key: cord-0813093-8mqewhht
authors: Chen, Junjie; Zhang, Yaya; Zeng, Dequan; Zhang, Bingchang; Ye, Xiaohong; Zeng, Zhiping; Zhang, Xiao-kun; Wang, Zhanxiang; Zhou, Hu
title: Merbromin is a mixed-type inhibitor of 3-chyomotrypsin like protease of SARS-CoV-2
date: 2021-12-30
journal: Biochem Biophys Res Commun
DOI: 10.1016/j.bbrc.2021.12.108
sha: 6a847f103159aa7b49d2009fbdb18ecc513ecae4
doc_id: 813093
cord_uid: 8mqewhht

3-chyomotrypsin like protease (3CLpro) has been considered as a promising target for developing anti-SARS-CoV-2 drugs. Herein, about 6000 compounds were analyzed by high-throughput screening using enzyme activity model, and Merbromin, an antibacterial agent, was identified as a potent inhibitor of 3CLpro. Merbromin strongly inhibited the proteolytic activity of 3CLpro but not the other three proteases Protease K, Trypsin and Papain. Michaelis-Menten kinetic analysis showed that Merbromin was a mixed-type inhibitor of 3CLpro, due to its ability of increasing the K(M) and decreasing the K(cat) of 3CLpro. The binding assays and molecular docking suggested that 3CLpro possessed two binding sites for Merbromin. Consistently, Merbromin showed a weak binding to the other three proteases. Together, these findings demonstrated that Merbromin is a selective inhibitor of 3CLpro and provided a scaffold to design effective inhibitors of SARS-CoV-2.

Coronavirus disease 2019 (COVID- 19) is an acute viral pneumonia, which was caused by infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Up till now, more than 258 million people were infected, and the number of the infection cases is still climbing fast. SARS-CoV-2 originally named 2019-nCoV was determined to be a new member of the coronavirus family [1] . Despite genetic differences among SARS-CoV-2, SARS-CoV, and Middle East Respiratory Syndrome coronavirus (MERS-CoV), these coronaviruses have conservative function proteins, such as 3-chyomotrypsin like protease (3CLpro, also known as M pro or nsp5 protease) [2, 3] . 3CLpro is essential for the maturation of PP1a and PP1ab during virus replication, enabling it as a valuable target for the development of anti-coronavirus drugs [4, 5] . Many compounds have been found as potent inhibitors of 3CLpro [6] , however, there are no specific and effective drugs targeting SARS-CoV-2's 3CLpro in clinical application for COVID-19 treatment so far. 3CLpro functions as a main protease responsible for the maturation of the viral nonstructural proteins (nsps) by cleaving the large polyprotein at 11 sites from nsp5 to nsp16. The junctions of those nsps show an extensive homology and provide a good recognition site for 3CLpro [7, 8] . Accordingly, synthetized peptides based on the amino acid sequences of the junctions were used as the substrate to evaluate the hydrolytic activity of 3CLpro in vitro [9] . This enzymatic assay system has been successfully applied for the screening of inhibitors that block the cleavage function of 3CLpro [6, 10] . In this study, an in vitro screening model based on proteolytic activity of 3CLpro was constructed and verified. After screening ~6000 compounds, an antibacterial agent Merbromin was identified as a potent and selective inhibitor of 3CLpro.

J o u r n a l P r e -p r o o f

Substrate of 3CLpro, MCA-AVLQ↓SGFR-Lys(Dnp)-Lys-NH2, was synthesized by GL Biochem Ltd (Shanghai, China). Amine Coupling Kit (Product code:BR100050) and Series S Sensor Chip CM5 (Product code:BR100530) used for SPR assay were purchased from Cytiva Ltd (United Kingdom). Glutathione-Sepharose 4B column and thrombin were purchased from Amersham Pharmacia Biotech (Sweden). Isopropyl-β-d-thiogalactoside (IPTG), Protease K, Papain and Trypsin were purchased from BBI Life Sciences Corporation (Shanghai, China).

