key: cord-0924147-w4dht8s7
authors: Malla, Tika R.; Tumber, Anthony; John, Tobias; Brewitz, Lennart; Strain-Damerell, Claire; Owen, C David; Lukacik, Petra; Chan, H. T. Henry; Maheswaran, Pratheesh; Salah, Eidarus; Duarte, Fernanda; Yang, Haitao; Rao, Zihe; Walsh, Martin A.; Schofield, Christopher J.
title: Mass spectrometry reveals potential of β-lactams as SARS-CoV-2 M(pro) inhibitors
date: 2021-02-11
journal: Chem Commun (Camb)
DOI: 10.1039/d0cc06870e
sha: a2a2235164161383d97c81abd0efc924febdb8c0
doc_id: 924147
cord_uid: w4dht8s7

The main viral protease (M(pro)) of SARS-CoV-2 is a nucleophilic cysteine hydrolase and a current target for anti-viral chemotherapy. We describe a high-throughput solid phase extraction coupled to mass spectrometry M(pro) assay. The results reveal some β-lactams, including penicillin esters, are active site reacting M(pro) inhibitors, thus highlighting the potential of acylating agents for M(pro) inhibition.

The reported pGFX-6p1-M pro plasmid 1 was used for M pro protein production. 2 In brief, the pGFX-6p1-M pro plasmid was transformed into a competent Escherichia coli expression cell line based on the BL21(DE3)-R3-pRARE. Multiple colonies were used to inoculate an LB media starter culture supplied with 100 µg/mL carbenicillin. The starter culture was incubated (37°C, 200 rpm) until an OD600 ≤ 2 was reached (~8 h). 1% (v/v) of the starter culture was used to inoculate pre-warmed autoinduction medium (Formedium, Terrific Broth base including trace elements) supplemented with 1% (v/v) glycerol and 50 µg/mL carbenicillin. Cells were grown at 37°C (200 rpm) for 5 h and at 18°C overnight (200 rpm). Cells were harvested by centrifugation (10 krpm, 30 min, 4°C).

The resulting cell pellet was suspended in lysis buffer (50 mM Tris, pH 8, 300 mM NaCl, 10 mM imidazole, 0.03 μg/ml benzonase) at a ratio of 1:3 (w/v). Cells were lysed using an Emulsiflex high-pressure homogeniser (Avestin Inc) (3 passes, 30 kpsi, 4°C). The lysate was centrifuged (50000 g, 30 min-60 min, 4°C) and filtered (0.22 μm). The filtrate was incubated with 10ml 50% (v/v) pre-equilibrated His60 Ni Superflow resin (Clontech Laboratories) (with agitation, 1 h, 4°C) and then applied to a gravity flow column. The column was washed (50 mM Tris pH 8, 300 mM NaCl, 25 mM imidazole), and the M pro protein was eluted with elution buffer (50 mM Tris, pH 8, 300 mM NaCl, 500 mM imidazole). To remove the M pro poly-histidine tag, the N-terminal His tagged HRV 3C protease was added to the eluted M pro fractions (1:10 (w/w) protease:M pro ). The mixture was dialysed overnight into 50 mM Tris pH 8, 300 mM NaCl, 0.5 mM TCEP at 4 °C. Mpro was purified by reverse His purification using a 5 mL HisTrap FF column (GE Healthcare) eluting with wash buffer (50 mM Tris pH 8, 300 mM NaCl, 25 mM imidazole). Purified M pro fractions were pooled and concentrated (< 5mL) using an Amicon® Ultra 10 MWCO centrifugal filter unit (Merck Millipore). The concentrate was loaded onto a S200 pg 16/60 size exclusion column (GE Healthcare), eluting at a flow rate of 1 mL/min (50 mM Tris pH 8, 300 mM NaCl). M pro fractions were concentrated to 36 mg/mL using a 10 kDa MWCO centrifugal filter. M pro was >95% pure by SDS PAGE analysis and had the anticipated mass of 33786 ± 1 Da (calculated: 33796.64 Da).

