Boroleucine-Derived Covalent Inhibitors of the ZIKV Protease

Abstract

The Zika virus (ZIKV) remains a potential threat to the public health due to the lack of both an approved vaccination or a specific treatment. In this work, a series of peptidic inhibitors of the ZIKV protease with boroleucine as P1 residue was synthesized. The highest affinities with K_i values down to 8 nM were observed for compounds with basic residues in both P2 and P3 position and at the N-terminus. The low potency of reference compounds containing leucine, leucine-amide or isopentylamide as P1 residue suggested a covalent binding mode of the boroleucine-derived inhibitors. This was finally proven by crystal structure determination of the most potent inhibitor from this series in complex with the ZIKV protease.

Keywords: NS2B-NS3 protease; Zika virus; boroleucine; crystal structure determination; drug design.

Figure 1

Design of acyclic NS2B‐NS3 inhibitors containing a neutral boroleucine as P1 residue.

Figure 2

Crystal structure of the ZIKV protease (bZiPro) in complex with inhibitor 6 (PDB: 7ZNO). The surface of NS2B is shown in pink, that of NS3 in gray, and water molecules as red spheres. Polar contacts between inhibitor 6 and bZiPro are depicted as black dashed lines with distances in Å. Inhibitor 6 and selected residues of the ZIKV protease are presented as stick model with boron in green, nitrogen in blue, oxygen in red and carbon either in gray (NS3), pink (NS2B) or yellow (inhibitor 6). (A) Binding mode of inhibitor 6 showing the simulated‐annealing F_o‐F_c omit map (green mesh) contoured at 3.0 σ, a well‐defined electron density was only obtained for the P2‐P1 segment and the backbone of the P3 residue. (B) Stick model of inhibitor 6 and the NS3 residues interacting with the P1 boronate moiety. The simulated‐annealing F_o‐F_c omit map of inhibitor 6 (green mesh) is contoured at 3.0 σ, the respective 2 F_o‐F_c omit map of the NS3 residues (blue mesh) at 1.0 σ. The electron density maps indicate a covalent contact between the boronate moiety and the side chain of S135. (C) Interactions of inhibitor 6 with residues of the ZIKV protease. (D) Superimposition with inhibitor 1 in complex with bZiPro (PDB: 6Y3B). Carbon atoms of inhibitor 1 and certain bZiPro residues from this complex are given in turquoise. Of the respective proteases, only residues F84 (NS2B), D129 (NS3) and S135 (NS3) are shown. A model of the complete inhibitor 6 in complex with the ZIKV protease is provided as Figure S5.

Scheme 1

Synthesis of inhibitor 6. a) Standard Fmoc‐SPPS on 2‐CTC resin loaded with Fmoc‐Lys(Boc)‐OH, followed by subsequent couplings of Fmoc‐Lys(Boc)‐OH, Fmoc‐3‐aminomethylphenylacetic acid and Fmoc‐d‐Ala‐OH. Before each coupling step and after the coupling of Fmoc‐d‐Ala‐OH, the Fmoc group was removed with 20 % (v/v) piperidine in DMF. All couplings were performed with a threefold excess of the respective amino acid in presence of 3.0 equiv. HATU and 6.0 equiv. DIPEA; b) 3.0 equiv. N,N’‐Di‐Boc‐1H‐pyrazole‐1‐carboxamidine, 6.0 equiv. DIPEA, DMF, 18 h; c) 1 % TFA (v/v) in dichloromethane (4×30 min); d) 1.15 equiv. (R)‐boroleucine‐(1S,2S,3R,5S)‐(+)‐2,3‐pinanediol ester trifluoroacetate, 1.15 equiv. HATU, 4.0 equiv. DIPEA, DMF, 18 h; e) 4 N HCl in 1,4‐dioxane, 2 h, precipitation in diethyl ether, preparative HPLC; f) 2.0 equiv. isobutylboronic acid, 1 m aq. HCl, methanol/pentane 1 : 1, preparative HPLC. All reactions were performed at ambient temperature.