key: cord-0689147-ijxh1ajg
authors: Lee, Hyun; Ren, Jinhong; Pesavento, Russell P.; Ojeda, Isabel; Rice, Amy J.; Lv, Haining; Kwon, Youngjin; Johnson, Michael E.
title: Identification and design of novel small molecule inhibitors against MERS-CoV papain-like protease via high-throughput screening and molecular modeling
date: 2019-05-15
journal: Bioorg Med Chem
DOI: 10.1016/j.bmc.2019.03.050
sha: 080c6709aaacf2b14083450c1227a9200e06c6b4
doc_id: 689147
cord_uid: ijxh1ajg

The development of new therapeutic agents against the coronavirus causing Middle East Respiratory Syndrome (MERS) is a continuing imperative. The initial MERS-CoV epidemic was contained entirely through public health measures, but episodic cases continue, as there are currently no therapeutic agents effective in the treatment of MERS-CoV, although multiple strategies have been proposed. In this study, we screened 30,000 compounds from three different compound libraries against one of the essential proteases, the papain-like protease (PL(pro)), using a fluorescence-based enzymatic assay followed by surface plasmon resonance (SPR) direct binding analysis for hit confirmation. Mode of inhibition assays and competition SPR studies revealed two compounds to be competitive inhibitors. To improve upon the inhibitory activity of the best hit compounds, a small fragment library consisting of 352 fragments was screened in the presence of each hit compound, resulting in one fragment that enhanced the IC(50) value of the best hit compound by 3-fold. Molecular docking and MM/PBSA binding energy calculations were used to predict potential binding sites, providing insight for design and synthesis of next-generation compounds.

The Middle East Respiratory Syndrome Coronavirus (MERS-CoV) was first documented in 2012 and has been found to be more challenging to control than the previously identified Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV) that emerged nearly a decade earlier. 1 The history and current therapeutic options for both SARS and MERS, as well as the less severe infections by human coronavirus strains 229E, OC43, NL63 and HKU1 have recently been reviewed in detail. 2 Briefly, following its initial identification, MERS spread within the Middle East and beyond, with cases continuing to arise on a sporadic basis, particularly in Saudi Arabia and the United Arab Emirates. 3, 4 There was also a major outbreak of MERS in South Korea in 2015. It was started from a returning traveler and spread through hospitals and infected a total of 186 people, 36 of which died. 5 Travel-related cases have been particularly challenging, as patients traveling from the Middle East present flu-like or pneumonia-like symptoms in other countries unprepared to recognize and treat MERS. 6 As of January 2019, the WHO reported 2279 confirmed cases of MERS, resulting in 806 fatalities across 27 countries, with continuing reports of new episodic cases. 7 The mortality rate of MERS is 35.5%, which is much higher than was the case for SARS (approximately 10%). Camel to human transmission appears to be the primary source of infection, 8, 9 with several studies implicating bats as the primary reservoir from which MERS-CoV originated. [10] [11] [12] Although an effective camel vaccine may ultimately be the best control strategy, 2,3 none appears likely in the near future. Person-to-person transmission occurs primarily from close contact in health-care settings or among stricken family members. 13, 14 Although therapeutic options for the treatment of MERS-CoV infections have been the subject of recent reviews, 4, 15, 16 the continuing emergence of new cases of MERS-CoV combined with the high fatality rate and the potential for future viral mutation makes development of effective therapeutics a continuing priority. 1 The multifunctional papain-like protease (PL pro ) isolated from MERS-CoV is recognized as a potentially druggable therapeutic target for the inhibition of viral replication. 17, 18 In addition to a role in the selective cleavage of viral replicase polyproteins, MERS-Cov PL pro (MERS-PL pro ) has been shown to contribute to MERS-CoV pathogenicity via deISGylase and deubiquitinase activity on host proteins. 19, 20 Recent studies further underscored the flexibility of MERS-PL pro when interacting with interferon stimulating gene product 15 (ISG15) from multiple species. 21 In this study, we identified new small molecule inhibitors of PL pro from high-throughput screening (HTS) of 30,000 compounds from three commercially available compound libraries including the FDA-approved drug library en route to potentially novel therapeutic alternatives in the future.