The 11-mer M pro substrate peptide TSAVLQ/SGFRK-NH2, corresponding to the N-terminal self-cleavage site of M pro on pp1a/b, was synthesised as a C-terminal amide by microwave-assisted solid-phase peptide synthesis (SPPS) as reported. 3 In brief, the M pro substrate peptide was synthesized on a 0.1 mmol scale from the C-to N-terminus on Rink amide-MBHA resin (100-200 mesh, 0.6-0.8 mmol. g −1 loading, AGTC Bioproducts) using a LibertyBlue peptide synthesizer (CEM) and N-Fmoc protected α-amino acids (CS Bio, Novabiochem, Sigma-Aldrich, TCI, Alfa Aesar, Merck or AGTC Bioproducts). A mixture of N,N'-diisopropylcarbodiimide (TCI Europe) and Oxyma Pure (Merck) in DMF was used for coupling and 20% (v/v) piperidine in DMF (peptide synthesis grade, AGTC Bioproducts) for Fmoc deprotection. Iterative cycles of peptide coupling and deprotection were performed using the manufacturer's protocol. After the final Fmoc-deprotection step, the resin was washed with CH2Cl2, dried in air, then treated with 5 mL of the deprotection solution (2.5%:2.5%:2.5%:92.5%) (v/v) of 1,3dimethoxybenzene, triisopropylsilane, MilliQ water and trifluoroacetic acid) (4 h at ambient temperature). The resulting mixture was filtered; the filtrate was diluted with cold Et2O (45 mL) to precipitate the peptide. The resulting suspension was centrifuged (4255 g, 10 min, 4.0 °C) and the supernatant decanted. The solid peptide was dried in air, dissolved in MilliQ water, then lyophilised. The peptide was purified by reverse phase HPLC (20 mL/min; linear gradient over 10 min: 2%→20% acetonitrile in water, each containing 0.1% (v/v) formic acid; tR = 8 min) using a Shimadzu HPLC purification system (composed of DGU-20A, 2 LC-20AR, CBM-20A, SPD-20A, and FRC-10A units) equipped with a NX-C18 LC column (250 × 21.2 mm, 110 Å; Phenomenex Gemini).

The purified peptide was dissolved in DMSO and its concentration was determined by 1 H NMR using an AVIII 700 machine (Bruker) equipped with a 5 mm 1 H( 13 C/ 15 N) inverse cryoprobe: 16 µL of the DMSO peptide solution were added to 143 µL DMSO-d6 and 1 µL of 1 mg/mL 3-(trimethylsilyl)propionic-2,2,3,3-d4 acid sodium salt (TSP; Apollo Scientific) (Acquisition parameters: d1 = 30 s, ns = 256). Peaks were assigned in MestreNova using the global spectrum deconvolution algorithm. The peptide concentration was then calculated using the following equation 4 :

where, is the molar ratio of TSP and the 11-mer substrate peptide respectively is the ratio of signal intensity corresponding to the methyl protons of TSP and the 11-mer substrate peptide (averaged for the Ala, Val, and Leu residues) respectively, and is the number of nuclei responsible for the signal intensity for 11-mer substrate (averaged for the Ala, Val, and Leu residues) and TSP respectively.

Peptide aliquots were prepared and stored at -20°C.

Syntheses of the M pro inhibitors N3, 1 1, 5 2, 6 3, 7 4, 8 and 5 9 have been reported.

M pro MS assays were performed using freshly thawed M pro ; thawing was not repeated because refreezing and thawing M pro reduces M pro activity, so affecting assay robustness and reproducibility. Freshly thawed M pro was diluted to 15 µM in freshly prepared 20 mM HEPES (Gibco) buffer pH 7.5, 300 mM NaCl for all MS assays with the exception of protein observe MS assays, for which M pro was diluted to 1 µM in freshly prepared reaction buffer (20 mM HEPES pH 7.5, 50 mM NaCl). LCMS (Merck, Supelco) grade solvents were used for MS analyses and to prepare buffers. All assays were performed in 384 well polypropylene plates (Greiner) in the reaction buffer. All compounds were purchased from Sigma Aldrich, Apollo and Fluorochem. All compounds were dry dispensed using an Echo 550 acoustic dispenser (LabCyte) unless otherwise stated. The substrate peptide and the enzyme was diluted in the reaction buffer to prepare stock solutions, 2 fold concentrated relative to the final concentration in the assay. M pro , peptide substrate, and formic acid solutions were all dispensed across 384 well plates using a Multidrop Combi dispenser (Thermo Scientific), the plates were then centrifuged (500 x g, 15s, Axygen Plate Spinner Centrifuge, Corning).

