Folding pathway mediated by an intramolecular chaperone: A functional peptide chaperone designed using sequence databases

Catalytic domains of several prokaryotic and eukaryotic protease families require dedicated N-terminal propeptide domains or "intramolecular chaperones" to facilitate correct folding. Amino acid sequence analysis of these families establishes three important characteristics: (i) propeptides are almost always less conserved than their cognate catalytic domains, (ii) they contain a large number of charged amino acids, and (iii) propeptides within different protease families display insignificant sequence similarity. The implications of these findings are, however, unclear. In this study, we have used subtilisin as our model to redesign a peptide chaperone using information databases. Our goal was to establish the minimum sequence requirements for a functional subtilisin propeptide, because such information could facilitate subsequent design of tailor-made chaperones. A decision-based computer algorithm that maintained conserved residues but varied all non-conserved residues from a multiple protein sequence alignment was developed and utilized to design a novel peptide sequence (ProD). Interestingly, despite a difference of 5 pH units between their isoelectric points and despite displaying only 16% sequence identity with the wild-type propeptide (ProWT), ProD chaperones folding and functions as a potent subtilisin inhibitor. The computed secondary structures and hydrophobic patterns within these two propeptides are similar. However, unlike ProWT, ProD adopts a well defined α - β conformation as an isolated peptide and forms a stoichiometric complex with mature subtilisin. The CD spectra of this complex is similar to ProWT·subtilisin. Our results establish that despite low sequence identity and dramatically different charge distribution, both propeptides adopt similar structural scaffolds. Hence, conserved scaffolds and hydrophobic patterns, but not absolute charge, dictate propeptide function.