A Quick Route to Multiple Highly Potent SARS-CoV-2 Main Protease Inhibitors**

Article abstract of DOI:10.1002/cmdc.202000924

The COVID-19 pathogen, SARS-CoV-2, requires its main protease (SC2M^Pro) to digest two of its translated long polypeptides to form a number of mature proteins that are essential for viral replication and pathogenesis. Inhibition of this vital pr

Full text of DOI:10.1002/cmdc.202000924

Accepted Article
Title: A Quick Route to Multiple Highly Potent SARS-CoV-2 Main
Protease Inhibitors
Authors: Wenshe Liu, Kai Yang, Xinyu Ma, Yuying Ma, Yugendar
Alugubelli, Danielle Scott, Erol vatansever, Aleksandra
Drelich, Banumathi Sankaran, Zhi Geng, Lauren Blankenship,
Hannah Ward, Yan Sheng, Jason Hsu, Kaci Kratch, Baoyu
Zhao, Hamed Hayatshahi, Jin Liu, Pingwei Li, Carol Fierke,
Chien-Te Tseng, and Shiqing Xu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.202000924
Link to VoR: https://doi.org/10.1002/cmdc.202000924
01/2020

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
A Quick Route to Multiple Highly Potent SARS-CoV-2 Main
Protease Inhibitors
Kai S. Yang,[a],† Xinyu R. Ma,[a],† Yuying Ma,[a],† Yugendar R. Alugubelli,[a],† Danielle A. Scott,[a],† Erol C.
Vatansever,[a],† Aleksandra K. Drelich,[b],† Banumathi Sankaran,[c] Zhi Z. Geng,[a] Lauren R.
Blankenship,[a] Hannah E. Ward,[a] Yan J. Sheng,[a] Jason C. Hsu,[b] Kaci C. Kratch,[a] Baoyu Zhao,[d]
Hamed S. Hayatshahi,[e] Jin Liu,[e] Pingwei Li,[d] Carol A. Fierke, *[a],[d] Chien-Te K. Tseng,*[b] Shiqing
Xu,*[a] and Wenshe Ray Liu, *[a],[d],[f],[g]
[a]
Dr. K. S. Yang, X. R. Ma, Dr. Y. Ma, Dr. Y. R. Alugubelli, D. A. Scott, E. C. Vatansever, Z. Z. Geng, L. R. Blankenship, H. E. Ward, Y. J. Sheng, K. C.
Kratch, Prof. Dr. C. A. Fierke, Prof. Dr. S. Xu, Prof. Dr. W. R. Liu
Department of Chemistry
Texas A&M University
College Station, TX 77843-3255, USA
E-mail: cafierke@tamu.edu
E-mail: shiqing.xu@tamu.edu
E-mail: wliu@chem.tamu.edu
[b]
A. K. Drelich, J. C. Hsu, Prof. Dr. C. K. Tseng
Department of Microbiology and Immunology
University of Texas Medical Branch
Galveston, TX 77555, USA
E-mail: sktseng@utmb.edu
[c]
[d]
[e]
[f]
Dr. B. Sankaran
Molecular Biophysics and Integrated Bioimaging, Berkeley Center for Structural Biology
Lawrence Berkeley National Laboratory
Berkeley, CA 94720, USA
Dr. B. Zhao, Prof. Dr. P. Wei, Prof. Dr. C. A. Fierke, Prof. Dr. W. R. Liu
Department of Biochemistry and Biophysics
Texas A&M University
College Station, TX 77843, USA
H. S. Hayatshahi, J. Liu
Department of Pharmaceutical Sciences
UNT Health Science Center
Fort Worth, TX 76107, USA
Prof. Dr. W. R. Liu
Department of Molecular and Cellular Medicine, College of Medicine
Texas A&M University
College Station, TX 77843, USA
[g]
[†]
Prof. Dr. W. R. Liu
Institute of Biosciences and Technology and Department of Translational Medical Sciences, College of Medicine
Texas A&M University
Houston, TX 77030, USA
These authors contributed equally to the paper.
Abstract: The COVID-19 pathogen, SARS-CoV-2, requires its main
protease (SC2MPro) to digest two of its translated long polypeptides to
form a number of mature proteins that are essential for viral replication
and pathogenesis. Inhibition of this vital proteolytic process is effective
in preventing the virus from replication in infected cells and therefore
provides a potential COVID-19 treatment option. Guided by previous
medicinal chemistry studies about SARS-CoV-1 main protease
(SC1MPro), we have designed and synthesized a series of SC2MPro
inhibitors that contain β-(S-2-oxopyrrolidin-3-yl)-alaninal (Opal) for the
formation of a reversible covalent bond with the SC2MPro active site
cysteine C145. All inhibitors display high potency with Ki values at or
below 100 nM. The most potent compound MPI3 has as a Ki value as
8.3 nM. Crystallographic analyses of SC2MPro bound to 7 inhibitors
indicated both formation of a covalent bond with C145 and structural
rearrangement from the apoenzyme to accommodate the inhibitors.