M pro activity was assayed using a RapidFire (RF) 365 high-throughput sampling robot (Agilent) connected to an iFunnel Agilent 6550 accurate mass quadrupole time-of-flight (Q-TOF) mass spectrometer. The spectrometer was operated in the positive ion ionisation mode with following operating parameters for all assays: capillary voltage (4000 V), nozzle voltage (1000 V), fragmentor voltage (365 V), drying gas temperature (280 °C), gas flow (13 L/min), sheath gas temperature (350 °C), sheath gas flow (12 L/min); except for protein observe MS assays, for which the gas temperature was reduced to 225 °C. The peptide/protein sample was loaded onto a solid-phase extraction (SPE) C4-cartridge, which was then washed with 0.1% (v/v) aqueous formic acid to remove non-volatile buffer salts (5.5 s, 1.5 mL/min) and with aqueous 85% (v/v) acetonitrile containing 0.1% (v/v) formic acid (5.5 s, 1.25 mL/min) to elute the peptides/protein. The cartridge was re-equilibrated with 0.1% (v/v) aqueous formic acid (0.5 s, 1.25 mL/min). Each sample aspiration step was followed by an aqueous wash before the next protein sample was injected onto the SPE cartridge.

The kinetic parameters of M pro were determined by monitoring M pro -catalysed cleavage of TSAVLQ/SGFRK-NH2 over time using 16 different peptide concentrations (the final DMSO concentration was >1% (v/v) in each experiment, Figure S1 ). The peptide was diluted (1:100) with the reaction buffer (20 mM HEPES, pH 7.5, 50 mM NaCl) to 0.5 mL reaction volume. Initially, a no-enzyme control was analysed using SPE-MS before M pro was added to the mixture (0.15 µM final concentration, based on the M pro monomer mass) using SPE-MS.

RapidFire integrator software (Agilent) was used to extract and integrate the m/z (+1) charge states of both the substrate peptide (1191.67 Da) and the N-terminal product peptide (TSAVLQ; 617.34 Da). The basic C-terminal cleavage product (SGFRK-NH2was not detected, most likely due to poor retention on SPE cartridge under these conditions and hence was not included in subsequent calculations. The percentage conversion (product peak integral/(product peak integral + substrate peak integral)*100) was calculated in Microsoft Excel. The slopes of the initial reaction rate was fitted with the Michaelis-Menten equation (Et= 0.15 µM) using non-linear regression (GraphPad Prism 8) to obtain the Km and kcat-values ( Figure S1 ). The experiments were performed in independent duplicates (n = 2; mean ± standard deviation, SD).

The final compound assay concentrations were 20 µM, 50 µM, or 100 µM; DMSO and ebselen were used as negative and positive inhibition controls, respectively. M pro (0.30 µM) was dispensed across the plate (25 µL/well); the resulting mixture was incubated for 30 or 60 minutes at ambient temperature. TSAVLQ/SGFRK-NH2 peptide (4 µM) was dispensed to the mixture (25 µL/well), which was incubated (10 minutes), then quenched by addition of 10% (v/v) aqueous formic acid (5 µL/well). The individual reactions were analysed using SPE-MS. Data were extracted and processed as described for the kinetic M pro assays above and the normalized percentage inhibition was determined in Microsoft Excel with respect to the positive and negative inhibition controls. Errors values were determined with respect to technical duplicates (n = 2; mean ± SD).

M pro inhibitors were dry dispensed in an 11 point 3 fold dilution series (100 µM top concentration). Ebselen and DMSO were used as positive and negative inhibition controls, respectively. M pro (0.30 µM) was dispensed across the plate (25 µL/well) and the resulting mixture was incubated (30 or 60 minutes) at ambient temperature. TSAVLQ/SGFRK-NH2 peptide solution (4 µM) was then dispensed to the mixture (25 µL/well). Reactions were incubated (10 minutes), then quenched by addition of 10% (v/v) aqueous formic acid (5 µL/well). Data were extracted and processed as described for the kinetic M pro assays above. IC50 curves were generated using non-linear regression and normalized with respect to the positive and negative inhibition controls (GraphPad Prism 8). IC50-values are reported as the mean of two independently determined IC50-curves each composed of technical duplicates (n = 2; mean ± SD). Signal to noise (S/N) and Z'-factor were calculated in Microsoft Excel for each plate analysed. 8

For the initial screening of β-lactams, β-lactam solutions (10 mM in DMSO) were transferred in quadruplicate from a 96 well plate to a 384 well plate (1 µL/well) using a CyBio Liquid Handler (Analytik Jena AG). M pro (1.0 µM) in the reaction was dispensed across the plate (100 µL/well) using a Multidrop Combi dispenser (Thermo Scientific) and the resulting mixture was incubated for a minimum of 1 h, 5 h, 19 h, and 24 h at ambient temperature before analysis using SPE-MS.

