αβ,α′β′-Diepoxyketones are mechanism-based inhibitors of nucleophilic cysteine enzymes

Epoxides are an established class of electrophilic alkylating agents that react with nucleophilic protein residues. We report αβ,α′β′-diepoxyketones (DEKs) as a new type of mechanism-based inhibitors of nucleophilic cysteine enzymes. Studies with the l,d-transpeptidase LdtMt2 from Mycobacterium tuberculosis and the main protease from SARS-CoV-2 (Mpro) reveal that following epoxide ring opening by a nucleophilic cysteine, further reactions can occur, leading to irreversible alkylation.


X-ray crystallography
Recombinant LdtMt2 (Δ1-55; with the N-terminal His6-Tag removed, in 50 mM tris, pH 8.0, 100 mM NaCl) was crystallised using sitting drop vapor diffusion at 4 °C, according to a reported procedure. 6The inhibitor was introduced to the crystals through soaking (1.5 mM, 24 h), after which time the crystals were cryocooled and stored in liquid nitrogen.Datasets were collected using the MX beamline I03 at the Diamond Light Source synchrotron (Harwell, United Kingdom).Structures were solved by molecular replacement using Phaser 7 , using PDB entry 6RRM 8 as the search model.Alternating cycles of refinement using PHENIX 9 and manual model building using COOT 10 were performed until Rwork and RFree converged.Data collection and refinement statistics can be found in Table S2.Ligands were visualised by mFo-DFc polder OMIT map. 11gle crystal X-ray diffraction data were collected for 1 at 150 K using a (Rigaku) Oxford Diffraction SuperNova diffractometer and CrysAlisPro.The structure was solved using 'Superflip' 12 before refinement with CRYSTALS 13,14 as described in the SI (CIF).The crystallographic data have been deposited with the Cambridge Crystallographic Data Centre (CCDC 2262059), and copies of these data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Preparation of recombinant BlaC protein
A codon-optimised synthetic gene (GeneArt, Thermo Fisher Scientific) encoding for BlaC Δ1-40 was amplified and cloned into the expression vector pCold using Sal1-HF (New England BioLabs) and Not1-HF (New England BioLabs) digestion and ligation using T4 DNA ligase (New England BioLabs) according to the manufacturer's protocol.The ampicillin resistance gene of the vector was exchanged for the kanamycin resistance gene using Gibson Assembly, 15 and transformed with Escherichia coli BL21(DE3).
The cell pellet was resuspended in HisTrap Buffer A (25 mM Tris-HCl pH 8.0, 500 mM NaCl, 0.5 mM tris(2-carboxyethyl)phosphine) (TCEP), 5% (v/v) glycerol, 20 mM imidazole) in the presence of DNase I, and lysed using a Continuous Flow Cell Disruptor (Constant Systems, 20 kpsi).The lysates were centrifuged (32,000 x g, 20 min), passed through a 0.45 µm filter, and loaded onto a 5 mL HisTrap column (GE Life Sciences) that had been pre-equilibrated in HisTrap Buffer A. The column was washed with HisTrap Buffer A, followed by a gradient running from 0 % to 100 % (v/v) HisTrap Buffer B (25 mM Tris-HCl pH 8.0, 500 mM NaCl, 0.5 mM TCEP, 5% (v/v) glycerol, 250 mM imidazole).Fractions containing BlaC (as observed by SDS-PAGE) were combined, the buffer was exchanged to HisTrap Buffer A, and the HisTag was cleaved using recombinant 3C protease at 4 °C, over 12 h.The HisTag cleaved BlaC was passed through a 5 mL HisTrap column (GE Life Sciences) and washed with HisTrap Buffer A. The BlaC containing fractions (as observed by NanoDrop ( Thermo Scientific) analysis) were concentrated and loaded onto a 300 mL Superdex 75 column (GE Life Sciences) pre-equilibrated in gel filtration buffer (25 mM Tris-HCl pH 8.0, 500 mM NaCl, 0.5 mM TCEP, 5% (v/v) glycerol).BlaC was eluted using the gel filtration buffer.Fractions containing BlaC (as observed by SDS-PAGE) were combined, concentrated, and frozen using liquid nitrogen.The identity and purity of BlaC was confirmed by mass spectrometry (calculated mass 28740 Da, observed deconvoluted mass 28740 Da) and SDS-PAGE (>95% purity).

Procedure for the synthesis of the diene precursors of 1 and 4-11 (General Procedure A)
A modified literature procedure was followed to prepare the αβ,α′β′-dienone precursors of 1 and 4-11, i.e. 13-18. 16To a neat solution of lithium perchlorate (LiClO4) (20 mmol, 2 equiv.),benzaldehyde (or a benzaldehyde derivative, as specified; 20 mmol, 2 equiv.), and the requisite ketone (10 mmol, 1 equiv.)was added triethylamine (Et3N) (0.3 mL, 2 mmol, 0.1 equiv.).The mixture was stirred at room temperature (rt) and the reaction progress was monitored by thin layer chromatography (TLC).Upon completion of the reaction, a saturated aqueous ammonium chloride (NH4Cl) solution was added, and the resulting mixture was extracted with dichloromethane.The organic extracts were dried over anhydrous sodium sulfate (Na2SO4) and concentrated under reduced pressure.The crude mixture was purified by flash column chromatography.