Virus inhibition assays revealed that several inhibitors have high
potency in inhibiting the SARS-CoV-2-induced cytopathogenic effect
in both Vero E6 and A549/ACE2 cells. Two inhibitors MPI5 and MPI8
completely prevented the SARS-CoV-2-induced cytopathogenic
effect in Vero E6 cells at 2.5-5 µM and A549/ACE2 cells at 0.16-0.31
µM. Their virus inhibition potency is much higher than some existing
molecules that are under preclinical and clinical investigations for the
treatment of COVID-19. Our study indicates that there is a large
chemical space that needs to be explored for the development of
SC2MPro inhibitors with ultra-high antiviral potency.
Introduction
Coronaviruses (CoVs) are a group of related RNA viruses that
cause diseases in a wide range of vertebrates including humans
and domestic animals.[1] Before 2003, there were only two CoVs,
HCoV-229E and HCoV-OC43, known as human pathogens.[2]
The SARS pandemic in 2003 led to the revelation of SARS-CoV-
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
1, a pathogen causing a severe respiratory infection.[3] The
subsequent surge in CoV research resulted in the discovery of
two additional human CoVs, HCoV-NL63 and HCoV-HKU1, that
are mildly pathogenic.[4] One addition to this group was MERS-
CoV that emerged in 2012 as a pathogen causing a severe
respiratory infection.[5] Although SARS-CoV-1 and MERS-CoV
are highly lethal pathogens, the public health, social, and
economic damages that they have caused are diminutive in
comparison to that from SARS-CoV-2, a newly emerged human
CoV pathogen that causes COVID-19.[6] Rival only to the 1918
influenza pandemic, the COVID-19 pandemic has led to
catastrophic impacts worldwide. As of July 13th, 2020, the total
global COVID-19 cases have surpassed 12 million with more than
570,000 deaths.[7] To alleviate catastrophic damages of COVID-
19 on public health, society and economy, finding timely treatment
options is of paramount importance.
entropy loss during the binding of SC1Mpro to the more rigid
lactam compared to the flexible Gln. Although converting the
scissile backbone amide to a Michael acceptor in a SC1MPro
ligand turns it into a covalent inhibitor, it eliminates the critical
hydrogen bond between the P1 α-carbonyl oxygen and SC1MPro
.
Therefore, most Michael acceptor inhibitors developed for
SC1MPro and recently for SC2MPro tend to have efficacy with low
micromolar or submicromolar IC50 values rather than low
nanomolar levels.[11a, 14] To maintain this critical hydrogen bond
and exploit a covalent interaction with the active site cysteine
C145 to form a hemiacetal for high affinity, both aldehyde and
ketoamide moieties have been used to replace the P1 C-side α-
amide to develop potent reversible covalent inhibitors for
SC1MPro. For aldehyde-based inhibitors, a typical potent inhibitor
contains Opal at the P1 site that consists of a β-S-2-oxopyrrolidine
side chain and an α-aldehyde for both taking advantage of strong
interactions with the SC1MPro P1-binding pocket and the
formation of a reversible covalent bond with C145 (Figure 1C).
Similar to all other CoVs, SARS-CoV-2 is an enveloped, positive-
sensed RNA virus with a genome of nearly 30 kb in size.[8] Its
genome encodes 10 open reading frames (ORFs). The largest
ORF, ORF1ab encompasses more than two thirds of the whole
genome. Its translated products, ORF1a (~500 kDa) and ORF1ab
(~800 kDa),[9] are very large polypeptides that undergo proteolytic
cleavage to form 15 mature proteins. These are nonstructural
proteins (Nsps) that are essential for the virus to modulate human
cell hosts for efficient viral protein expression, viral genome
replication, virion packaging, and viral genomic RNA processing.
The proteolytic cleavage of ORF1a and ORF1ab is an
autocatalytic process. Two internal polypeptide regions, Nsp3 and
Nsp5, possess cysteine protease activity that cleaves
themselves, and all other Nsps, from the two polypeptides. Nsp3
is commonly referred to as papain-like protease (PLPro), and Nsp5
as 3C-like protease (3CLPro) or, more recently, main protease
(MPro).[10] Although we have yet to understand SARS-CoV-2
biology and COVID-19 pathogenesis, previous studies of SARS-
CoV-1 have established that activity of both PLPro and MPro is
essential to viral replication and pathogenesis. Of the two
proteases, MPro processes 12 out of the total 15 Nsps; inhibition
of this enzyme is anticipated to have more significant impacts on
the viral biology than that of PLPro. Therefore, small molecule
medicines that potently inhibit SARS-CoV-2 MPro (SC2MPro) are
potentially effective treatment options for COVID-19.[11] In this
work we report our progress in the development of potent
SC2MPro inhibitors.
