Improving the Selectivity of Engineered Protease Inhibitors: Optimizing the P2 Prime Residue Using a Versatile Cyclic Peptide Library

Abstract

Standard mechanism inhibitors are attractive design templates for engineering reversible serine protease inhibitors. When optimizing interactions between the inhibitor and target protease, many studies focus on the nonprimed segment of the inhibitor's binding loop (encompassing the contact β-strand). However, there are currently few methods for screening residues on the primed segment. Here, we designed a synthetic inhibitor library (based on sunflower trypsin inhibitor-1) for characterizing the P2' specificity of various serine proteases. Screening the library against 13 different proteases revealed unique P2' preferences for trypsin, chymotrypsin, matriptase, plasmin, thrombin, four kallikrein-related peptidases, and several clotting factors. Using this information to modify existing engineered inhibitors yielded new variants that showed considerably improved selectivity, reaching up to 7000-fold selectivity over certain off-target proteases. Our study demonstrates the importance of the P2' residue in standard mechanism inhibition and unveils a new approach for screening P2' substitutions that will benefit future inhibitor engineering studies.

A potent, proteolysis-resistant inhibitor of kallikrein-related peptidase 6 (KLK6) for cancer therapy, developed by combinatorial engineering

“…Notably, these subsites are largely shaped by KLK6 residues 39-41, representing a stretch of sequence that is highly unique to KLK6 and is poorly conserved between KLKs and other trypsin-like proteases. Indeed, the improved affinity of APPI-4M toward KLK6 versus other KLKs suggests that an optimization of the S2′-S4′ subsite contacts may represent a general strategy for developing selective KLK6 inhibitors (71)(72)(73).…”

Inhibitors of kallikrein‐related peptidases: An overview

“…By exchanging individual amino acids, it was possible to alter target specificity and design potent inhibitors of several KLK‐related peptidases. Whereas the natural SFTI‐1 possesses only weak to moderate inhibitory activities against KLKs, several groups reported that substituting preferred substrate sequences into the P4, P2, P1, and P2’ residues of SFTI‐1 (Table ) yielded several potent inhibitors of KLK5 ( K i = 2.1 nM), KLK4 ( K i = 39 pM), KLK7 ( K i = 0.8 nM), and KLK14 (0.4 nM) . Swedberg et al.…”

Design and Development of a Cyclic Decapeptide Scaffold with Suitable Properties for Bioavailability and Oral Exposure

“…Even if these molecules fall outside the Lipinski “rules of 5” and therefore represent a challenge regarding the optimization of their ADME properties, their size might be inadequate to tackle difficult targets and modes of action such as protein–protein interactions. Indeed, hexamers might have an appropriate molecular size if the target contains a deep pocket for ligand binding; however, if the activity‐relevant interface of the target consists of a flat region, their size might be a true limiting factor for high affinity ,. Therefore, we decided to investigate cyclic decapeptides, which we believed would be better able to interfere with such flat and large interfaces.…”