A total of 30,000 compounds from three screening libraries (FDAapproved Prestwick, Maybridge and Chembridge libraries) were screened against MERS-PL pro by fluorescence intensity based enzymatic end-point assays followed by IC 50 value determination by dose-response. The overall systematic screening process is summarized in Fig. 1A , with compound numbers from each library listed in Table 1 . The primary screens were done in duplicate using a Tecan liquid handling robot. The Z'-factors varied between 0.53 and 0.80 (average 0.67), indicating high quality of the screening campaign. The replicate plot of percent inhibition from duplicate data shown in Fig. 1B also illustrates the good quality of the screens. A total of 48 primary hit compounds exhibited greater than 35% inhibition at 50 µM compound concentration (shown in the red square in Fig. 1B) . After careful examination of each hit compound structure, compounds that were either toxic or contained reactive functional groups were eliminated. Thirty compounds were cherry-picked and retested in triplicate for their percent inhibition by continuous enzymatic assays. Based upon reproducible percent inhibition results, 23 compounds were selected and re-ordered from their commercial vendors for further hit validation.

The half maximal inhibitory concentration (IC 50 ) values were determined for 23 re-purchased compounds; dose-response curves of the two best compounds are shown in Fig. 2A . The IC 50 values in this work refer to the inhibition of the MERS-PL pro protease activity rather than inhibitory activity of MERS-CoV replication. The IC 50 values of compounds 1 and 6 were 25 µM and 20 µM, respectively. This reduced the number of hits to 12 compounds with IC 50 values below 55 µM. It is known that the enzymatic functional assay alone can be biased in selecting true positive hits. Consequently, the twelve selected compounds were subjected to a direct binding analysis with the MERS-PL pro enzyme by surface plasmon resonance (SPR) for further validation. The SPR sensorgrams of compound 6 at a series of increasing concentrations are shown in Fig. 2B as an example. Compound 6 possesses slow association and slow dissociation rates at 3.23 × 10 3 M −1 s −1 and 3.00 × 10 −2 s −1 , respectively, resulting in a binding affinity (K D ) of 9.3 µM, and suggests that binding may be controlled through a gating mechanism. Of the initial 12 compounds, seven showed direct binding to the MERS-PL pro protein with a varying range of binding affinity (K D ) between 7.6 and 112 µM. Most of the seven compounds had K D values comparable to IC 50 values within 0.5 to 2-fold ranges (Fig. 2C ). Of the seven validated hits, two compounds came out of a library of FDA-approved drugs, which are Pranlukast (compound 1) and Sulfasalazine (compound 2). Pranlukast is a cysteinyl leukotriene receptor-1 antagonist used for antagonism of bronchospasms caused by an allergic reaction to allergens in asthmatic patients and Sulfasalazine is being used to treat multiple diseases including rheumatoid arthritis, Crohn's [1981] [1982] [1983] [1984] [1985] [1986] [1987] [1988] [1989] disease and ulcerative colitis. Two hits (compounds 3 & 4) were from the diverse sets of the Maybridge library, and the Chembridge library produced three hits (compounds 5-7).

To characterize potential mechanism of inhibition of the lead compounds (i.e., 1, 6), enzymatic mode of inhibition studies were carried out varying both the inhibitor (0-200 µM) and substrate concentrations (100-300 µM). The best fit model was determined to be competitive inhibition for both compounds 1 and 6 with Ki values of 127 µM and 36.6 µM, respectively (a Dixon plot of 6 is shown in Fig. 3A as an example). The IC 50 values of 1 and 6 were determined utilizing a fixed substrate concentration of 50 µM and remained similar at both 25 and 20 µM, differing significantly from the Ki values listed above when both the substrate and inhibitor concentrations were varied, (Fig. 3B ). It appears that compound 1 does not inhibit MERS-PL pro at higher substrate concentration. This may suggest that the substrate can compete out compound 1 easier than compound 6, as confirmed by competition SPR. The K D values of compounds 1 and 6 alone were 14.1 µM and 7.6 µM. respectively. In the presence of 500 µM substrate, the binding of compound 1 was completely abolished, while that of compound 6 was 7-fold weaker (Fig. 3C ). This indicates that the substrate competes with both compounds, with compound 1 more easily substituted by the substrate than compound 6, in agreement with enzymatic assay results.