For selected M pro inhibitors identified from the initial screening of β-lactams or from the screening of small-molecule libraries (LOPAC 1280 (Sigma Aldrich) or the small-molecule library containing FDA approved compounds (Pharmacon) for human use), inhibitor solutions (20 mM in DMSO) were diluted to 1 mM with the reaction buffer and transferred to a 384 well plate (2 µL/well). M pro (1.0 µM) was dispensed across the plate (100 µL/well) using a Multidrop Combi dispenser (Thermo Scientific) and the resulting mixture was incubated for a minimum of (5 min and 1 h) or (25 min, 1 h , 16 h and 24 h) at ambient temperature before analysis using SPE-MS.

To investigate whether the identified M pro inhibitors selectively modify the M pro active site Cys-145, M pro (1 µM) was incubated with the reported Cys-145-selective M pro inhibitor N3 1 (3 µM) on ice for 3 h in 7.5 mL reaction buffer (20 mM HEPES, pH 7.5, 50 mM NaCl). The inhibitor of interest (20 mM in DMSO) was diluted to 1 mM with the reaction buffer and transferred to a 384 well plate (2 µL/well). M pro /N3 mixture was dispensed to the inhibitor of interest on a 384 well plate (100 µL/well) and the resulting mixture was incubated at ambient temperature for 5 min and 16 h, respectively, before analysis using SPE-MS.

Protein observed mass spectra were acquired (at least) in independent duplicates using SPE-MS with the instrument parameters as stated above. Only one representative of the data are shown as all duplicates were apparently identical. The reactions were not quenched, thus, each subsequent sample was injected with a delay from the primary injection time (stated in the individual figures). Protein spectra were deconvoluted (m/z range: 10-60 kDa, m/z range 850-1350 Da, mass step: 1 Da) using the MaxEnt1 function in Agilent MassHunter Version 7. The deconvoluted files were extracted, normalised and plotted (GraphPad Prism 8). µM. Each data point represents the mean of two independent repeats (n=2 ± standard deviation (SD)). Note that interconversion between monomeric and dimeric forms of M pro may cause complexity in its kinetics. Refer to Experimental Section for assay details. Table S1 . Comparison of the M pro kinetic parameters obtained using SPE-MS with those reported. The kinetic parameters of SARS-CoV-2 M pro determined using SPE-MS (Entries 12 and 13) are in the range of those recently reported for SARS-CoV-2 M pro . Note that the catalytic efficiency of SARS-CoV-2 Mpro is slightly higher than that for SARS-CoV, possibly due to the apparent close packing of M pro domain III in SARS-CoV-2 M pro . 10 SARS-CoV and SARS-CoV-2 M pro share 98% sequence identity. 11 Note that the N-terminal residue (Ser-1) has been shown to be important for M pro dimerization. 

Cys-38

Cys-85

Cys-128

Cys-160

Cys-300 RMSD 0.08 0.13 0.04 0.14 0. RMSD values were calculated for specified residue using least square fitting for all atoms of the specified residue in COOT (Table S2) . factors were calculated for all 5 assay plates according to the literature. 18 Both values indicate that the assay is of high quality and robustness. Compounds that exhibited ≥80% inhibition in LOPAC 1280 library and library of FDA approved small molecules are shown in Table S2 and Table S3 . Table S5 . Several β-lactam M pro inhibitors identified from library screen ( Figure S7) . The observed inhibition of M pro by cephalosporin C zinc salt is likely Zn(II) ion mediated(at least predominantly) ( Figure S16) . 24 . For cefotetan +576 Da adducts, cefbuperazone +627 Da adducts and cefoxitin +427 Da adducts, evidence for more than one reaction was accrued. For cefazolin, an additional +338 Da adduct was observed. For M pro incubated with moxalactam, an additional peak with mass shift (+476 Da) corresponding to -45 Da loss relative to the moxalactam adduct (+520 Da) was observed, possibly due to decarboxylation of moxalactam. as reported in MS analysis of carbapenems 28 and other β-lactams 29 . The structures of the inhibitors and possible outcomes of reaction with a nucleophilic cysteine are shown. Refer to Experimental Section for assay details.

Enzyme Structure