Evolution in the structure and function of carboxyl proteases

Abstract

A model for the structure and function of extracellular carboxyl (acid) proteases can be established from three amino acid sequences and four crystal structures of these enzymes. The carboxyl proteases from gastric and fungal origins are very homologous in both primary and tertiary structures. The molecules consist of about 320 residues organized with a secondary structure which is primarily comprised of beta-strands and very similar tertiary structures. An apparent binding cleft, which can accommodate a substrate with about eight amino acid residues, contains near its midpoint the active center residues Asp-215, Asp-32, and Ser-35. These three residues are hydrogen bonded to each other. An intracellular carboxyl protease, cathepsin D, is very homologous to the extracellular enzymes in N-terminal amino acid sequence and primary structure location of active center residues. The tertiary structure of cathepsin D is probably similar, as well. However, cathepsin D contains a unique hydrophobic "tail" made up of about 100 residues added on the C-terminal side. Cathepsin D precursor is over 100,000 daltons in molecular weights, as contrasted to the gastric carboxyl protease zymogens, which are about 40,000 daltons. Carboxyl proteases contain two lobes symmetrical in peptide chain conformations. Each of the lobes also consists of two homologous structural units. These structural characteristics suggest that the original gene was coded for only about eighty amino acid residues and that gene duplication and fusion has taken place twice to produce a single chain carboxyl protease with four basic structural units in two symmetrical lobes. The formation of the zymogens and the cathepsin D "tail" must have resulted from various gene fusions. Partial sequence comparisons also suggest that cathepsin D may be an evolutionary ancestral chain for gastric carboxyl proteases.