Application of Dually Activated Michael Acceptor to the Rational Design of Reversible Covalent Inhibitor for Enterovirus 71 3C Protease

Article abstract of DOI:10.1021/acs.jmedchem.9b00387

Targeted covalent inhibitors (TCIs) have attracted growing attention from the pharmaceutical industry in recent decades because they have potential advantages in terms of efficacy, selectivity, and safety. TCIs have recently evolved into a new version with reversibility that can be systematically modulated. This feature may diminish the risk of haptenization and help optimize the drug-target residence time as needed. The enteroviral 3C protease (3C^pro) is a valuable therapeutic target, but the development of 3C^pro inhibitors is far from satisfactory. Therefore, we aimed to apply a reversible TCI approach to the design of novel 3C^pro inhibitors. The introduction of various substituents onto the α-carbon of classical Michael acceptors yielded inhibitors bearing several classes of warheads. Using steady-state kinetics and biomolecular mass spectrometry, we confirmed the mode of reversible covalent inhibition and elucidated the mechanism by which the potency and reversibility were affected by electronic and steric factors. This research produced several potent inhibitors with good selectivity and suitable reversibility; moreover, it validated the reversible TCI approach in the field of viral infection, suggesting broader applications in the design of reversible covalent inhibitors for other proteases.

Full text of DOI:10.1021/acs.jmedchem.9b00387

Subscriber access provided by Stockholm University Library
Article
Application of Dually Activated Michael Acceptor to the Rational
Design of Reversible Covalent Inhibitor for Enterovirus 71 3C Protease
Yuying Ma, Linfeng Li, Shuai He, Chengyou Shang, Yang Sun,
Ning Liu, Thomas D. Meek, Yaxin Wang, and Luqing Shang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00387 • Publication Date (Web): 11 Jun 2019
Downloaded from http://pubs.acs.org on June 12, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Application of Dually Activated Michael
Acceptor to the Rational Design of Reversible
Covalent Inhibitor for Enterovirus 71 3C
Protease
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
‡
†
†
||
†
Yuying Ma, Linfeng Li, Shuai He, Chengyou Shang, Yang Sun, Ning Liu, Thomas
‡
*,§
*,†
D. Meek, Yaxin Wang, and Luqing Shang
†College of Pharmacy, State Key Laboratory of Medicinal Chemical Biology and Tianjin
Key Laboratory of Molecular Drug Research, Nankai University, Tianjin 300350, China
‡Department of Biochemistry and Biophysics, Texas A&M University, College Station,
Texas 77843, United States
||Center of Basic Molecular Science, Department of Chemistry, Tsinghua University,
Beijing 100084, China
§School of Life Sciences, Tianjin University, Tianjin 300110, China
KEYWORDS: Enterovirus 71, 3C protease, reversible targeted covalent inhibitor,
Michael acceptor, time-dependent inhibition.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: Targeted covalent inhibitors (TCIs) have attracted growing attention from
the pharmaceutical industry in recent decades because they have potential advantages
in terms of efficacy, selectivity and safety. TCIs have recently evolved into a new version
with reversibility that can be systematically modulated. This feature may diminish the risk
of haptenization and help optimize the drug-target residence time as needed. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pro
enteroviral 3C protease (3C ) is a valuable therapeutic target, but the development of
3
Cpro inhibitors is far from satisfactory. Therefore, we aimed to apply a reversible TCI
approach to the design of novel 3Cpro inhibitors. The introduction of various substituents
onto the α-carbon of classical Michael acceptors yielded inhibitors bearing several
classes of warheads. Using steady-state kinetics and biomolecular mass spectrometry,
we confirmed the mode of reversible covalent inhibition and elucidated the mechanism
by which the potency and reversibility were affected by electronic and steric factors. This
research produced several potent inhibitors with good selectivity and suitable reversibility;
moreover, it validated the reversible TCI approach in the field of viral infection, suggesting
broader applications in the design of reversible covalent inhibitors for other proteases.
INTRODUCTION
Most of the early covalent drugs such as penicillin, omeprazole and aspirin were
1
discovered by serendipity or resulted from phenotypic screening. Although these drugs
proved to be great successes and are being widely used today, the pharmaceutical
industry has long been reluctant to develop covalent drugs because of safety concerns
that are associated with the nonspecificity and irreversibility of covalent inhibition. In
recent years, the growing fields of structural biology and bioinformatics have greatly
promoted the rational design of targeted covalent inhibitors (TCIs).1-2 Over the past two
decades, several TCIs have been approved by the FDA, while more are in clinical trials
3
or being used as chemical tools in drug discovery for various classes of proteins. Among
the benefits specific to these TCIs are high biochemical potency, improved selectivity and
reduced drug resistance.2, 4
2
ACS Paragon Plus Environment

Page 3 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In general, the mechanism by which a TCI inhibits its target occurs in two steps.
The inhibitor initially recognizes and noncovalently binds to the target site, then forms
5
bond with certain nucleophilic residue (Figure 1A). Accordingly, the optimization of a TCI
takes into account both the initial binding affinity (K) and the chemical reactivity (kinact).
i
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Sometimes the chemical reaction may be reversible, resulting in two types of reversible
TCIs. The first type of inhibitor, bearing warhead like aldehyde or nitrile, can form unstable
2
adduct with target which readily break down on the biological timescale. In contrast, the
adduct formed by the second type of inhibitor is metastable. The reversal of such adduct
is relatively slow and can be systematically modulated, as exemplified by some recently
6
developed electron-deficient alkenes. These inhibitors were thought to possess a
reduced risk of hapten formation and to enable fine-tuning of the inhibitor-target residence
time. The potency of a reversible TCI is determined by the overall inhibition constant Ki*
that depends not only on the K but also on the relative magnitudes of the values for k
i
5
and k (Figure 1A).
6
Figure 1. (A) Generic two-step mechanism of irreversible/reversible covalent inhibition. kinact/K
usually used for evaluation of irreversible inhibitors, while K
inhibitor that conforms to a two-step mechanism. The relationship between K
B) Structures and bioactivity data of rupintrivir and compound 1. Apparent IC50 or K
i
is
i
* measures the potency of reversible covalent
, K*, k and k are described.
* values are indicative
i
i
5
6
(
i
of their inhibitory effects on EV71 3Cpro. EC50 values represent their antiviral activity in EV71-infected human
rhabdomyosarcoma (RD) cells.
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Enterovirus 71 (EV71) is one of the causative agents of hand, foot and mouth
disease (HFMD), which tends to infect children and may be associated with severe
7
pro
neurological and cardiopulmonary disorders. The 3C cysteine protease (3C ) is an
essential viral enzyme that is now considered an attractive anti-EV71 target. Currently,
noncovalent inhibitors obtained from screening methods produced disappointing
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
inhibitory effects. For peptidyl inhibitors, the existence of an appropriate electrophile at
the P1’ position is critical for the potency, based on cumulative knowledge of enteroviral
pro 9
3C . The most studied inhibitor is rupintrivir, originally a covalent protease inhibitor of
human rhinovirus (HRV) which is homologous to EV71. This inhibitor employs a classical
Michael acceptor (α, β-unsaturated ethyl ester) as its warhead that reacts irreversibly with
the active-site cysteine (Figure 1B).10 However, rupintrivir is less potent against EV71
3Cpro than it is against HRV.11 To develop a better TCI for EV71 3C , we previously
synthesized a number of peptidyl inhibitors.9b, 12 The binding affinities of these inhibitors
were improved by modifying the peptidomimetic scaffold; however, the inhibitors had
druglikeness and potency issues, presumably because of their aldehyde and α-
ketoamide-based warheads.
pro
The aforementioned slow reversible TCI presents a prospective approach for the
development of a potent 3Cpro inhibitor with fewer disadvantages. Here our principal goal
is to adapt novel Michael acceptors as warheads with suitable reactivities that can form
pro
covalent—ideally slowly reversible—adducts with the catalytic Cys147 of EV71 3C . To
this end, a variety of gem-disubstituted alkenes were attached to a common
peptidomimetic scaffold. By means of steady-state kinetics and mass spectrometry, we
assessed the capability of these warheads to provide optimal chemical reactivity and
reversibility. Our assessment not only generated several potent inhibitors with desired
reversibility and high selectivity, more importantly, it also developed a systematic insight
into the complex relationship between the chemical properties of gem-disubstituted
alkenes and the bioactivities of their corresponding inhibitors.
RESULTS AND DISCUSSION
4
ACS Paragon Plus Environment