The hydrolytic activity of 3CLpro was detected by using substrate MCA-AVLQ↓SGFR-Lys(Dnp)-Lys-NH2 [6, 10, 11] . Experiments were performed on an opti384-well plate (Black, PE) with 60 μL of assay buffer (50 mM HEPES, pH 7.5, 0.1 mg/ml BSA, and 0.01% Triton X-100). Substrate of 3CLpro was used for the detection of Protease K, Trypsin, Papain.

Enzyme concentration was initially designed to be 0. 2 [11, 12] .

Herein, KD represents the dissociation constant of the dimer to monomer, and is defined by measured. The concentration of substrate was more than 3 times of KM in order to generate the Vmax in the initial phase of reaction. The value of KD was calculated by using the nonlinear regression according to the plot of Vmax to Ct. IC50 was obtained according to the following logistic regression function [13, 14] , Eqn.3:

where A1 is the inhibition rate without inhibitor, A2 is the ideal constant of complete inhibition, x is the concentration of the inhibitor, y is the inhibition rate, and p is the Hill factor. 

where Alpha, Ki and Ic represent the mixed constant, inhibition constant and concentration of inhibitor, respectively.

ITC assay was performed on a ITC200 (MicroCal) at 25 ℃. HEPES buffer (10 mM, pH 7.4) with 2% DMSO was used as the reaction buffer. Initially, the solution of compound (2mM) was pre-loaded on the syringe, and was then sequentially injected into the reaction chamber containing 40 μM of proteases (3CLpro, Protease K, Papain or Trypsin) by the automatic titration apparatus. Interval between each titration was set to 120 s to allow the heat signals back to baseline. Control assay was performed with the identical titration process except for the titrated buffer without proteases. All the experimental data were calibrated by subtracting the data of control, and the data fittings were executed on the commercial analysis software of ITC200.

SPR assay was performed on a BIAcore T200 machine (General Electric). 3CLpro protein (20 μg/mL) prepared in sodium acetate buffer (10 mM, pH 4.0) was immobilized on a CM5 chip with 16,000 RU by amino coupling method. Merbromin was diluted in the running buffer Fitting of kinetic mode in the software of the BIAcore T200.

The fluorescence titration was performed on an Eclipse fluorescence spectrophotometer (Varian). The excitation wavelength was set to 280 nm and the emission wavelengths were set between 290 and 480 nm. Protein solution (1.0 μM, 2 mL) prepared in HEPES buffer (10 mM, pH 7.4) was disposed by a continuous injection of Merbromin, with the final ratios of Merbromin to protein from 1 to 12. Data were analyzed according to the previous report [16] .

Molecular docking studies of the binding between Merbromin and 3CLpro were performed by using the Molecular Operating Environment (MOE) software. Compounds were prepared according to partial charge, energy minimization, and geometry optimization before simulate docking. The 3D structure of 3CLpro (pdb 6m2n) downloaded from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB) was processed by "ignore waters", "build hydrogens" and "protonate 3D" under the condition of pH of 7.0 and salt value of 0.1. Two docking regions consisting of the catalytic area and junction of dimer were predicted by the "site finder" procedure before docking. Docking process was launched by "dock" procedure with "induced fit" mode. Briefly, placement was set to "triangle matcher" J o u r n a l P r e -p r o o f mode, and london dG was used to rescoring. To gain more precision results, forcefield was also selected for refining, and affinity dG was used for rescoring. Finally, thirty docking poses were generated, and the best docking was selected as an outcome.

The procaryotic expression plasmid of 3CLpro was constructed and transformed into E.coli BL21. As shown in Supplemental Figure 1 , the expression of GST-3CLpro fusion protein was highly induced by IPTG in E.coli. After purification by affinity chromatograph and cleavage by thrombin, 3CLpro protein with more than 95% purity was obtained.

In the enzyme reaction system containing 0.2 μM of 3CLpro and 0-30 μM of substrate, the hydrolysis rate of 3CLpro at low substrate concentrations increased almost in a linear fashion with increasing substrate concentration, but gradually reached to a steady-state at very high substrate concentrations, exhibiting a typical Michaelis-Menten (MM) curve (Fig. 1A) .