Figure 1. The design of SC2MPro inhibitors based on medicinal chemistry
learned from SC1MPro studies. (A) The structure of SC1MPro complexed with a
peptide substrate (based on the pdb entry 5b6o). Active site cavities that bind
P1, P2, P4, and P3’ residues in the substrate are labeled. (B) A sche-matic
diagram that shows interactions between SC1MPro and a substrate. (C) A
scheme in which a substrate P1 residue is converted to glutaminal and then β-
(S-2-oxopyrrolidin-3-yl)-alaninal (Opal) to form a reversible covalent inhibitor
that reacts with the SC1MPro active site cysteine C145. (D) Scaffold structures
of Opal-based inhibitors designed for SC2MPro
.
Typical examples of this design include GC376 that was originally
developed for MPro from feline infectious peritonitis (FIP) CoV and
two inhibitors, 11a and 11b, that were recently developed for
SC2MPro [15]
Given its relative simplicity, we have followed a
.
Results
similar scheme according to structure diagrams shown in Figure
1D to design and synthesize reversible covalent inhibitors for
SC2MPro and pursued structural variations at P2, P3, and R
positions for improved potency.
The design of β-(S-2-oxopyrrolidin-3-yl)-alaninal (Opal)-
based, reversible covalent inhibitors for SC2MPro. Although we
are at the inaugural stage of learning medicinal chemistry to inhibit
SC2MPro, much has been learned from studies of SARS-CoV-1
Synthesis and IC50 characterization of SC2MPro inhibitors
MPro (SC1MPro
SC2MPro [12]
)
that shares 96% sequence identity with
(MPIs). GC376 (Figure 2A) has confirmed potency against
.
SC1MPro has a large active site that consists of
SC1MPro [16] We purchased it as a potential SC2MPro inhibitor. We
.
several smaller pockets for the recognition of residues at P1, P2,
P4, and P3’ positions in a protein substrate (Figure 1A).[13] P4 is
typically a small hydrophobic residue while P2 and P3’ are large.
For all Nsps that are processed by SC1MPro and SC2MPro, Gln is
the P1 residue at their cleavage sites. In order to bind the P1 Gln,
SC1MPro forms strong van der Waals interactions with the Gln side
chain, and also utilizes two hydrogen bonds with the Gln side
chain amide oxygen and α-carbonyl oxygen atoms (Figure 1B).
Previous efforts in the development of irreversible covalent
inhibitors for SC1MPro primarily focused on fixing the P1 residue
as a more potent β-S-2-oxopyrrolidine-containing Gln analog, and
changing the scissile backbone amide to an alkene Michael
acceptor in order to react with the active site cysteine C145, as
well as varying substituents on two sides to improve potency.[14]
The enhanced potency from the use of the β-S-2-oxopyrrolidine-
containing Gln analog is most probably due to the reduction of
designed two similar dipeptidyl compounds MPI1-2 (Figure 2A)
and synthesized them according to a synthetic scheme shown in
Supplementary Scheme 1. Both MPI1 and MPI2 have Phe at the
P2 site which was previously shown to contribute to strong
bonding to SC1MPro [14]
.
MPI2 has also an o-fluoro-p-
chlorocinnamyl group as an N-terminal cap. This group is more
rigid than the CBZ group and therefore possibly introduces a
strong interaction with the P4-binding pocket in SC2MPro [17]
To
.
characterize IC50 values of all three molecules for inhibition of
SC2MPro, we recombinantly expressed a 6×His-SUMO-SC2MPro
fusion protein in E. coli and purified and digested this protein with
SUMO protease to obtain intact SC2MPro with more than 95%
purity. We used a previously described fluorescent peptide assay
(see supplementary data) to measure the IC50 values for GC376,
MPI1, and MPI2 as 31 ± 4, 100 ± 23, and 103 ± 14 nM,
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
respectively (Figure 2B).[11c] Our determined IC50 value for GC376
agrees well with that from Ma et al.[18] In the light of the publication
of inhibitors 11a and 11b that showed similar IC50 values as 53 ±
5 and 40 ± 2 nM, respectively,[15b] we shifted our focus from the
synthesis of bipeptidyl inhibitors to that of tripeptidyl inhibitors. By
adding one more residue to the design of inhibitors, additional
interactions with SC2MPro might be achieved to improve potency.