Page 5 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and characterization of compounds 2-33
In this article, 32 new compounds were synthesized and characterized. All target
compounds started from aldehydic compound 1, for which a synthetic procedure was
13
previously reported. The key step of the synthetic work was to build a double bond, and
for this purpose, we mainly used two reactions: Horner-Wadsworth-Emmons (HWE)
reaction and Knoevenagel condensation.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of intermediates 34-45 and target compounds 2-20a.
O
O
P
R1
a
b
O
O
P
O
O
R1
R2
34-45
O
P
R2
R2
+
R1
R2
R
H
c/d
e
O
O
Cl
R
R1
R
R1
O
3
3
4: R1 = CN, R2 = CH3
1
2-20
18
5: R1 = CN, R2 = Cl
36: R1 = CN, R2 = Br
2: (E) R1 = COOCH , R = H
2
12: (Z/E) R1 = COOCH CH , R = Br
2
H
3
2
3
O
N
7: R1 = COOCH , R = Cl
2
3: (E) R1 = COOCH CH , R = H 13: (Z) R1 = CN, R2 = Ph
2
3
3
3
2 3
3
8: R1 = COOCH , R = Br
2
4: (E) R1 = CONHCH , R = H
2
14: (Z) R1 = CN, R2 = Naph
3
3
O
9: R1 = COOCH CH , R2 = Cl
5: (E) R1 = CN, R2 = H
15: (Z) R1 = CN, R2 = 4-(CF )-Ph
2
3
3
H
O
6: (Z/E) R1 = CN, R2 = CH3
16: (E) R1 = 2-Py, R2 = CN
17: (Z) R1 = CN, R2 = 2-Py
18: -- R1 = CN, R2 = 2-Py
N
40: R1 = COOCH CH , R = Br
2
R =
2
3
N
N
41: R1 = CN, R2 = Ph
7: (Z/E) R1 = CN, R2 = Cl
8: (Z/E) R1 = CN, R2 = Br
H
O
e
4
4
4
4
2: R1 = CN, R2 = Naph
3: R1 = CN, R2 = (4-CF )-Ph
9: (Z/E) R = COOCH , R = Cl
1
2
19: (E) R1 = 3-Thio, R2 = CN
20: (Z) R1 = CN, R2 = 3-Thio
3
3
F
4: R1 = CN, R2 = 2-Py
5: R1 = CN, R2 = 3-Thio
10: (Z/E) R1 = COOCH , R = Br
2
3
11: (Z/E) R1 = COOCH CH , R = Cl
2
2
3
aReagents and conditions: (a) LiHMDS, NBS or NCS, THF, 0 °C to RT, 22-52% (expect for 34 prepared
with K CO , MeI, DMF, RT, 62%); (b) LiHMDS, THF, N , -78 °C to 0 °C, 48-67%; (c) NaHMDS, THF, N
8 °C to 0 °C, 80-83% (for compound 2 and 3); (d) LiCl, DBU, MeCN, RT, 70-85%; (e) Pd/C (10 wt %
Pd), H , MeOH, 40 °C, 38%.
2
3
2
2
, -
7
2
The synthetic routes to singly activated alkenes 2-5, methyl acrylonitrile 6,
halogenated acrylates/acrylonitriles 7-12, aryl/heteroaryl acrylonitrile derivatives 13-20
are outlined in Scheme 1 where the HWE reaction is the central step. Alkenes 2-5 were
obtained by reacting 1 with corresponding commercially available phosphonates using
NaHMDS or LiCl/DBU as the base according to a protocol modified by Blanchette et al,14
while the preparation of 6-17 and 19-20 required the in-lab synthesis of 34-45.
Phosphonate 34 was prepared by treating diethyl cyanomethylphosphonate with
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
iodomethane (MeI) as the methylating agent and potassium carbonate (K CO ) as the
2
3
base. Phosphonates 35-40 were produced from commercially obtained phosphonates
that underwent halogenation with N-chlorosuccinimide (NCS) or N-bromosuccinimide
15
(NBS). The bromo derivatives (36, 38 and 40) were particularly unstable under ambient
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
conditions and had to be used in the next step immediately. Phosphonates 41-45 were
16
obtained by electrophilic phosphorylation of lithiated nitriles. It was found that 44 existed
17
as tautomers due to spontaneous proton transfer. Finally, the saturated compound 18
was produced by Pd/C-catalyzed hydrogenation of 17.
Scheme 2. Synthesis of compounds 21-33a
R3
R3
R
H
f
e
+
R3 R4
R
R
R4
R4
O
1
21-33
26
3
4
3
4
21: (E) R = CN, R = COOCH
31: (E) R = CN, R = CON(Boc)CH CH
3
2
3
e
g
3
4
3
4
2
2
2
2
2
3
2
6: -- R = CN, R = COOCH
28: (E) R = CN, R = CONHCH CH
3
2 3
3
4
3
4
2: (E) R = CN, R = COOCH CH
32: (E) R = CN, R = CON(Boc)CH CH CH
2 2 3
2
3
g
3
4
3
4
3: (E) R = CN, R = COOCH CH CH
29: (E) R = CN, R = CONHCH CH CH
2 2 3
2
2
3
3
4
3
4
4: (E) R = CN, R = COOCH(CH )
33: (E) R = CN, R = CON(CH )
3 2
3
2
3
4
5: (E) R = CN, R = COOC(CH )
3
3
3
4
0: (E) R = CN, R = CON(Boc)CH
3
g
3
4
7: (E) R = CN, R = CONHCH
3
aReagents and conditions: (e) Pd/C (10 wt % Pd), H
, MeOH, RT, 57%; (f) piperidine, DCM, RT or heated,
2
26-55%; (g) TFA, DCM, RT, 26-35%. Structure of 1 was drawn in Scheme 1.
The synthetic routes to cyanoacrylate derivatives 21-26 and cyanoacrylamides 27-
33 are outlined in Scheme 2, where Knoevenagel condensation is the central step. The
cyanoacrylates 21-26 and Boc-protected cyanoarylamides 30-33 were obtained by
reaction of 1 with corresponding active methylene compounds under Knoevenagel
conditions. The cyanoacrylamides 27-29 were obtained by removing the Boc groups of
30-32 with TFA at an ambient temperature. Like the saturated 18, compound 26 was
converted from 21.
The double bonds of 2-5 in Scheme 1 and all compounds in Scheme 2 were
determined to be exclusively in the E form, while those of the aryl acrylonitriles 13-15
6
ACS Paragon Plus Environment