Through the nonlinear fitting procedure based on MM equation, the enzymatic KM and Kcat were determined to be 8.9 ± 0.8 μM and 0.28 ± 0.04 min -1 , respectively.

It has been reported that the dimer rather than the monomer is the active form of 3CLpro [12, 17] . We then determined the composition of 3CLpro dimer in our reaction system. The proteolytic activity was measured in the presence of 30 μM substrate and different concentrations of 3CLpro. As shown in Figure 1B 

The feasibility of the model used for the high-throughput screening (HTS) assay was determined by evaluating the signal to noise ratio (SNR) and the Z′ factor. A statistical analysis between the blank (negative) and 3CLpro (positive) groups was performed under the reaction condition of 0.2 μM of enzyme and 20 μM of substrate according to the previous report [6] .

Our result showed that the model had a reproducible SNR of 9.1-12.5 and a feasible Z′ factor (0.62-0.9) [18] , meeting the requirements for HTS assay (Supplemental Figure 2A) . The Z' factor between the 3CLpro group and the group of 3CLpro with Ebselen, a reported inhibitor of 3CLpro [6] , was 0.58-0.71, indicating that Ebselen can be used as a positive control for screening 3CLpro inhibitors in this assay [18, 19] . Based on our above investigation and the previous report [6] , the reaction condition of 0.2 μM of 3CLpro and 20 μM of substrate was chosen for the subsequent HTS experiment. After screening ~6,000 compounds, we found 33 compounds as potential 3CLpro inhibitors with more than 95% inhibition rate at 16.7 μM (Supplemental Figure 2B) . Consistent with previous reports [6] , Ebselen, Disulfiram, PX-12, Shikonin, Tideglusib and TDZD-8 were screened out, indicating the reliability of our HTS system.

Among the screened potential inhibitors, Merbromin as a clinic drug intrigued us, and its IC50 was determined to be 2.7 ± 0.17 μΜ (Fig. 2A) , comparable to the reported IC50 of PX-12 (0.73 μΜ) and (+)-Shikonin (1.78 μΜ) [6] . However, Merbromin exhibited weak inhibiting effect on Papain with an IC50 of 8.03 μΜ, and no significant effects on Trypsin and Protease K with the estimated IC50 values more than 70 μΜ (Fig. 2B-D) . Thus, Merbromin is of certain selectivity in inhibiting 3CLpro.

We further investigated the inhibition type of Merbromin. As shown from the scatterplots in 

We further investigated the binding of Merbromin to 3CLpro by ITC assay. The ITC reaction 1 (upper panel, Fig. 3A ). This kind of thermal trend generally reflected the interaction between the two titrating molecules. The integral of each exothermic change against the ratio of Merbromin/3CLpro was plotted (lower panel, Fig. 3A) . Interestingly, although the scatters presented a sigmoid type, they could not gain a suitable fitting by using traditional one-site binding model. We found that the exothermic change was not a simple decreasing process during the whole titration and it showed a slightly increasing tendency at the first five titrations. This implied that there were two binding sites of 3CLpro for Merbromin.

Fitting results obtained by using "two sets of independent sites" model revealed that the dissociation constant KD of the main binding between 3CLpro and Merbromin was 1.33 μM, while the KD of the secondary binding was 67.8 μM. Besides, the thermodynamic process for Merbromin binding to both sites were negative enthalpies and positive entropy, implying that the main binding forces between 3CLpro and Merbromin were hydrogen bond and Van der Waals' interaction [20] .