In the design of SC1MPro inhibitors, Leu, Phe, and Cha
(cyclohexylalanine) are three residues used frequently at the P2
site and Val and Thr(tBu) (O-tert-butyl-threonine) are two residues
Structural characterization of SC2MPro interactions with Opal-
based inhibitors. In order to understand how our designed
inhibitors interact with SC2MPro at its active site, we screened
crystallization conditions for apo-SC2MPro, soaked apo-SC2MPro
crystals with different inhibitors, and determined the crystal
structures of these inhibitors in complex with SC2MPro. We used
Hampton Research Crystal Screen and Index kits to perform initial
screening and identified several conditions that yielded single
crystals of apo-SC2MPro. For all conditions, crystals were in a thin
plate shape (Supplementary Figure S1). The best crystallization
condition contained 0.2 M dibasic ammonium phosphate and 17%
PEG 3,350. We refined the structure of apo-SC2MPro against
diffraction data to 1.6 Å resolution (PDB: 7JPY). In the apoenzyme
crystals, SC2MPro existed as a monomer in the crystallographic
asymmetric unit and packed relatively densely. The active site of
each monomer stacked upon another monomer (two
representative monomers are shown in red and blue respectively
in Figure 3A). This close contact and dense protein packing made
the diffusion of inhibitors to the active site quite slow. We soaked
apo SC2MPro crystals with all 9 inhibitors that we synthesized and
Figure 3. X-ray crystallography analysis of SC2MPro in its apo-form and
complexes with different inhibitors. (A) The packing of apo-SC2MPro in its
crystals. An asymmetric unit monomer is colored in red in the center. Its active
site is presented as a concaved surface. Another monomer that stacks upon the
active site of the red monomer is colored in blue. (B) A contoured 2Fo-Fc map
at the 1σlevel around MPI3 and C145 in the active site of SC2MPro. A covalent
bond between MPI3 and C145 is observable. (C) The structure overlay between
apo-SC2MPro and the SC2MPro-MPI3 complex. A black arrow points to a region
that undergoes structure rearrangement in the SC2MPro-MPI3 complex from
apoenzyme to accommodate MPI3. (D) The occupation of the active site cavity
of SC2MPro by MPI3. The enzyme is shown in its surface presentation mode.
(E) Extensive hydrogen bonding and van del Waals interactions between
SC2MPro and MPI3. The backbone of SC2MPro is colored in marine blue and
side chain carbon atoms in orange. Hydro-gen bonds between MPI3 and
SC2MPro are depicted as yellow dashed lines. (F) The overlay of 7 Opal-based
inhibitors at the active site of SC2MPro. Inhibitors are colored according to their
color-coded names shown in the Figure. All images were made using the
program PyMOL. The PDB entry codes for SC2MPro in its apo-form and
complexes with inhibitors are 7JPY (apo), 7JPZ (MPI1), 7JQ0 (MPI3), 7JQ1
(MPI4), 7JQ2 (MPI5), 7JQ3 (MPI6), 7JQ4 (MPI7), 7JQ5 (MPI8).
Figure 2. SC2MPro inhibitors and their determined IC50 values. (A) Structures of
GC376 and 10 Opal-based inhibitors. (B) The inhibition curves of all 11 inhibitors
toward SC2MPro. Triplicate experiments were performed for each compound.
The determined IC50 values and Ki values are presented as mean ± standard
error (SE) in the associated table.
used frequently at the P3 site.[14] Installation of these residues at
two sites and including CBZ as a N-terminal cap led to the design
of six compounds MPI3-8 (Figure 2A). We added one additional
compound MPI9 that has an o-fluoro-p-chlorocinnamyl cap to this
series to compare the effect of the two N-terminal caps on the
inhibitor potency for SC2MPro. We synthesized all 7 compounds
according to a synthetic scheme presented in Supplementary
Scheme 2 and characterized their IC50 values using the
fluorescent peptide assay. As shown in Figure 2B, all inhibitors
have IC50 values below 100 nM, except for MPI8 that has an IC50
value as 105 ± 22 nM. The most potent compound is MPI3 with
an IC50 value as 8.5 ± 1.5 nM, followed by MPI4 and MPI5 with
IC50 values as 15 ± 5 and 33 ± 2 nM, respectively. We also
synthesized 11a (named as MPI10 in our series) according to the
procedure in Dai et al. and used it as a positive control in our
enzyme and viral inhibition analyses.[15b] Using our fluorescent
peptide assay, we determined the IC50 value of 11a as 31 ± 3 nM.
As far as we know, MPI3 is the most potent SC2MPro inhibitor that
has been reported so far. From the perspective of enzyme
inhibition, Leu and Val are optimal residues at P2 and P3 sites in
an inhibitor for improved affinity for SC2MPro and CBZ also
enhances affinity compared to the o-fluoro-p-chlorocinnamyl as a
N-terminal capping group.
collected and processed their X-ray diffraction data for structural
determination. For crystals that we soaked with the inhibitors for
just 2 h, we did not find observable ligand electron density at the
enzyme active site. For 7 inhibitors including MPI1 and MPI3-8,
we performed two-day soaking and observed clear electron
density in the difference maps in the active site of the enzyme.
For MPI2 and MPI9, we were not able to determine structures of
their complexes with SC2MPro due to cracking of the crystals upon
soaking with the inhibitors. For MPI3, the electron density around
the P1, P2, and P3 residues were well defined, and the covalent
3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
interaction between the C145 side chain thiolate and the Opal
aldehyde to form a hemiacetal was clearly observable (Figure 3B)
(PDB: 7JQ0). The electron density around CBZ was very weak
indicating flexible CBZ binding around the enzyme P4-binding
pocket. Figure 3C shows the superposition of apo SC2MPro and
the SC2MPro-MPI3 complex structures. The two structures display
very little overall variation with RMSD as 0.2 Å. Around the active
site in the two structures, large structural rearrangements exist for
residues M49 and N142 and the loop region that contains P168.