Page 7 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were exclusively in the Z form. Compounds 6-12 were obtained as mixtures of E/Z
isomers and were used directly in kinetic evaluations, whereas we were able to separate
the geometric isomers of heteroaryl acrylonitriles 16, 17, 19 and 20. Two methods were
employed to identify the configuration of trisubstituted alkenes. For nitrile-containing
compounds, the double bond configuration could be inferred by the three-bond
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
18
heteronuclear coupling constant ( J ) between the vinyl proton and the cyano carbon.
CH
For ester derivatives 9-12, the configuration was determined via a NOESY technique. The
determination of double bond configuration by NMR is detailed in the Supporting
Information (Figures S1-S5).
Classical Michael acceptors
Based on the structure-activity relationships that we reported previously,9b, 12 the
peptidomimetic scaffold shown in Figure 1B was adopted throughout our present study
to offer optimal specificity for enzyme-inhibitor recognition. We commenced our survey of
warheads by using classical Michael-acceptor acrylates (2, 3), acrylamide (4) and the
rarely used acrylonitrile (5). EV71 3Cpro was preincubated with 2, 4 or 5, and then
ultrafiltered to remove unbound inhibitors. The sample was analyzed by ESI-MS to
examine the formation of an enzyme-inhibitor adduct. The degree of covalent modification
of protein, expressed as the ratio of peak intensity of covalently modified to unmodified
protein (I /I ), was >100 for three samples (Figure 2A), indicating the formation of a
add pro
stable covalent bond. Notably, longer pre-incubation period and higher concentration of
inhibitor were required for 5 to achieve a level of covalent modification that was
comparable to 2 and 4, suggesting lower intrinsic reactivity of the acrylonitrile for the
addition of thiol.
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. (A) Deconvoluted ESI-MS spectra of 3Cpro and its adduct with classical Michael acceptors. From
top to bottom are—unmodified 3Cpro, 3Cpro incubated with 10 equiv 2, 4, 5 for 30 min and 3Cpro incubated
with 20 equiv 5 for 1.0 h, respectively. (B) Time-course inhibition of 3Cpro at variable concentrations of 2.
Lines drawn through the time courses result from fitting of data to equation 2 in Experiment Section. (C)
Replotting the calculated kobs values over concentration of 2 (equation 3 in Experiment Section) yields a
partial hyperbola. Error bar represents standard error of the mean (SEM, n = 3). (D) Time-course inhibition
of 3Cpro at variable concentrations of 5.
The time-course of 3Cpro inactivation by compounds 2-4 displayed a time-
dependent behavior and a steady-state velocity approaching zero (Figure 2B, Figures
S6A and C), both of which are features of a two-step irreversible covalent mechanism.19
According to the data processing method described in the Experimental Section, the
observed rate constants (k ) fitted from these curves exhibited a hyperbolic dependence
obs
on the inhibitor concentration (Figure 2C, Figures S6B and D), which further supported
the two-step mechanism. The second-order rate constants of inactivation (kinact/K) that
i
are listed in Table 1 were used to compare the potency of our compounds. Compounds
8
ACS Paragon Plus Environment

Page 9 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
-
1
-1
2
-4 showed similarly modest activity (kinact/K= 34-72 M ·s , EC = 17-21 µM), whereas
i
50
compound 5 produced discernable inhibition only at the maximal concentration (64 µM,
Figure 2D), which agreed with the conclusion from the MS experiments that acrylonitrile
had a much poorer reactivity toward Cys147 addition. Overall, this series of classical
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pro
Michael acceptors could irreversibly inactivate 3C , yet their inhibitory effects were
barely satisfactory.
Table 1. Inhibition data of compounds 2-5.
H
O
N
O
X
H
N
O
N
N
H
Y
O
F
scaffold A
NO
.
K
inact/K
i
EC50
Cytotoxicity
Config.a
Scaffold
X
Y
-1 -1 b,c
(nM)b
at 100 μM)b,c
(
M ·s )
(
18,300 ±
2
3
4
5
A
A
A
A
H
H
H
H
COOCH3
E
E
E
E
34 ± 5
72 ± 14
46 ± 4
N.D.
5,100
COOCH CH
17,000 ±
3,900
2
N.D.
N.D.
N.D.
3
21,000 ±
CONHCH3
5
,700
7,000 ±
,200
3
CN
N.D.
4
a
1
b
The configuration was determined by H-NMR spectrum. Each value represented the average result
c
from three independent experiments (SEM indicated). N.D.: Not detected under the experimental
condition.
Michael acceptors activated with weak EWGs
Taunton and coworkers described reversible covalent inhibitors in which the
introduction of a second electron-withdrawing group (EWG) on the α-carbon of a Michael
acceptor rendered the dually activated double bond intrinsically more reactive, while due
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to concomitant increased acidity of the α-proton, the thioadduct of such Michael acceptor
was readily reversible via an E1cB elimination.20 Therefore, based on a preliminary
21
investigation of the effect of EWGs on alkenes, we started to identify appropriate second
substituents, either to render the acrylate/acrylamide more reversible or to activate the
inert acrylonitrile. As will be discussed later, these dually activated alkenes are basically
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
reversible, so their potencies are measured by their overall inhibition constant K* instead
i
of kinact/Ki.
As shown in Table 2, to characterize a full spectrum of electronic properties, an
electron-releasing methyl group was first introduced (6), which unsurprisingly, made no
difference to the inhibitory behavior of acrylonitrile. Halogen atoms, which are considered
to be relatively weak EWGs, were then incorporated into 2, 3 and 5. Halogenated
acrylonitriles (7, 8) also exhibited no inhibition against the protease or the virus, thereby
demonstrating insufficient electron-withdrawing capability by the halogens for the
activation of acrylonitrile. In contrast, halogenation of acrylates (9-12) led to considerable
potency with K* of 115-164 nM and antiviral activity (EC ) of 290-670 nM. Unfortunately,
i
50
these halogenated derivatives were accompanied by substantial cytotoxicity and, thus,
were not pursued further.
To continue searching for the methods for activating the acrylonitrile warhead, we
next replaced halogens with several aryl or heteroaryl activating groups (13-20). The
introduction of a phenyl or a naphthyl group (13, 14) made little contribution, similar to the
effect of electron-releasing methyl group with compound 6. In an attempt to increase the
electron withdrawing capability, incorporation of p-trifluoromethyl (15) resulted in only a
slightly improved cell-based antiviral activity (EC = 1,400 nM). In fact, this antiviral effect
50
may be nonspecific since significant cytotoxicity was detected at a relatively low dose (33
µM). Unlike aryl hydrocarbons, compounds 16 and 17, containing an electron-deficient o-
pyridyl group, demonstrated measurable values of K* (693 nM and 119 nM, respectively)
i
and acceptable cell-based antiviral activity (EC : 2,700 nM and 460 nM, respectively).
50
Unfortunately, these compounds also exhibited cellular toxicity. For compound 17, the
10
ACS Paragon Plus Environment