Both compounds 1 and 6 exhibited similar inhibitory activity and binding affinity to MERS-PL pro ; however, 6 is a more suitable lead candidate for further optimization because 6 has a lower molecular weight and also has a lower Ki value than 1. Hence, compound 6 was selected for further characterization, including reversibility studies. In addition to the observed dissociation in Fig. 2B , the reversibility of compound 6 was investigated via three different methods: dialysis, desalting and dilution. After complete inhibition of MERS-PL pro enzyme by compound 6, both inhibited and control samples (DMSO) were dialyzed overnight in order to remove compound 6 and recover enzyme activity. Unfortunately, the activity of the active enzyme control was completely lost after dialysis. Second, we utilized a desalting column to remove compound 6, also resulting in the deactivation of active control sample. Thirdly, we set up both enzyme and compound 6 at very high concentrations and gradually diluted both active control and inhibited sample in the exact same manner and measured enzyme activity at each dilution step. If compound 6 behaves as an irreversible inhibitor, the enzyme activity should not recover once it is fully inhibited. (However deleted) The enzyme activity gradually recovered, providing support for compound 6 being a reversible inhibitor (Fig. 3D ). The interaction between MERS-PL pro and compound 6 was also monitored by a 2D Transverse Relaxation Optimized Spectroscopy TROSY [ 1 H-15 N] NMR spectra using uniformly 15 N-labeled His-tagged MERS-PL pro (Fig. 4) . Although sequential assignment has not been carried out on apo-MERS-PL pro , we observed more than thirteen peaks that either shifted (green dotted boxes with arrows) or disappeared (green solid boxes) upon compound 6 binding to MERS-PL pro . The loss of 1 H-15 N resonances may be attributed to either exchange broadening as compound 6 binds and dissociates or a conformational change in the MERS-PL pro backbone.

The next step was to improve inhibitory activity of our lead compound 6. We have searched analogs of 6, but only the four analogs of compound 6 shown in Fig. 5A were commercially available. Based on activities of the five structural analogues, a preliminary structure-activity relationship (SAR) was obtained (Fig. 5B ). The core functionality of our new lead scaffold is a benzene sulfonamide, and a separate sixmembered nitrogen-containing ring is bound to the sulfonamide moiety at the R 1 location highlighted in blue. The pyrimidin-2 with two methyl groups at ortho and para-positions has better inhibitory activity (6, 73% inhibition) than piperazine-2,6-dione (6a, 16% inhibition) or piperidine (6b, 7% inhibition) moieties. However, it is unclear if either the methyl on the pyrimidin-2 or the heterocycle core itself is crucial for the maintenance of the activities. At the R 2 location shown in pink, the amino group at the para-position of benzene is essential for activity. When the amino group was replaced by methyl group (6c), the inhibitory activity decreased significantly to 11%, and the inhibitory activity was almost completely lost (2% inhibition) when the amino group was acetylated (6d). Given the limited number of structural analogs of 6, the preliminary SAR simply underscores key structural features integral to the inhibitory activity. 