The binding kinetics were analyzed by Surface Plasmon Resonance (SPR) assay. As shown in Figure 3B , Merbromin significantly increased the SPR response of 3CLpro in a dosedependent manner. The Rmax obtained from kinetic analysis reached up to approximate 600 RU that was 1.5 times of the theoretical Rmax (420 RU), implicating the stoichiometric ratio of the binding between Merbromin and 3CLpro might be more than "1", in accord with the result of our ITC assay. The kinetics of binding process was then evaluated by the heterogeneous ligand model to obtain two sets of binding parameters (Ka1=694.6 ± 2.61 L·mol -1 ·s -1 , Kd1=1.25 ± 0.01×10 -3 s -1 , KD2=1.80 ± 0.01 μM, and Ka2=646.7 ± 3.98 L·mol -1 ·s -1 , Kd2=16.25 ± 0.08×10 -3 s -

The binding of Merbromin to 3CLpro was further confirmed by fluorescence titration assay (Fig. 3C) . The endogenic fluorescence of 3CLpro could be dose-dependently quenched by Merbromin, implying the binding of Merbromin to 3CLpro. The data fitting procedure obtained an apparent KD of 11.76 ± 0.98 μM, which fell in between the two KD values of the two bindings obtained from either SPR or ITC assays.

We further evaluated the binding of Merbromin to three other proteases including Protease K, Papain and Trypsin whose activities were not significantly affected by Merbromin. The visible exothermic heat changes could be observed during the titration process for all the three proteases (Supplemental Figure 3A-C) , indicating the potential interactions between the three proteases and Merbromin since the dilution heat of Merbromin was endothermic (Supplemental Figure 3D ). However, these interactions, if existed, were extremely weak because no convergence of heat changing tendencies was observed under the same condition of Merbromin-3CLpro titration. The KD of Merbromin to Protease K, Papain and Trypsin were roughly estimated to be 3 mM, 39 mM and 700 μM, respectively. Together, these results indicated that Merbromin selectively binds to 3CLpro.

To gain a further insight into the binding modes, the molecular docking simulation was performed (Fig. 4) . As shown from the crystal structure (pdb:6m2n) [21] , 3CLpro exists in a form of dimer (Fig. 4A) . Several amino acid residues (Gly2, Phe3, Gly283, Arg4, Leu282, Lys5, J o u r n a l P r e -p r o o f Leu286, Glu288, Lys137, and Gly138) around the dimerization interface gathered by hydrogen bonds to form a pocket. This pocket was a potential binding site predicted by the "site finder" procedure of Molecular Operating Environment (MOE). Indeed, in our simulating docking, Merbromin bound to this site by forming hydrogen bonds with Glu288, Ser284 and Lys137 (Fig. 4C) and yielded the gliding energy of -6.68. This also suggested that Merbromin may regulate the formation of dimer to inhibit the activity of 3CLpro. In addition, Merbromin could also occupy the catalytic area with the gliding energy of -5.79. The sulfhydryl group of Cys145, a core residue in catalytic center, was forced to form hydrogen bonds with the carboxyl group of Merbromin (Fig. 4B) . Besides, the pi-H bond formed between the benzene ring of Merbromin and the methylene of Glu166 further stabilized the binding. Thus, our docking study revealed the two binding sites in 3CLpro for Merbromin, consistent with the results obtained from our SPR and ITC binding assays.

COVID-19 caused by SARS-CoV-2 infection has become a serious public health issue in the world. Until now, no effective drugs are available for inhibiting SARS-CoV-2, and only some potential drugs including chloroquine and remdesivir were prudently recommended to clinical therapeutics [22] [23] [24] [25] . Among the potential drug targets, 3CLpro has been extensively studied, due to its indispensable role in the maturity of function proteins during the virus replication process [4, [26] [27] [28] . In this study, we constructed the in vitro screening model of 3CLpro and used it to screen ~6,000 compounds by fluorescence-based HTS [9] , leading to identifying inhibiting 3CLpro, showing from its selectivity in inhibiting the proteolytic activity of 3CLpro and binding to 3CLpro.

Analysis of inhibition type based on enzymatic reaction showed that Merbromin was a mixed-type inhibitor of 3CLpro. Interestingly, binding data from both SPR and ITC assays suggested that 3CLpro possessed two sites for Merbromin binding. Consistently, molecular docking also revealed the two binding sites in 3CLpro. Merbromin could directly bind to the catalytic site of 3CLpro by forming hydrogen bond with Cys145, from which Merbromin directly inhibited the catalytic action of 3CLpro. Besides, Merbromin could also bind to the interaction interface formed by two monomers, by which Merbromin may function through modulating the dimer formation of 3CLpro. Thus, Merbromin binds to distinct sites of 3CLpro to employ two different mechanisms of inhibition, which may also render it as a mixed-type inhibitor of 3CLpro.