Page 11 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
lack of inhibitory activity of its alkane counterpart 18 suggested the importance of covalent
bond formation for enzyme inhibition. Replacing the pyridyl group (17) with an electron-
rich thiophenyl group (20) decreased the inhibitory activity by approximately 5-fold in both
enzyme and cell-based assays. It is noteworthy that the Z configuration is generally the
active form for heteroaryl acrylonitriles (17 and 20), which geometrically corresponds to
the E configuration of the cyanoacrylates and cyanoacrylamides that are discussed below.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 2. Inhibition data of compounds 6-20.
H
H
O
N
O
N
O
X
O
X
H
H
O
O
N
N
N
N
Y
N
N
Y
H
H
O
O
F
F
scaffold A
scaffold B
Ki*
EC50
Cytotoxicity
NO. Scaffold
X
Y
Config.a
(nM)b,c
(nM)b
(at 100 μM)b,c
E:Z =
6
7
8
9
A
A
A
A
A
A
A
CH3
Cl
CN
CN
CN
N.D.
>100,000
>100,000
>100,000
670 ± 160
290 ± 60
380 ± 140
350 ± 80
N.D.
2:1
E:Z =
N.D.
N.D.
100%
100%
100%
100%
100%
100%
4:5
E:Z =
Br
Cl
2
:1
E:Z =
:2
COOCH3
COOCH3
152 ± 24
115 ± 11
164 ± 21
117 ± 17
1
E:Z =
10
Br
Cl
3:5
E:Z =
11
COOCH CH
2
3
3
1
:1
E:Z =
:5
12
Br
COOCH CH
2
2
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
1
1
1
3
4
5
6
A
A
A
A
A
B
A
A
CN
CN
Ph
Z
Z
Z
E
Z
--
E
Z
N.D.
N.D.
>100,000
13,000 ± 2,100
1,400 ± 400
2,700 ± 700
460 ± 100
N.D.
N.D.
Naph
CN
(4-CF )-Ph
N.D.
100%
90%
3
2-Py
CN
CN
2-Py
2-Py
CN
693 ± 95
119 ± 15
N.D.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
17
18
100%
50%.
N.D.
CN
>100,000
1
9
0
3-Thio
CN
N.D.
19,500 ± 4,200
3,100 ± 400
2
3-Thio
516 ± 42
N.D.
a
1
b
The ratio of E/Z isomers was determined by H-NMR spectrum. Each value represented the average
c
results from three independent experiments (SEM indicated). N.D.: Not detected under the experimental
condition.
Michael acceptors activated with strong EWGs
Next, we synthesized two series of compounds: cyanoacrylates 21-25 and
cyanoacrylamides 27-33. Because carbonyl (ester or amide) and cyano groups are both
strongly electron-withdrawing, their combination synergistically activated the double bond
and thus conferred the new Michael acceptors with excellent potency, as shown in Table
. The inhibitory effects of cyanoacrylates 21-25 on 3Cpro were at virtually the same level
6.7-16.1 nM), regardless of the size or shape of the ester sidechains. Furthermore, the
3
(
antiviral activity of these inhibitors also fell into a satisfactory range (30-77 nM), which
correlated well with the enzyme inactivation data and was comparable to rupintrivir (EC50
11
pro
=
14-41 nM). To gain an insight into their interaction with 3C , a molecular modeling
study of compound 25 was carried out based on a covalent-docking protocol. The top-
scoring pose of 25 (ΔGcalc = -7.22 kcal/mol) aligned well with rupintrivir that was
determined by X-ray crystallography (Figure 3A). The hydrogen bonding patterns of their
scaffolds were nearly conserved, while a slight difference was observed at the P1’
position. In addition to the fact that the carbonyl oxygens in both 25 and rupintrivir served
as hydrogen bond acceptors, the cyano group of the former could also form another
hydrogen bond with His . The tert-butyl group of 25, being the bulkiest group in this series
40
12
ACS Paragon Plus Environment

Page 13 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of compounds, fitted snugly into the S1’ pocket, probably because of an entropy-driven
expulsion of bound water from the active site. All of these determinants resulted in tight
binding of these cyanoacrylates. In contrast to the curvilinear time-courses obtained with
the singly activated Michael acceptors, the progress curves for the cyanoacrylate series
were essentially linear (Figure 3B and Figure S7) as their values of kobs were too large to
be measured by steady state kinetic method, which demonstrated a negligible steady-
state level of the initial E·I complex and rapid equilibrium for the E·I* complex. In other
words, the good initial binding of inhibitors presumably oriented the double bond toward
Cys , and dual activation by strong EWGs further accelerated formation of the adduct.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
147
Figure 3. (A) Top-scoring docking pose of 25 (yellow) was superposed on rupintrivir (cyan) from crystal
structure (PDB ID: 3SJO). Hydrogen bonds were drawn as dashed lines. The bound water was labeled as
solid sphere. Residues at S1’ pocket were labeled. (B) Time-course inhibition of 3Cpro at variable
concentrations of 25. Linear progress curves suggest rapid equilibrium of enzyme-inhibitor complex. (C)
Raw ESI-MS spectrum of 3Cpro incubated with 25. The appearance of molecular ion peak of 25 implied that
it came along with 3Cpro but detached during or after ionization.
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
MS analysis was also carried out for cyanoacrylates. Although the MS results for
2
1 (Figure S8) or 25 (Figure 3C) did not explicitly demonstrate the formation of a covalent
bond, the molecular ion peaks of both small molecules did appear in their raw spectra.
Similar to the treatments with classical Michael acceptors, ultrafiltration was performed
three times to assure the removal of any free small molecules before analysis by mass
spectrometry; therefore, we could safely conclude that the observed molecular ion was
not a contaminant but was cleaved from 3Cpro during or after ionization. Compared with
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pro
pro
the more stable 3C -2 complex, the apparently less stable 3C -21/25 complex implied
the increased reversibility of this complex due to the electron-withdrawal from the added
cyano group. Saturation of the double bond (26) caused total loss of inhibitory activity,
further corroborating the essentiality of Michael addition in order to achieve enzyme
inhibition.
Table 3. Inhibition data of compounds 21-33.
Cytotoxicity
Ki*
EC50
NO. Scaffolda
X
Y
Config.b
(at 100
μM)c,d
(nM)c,d
(nM)c
9
.4 ±
2
1
2
A
A
A
A
CN
CN
CN
CN
COOCH3
E
E
E
E
53 ± 9
72 ± 13
77 ± 20
58 ± 10
N.D.
N.D.
N.D.
N.D.
5.0
13.1 ±
2
COOCH CH
2
3
4
.0
0.3 ±
.2
6.1 ±
.7
6.7 ±
.0
N.D.
1
1
23
COOCH CH CH
2
2
3
3
24
COOCH(CH3)2
4
2
5
6
A
B
CN
CN
COOC(CH3)3
COOCH3
E
30 ± 8
N.D.
N.D.
2
2
--
>100,000
14
ACS Paragon Plus Environment