None of the tested commercial analogs showed better inhibitory activity against MERS-PL pro than the original lead compound 6. This led us to screen a 352-compound Zenobia small fragment library in the presence of the lead compound 6 in an effort to increase inhibitory activity. The same fluorescence-base enzyme assay was used to screen the Zenobia library in attempts to search additional small fragments that can enhance inhibitory activity of 6. Of the 352 fragment compounds, 11 fragments exhibited over 60% inhibition (> 10% inhibition enhancement) at 50 µM concentration in the presence of 20 µM compound 6. Some fragments showed the opposite effect, decreasing the inhibitory activity of compound 6, probably by competing for the same enzymatic binding site. Five fragments among the 11 selected fragments did improve the apparent IC 50 values of compound 6, with improvements ranging from 1.3-fold to 3-fold better ( Table 2 ). The IC 50 value of each fragment alone was also determined without compound 6 present, and they varied significantly between 71.6 µM (ZT834) and 239 µM (ZT626). Fragments ZT426 and ZT470 exhibited moderate enhancement with 1.3-fold and 1.4-fold improved IC 50 values for compound 6, while another two fragments, ZT537 and ZT834, enhanced the inhibitory activity of 6 slightly better, at almost 2-fold. Interestingly, the least effective fragment ZT626 by itself had the best activity enhancement for compound 6, reducing the IC 50 value by 3fold to 6.6 µM. This suggests compound 6 and ZT626 might inhibit MERS-PL pro additively. In addition, the binding affinity of compound 6 in the presence of constant concentration of ZT626 (200 µM) was determined to be similar (10.8 µM) to compound 6 alone (9.3 µM), suggesting that ZT626 probably binds to a different location from the 6 binding site (Figs. 6 and 2B) . Fig. 6A also shows that ZT626 readily dissociates from MERS-PL pro , indicating that its binding is also reversible.

Compound 6 is a competitive inhibitor against MERS-PL pro enzyme and competes with the substrate in the catalytic site. Molecular docking was performed for compound 6 in the catalytic site of MERS-PL pro with GOLD5.2.2, 22 and three different binding poses for compound 6 were selected and shown in Fig. 7A . In order to investigate the most probable binding pose, 5 ns molecular dynamic simulations were performed on the three conformations of compound 6 binding with MERS-PL pro , followed by MM/PBSA binding affinity calculations using the AMBER14 package. 23 The calculated binding energies of the three binding poses ranged from -10.6 kcal mol −1 to -15.8 kcal mol −1 (Table 3) . Pose 1 (shown in cyan in Fig. 7A ) has the lowest calculated binding energy at -15.8 kcal mol −1 among the three potential binding conformations and corresponds to relatively stable binding interactions. Therefore, pose 1 was selected as the most probable binding conformation for compound 6.

Another round of molecular docking, MD simulations and MM/ PBSA binding affinity calculations was carried out to predict the binding pose of the additive fragment ZT626. The docking of ZT626 was performed in a 10 Å radius area around the compound 6 binding pose-1, keeping the position of compound 6 constant during fragment docking. Two different binding poses of ZT626 were obtained from GOLD (Fig. 7B) . Then 5 ns MD simulations and MM/PBSA calculations were performed on these ternary complexes (MERS-PL pro /compound 6/ ZT626), using the same parameters as used for determination of binding poses for compound 6. The binding energy of compound 6 was improved by the addition of ZT626/pose-1 to -17.4 kcal mol −1 in comparison to compound 6 binding alone (-15.8 kcal mol −1 ). On the other hand, the ZT626 binding pose-2 showed an opposite effect, decreasing the overall binding energy to -13.8 kcal mol −1 . Therefore, pose-1 for ZT626 was selected as the most probable binding conformation.

MM/PBSA binding energy calculations led us to choose pose-1 for both compound 6 and ZT626 in MERS-PL pro . The overall MERS-PL pro enzyme structure and the potential binding poses of both compounds are shown in Fig. 8A . The carbonyl group of compound 6 could form an H-bond with the sidechain of Y279, and this residue could also form Hbonds with two phenolic hydroxyls of ZT626 (pink circle in Fig. 8A ). Hence, Y279 may play an important role for the binding of both compound 6 and ZT626. Meanwhile, the sidechain of S167 could also form an H-bond with one phenolic hydroxyl of the fragment compound. In addition, many hydrophobic interactions with the surrounding residues such as P163, D164, D165, G248, T249, P250, F269, E273, A275, V276, G277 and T308 are also important for the binding of compounds 6 and ZT626 (Fig. 8B ).