Procedure for the synthesis of DEKs 1 and 4-11 (General Procedure B)
A modified literature procedure was followed to prepare αβ,α′β′-diepoxide ketones 1 and 4-11. 17To a stirred suspension of potassium fluoride supported on alumina (KF-Al2O3) (prepared as described 18 ) in tert-butyl hydroperoxide ( t BuOOH; 5.0-6.0M in decane, 3.8 mL, ~21 mmol) under N2 was added a solution of the specified αβ,α′β′-dienone (2 mmol) in anhydrous acetonitrile (10 mL).The mixture was stirred at rt, and the reaction progress was monitored by TLC.Upon completion of the reaction, the mixture was filtered with a sintered funnel in vacuo.The filtrate was washed with brine and extracted with ethyl acetate.The combined organic extracts were dried over anhydrous Na2SO4 and concentrated in vacuo.The crude residue was purified by recrystallisation from ethanol to afford the trans/trans-αβ,α′β′-diepoxide ketone.
b Diastereomeric ratio (trans,trans:cis,cis) following purification, as determined by 1 H NMR analysis.

356
Average (kinact/KI)inhibitor values and compound structures are given in Table S3.S3.  followed by retro-aldol reaction (blue outline).Initial reaction at the DEK carbonyl followed by rearrangement is also possible (Figure 2C).The fragmentation product 3 (Figure 2) obtained after retro-aldol reaction was sometimes (as indicated by compound numbers) observed (by MS analysis) to further react to give: (i) the corresponding hydrolysis product, likely through hydrolytic ring opening of its epoxide (yellow outline), or (ii) a dehydroalanine (Dha) residue (green outline).Another unassigned product corresponding to a mass shift of +57 Da compared to the unmodified enzyme was (sometimes) observed (red outline).Among other possibilities, the +57 Da fragmentation product(s) may arise through reaction with another nucleophilic residue in the active site, leading to a cross-linked adduct.S1 for 1 and 9 for reaction with LdtMt2.S3.

Figure S2 .
Figure S2.Protein observed SPE-MS analysis for the reaction of LdtMt2 with DEKs 1 and 4-11 and mono-epoxide ketone 12. 1 µM LdtMt2 was incubated with the inhibitors (20 µM for 1 and 4 -11, 100 µM for 12) at rt in 50 mM Tris, pH 7.5.Samples were analysed after the indicated times.Deconvoluted spectra, obtained using the maximum entropy algorithm in the MassHunter Workstation Qualitative Analysis B.07.00 program (Agilent), are shown.Mass shifts and assignments are given in TableS1.

Figure S3 .
Figure S3.Stability of DEK 1 in aqueous solution.A solution of DEK 1 (250 µM) in 50 mM tris-d11, pH 7.5, 10% D2O, was analysed by 1 H-NMR (950 MHz) for up to 12 h.No changes in peaks corresponding to 1 (in blue and green) in spectra were observed.Buffer and solvent peaks are in grey.

Final 1 (
Figure S6.Dose-response curves for the pure trans,trans isomer of 1 and an isomeric mixture of 1 (trans,trans:cis,cis of ~1:3).Inhibition assays were carried out using 100 nM LdtMt2 and 25 µM Probe 1 with 15 min pre-incubation at rt in 50 mM HEPES, pH 7.2 with 0.01% (v/v) Triton X-100.Error bars represent the standard deviation (n=4).Note that pIC50 values are similar, but imply that the pure trans,trans isomer of 1 is the most active stereoisomer.

362 363Figure S9 .
Figure S9.LCMS studies of the reaction between GSH and 1. A. A solution of 1 (250 µM) and GSH (250 µM) in 50 mM tris pH 7.5 was incubated for 16 h in the presence of TCEP (250 µM), then analysed by LCMS operating in the positive ion mode.B. The compounds eluting at 0.78 min and 2.82 min correspond to unreacted GSH and GSH reacted with 1 (apparently leading to the fragmented species analogous to 3), respectively.

Figure S10 .
Figure S10.Summary of the reactions of DEKs with nucleophilic cysteine enzymes.The DEKs are proposed to inhibit nucleophilic cysteine enzymes via initial reaction of the nucleophilic cysteine with one of the epoxides,

Figure
Figure S11.Protein-observed SPE-MS based Cys354 selectivity assays.LdtMt2 (1 µM) was preincubated with ebselen (a reported LdtMt2 inhibitor8 which reacts with the nucleophilic Cys354, 10 µM) for 1 h in 50 mM Tris, pH 7.5.Inhibitors 1 and 4 -10 (100 µM) were then added and samples were analysed after an additional 24 h incubation at room temperature using SPE-MS.The spectrum in white corresponds to LdtMt2 reacted with ebselen.The spectrum in light grey corresponds to LdtMt2 reacted with the specified inhibitor, following preincubation with ebselen.The spectrum in dark grey corresponds to LdtMt2 reacted with the specified inhibitor.Deconvoluted spectra, obtained using the maximum entropy algorithm in the MassHunter Workstation Qualitative Analysis B.07.00 program (Agilent), are shown.

Figure S13 .
Figure S13.Protein observed SPE-MS analysis for the reaction of SARS-CoV-2 M pro with DEKs 1 and 9. M pro (2 µM) was incubated with DEKs 1 and 9 (20 µM) at rt in 20 mM HEPES, pH 7.5.Samples were analysed after the indicated times.Deconvoluted spectra, obtained using the maximum entropy algorithm in the MassHunter Workstation Qualitative Analysis B.07.00 program (Agilent), are shown.Mass shifts and assignments are analogous to those described in TableS1for 1 and 9 for reaction with LdtMt2.

Table S1 . Calculated and observed masses (Da) and mass shifts (Da) for protein-observed SPE-MS experiments 309 with LdtMt2 and inhibitors 1 and 4-12, and their assignments.
Mass shifts are relative to unmodified LdtMt2.The

Table S2 . Data collection and refinement statistics for the crystal structure of LdtMt2 reacted with 1.
$ RMS = root mean square.*Highest resolution shell in parentheses.