Page 15 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
04 ±
2
2
2
7
8
9
A
A
A
A
CN
CN
CN
CONHCH3
E
E
E
E
400 ± 80
510 ± 50
990 ± 110
51 ± 4
N.D.
N.D.
N.D.
N.D.
20
332 ±
CONHCH CH
2
3
21
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
752 ±
CONHCH CH CH
2
2
3
32
9
.3 ±
3
0
1
CN
CN
CON(Boc)CH3
4.3
3
A
A
A
CON(Boc)CH CH
E
E
E
24 ± 7.8
18 ± 8.4
67 ± 10
89 ± 20
110 ± 6
N.D.
N.D.
N.D.
2
3
32
33
CN CON(Boc)CH CH CH
2
2
3
CN
CON(CH3)2
94 ± 10
a
b
The structures of scaffold A and B were described in Table 2. The method for the configuration
c
identification was described in Supporting Information. Each value represented the average results from
d
three independent experiments (SEM indicated). N.D.: Not detected under the experimental condition.
The cyanoacrylamide series (27-33) constituted a more complicated yet intriguing
scenario. Although our primary goal was to obtain N-monosubstituted cyanoacrylamides
2
7-29 bearing structural similarity to their corresponding cyanoacrylates, these inhibitors
demonstrated only modest inhibition of 3Cpro with respective K* values of 304, 332 and
i
7
52 nM. To our surprise, the Boc-protected synthetic intermediates (30-32) of these
inhibitors exhibited on average an ~30-fold improvement in inhibitory activity (Table 3).
According to the docking result of 30, unlike 25, its warhead seemed too large to be
accommodated by the S1’ pocket (Figure 4A). Despite the formation of two hydrogen
bonds between warhead and active site residues (His and Gly ), the calculated binding
4
0
145
affinity of 30 (ΔGcalc = -6.09 kcal/mol) was less than that of 25 (ΔGcalc = -7.22 kcal/mol).
Furthermore, as shown in Figure 4B, the obviously bending progress curve of Boc-
protected 31 demonstrated a slow-binding behavior compared with cyanoacrylate (21,
2
5) or N-monosubstituted acrylamide (28), which indicated that establishment of the
binding equilibrium was retarded.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. (A) Top-scoring docking pose of 30 (green). Hydrogen bonds were drawn as dashed lines. Red
colored area represented hydrophobic protein surface. The arrow pointed to α-proton of Michael acceptor,
which resided in a hydrophobic region and was shielded by the P1’. (B) Time-course inhibition of 3Cpro at
variable concentrations of 28 (left) and 31 (right). 31 displayed a more obvious slow-binding behavior (time-
dependency) than 28. (C) Deconvoluted ESI-MS spectra of 3Cpro incubated with different
cyanoacrylamides. Upper left: unmodified 3Cpro; lower left: 3Cpro-27; upper right: 3Cpro-30; lower right: 3Cpro
3. (D) Putative reaction mechanism for the elimination of 3Cpro-31 adduct. (E) Top-scoring docking pose
of 33 (magenta). Other representations were similar with (A). The α-proton was solvent-exposed.
-
3
Both computational and kinetic results indicated that the Boc-protected
pro
cyanoacrylamides should have difficulty in the initial binding to 3C , thus resulting in low
affinity (high K value). Therefore, there must be other factors that contribute to the
i
improvement of the overall potency (low K* value). Based on the relationship between Ki
i
and K* (Figure 1A), we inferred that the reversal of the covalent complex E·I* must in
i
comparison be slowed down even more than its rate of formation (k << k ). This
6
5
16
ACS Paragon Plus Environment

Page 17 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
hypothesis was supported by the mass spectrometric results (Figure 4C), where the
remarkable modification of 3Cpro by 30 (I /I =1.5) indicated a more durable covalent
add pro
bond than that of 27 (I /I =0.1). One plausible explanation for these facts is that the α-
add pro
proton of Boc-protected thioadduct was shielded by the two N-substituents to the extent
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
that the E1cB elimination reaction was impaired (Figure 4D).
By reexamining the binding mode of 30, it could be seen that the α-proton resided
in a hydrophobic area under the sheltering of the P1’ structure. To further verify this
hypothesis, we synthesized 33, which contained a second methyl group in place of the
Boc group. This compound was three times more potent than 27 but less potent than 30,
which is consistent with an inhibition mechanism in which the second methyl group
impedes the abstraction of the α-proton but to a lesser extent than the bulkier Boc group.
The degree of covalent modification by 33 (Iadd/Ipro =1.0, Figure 4D) was between those
observed by 27 and 30, which also agreed well with our proposed mechanism.
Furthermore, the docking model of 33 offered a more straightforward elucidation (Figure
4E). On one hand, compound 33 bound favorably to 3Cpro with high binding affinity (ΔGcalc
=
-7.06 kcal/mol). On the other hand, unlike 30, the α-proton of the thioadduct of 33 was
solvent-exposed and susceptible to base attack, leading to a more rapid E1cB reaction
large k ). The final K* was a result of these counteracting effects.
(
6
i
Reversibility study
In the preceding experiments, ultrafiltration was performed prior to the MS
analyses. In fact, we also determined 3Cpro activity with the ultrafiltered samples (Figure
A). The inhibitor concentrations in these samples were diluted by approximately 103-
5
6
and 10 -fold, respectively. As a classical Michael acceptor, no restoration of proteolytic
pro
activity was detected in the ultrafiltered 3C -2 complex, indicative of an irreversibly
modified catalytic Cys . The cyanoacrylates 21 and 25 exhibited approximately 65-85%
147
recovery of activity, demonstrating decent reversibility for this series. There was
significant recovery of activity with cyanoacrylamides 27 and 33, whereas only 25% of the
activity was restored in Boc-protected 30, providing another piece of evidence for the
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
“
proton-shielding” mechanism.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. (A) Enzyme activity recovery after ultrafiltration of preincubated 3Cpro-inhibitor complex. The
3
6
concentration of given inhibitor was diluted by approximately 10 - and 10 -fold using ultrafiltration method.
NS, not significant; P(***)<0.001. (B) Enzyme activity recovery after dialysis of 3Cpro pretreated with the
indicated inhibitors or DMSO. Aliquots were taken at indicated time points, tested for the enzyme activity
and normalized to a DMSO control. Error bar represents SEM (n = 3) in both (A) and (B).
The removal of covalently bound inhibitor by ultrafiltration might introduce a
mechanical force that may disrupt the adduct. To provide a more authentic assessment
of the reversibility of the covalent inhibitor, we evaluated the recovery of 3Cpro activity
following dialysis instead of ultrafiltration, because dialysis mimics intracellular diffusion
20
to some degree (Figure 5B). After dialyzing for 6, 24, 48 and 72 h, the remaining activity
of 3Cpro was measured at each time point and normalized to control values. 3C -21 or
pro
pro
3
C -27 gradually recovered to approximately 50% of the original proteolytic activity,
pro
while 3C -30 demonstrated only 30% recovery because of its slowed elimination.
Evaluation of selectivity
Physiological nucleophiles present a challenge for the design of covalent
18
ACS Paragon Plus Environment