Based on the synergy effects for the fragment of ZT626 with compound 6, as well as their predicted binding conformations, we designed candidates with both structural moieties of ZT626 and compound 6 via appropriate linkers for future synthesis. Our fragment-linking strategy involved in the replacement of sulfonyl group with chiral carbon to get the linking and extending position in compound 6. Introducing a flexible linker with proper length is important for keeping the binding poses for both compound 6 and ZT626. Two atom, three atom and four atom linkers as alkyl or ester group between the newly introduced chiral carbon in compound 6 and the 3-hyroxy of ZT626 were used. The molecular docking studies were carried out for these new compounds with different linkers to verify which structures recapitulate the key binding features of the compound 6 and fragment ZT626. From the molecular docking results, the two carbon linker (L1) appears inadequate for the proper binding of ZT626 moiety (Fig. 9A) . Therefore, we extended the linker to three or four atoms in compounds L2 and L3 (Fig. 9B & C) , and the fragment moiety matches better than a twocarbon linker compound. This provides better direction for future compound synthesis, and also highlights interactions that contribute to inhibitor binding.

Although repurposing of HIV protease inhibitor, lopinavir/ritonavir, [1981] [1982] [1983] [1984] [1985] [1986] [1987] [1988] [1989] has been found to be somewhat effective against MERS-CoV and could be a backup for the treatment of MERS, development of specific inhibitors targeting MERS-CoV are still needed. 5, 26, 27 We report a new small molecule scaffold effective as a competitive inhibitor against MERS-PL pro . This scaffold was identified from screening three structurally diverse compound libraries including the FDA-approved drug library called Prestwick. A thorough hit validation and confirmation strategy was applied, and produced two lead compounds with moderate inhibitory activity and binding affinity. These two compounds were characterized as competitive inhibitors based on both enzymatic mechanism of inhibition and competition SPR studies, one of which was selected as a better lead candidate for further characterization and development. Due to lack of commercially available analogs of lead compound 6, only a very limited preliminary SAR was obtained. Hence, we explored a fragment-linking strategy to improve our lead compound activity. Additional small fragment library screening in the presence of the lead compound produced an additive fragment that can enhance the IC 50 value of the lead compound by 3-fold. Molecular docking and MM/ PBSA binding energy calculations were performed to predict the potential binding sites of compounds 6 and ZT626. This provides direction for the future design of fragment-linked inhibitors to improve its potency in developing inhibitors against MERS-PL pro .

We screened three commercially available compound libraries purchased from the FDA-approved Prestwick, Maybridge, and Chembridge libraries, which consisted of 1200, 14,400 and 14,400 compounds, respectively. All compounds were dissolved in 100% DMSO and stored as 10 mM stock solutions in desiccated condition at −30°C. The primary HTS assay was performed by a Tecan Freedom EVO 200 robot equipped with a Te-Mo 3 × 3 96-channel Liquid Handler dispenser and a 384-pin stainless steel pin tool (V&P Scientific) with a 200 nL capillary capacity. All assays were done in duplicate in black 384-well plates (Matrix Technologies) at room temperature. The MERS-PL pro enzyme was purified as described 28 and was prepared in assay buffer (50 mM HEPES, pH 7.5, 0.01% Triton X-100 (v/v), 0.1 mg mL −1 BSA, and 2 mM DTT). 30 µL of enzyme solution (400 nM final concentration) was dispensed into wells, and then 200 nL of 10 mM compound (50 µM final concentrations) were added and incubated for 5 min. Enzyme reactions were initiated by adding 10 µL of substrate Z-Arg-Leu-Arg-Gly-Gly-AMC (Bachem Bioscience) at 50 µM final concentration dissolved in assay buffer and incubated for 6 min. Enzyme reactions were stopped by adding 10 µL of 10% SDS as a stop solution, and fluorescence intensity was monitored at 360 nm (excitation) and 450 nm (emission) with a Tecan Genios Pro microplate reader. Each plate contained a total of 32 positive and 32 negative controls.

Structures of the initial hit compounds from the HTS were examined, cherry-picked and reanalyzed by continuous kinetic assay by hand for confirmation. For those that showed over 50% inhibition at 50 µM concentration in the confirmation assay, IC 50 values were measured using the same assay conditions as the primary screen by hand in triplicate. A series of increasing concentrations (0-200 µM final concentration at 2-fold serial dilution) in 100% DMSO were prepared in a 384-well plate. 7 µL of 600 nM (3X) enzyme solution was distributed into wells, and 7 µL of varying concentration of 3X compounds were added and incubated for 5 min. The enzyme reaction was initiated by adding 7 µL of the 150 µM (3X) substrate, and its activity was continuously monitored for at least 10 min. The IC 50 values were calculated by fitting the data with the Hill equation (1), using SigmaPlot v12.0 where y is percent inhibition, x is inhibitor concentration, n is the slope of the concentration-response curve (Hill slope), and V max is maximal inhibition from three to four independent assays.