In apoenzyme, the side chain of M49 folds into the P2-binding
pocket. It flips toward the solvent to make space available for the
binding of the P2 Leu in MPI3. The side chain of N142 rotates by
conducted a live virus-based microneutralization assay in Vero E6
cells. Vero E6 is a kidney epithelial cell line isolated from African
Green Monkey. It has been used widely as a model system for
human CoV studies.[19] We tested 10 molecules including GC376,
MPI1-8, and 11a in a concentration range from 80 nM to 10 µM
and recorded cytopathogenic effect (CPE) observed in SARS-
CoV-2-infected Vero E6 cells that we cultured in the presence of
different concentrations of inhibitors. 11a was included as a
positive control. For each condition, we conducted two repeats.
Although it was disappointing that MPI3 was not able to
completely prevent CPE at all tested concentrations, several
inhibitors abolished CPE: GC376, MPI2, MPI6, and 11a at 10 µM,
MPI5 at 5 µM, MPI7 at 2.5-5 µM, and MPI8 at 2.5 µM (Figure 4A).
Three compounds MPI5, MPI7 and MPI8 performed better than
GC376 that has been recently explored by Anivive Lifesciences
for the treatment of COVID-19 and 11a that has been considered
for COVID-19 clinical studies.[11c] Since we only recorded
complete abolition of CPE, the real EC50 values for these
compounds are expected to be much lower than lowest observed
concentrations for CPE abolishment. Encouraged by our results
in Vero E6 cells, we tested the three most potent compounds
MPI5, MPI7, and MPI8 and also 11a in A549/ACE2 cells. The
A549/ACE2 cell line was derived from human alveolar epithelial
cells. It mimics the SARS-CoV-2 infection of the human
respiratory tract system better than Vero E6.[20] We tested a same
concentration range for all four compounds. MPI7 was not able to
almost 180° between the two structures and adopts
a
conformation in the SC2MPro-MPI3 complex that closely caps the
P1-binding site for strong van der Waals interactions with the Opal
residue in MPI3. In the SC2MPro-MPI3 complex, the P168-
containing loop is pushed away from its original position in the
apoenzyme, probably by interaction with the CBZ group, which
triggers a position shift for the whole loop. Except for M49, N142,
and the P168-containing loop, structural orientations of all other
residues at the active site closely resemble each other in the two
structures. In the active site, MPI3 occupies the P1, P2, and P4-
binding pockets and leaves the large P3’-binding pocket empty
(Figure 3D). Extensive hydrogen bonding and van der Waals
interactions in addition to the covalent interaction with C145
contribute to the strong binding of MPI3 to SC2MPro (Figure 3E).
Residues F140, N142, H163, E166, and H172 form a small cage
to accommodate the Opal side chain. Three hydrogen bonds form
between the Opal lactam amide and the E166 side chain
carboxylate, H163 imidazole, and F140 backbone carbonyl
oxygen. The precise fitting of Opal into the P1-binding pocket and
the formation of three hydrogen bonds explain the preferential
binding of the Opal side chain to this pocket. In the SC2MPro-MPI3
complex, M49 flips from the P2-binding pocket to leave space for
the binding of the P2 Leu in MPI3. Residues H41, M49, M165 and
D187, backbones of the M165-containing strand, and the D-187-
containing loop form a hydrophobic pocket that is in a close range
of van der Waals interactions with the P2 Leu in MPI3. We
observe Leu as the best residue in this position probably due to
this close van der Waals interaction range for the recognition of
the P2 Leu side chain. The enzyme has no P3-binding pocket.
However, the P3 Val in MPI3 positions its side chain in van der
Waals interaction distance to E166 and P168. In the structure,
CBZ narrowly fits into the P4-binding pocket and the channel
formed between the P168- and Q192-containing loops. The P168
loop rearranges its position from that in apoenzyme to
accommodate the CBZ group. The CBZ group also has weak
electron density. These observations indicate that CBZ is not an
optimal structural moiety for interaction at these sites. Besides
interactions involving side chains and the CBZ group in MPI3, its
two backbone amides and carbamate form 6 hydrogen bonds with
the enzyme. Two of them are formed between the P3 Val in MPI3
and the backbone amino and carbonyl groups of E166 in
SC2MPro. One water molecule mediates a hydrogen bond bridge
between the P2 Leu amino group in MPI3 and the Q189 side
chain amide in SC2MPro. For the P1 Opal residue in MPI3, its α-
amino group forms a hydrogen bond with the H164 α-carbonyl
oxygen in the enzyme. The original aldehyde oxygen in MPI3
forms two hydrogen bonding interactions, one with the α-amino
group of G143 and the other the C145 α-amine in SC2MPro. The
two hydrogen bonds are probably the reason that Opal-based
reversible covalent inhibitors are typically stronger than Michael
acceptor inhibitors, in which the original scissile amide is replaced
with an alkene, for inhibition of MPro enzymes. In the structures of
SC2MPro complexes with the other 6 inhibitors, we observed
similar structure rearrangements at M49, N142, and the P168-
containing loop to accommodate inhibitors and a covalent
interaction (PDB: 7JPZ, 7JQ1, 7JQ2, 7JQ3, 7JQ4, 7JQ5).