Page 19 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
2
inhibitors. These include: 1) small molecules such as glutathione (GSH), ethanolamine
or lysine and 2) proteins with exposed nucleophilic centers, especially serine and cysteine
proteases. To assess the performance of dually activated Michael acceptors under
intracellular redox conditions, representative compounds (21, 30, 33) were tested against
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
Cpro in the presence of 5 mM GSH (Figure S9). A reduction of protease inhibition
occurred in all cases and was inhibitor-dependent. However, the addition of GSH and
other potential cellular thiols to these inhibitors did not weaken their cell-based activity.
Due to high potency of these inhibitors, low dosage is expected to be used in practice,
and hence there is less chance of cellular GSH depletion. Nonetheless, the hepatotoxicity
remains to be evaluated. Moreover, ethanolamine and lysine had no effect on the activity
of these inhibitors.
Figure 6. Selectivity study. All three enzymes were inhibited by 50 μM of indicated compounds. Inhibitory
effects were calculated from fractional initial velocity as detailed in Experiment Section. Error bar represents
SEM (n = 2).
To evaluate the selectivity of the above 3Cpro inhibitors relative to other common
mammalian proteases, several compounds (Figure 6) were assayed with two cysteine
proteases (cathepsin K and calpain-1) and one serine protease (chymotrypsin) at a
specified concentration of 50 µM, which represented up to 500 times the apparent Ki*
value of any given inhibitor. None of the tested new compounds exhibited measurable
inhibition against chymotrypsin, demonstrating significant indifference to this serine
protease. For the two cysteine proteases, although some of these compounds showed
partial inhibition, they had acceptable selectivity in general. With the exception of 17 (65%
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
inhibition on cathepsin K), they all exhibited <50% inhibition against cathepsin K and
calpain-1, while our previously reported inhibitors 1 and zyy16 that contained, respectively,
aldehyde and cyanohydrin warheads, almost completely inhibited both enzymes.9b, 23
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CONCLUSION
An alkene conjugated with two EWGs, referred to as a dually activated Michael
acceptor, could serve as a novel warhead in the design of covalent inhibitors to attain
improved reactivity and potential reversibility. Here, we validated this concept for the
pro
design of novel inhibitors of EV71 3C , a promising therapeutic target. Generally, the
combination of a cyano group with different carbonyl groups provided these novel Michael
acceptors with improved potency and selectivity. Moreover, the N-disubstituted
cyanoacrylamides possessed unexpectedly high inhibitory activity. We proposed a
hypothesis in which the spatially shielded α-proton in the enzyme-inhibitor adduct resulted
in a reduced off-rate that eventually lowered the overall inhibition constant. From steady-
state kinetics, MS analyses and computational studies, we not only confirmed the
covalent mode of inhibition and investigated the reversibility but also provided some
evidence for the “proton-shielding” hypothesis. The exploration of these dually activated
Michael acceptors may lead to novel anti-EV71 candidates and could find potential
application in the design of reversible covalent inhibitors for other proteases.
EXPERIMENTAL SECTION
General chemical information. Unless otherwise specified, all reagents were
purchased from commercial suppliers and were used as received. All solvents were of
reagent grade and, when necessary, were purified and dried by standard methods.
Reaction progress was monitored by thin-layer chromatography (TLC) that was
performed on glass-backed silica gel plates, and spots were visualized using a UV lamp
(
λ = 254, 365 nm), iodine vapor or by KMnO staining. Reaction products were purified
4
by silica gel column chromatography or as otherwise noted. NMR spectra were recorded
1
13
on a Bruker AVANCE-400 NMR spectrometer (400 MHz for H and 100 MHz for C) in
CDCl -d, CD OD-d or DMSO-d solutions with TMS as the internal standard. Chemical
3
3
4
6
20
ACS Paragon Plus Environment

Page 21 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
shifts were reported in ppm with residual solvent as the internal standard. Standard
abbreviations were used for denoting signal multiplicities. All coupling constants were
given in Hz. Low-resolution mass spectra were obtained using a Shimadzu LCMS-2020
(Shimadzu, Kyoto, Japan). High-resolution mass spectra (HRMS) were obtained using a
high-resolution ESI-FTICR mass spectrometer (Varian 7.0 T, Varian, USA). Optical
rotations were measured with an Insmark IP-120 automatic polarimeter (Insmark,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Shanghai, China). Measurements were collected at 20 °C in DCM at 589 nm. [α] values
D
were given in unit of (deg × mL)/(g × dm). HPLC was performed on a Dionex UltiMate
3000 (Germany). All target compounds possessed a purity of ≥ 95% as determined by
HPLC.
Preparation of methyl (S,E)-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-3-
carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (2). Sodium
bis(trimethylsilyl) amide (0.3 mL of 1.0 M solution in THF, 0.3 mmol, 1.2 equiv) was added
to a solution of methyl diethylphosphonoacetate (56 mg, 0.3 mmol, 1.2 equiv) in
anhydrous THF (20 mL) under a nitrogen atmosphere at -78 °C and stirred for 30 min. 1
(100 mg, 0.23 mmol, 1.0 equiv) in anhydrous THF (5 mL) was added dropwise via cannula,
then the reaction mixture was stirred at -78 °C for 30 min until TLC indicated full
consumption of starting material. Then the reaction was warmed to room temperature (RT)
and quenched with 1.0 M HCl solution (5 mL). Subsequently, THF was removed, and
extractive workup was done with DCM (3×). The combined organic layer was dried over
anhydrous sodium sulfate and concentrated, and the residue was purified via silica gel
column chromatography (MeOH:DCM = 1:70 v/v) to afford the pure product 2 as a white
2
0
1
solid (90 mg, yield 80%). [훼] = −14.44 (c = 0.097, DCM). H NMR (CDCl -d) δ 7.2 – 7.1
퐷
3
(
m, 2H), 6.9 (t, J = 8.6 Hz, 2H), 6.7 (d, J = 5.7 Hz, 1H), 6.3 (s, 1H), 5.6 (dd, J = 15.6, 1.5
Hz, 1H), 4.9 (hept, J = 7.3 Hz, 1H), 4.5 (s, 1H), 3.7 (s, 3H), 3.2 (tq, J = 12.5, 7.2 Hz, 2H),
.1 (s, 2H), 2.4 (s, 3H), 2.2 (qd, J = 9.4, 7.9, 3.1 Hz, 1H), 2.1 (ddd, J = 14.1, 12.0, 4.4 Hz,
1H), 2.0 (dt, J = 13.7, 5.0 Hz, 1H), 1.9 – 1.8 (m, 1H), 1.7 – 1.6 (m, 1H), 1.5 (dddd, J =
3
1
3
2
3.7, 20.9, 9.1, 3.3 Hz, 2H). C NMR (CDCl -d) δ175.0, 171.2, 170.4, 166.7, 162.0 (d, J
3
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
=
1
244.8 Hz), 158.9, 158.4, 147.6, 132.0 (d, J = 3.1 Hz), 131.1, 131.0, 120.7, 115.5, 115.3,
01.5, 54.6, 51.7, 48.4, 42.3, 38.1, 38.0, 35.7, 26.8, 21.4, 12.3. HRMS (ESMS):
+
C H FN NaO (M + Na) , calcd. 523.1963, found 523.1968. Purity: >99.0%, HPLC t =
25
29
4
6
ret
6.57 min.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Preparation of ethyl (S,E)-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-3-
carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (3). 3 was obtained
from 1 (100 mg) and triethyl phosphonoacetate (55 μL) following similar procedure with
2. The crude product was purified by silica gel column chromatography (MeOH:DCM =
2
0
1:50 v/v) to afford the pure product as a white solid 3 (96 mg, 83%). [훼] = −11.41 (c =
퐷
1
0.23, DCM). H NMR (CDCl -d) δ 7.2 (dd, J = 8.4, 5.5 Hz, 2H), 6.9 (t, J = 8.7 Hz, 2H), 6.7
3
(dd, J = 15.6, 5.7 Hz, 1H), 6.3 (s, 1H), 5.7 (dd, J = 15.7, 1.4 Hz, 1H), 5.0 – 4.9 (m, 1H),
4.5 (t, J = 6.2 Hz, 1H), 4.2 (q, J = 7.1 Hz, 2H), 3.3 (d, J = 8.7 Hz, 2H), 3.1 (d, J = 6.7 Hz,
2
H), 2.4 (s, 3H), 2.2 (q, J = 5.6 Hz, 1H), 2.1 – 2.1 (m, 1H), 2.0 (td, J = 10.7, 8.5, 5.0 Hz,
H), 1.8 (dt, J = 13.7, 4.1 Hz, 1H), 1.7 (dt, J = 13.6, 7.6 Hz, 1H), 1.5 (tdd, J = 14.7, 5.7,
1
13
2.6 Hz, 2H), 1.3 – 1.3 (m, 3H). C NMR (CDCl -d) δ 175.0, 171.3, 170.4, 166.3, 162.0
3
(d, J = 245.0 Hz), 158.9, 158.4, 147.2, 132.0 (d, J = 3.0 Hz), 131.2, 131.1, 121.3, 115.6,
115.4, 101.5, 60.6, 54.6, 48.9, 42.4, 38.4, 38.2, 35.7, 27.3, 21.6, 14.3, 12.4. HRMS
+
(ESMS): C H FN NaO (M + Na) , calcd. 537.2120, found 537.2125. Purity: >99%,
26
31
4
6
HPLC t = 7.35 min.
ret
Preparation of N-((S)-3-(4-fluorophenyl)-1-(((S,E)-5-(methylamino)-5-oxo-1-((S)-
2-oxopiperidin-3-yl)pent-3-en-2-yl)amino)-1-oxopropan-2-yl)-5-methylisoxazole-3-
carboxamide (4). To a solution of 1 (100 mg, 0.23 mmol, 1.0 equiv) in dry acetonitrile (20
mL) at RT were sequentially added diethyl (2-(methylamino)-2-oxoethyl)phosphonate (95
mg, 0.46 mmol, 2.0 equiv), lithium chloride (14 mg, 0.35 mmol, 1.5 equiv) and 1, 8-
diazabicyclo[5.4.0]undec-7-ene (40 mg, 0.28 mmol, 1.2 equiv). The reaction mixture was
stirred at RT and quenched with saturated NH Cl after complete conversion.
4
Subsequently, acetonitrile was removed, and extractive workup was done with DCM (40
mL×3), followed by washing with H O (40 mL×2), saturated citric acid solution (40 mL×2),
2
22
ACS Paragon Plus Environment

