Positional-scanning combinatorial libraries of fluorescence resonance energy transfer peptides for defining substrate specificity of the angiotensin I-converting enzyme and development of selective C-domain substrates

Positional-scanning combinatorial libraries of fluorescence resonance energy transfer peptides were used for the analyses of the S3 to S1′ subsites of the somatic angiotensin I-converting enzyme (ACE). Substrate specificity of ACE catalytic domains (C- and N-domains) was assessed in an effort to design selective substrates for the C-domain. Initially, we defined the S1 specificity by preparing a library with the general structure Abz-GXXZXK(Dnp)-OH [Abz = o-aminobenzoic acid, K(Dnp) = Nε-2,4-dinitrophenyllysine, and X is a random residue], where Z was successively occupied with one of the 19 natural amino acids with the exception of Cys. The peptides containing Arg and Leu in the P1 position had higher C-domain selectivity. In the sublibraries Abz-GXXRZK(Dnp)-OH, Abz-GXZRXK(Dnp)-OH, and Abz-GZXRXK(Dnp)-OH, Arg was fixed at P1 so we could define the C-domain selectivity of the S 1′, S2, and S3 subsites. On the basis of the results from these libraries, we synthesized peptides Abz-GVIRFK(Dnp)-OH and Abz-GVILFK(Dnp)-OH which contain the most favorable residues for C-domain selectivity. Systematic reduction of the length of these two peptides resulted in Abz-LFK(Dnp)-OH, which demonstrated the highest selectivity for the recombinant ACE C-domain (kcat/Km = 36.7 μM -1 s-1) versus the N-domain (kcat/Km = 0.51 μM-1 s-1). The substrate binding of Abz-LFK(Dnp)-OH with testis ACE using a combination of conformational analysis and molecular docking was examined, and the results shed new light on the binding characteristics of the enzyme.