Figure 4. The SARS-CoV-2 viral inhibition results of selected inhibitors in (A)
Vero E6 and (B) A549/ACE2 cells. CPE is an abbreviation for cyto-pathogenic
effect.
completely abolish CPE at all tested conditions. However, both
MPI5 and MPI8 performed much better than in Vero E6 cells with
complete abolition of CPE at 160-310 nM and much better than
11a (Figure 4B). 11a displayed potency similar to that shown in
Vero E6 cells. Given that real EC50 values are expected to be
lower than the lowest observed concentration for CPE
abolishment, MPI5 and MPI8 are, as far as we know, the most
potent anti-SARS-CoV-2 small molecules in infected cells that
have been reported so far.
Discussion and Conclusion
Guided by previous medicinal chemistry studies about SC1MPro
,
we designed and synthesized a number of Opal-based dipeptidyl
and tripeptidyl inhibitors that potently inhibit SC2MPro, an essential
enzyme for SARS-CoV-2, the pathogen of COVID-19. As the
most potent inhibitor of SC2MPro, MPI3 displayed an IC50 value of
8.5 nM. As far as we know, this is the lowest reported IC50 for
known SC2MPro inhibitors. During the search of optimal conditions
for IC50 characterizations, we noticed that 10 nM was the lowest
SC2MPro concentration that could provide reliable activity.[11c] To
SARS-CoV-2 inhibition analysis of GC376, MPI1-8, and 11a.
To evaluate our molecules’ ability to inhibit SARS-CoV-2, we
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
characterize the Ki value of an inhibitor, the Km value for a used
substrate needs to be determined. The substrate we used for the
SC2MPro. This is expected since cellular stability and other
features of these inhibitors are very different. However,
information regarding both enzyme inhibition IC50 values and anti-
SARS-CoV-2 activity is critical for the design of a new generation
of inhibitors that perform excellent in both aspects. Given that
MPI3 has already reached a single digit nanomolar IC50 value and
MPI5 and MPI8 display high potency in inhibiting SARS-CoV-2,
merging features of the three molecules will lead to inhibitors with
extreme potency in inhibiting the virus. Our antiviral assays
indicated that MPI5 and MPI8 performed much better than GC376
and 11a, two molecules that have been explored for COVID-19
preclinical and clinical tests. These two molecules are ready for
preclinical analysis that we are actively exploring. We noticed in
our antiviral assays that MPI5 and MPI8 have much higher
potency in A549/ACE2 cells than in Vero E6 cells. These two cell
lines have different host protease proteomes. It is likely that MPI5
and MPI8 inhibit some host proteases that serve critical functions
in the SARS-CoV-2 entry into and replication in host cells and
therefore exert different SARS-CoV-2 inhibition.
inhibitor characterization has
a
sequence as Dabcyl-
KTSAVLQSGFRKME-Edans. At a fixed enzyme concentration at
20 nM, the substrate cleavage rate was roughly proportional to
the added substrate concentration up to 200 ꢀM (Figure S2).
Above 200 ꢀM, the substrate had a significant quenching effect
and was also not well soluble. Based on our results, the Km value
is about 422.4 ꢀM. Based on this determined Km value, Ki for all
the inhibitors are calculated and presented in Figure 2B. X-ray
crystallography analysis of the SC2M-MPI3 complex revealed
that MPI3 fits precisely in the P1- and P2-binding pockets at the
SC2MPro active site (PDB: 7JQ0). Strong van der Waals
interactions at the P1- and P2-binding pockets, 9 hydrogen bonds
with active site residues, and the covalent inter-action with C145
necessitate high affinity of MPI3 to SC2MPro. The N-terminal
capping group of MPI3 and other inhibitors are not well defined in
the crystal structures, indicating an unfitting size for this group or
relatively loosely bound pattern in P4-binding pocket.
Optimization on size or ligand-protein interacting to introduce
stronger interaction between ligand and SC2Mpro at this site would
contribute to the generation of more potent inhibitors in the future.
Although MPI3 is the most potent inhibitor for the enzyme, its
cellular activity in inhibiting SARS-CoV-2 is much lower than
several other inhibitors we have generated. A likely reason is its
lower cellular stability. MPI3 has Leu and Val at its P2 and P3
sites respectively. Both are naturally-occurring amino acids that
are expected to be targeted by both extracellular and cellular
proteases. Since Leu and Val are optimal residues at two sites,
modest changes based on these structures will be necessary for
both maintaining high potency in inhibiting SC2MPro and improving
cellular stability for enhanced cellular activity in inhibiting the virus.