The structure of Merbromin contains mercury, which limits its usage due to the potential toxicity of mercury. However, it is worthwhile to trade-off the benefit of Merbromin on COVID-19 treatment and the risk of the side effects. Nevertheless, in view of its selective and potent effect on inhibiting 3CLpro, Merbromin can serve well as a lead compound for the development of 3CLpro inhibitors with strong efficacy and low toxicity. 

The authors declare no conflicts of interest in regards to this manuscript. potential anti-SARS-CoV-2 agents, Commun Biol, 4 (2021) 93.DOI: 10.1038/s42003-020-01577-x. 

Human Coronaviruses SARS-CoV, MERS-CoV, and SARS-CoV-2 in Children

From SARS and MERS CoVs to SARS-CoV-2: Moving toward more biased codon usage in viral structural and nonstructural genes

Comparison of the SARS-CoV-2 (2019-nCoV) M protein with its counterparts of SARS-CoV and MERS-CoV species

SARS-CoV-2 M(pro) inhibitors with antiviral activity in a transgenic mouse model

Identification of SARS-CoV-2 3CL Protease Inhibitors by a Quantitative High-Throughput Screening

Structure of M(pro) from SARS-CoV-2 and discovery of its inhibitors

Potential 3-chymotrypsin-like cysteine protease cleavage sites in the coronavirus polyproteins pp1a and pp1ab and their possible relevance to COVID-19 vaccine and drug development

N-Terminal Finger Stabilizes the S1 Pocket for the Reversible Feline Drug GC376 in the SARS-CoV-2 M(pro) Dimer

SARS-CoV 3CL protease cleaves its C-terminal autoprocessing site by novel subsite cooperativity

Structure of Main Protease from Human Coronavirus NL63: Insights for Wide Spectrum Anti-Coronavirus Drug Design

Evaluating the 3C-like protease activity of SARS-Coronavirus: recommendations for standardized assays for drug discovery

Ligand-induced Dimerization of Middle East Respiratory Syndrome J o u r n a l P r e -p r o o f (MERS) Coronavirus nsp5 Protease (3CLpro): IMPLICATIONS FOR nsp5 REGULATION AND THE DEVELOPMENT OF ANTIVIRALS

The five-parameter logistic: a characterization and comparison with the four-parameter logistic

Use of different parameters and equations for calculation of IC(5)(0) values in efflux assays: potential sources of variability in IC(5)(0) determination

Evaluation of enzyme inhibitors in drug discovery. A guide for medicinal chemists and pharmacologists

A novel spectroscopic titration method for determining the dissociation constant and stoichiometry of protein-ligand complex

A monomer-dimer nanoswitch that mimics the working principle of the SARS-CoV 3CLpro enzyme controls copper-catalysed cyclopropanation

A Simple Statistical Parameter for Use in Evaluation and Validation of High Throughput Screening Assays

Alternative statistical parameter for high-throughput screening assay quality assessment

Thermodynamics of protein association reactions: forces contributing to stability

Anti-SARS-CoV-2 activities in vitro of Shuanghuanglian preparations and bioactive ingredients

Persistent replication of SARS-CoV-2 in a severely immunocompromised patient treated with several courses of remdesivir

Chloroquine does not inhibit infection of human lung cells with SARS-CoV-2

Clinical benefit of remdesivir in rhesus macaques infected with SARS-CoV-2

Are hydroxychloroquine and chloroquine effective in the treatment of SARS-COV-2 (COVID-19)?

Development of a Fluorescence-Based, High-Throughput SARS-CoV-2 3CL(pro) Reporter Assay

Screening potential FDA-approved inhibitors of the SARS-CoV-2 major protease 3CL(pro) through high-throughput virtual screening and molecular dynamics simulation

Identification of 3-chymotrypsin like protease (3CLPro) inhibitors as

We thank Professor Sheng-ce Tao