(1)

The MERS-PL pro enzyme was diluted to 50 µg/mL with 10 mM sodium acetate (pH 5.5) and immobilized on a CM5 sensor chip by standard amine-coupling with running buffer PBSP (10 mM phosphate, pH 7.4, 2.7 mM KCl, 137 mM NaCl, 0.05% surfactant P-20) using a Biacore T200 instrument. MERS-PL pro enzyme was immobilized to flow cells 2 and 4, and immobilization levels were ∼10,000 RU for both flow cells. Unmodified flow cells 1 and 3 were used as controls. Compound solutions with a series of increasing concentrations (0-50 µM at 2-fold dilution) were applied to all four channels in SPR binding buffer (PBSP + 0.5 mM TCEP and 2% DMSO) at a 30 µL/min flow rate at 25°C. Data were double-referenced with both reference cell RU values and zero concentration (2% DMSO) signals, and sensorgrams were analyzed using the Biacore T200 evaluation software 3.0. Response units at each concentration were measured during the equilibration phase for steady-state affinity fittings, and the K D values were determined by fitting the data to a single rectangular hyperbolic curve Eq.

(2), where y is the response, y max is the maximum response and x is the compound concentration.

Kinetic rate constants were determined by fitting globally to the 1:1 Langmuir model embedded in the Biacore T200 evaluation software v3.0.

Enzymatic activities of MERS-PL pro were investigated in the same way as the IC 50 value determination by varying concentration of both substrate (0-300 µM) and inhibitors (0-200 µM). The data were fit to four equations (3-6) using SigmaPlot Enzyme Kinetics Module 1.3 in order to determine the best fit inhibition mechanism and kinetic parameters for each compound. 

where V is the reaction rate, V max is the maximum rate of the reaction, K m is the Michaelis-Menten constant for the substrate, [S] is the substrate concentration, [I] is the inhibitor concentration, K i is the dissociation constant of the inhibitor I to the free enzyme and αK i is the dissociation constant for the inhibitor I to the ES complex.

4 µM MERS-PL pro was incubated with compound 6 at 125X the concentration of the IC 50 for 1 h at room temperature in assay buffer containing 50 mM HEPES (pH 7.5), 2 mM DTT, 0.1 mg/ml BSA, and 0.01% Triton X-100 in a final volume of 200 µL. Control MERS-PL pro without any compound was also prepared in the same way with the same volume of DMSO. Then, both samples were diluted by 2-fold using assay buffer (Supplementary Fig. S1 ). MERS-PL pro enzyme activity of both samples were measured. Seven additional 2-fold dilutions were done followed by enzyme activity measurement. MERS-PL pro activity was measured in the same way as IC 50 measurements.

The Zenobia fragment library consisting of 352 compounds was screened in a similar way as the primary HTS. The original stock concentration of all fragments was 200 mM dissolved in 100% DMSO and they were diluted do 20 mM in 100% DMSO. Compound 6 was added to assay buffer at 20 µM final concentration for screening wells, and 16 positive and 16 negative controls contained the same amount of just DMSO. 30 µL of enzyme solution (400 nM final concentration) was dispensed into wells, and then 200 nL of 20 mM fragment (100 µM final concentrations) were added and incubated for 5 min. Enzyme reactions were initiated with 10 µL of substrate (50 µM final concentration) dissolved in assay buffer and incubated for 6 min. Enzyme reactions were continuously monitored for 10 min at 360 nm (excitation) and 450 nm (emission) with a Tecan Genios Pro microplate reader.