As such, Val and Leu analogs at these two sites need to be
explored. Since both MPI5 and MPI8 show high anti-SARS-CoV-
2 activity in both Vero E6 and ACE2+ A549 cells and each has
Cha at their P2 site, we suggest maintaining Opal and Cha at P1
and P2 sites and varying the residue at P3 and the N-terminal
capping moiety to improve anti-SARS-CoV-2 activity in cells.
Based on our structures of SC2MPro complexes with 7 inhibitors,
the P1 Opal occupies precisely the P1-binding pocket in SC2MPro
and three hydrogen bonds to the Opal lactam amide are critical in
maintaining strong binding to SC2MPro. Chemical space to
manipulate the P1 residue in an inhibitor for improved binding to
SC2MPro is minimal. But one direction that may be explored is to
introduce additional heteroatom(s) to Opal for the formation of
hydrogen bond(s) with the N142 side chain am-ide. In the
SC2MPro-MPI3 complex, the N142 side chain flips by about 180°
from its position in apoenzyme to form a closed P1-binding pocket.
However, only van der Waals inter-actions with Opal are involved
with N142. Given the close distance between the Opal side chain
and the side chain amide of N142, some hydrogen bonds may be
designed for improved potency. In all our designed inhibitors, an
Opal aldehyde is involved in the formation of a covalent
interaction with C145. This design, although necessary for the
formation of a hemiacetal covalent complex, effectively excludes
the exploration of the P3’-binding site in SC2MPro for improved
potency in a designed inhibitor. Figure 4D illustrates that the P3’-
binding pocket is completely empty. In our early discussion, we
argued that it is critical to maintain the hydrogen bond between
the scissile amide oxygen in a substrate and SC2MPro for high
affinity. Changing the scissile amide to an aldehyde in an inhibitor
is effective in maintaining this hydrogen bond and allows a
covalent interaction with C145. Two hydrogen bonds formed
between the hemiacetal alcohol and SC2MPro contribute to high
potency of this group of molecules.
Acknowledgements
This work was supported in part by National Institutes of Health
(R01GM121584 to W. R. Liu and grant R01AI145287 to P. Li),
Welch Foundation (grant A-1715 to W.R. Liu and grant A-1987 to
C. A. Fierke), Texas A&M University X Grant Mechanism, Texas
A&M University Translational Investment Fund, the Texas A&M
University President’s Excellence Fund, and Texas A&M
University Strategic Transformative Research Program. We are
grateful to Prof. Thomas Meek for allowing us to use the LC-MS
system in his group for purification and characterization of some
of our compounds. The ALS-ENABLE beam-lines are supported
in part by the National Institutes of Health, National Institute of
General Medical Sciences, grant P30 GM124169-01 and the
Howard Hughes Medical Institute. The Advanced Light Source is
a Department of Energy Office of Science User Facility under
Contract No. DE-AC02-05CH11231.
Keywords: COVID-19 • SARS-CoV-2 • main protease • 3C-like
protease • reversible covalent inhibitors • antivirals
[1]
[2]
S. Adachi, T. Koma, N. Doi, M. Nomaguchi, A. Adachi, Frontiers in
immunology 2020, 11, 811.
a) K. McIntosh, J. H. Dees, W. B. Becker, A. Z. Kapikian, R. M.
Chanock, Proc Natl Acad Sci U S A 1967, 57, 933-940; b) D.
Hamre, J. J. Procknow, Proc Soc Exp Biol Med 1966, 121, 190-
193.
[3]
[4]
C. Drosten, S. Gunther, W. Preiser, S. van der Werf, H. R. Brodt,
S. Becker, H. Rabenau, M. Panning, L. Kolesnikova, R. A.
Fouchier, A. Berger, A. M. Burguiere, J. Cinatl, M. Eickmann, N.
Escriou, K. Grywna, S. Kramme, J. C. Manuguerra, S. Muller, V.
Rickerts, M. Sturmer, S. Vieth, H. D. Klenk, A. D. Osterhaus, H.
Schmitz, H. W. Doerr, N Engl J Med 2003, 348, 1967-1976.
a) L. van der Hoek, K. Pyrc, M. F. Jebbink, W. Vermeulen-Oost, R.
J. Berkhout, K. C. Wolthers, P. M. Wertheim-van Dillen, J.
Kaandorp, J. Spaargaren, B. Berkhout, Nat Med 2004, 10, 368-
373; b) P. C. Woo, S. K. Lau, C. M. Chu, K. H. Chan, H. W. Tsoi,
Y. Huang, B. H. Wong, R. W. Poon, J. J. Cai, W. K. Luk, L. L.
Poon, S. S. Wong, Y. Guan, J. S. Peiris, K. Y. Yuen, J Virol 2005,
79, 884-895.