Page 23 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
saturated NaHCO solution (40 mL×2) and brine (40 mL×2). The organic phase was dried
3
over sodium sulfate and concentrated, and the crude product was purified by silica gel
column chromatography (MeOH:DCM = 1:30 v/v) to afford the pure product as a white
2
0
1
solid 4 (76 mg, yield 68%). [훼] = −31.41 (c = 0.43, DCM). H NMR (CDCl -d) δ 7.2 – 7.1
퐷
3
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(m, 2H), 6.9 (t, J = 8.6 Hz, 2H), 6.7 (dd, J = 15.7, 5.6 Hz, 1H), 6.3 (s, 1H), 5.7 (d, J = 15.6
Hz, 1H), 4.9 (d, J = 7.8 Hz, 1H), 4.5 (s, 1H), 3.7 (s, 3H), 3.2 (d, J = 11.4 Hz, 2H), 3.1 (d,
J = 6.8 Hz, 2H), 2.4 (s, 3H), 2.2 (s, 1H), 2.1 (s, 1H), 2.0 (d, J = 13.8 Hz, 1H), 1.8 (d, J =
13
13.5 Hz, 1H), 1.7 – 1.6 (m, 1H), 1.5 (p, J = 11.0 Hz, 2H). C NMR (CDCl -d) δ 175.0,
3
1
1
2
71.3, 170.4, 166.8, 162.0 (d, J = 244.9 Hz), 158.9, 158.4, 147.6, 132.0 (d, J = 3.0 Hz),
31.1, 131.0, 120.7, 115.6, 115.4, 101.5, 54.6, 51.7, 48.6, 42.4, 38.2, 38.2, 35.7, 27.1,
+
1.5, 12.4. HRMS (ESMS): C H FN NaO (M + Na) , calcd. 522.2123, found 522.2125.
25
30
5
5
Purity: 95.0%, HPLC t = 5.37 min.
ret
Preparation of N-((S)-1-(((S,E)-4-cyano-1-((S)-2-oxopiperidin-3-yl)but-3-en-2-
yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-carboxamide (5). 5
was obtained from 1 (100 mg) and diethyl cyanomethylphosphonate (75 μL) following
similar procedure with 4 at RT for 2 h. The crude product was purified by silica gel column
chromatography (MeOH:DCM = 1:70 v/v) to afford the pure product as a white solid 5 (86
20
1
mg, yield 82%). [훼] = −11.52 (c = 0.14, DCM). H NMR (CDCl -d) δ 7.2 (dd, J = 8.3, 5.4
퐷
3
Hz, 2H), 7.0 (t, J = 8.6 Hz, 2H), 6.4 (dd, J = 16.3, 5.4 Hz, 1H), 6.3 (s, 1H), 5.1 (d, J = 16.3
Hz, 1H), 4.9 (q, J = 7.4 Hz, 1H), 4.4 (dq, J = 10.8, 5.0 Hz, 1H), 3.3 (qt, J = 11.6, 4.6 Hz,
2H), 3.1 (qd, J = 13.7, 6.9 Hz, 2H), 2.4 (s, 3H), 2.2 (dq, J = 11.6, 6.0 Hz, 1H), 2.1 – 1.9
(m, 2H), 1.8 (dq, J = 13.8, 4.5 Hz, 1H), 1.8 – 1.6 (m, 1H), 1.5 (ddt, J = 13.6, 10.5, 3.4 Hz,
13
2H). C NMR (CDCl -d) δ 174.8, 171.5, 170.6, 162.1 (d, J = 245.9 Hz), 159.0, 158.3,
3
1
4
4
53.6, 131.9 (d, J = 3.1 Hz), 131.2, 131.1, 117.0, 115.7, 115.5, 101.5, 100.2, 54.6, 49.4,
+
2.3, 38.2, 38.0, 35.1, 27.2, 21.6, 12.4. HRMS (ESMS): C H FN NaO (M + Na) , calcd.
24
26
5
4
90.1861, found 490.1865. Purity: >99.0%, HPLC t = 5.97 min.
ret
Preparation of N-((S)-1-(((S)-4-cyano-1-((S)-2-oxopiperidin-3-yl)pent-3-en-2-
yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-carboxamide (6). 6
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was obtained from 1 (100 mg) and 34 (86 mg) following similar procedure with 4 at RT for
1
1
h. The crude product was purified by silica gel column chromatography (MeOH:DCM =
:65 v/v) to afford the pure product as a white solid 6 (85 mg, yield 78%). The E/Z isomers
2
0
were not separated using the given chromatographic conditions. [훼] = −4.14 (c = 0.34,
퐷
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
DCM). H NMR (CDCl -d) δ 7.1 (ddd, J = 13.3, 8.2, 5.4 Hz, 2H), 6.9 (dt, J = 15.4, 8.5 Hz,
3
2
H), 6.3 (d, J = 3.1 Hz, 1H), 5.8 – 5.5 (m, 1H), 4.9 (dq, J = 22.6, 7.4 Hz, 1H), 4.7 – 4.5 (m,
H), 3.2 (d, J = 10.4 Hz, 2H), 3.1 (h, J = 7.2 Hz, 2H), 2.4 (d, J = 1.8 Hz, 3H), 2.2 (ddt, J =
1
16.5, 12.4, 5.9 Hz, 2H), 2.0 (dt, J = 9.2, 4.7 Hz, 1H), 2.0 – 1.9 (m, 3H), 1.8 (dq, J = 9.2,
4
3
1
.6 Hz, 1H), 1.7 – 1.6 (m, 1H), 1.4 (ddd, J = 13.6, 8.7, 3.1 Hz, 1H), 1.3 (ddd, J = 14.1, 7.2,
13
.8 Hz, 1H). C NMR (CDCl -d) δ 174.6, 171.3, 170.3, 162.0 (d, J = 245.3 Hz), 158.8,
3
58.3, 146.8, 131.9 (d, J = 3.0 Hz), 131.0, 131.0, 120.0, 115.5, 115.3, 110.8, 101.3, 54.3,
49.0, 42.3, 38.3, 37.8, 35.0, 27.1, 21.5, 15.3, 12.3. HRMS (ESMS): C H FN NaO (M +
25
28
5
4
+
Na) , calcd. 504.2018, found 504.2022. Purity: 97.6%, HPLC t = 18.41 min (E-isomer)
ret
and 19.86 (Z-isomer).
Preparation of N-((S)-1-(((S)-4-chloro-4-cyano-1-((S)-2-oxopiperidin-3-yl)but-3-
en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-carboxamide
(7). 7 was obtained from 1 (100 mg) and 35 (95 mg) following similar procedure with 4 at
RT for 1 h. The crude product was purified by silica gel column chromatography
(MeOH:DCM = 1:65 v/v) to afford the pure product as a white solid 7 (92 mg, yield 82%).
2
0
The E/Z isomers were not separated using the given chromatographic conditions. [훼] =
퐷
1
−
27.12 (c = 0.081, DCM). H NMR (CDCl -d) δ 7.1 (dt, J = 8.7, 5.2 Hz, 2H), 7.0 (q, J =
3
7.9 Hz, 2H), 6.3 (s, 1H), 6.1 (d, J = 8.2 Hz, 1H), 4.9 (q, J = 7.6 Hz, 1H), 4.8 – 4.5 (m, 1H),
3.3 (d, J = 10.2 Hz, 2H), 3.1 (qd, J = 13.7, 6.7 Hz, 2H), 2.4 (s, 3H), 2.3 – 2.1 (m, 2H), 2.0
d, J = 15.2 Hz, 1H), 1.9 (d, J = 13.8 Hz, 1H), 1.8 – 1.6 (m, 1H), 1.5 (dd, J = 25.0, 12.3
(
13
Hz, 2H). C NMR (CDCl -d) δ 174.7, 171.4, 170.5, 162.0 (d, J = 245.5 Hz), 158.9, 158.3,
3
148.3, 132.0 (d, J = 3.6 Hz), 131.2, 131.1, 115.6, 115.6, 113.0, 104.4, 101.4, 54.3, 49.1,
+
42.4, 38.3, 35.1, 33.8, 27.2, 21.6, 12.4. HRMS (ESMS): C H ClFN NaO (M + Na) ,
24
25
5
4
calcd. 524.1471, found 524.1475. Purity: 97.9%, HPLC tret = 18.51 min (Z isomer) and
24
ACS Paragon Plus Environment