The crystal structure of the MERS-PL pro in complex with ubiquitin (PDB code 4RF1 29 with resolution of 2.15 Å) was selected to perform molecular docking. The MERS-PL pro structure was optimized through the Protein Preparation Wizard in the Schrödinger Suite. 30 All hydrogens and charges were added in the OPLS3 force field. Restrained minimization was performed on the added hydrogens. Meanwhile, the LigPrep module in the Schrödinger Suite 31 was used to create the 3D structures of compound 6 as well as to perform the geometric optimization. Molecular docking was performed by GOLD v5.2.2 22 using the above prepared MERS-PL pro and compound 6. Ubiquitin was extracted before performing docking, and the active site for MERS-PL pro was defined as being within a 10 Å radius around the catalytic residue Cys111 for the docking of compound 6. The MERS-PL pro was maintained static, while compound 6 was flexible with "flip amide bonds", "detect internal H bonds", and "flip ring corners" set to on during ligand conformations searching. Other parameters applied default values in GOLD. Then three different binding poses for compound 6 were chosen for the following MD simulations and subsequent MM/PBSA calculation to predict binding affinities using the AMBER14 suite of programs. 23 Preparation of the MD simulations included assigning Restrained Electrostatic Potential (RESP) atomic partial charges to compound 6 using HF/6-31G* and Gaussian 09 in the R.E.D. online server, 32 and then assigning the general AMBER force field 33 (GAFF) parameters to the above prepared ligand. A 10 Å octahedral TIP3P water molecule box was added to each of the complex systems using Clcounter-ions to neutralize the system. The Amber FF14SB force field was applied for the system. The systems were first minimized using 10,000 steps of steepest descent minimization followed by another 10,000 steps of conjugate gradient minimization. After heating from 0 K to 300 K over 100 ps, the systems were equilibrated over 100 ps at constant pressure (1 bar) and temperature (300 K) with a restraint of 10 kcal mol −1 Å −2 on the complex and then subsequently without any restraint. Then 5 ns NPT production runs without positional restraints were performed at 300 K and 1 bar for the systems. The atomic coordinates were saved every 2.5 ps. During the MD simulation, all bonds involving hydrogen atoms were constrained to their equilibrium distance using the SHAKE algorithm, 34 and a time step of 2 fs. The python script, MMPBSA.py, 35 included in AMBER 14, was used to perform the MM/PBSA calculations using the 1200 frames from the last 3 ns of the 5 ns simulation Fig. 9 . The designed compounds from fragment linking strategy and their predicted binding poses. Among three designed compounds for synthesis, two compounds (L2 and L3) nearly recapitulate the binding pose of compound 6 and ZT626 after molecular docking. The linkers were colored in red in the structures. Compound 6 is colored in sky blue, ZT626 is colored in dark blue and the MERS-PL pro enzyme is shown in grey surface. trajectory file. Default parameters were applied in binding affinity calculations.

The binding site for the fragment of ZT626 was selected as being 10 Å around the above predicted binding position of compound 6. After docking, two different binding poses of ZT626 were picked to perform the subsequent 5 ns MD simulations and the binding energy calculation for compound 6. Consistent parameters were applied to MD simulations for the ternary complexes of MERS-PL pro /compound 6/ZT626 and that of the binary complex of MERS-PL pro /compound 6, in order to compare the binding affinity difference for compound 6 by incorporating ZT626.

2D NMR spectra of uniformly 15 N-labeled His-tagged Apo-MERS-PL pro , in the presence and absence of excess compound 6, were recorded on a Bruker 900 MHz NMR spectrometer. Each NMR solution contained 15 N-labeled HIS-tagged MERS-PL pro (240 µM), 25 mM KH 2 PO 4 (pH 7.4), 25 mM KCl, 1 mM DTT and 10% D 2 O. The MERS-PL pro + excess inhibitor spectrum was generated by adding four equivalents of compound 6 (in d 6 -DMSO) to Apo-MERS-PL pro , producing an aqueous buffer with 3% DMSO. Each solution was gently vortexed, transferred to a 5 mm OD 7″L glass NMR tube, and the Transverse Relaxation Optimized Spectroscopy TROSY [ 1 H-15 N] spectrum was acquired overnight at RT. SPARKY was utilized to process the NMR data and produce the images.

