Sequence modeling using semisynthetic ribonuclease S.

We have designed and synthesized a model pentadecapeptide predicted to have the essential sequence information needed to form a stable and enzymatically active noncovalent complex with bovine pancreatic ribonuclease S-protein. The model peptide sequence, based on the conformational approach of simplifying the native sequence in a manner consistent with retention of essential noncovalent contacts and of secondary structure features, contained alanine at all positions except for Glu 2, Lys 7, Phe 8, Arg 10, His 12, and Met 13. The peptide was synthesized by the Merrifield solid phase method. The circular dichroism spectra of the purified model peptide in water and trifluoroethanol indicated a tendency to form an alpha-helical structure similar to that found for native S-peptide. The model peptide formed a stable complex with ribonuclease S-protein. With 12-fold excess of the peptide, the complex exhibited 36% of the specific activity of fully native ribonuclease S against the substrate cyclic cytidine 2':3'-monophosphate at pH 7.15. The dissociation constant of the model peptide for S-protein was found to be 1.1 x 10(-6) M, compared with 0.1 x 10(-6) M for native S-peptide. Crystals grown of the model peptide-S-protein complex were found to be isomorphous with those of native complex. The activity, stability, and structural integrity of the model complex verify the deductions made about essential sequence information in the NH2-terminal region of ribonuclease. Further, the results emphasize the general usefulness of the conformational approach in designing simplified sequences for other peptides and proteins.