursors for metalloproteinases and disintegrins. The proposed common precursor and related proteins have four distinct domains (A-D). Domain B contains the metal binding site and the catalytic Glu residue, which comprise the active site of metalloproteinases. Domain C contains the Arg-Gly-Asp sequence and hence the ability to inhibit the activity of integrins. Domains A and D are unique and their biochemical or biological activity is unknown. The proposed precursor can be proteolytically cleaved at several interdomain sites, releasing the disintegrins and metalloproteinases. A survey of more than 100 venom metalloproteinases and disintegrins strongly supports the existence of precursor proteins and their structural domains. This is also upheld by the co-occurrence occurrence of metalloproteinases and disintegrins in the venoms of several genera of crotalid and viperid snakes. The likelihood of intradomain disulfide bridges, and accessibility of all interdomain cleavage sites also supports our contention. The susceptibility of the cleavage sites appears to be determined by nearby disulfide bridges and glycosylation. Recognition of the proposed structural domains of venom proteinases should help clarify the structure-function relationships of several related proteins, and influence the synthesis of recombinant disintegrins, metalloproteinases and related polypeptides.