MERS-CoV: understanding the latest human coronavirus threat

Coronaviruses -drug discovery and therapeutic options

Middle East respiratory syndrome coronavirus: another zoonotic betacoronavirus causing SARS-like disease

Update on therapeutic options for middle east respiratory syndrome coronavirus (MERS-CoV)

Combination therapy with lopinavir/ ritonavir, ribavirin and interferon-alpha for Middle East respiratory syndrome

Spread of MERS to South Korea and China

Middle East Respiratory Syndrome Coronavirus

Middle East respiratory syndrome coronavirus in dromedary camels: an outbreak investigation

Middle East respiratory syndrome coronavirus neutralising serum antibodies in dromedary camels: a comparative serological study

Rooting the phylogenetic tree of middle East respiratory syndrome coronavirus by characterization of a conspecific virus from an African bat

Genetic characterization of Betacoronavirus lineage C viruses in bats reveals marked sequence divergence in the spike protein of pipistrellus bat coronavirus HKU5 in Japanese pipistrelle: implications for the origin of the novel Middle East respiratory syndrome coronavirus

Bat origins of MERS-CoV supported by bat coronavirus HKU4 usage of human receptor CD26

Hospital-associated outbreak of Middle East respiratory syndrome coronavirus: a serologic, epidemiologic, and clinical description

Hospital outbreak of Middle East respiratory syndrome coronavirus

Advances in MERS-CoV vaccines and therapeutics based on the receptor-binding domain

Development of small-molecule MERS-CoV inhibitors

Crystal structure of the papain-like protease of MERS coronavirus reveals unusual, potentially druggable active-site features

MERS-related betacoronavirus in Vespertilio superans bats

Structurally guided removal of DeISGylase biochemical activity from papain-like protease originating from middle east respiratory syndrome coronavirus

MERS-CoV papainlike protease has deISGylating and deubiquitinating activities

Structural Insights into the interaction of coronavirus papain-like proteases and interferon-stimulated gene product 15 from different species

Improved protein-ligand docking using GOLD

UCSF Chimera-a visualization system for exploratory research and analysis

LigPlot+: multiple ligand-protein interaction diagrams for drug discovery

Treatment of middle east respiratory syndrome with a combination of lopinavir-ritonavir and interferon-β1b (MIRACLE trial): study protocol for a randomized controlled trial

With lopinavir/ritonavir or interferon-β1b improves outcome of MERS-CoV infection in a nonhuman primate model of common marmoset

Inhibitor recognition specificity of MERS-CoV papain-like protease may differ from that of SARS-CoV

Crystal structure of the middle east respiratory syndrome coronavirus (MERS-CoV) papain-like protease bound to ubiquitin facilitates targeted disruption of deubiquitinating activity to demonstrate its role in innate immune suppression

Schrödinger Release 2016-1: Schrödinger Suite 2016-1 Protein Preparation Wizard

Epik Version 3.5, Schrödinger, LLC

Schrödinger Release 2016-1: Ligprep, Version 3.7. Schrödinger, LLC

a web service for deriving RESP and ESP charges and building force field libraries for new molecules and molecular fragments

Development and testing of a general amber force field

A fast SHAKE algorithm to solve distance constraint equations for small molecules in molecular dynamics simulations

MMPBSA.py: an efficient program for end-state free energy calculations

This work was supported in part by National Institutes of Health Grants R56 AI089535. We thank K. Ratia for performing HTS and primary screening data analysis. We also thank Benjamin Ramirez for 2D NMR data acquisition and Seon Beom Kim for HPLC analysis of two hit compounds. This work used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by National Science Foundation grant number OCI-1053575. We also thank ChemAxon for access to JChem for Excel, which was used for compound data management. Molecular graphics and analyses were performed with the UCSF Chimera package. Chimera was developed by the Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco (supported by NIGMS P41-GM103311).

HL performed all experiments with assistance of AJR, IO, HLv and YK. JR performed computational studies. RPP did 2D NMR studies. HL, JR, RPP and MEJ designed the experiments and wrote the manuscript.

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bmc.2019.03.050.