[5]
[6]
J. F. Chan, K. S. Li, K. K. To, V. C. Cheng, H. Chen, K. Y. Yuen, J
Infect 2012, 65, 477-489.
C. Sohrabi, Z. Alsafi, N. O'Neill, M. Khan, A. Kerwan, A. Al-Jabir,
C. Iosifidis, R. Agha, Int J Surg 2020, 76, 71-76.
Wikipedia, 2020.
[7]
[8]
R. A. Khailany, M. Safdar, M. Ozaslan, Gene Rep 2020, 19,
100682.
[9]
[10]
T. Phan, Infect Genet Evol 2020, 81, 104260.
Y. W. Chen, C. B. Yiu, K. Y. Wong, F1000Res 2020, 9, 129.
In our study, cell-based anti-SARS-CoV-2 activity of our de-
signed inhibitors do not correlate with their IC50 values in inhibiting
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
[11]
a) Z. Jin, X. Du, Y. Xu, Y. Deng, M. Liu, Y. Zhao, B. Zhang, X. Li,
L. Zhang, C. Peng, Y. Duan, J. Yu, L. Wang, K. Yang, F. Liu, R.
Jiang, X. Yang, T. You, X. Liu, X. Yang, F. Bai, H. Liu, X. Liu, L. W.
Guddat, W. Xu, G. Xiao, C. Qin, Z. Shi, H. Jiang, Z. Rao, H. Yang,
Nature 2020, 582, 289-293; b) L. Zhang, D. Lin, X. Sun, U. Curth,
C. Drosten, L. Sauerhering, S. Becker, K. Rox, R. Hilgenfeld,
Science 2020, 368, 409-412; c) E. C. Vatansever, K. Yang, K. C.
Kratch, A. Drelich, C. C. Cho, D. M. Mellot, S. Xu, C. K. Tseng, W.
R. Liu, bioRxiv 2020, 382, 1677, preprint.
[12]
[13]
J. S. Morse, T. Lalonde, S. Xu, W. R. Liu, Chembiochem 2020, 21,
730-738.
T. Muramatsu, C. Takemoto, Y. T. Kim, H. Wang, W. Nishii, T.
Terada, M. Shirouzu, S. Yokoyama, Proc Natl Acad Sci U S A
2016, 113, 12997-13002.
[14]
[15]
T. Pillaiyar, M. Manickam, V. Namasivayam, Y. Hayashi, S. H.
Jung, J Med Chem 2016, 59, 6595-6628.
a) N. C. Pedersen, Y. Kim, H. Liu, A. C. Galasiti Kankanamalage,
C. Eckstrand, W. C. Groutas, M. Bannasch, J. M. Meadows, K. O.
Chang, J Feline Med Surg 2018, 20, 378-392; b) W. Dai, B. Zhang,
X. M. Jiang, H. Su, J. Li, Y. Zhao, X. Xie, Z. Jin, J. Peng, F. Liu, C.
Li, Y. Li, F. Bai, H. Wang, X. Cheng, X. Cen, S. Hu, X. Yang, J.
Wang, X. Liu, G. Xiao, H. Jiang, Z. Rao, L. K. Zhang, Y. Xu, H.
Yang, H. Liu, Science 2020, 368, 1331-1335.
[16]
[17]
[18]
Y. Kim, S. Lovell, K. C. Tiew, S. R. Mandadapu, K. R. Alliston, K.
P. Battaile, W. C. Groutas, K. O. Chang, J Virol 2012, 86, 11754-
11762.
A. K. Ghosh, G. Gong, V. Grum-Tokars, D. C. Mulhearn, S. C.
Baker, M. Coughlin, B. S. Prabhakar, K. Sleeman, M. E. Johnson,
A. D. Mesecar, Bioorg Med Chem Lett 2008, 18, 5684-5688.
C. Ma, M. D. Sacco, B. Hurst, J. A. Townsend, Y. Hu, T. Szeto, X.
Zhang, B. Tarbet, M. T. Marty, Y. Chen, J. Wang, Cell research
2020, 30, 678-692.
[19]
[20]
M. L. Ng, S. H. Tan, E. E. See, E. E. Ooi, A. E. Ling, J Gen Virol
2003, 84, 3291-3303.
Y. T. Yen, F. Liao, C. H. Hsiao, C. L. Kao, Y. C. Chen, B. A. Wu-
Hsieh, J Virol 2006, 80, 2684-2693.
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.202000924
ChemMedChem
FULL PAPER
Entry for the Table of Contents
We designed and synthesized a series of SARS-CoV-2 MPro covalent inhibitors that exhibited excellent inhibitory activity. Protein
crystallography analysis and a live virus-based microneutralization assay found two most potent anti-SARS-CoV-2 small molecules
so far. Due to the urgent matter of the COVID-19 pandemic, these two inhibitors may be quickly advanced to preclinical and clinical
tests for COVID-19.
Researcher Twitter usernames: liulab@tamu
7
This article is protected by copyright. All rights reserved.