Page 25 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
9.41 (E isomer).
Preparation of N-((S)-1-(((S)-4-bromo-4-cyano-1-((S)-2-oxopiperidin-3-yl)but-3-
en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-carboxamide
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(8). 8 was obtained from 1 (100 mg) and 36 (115 mg) following similar procedure with 4.
The crude product was purified by silica gel column chromatography (MeOH:DCM = 1:65
20
v/v) to afford the pure product as a light yellow solid 8 (98 mg, 80%). [훼] = −44.79 (c =
퐷
1
0.067, DCM). H NMR (CDCl -d) δ 7.2 (dt, J = 8.4, 4.7 Hz, 2H), 7.0 (td, J = 8.6, 2.7 Hz,
3
2H), 6.3 (s, 1H), 6.2 (d, J = 9.0 Hz, 1H), 4.9 (q, J = 7.4 Hz, 1H), 4.6 – 4.5 (m, 1H), 3.3 (s,
2H), 3.2 – 3.0 (m, 2H), 2.5 (s, 3H), 2.2 – 2.0 (m, 3H), 1.9 (dt, J = 13.3, 4.3 Hz, 1H), 1.7
13
(dd, J = 15.1, 6.9 Hz, 1H), 1.6 – 1.4 (m, 2H). C NMR (CDCl -d) δ 174.8, 171.2, 170.4,
3
161.9 (d, J = 245.5 Hz), 158.9, 158.2, 152.4, 150.9, 131.9 (d, J = 3.3 Hz), 131.1, 131.0,
115.5, 115.3, 113.6, 101.3, 88.3, 54.3, 50.7, 42.3, 38.3, 38.1, 34.8, 27.5, 21.5, 12.3.
+
HRMS (ESMS): C H BrFN NaO (M + Na) , calcd. 568.0966, found 568.0970. Purity:
2
4
25
5
4
95.9%, HPLC t = 18.75 min (Z-isomer) and 19.53 (E-isomer).
ret
Preparation of methyl (S)-2-chloro-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-
3-carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (9).
9
was
obtained from 1 (100 mg) and 37 (110 mg) following similar procedure with 4. The crude
product was purified by silica gel column chromatography (MeOH:DCM = 1:65 v/v) to
afford the pure product as a white solid 9 (93 mg, yield 78%). [훼]20 = −3.03 (c = 0.13,
퐷
1
DCM). H NMR (CDCl -d) δ 7.1 (dt, J = 8.6, 4.9 Hz, 2H), 6.9 (dt, J = 11.3, 8.4 Hz, 2H),
3
6
.6 (d, J = 8.3 Hz, 1H), 6.4 (d, J = 27.8 Hz, 1H), 5.3 – 4.6 (m, 2H), 3.8 (m, 3H), 3.3 (d, J
6.3 Hz, 2H), 3.2 – 2.9 (m, 2H), 2.4 (s, 3H), 2.2 (s, 1H), 2.1 (dd, J = 15.5, 8.1 Hz, 2H),
=
13
1.9 – 1.8 (m, 1H), 1.8 – 1.6 (m, 1H), 1.6 – 1.4 (m, 2H). C NMR (CDCl -d) δ 174.9, 171.3,
3
170.4, 162.5, 162.0 (d, J = 245.0 Hz), 158.8, 158.4, 145.2, 132.1 (d, J = 3.8 Hz), 131.2,
131.1, 124.3, 115.5, 115.3, 101.4, 54.4, 53.3, 47.7, 42.4, 38.5, 38.2, 34.2, 27.0, 21.5, 12.4.
+
HRMS (ESMS): C H ClFN NaO (M + Na) , calcd. 557.1574, found 557.1577. Purity:
2
5
28
4
6
97.1%, HPLC t = 17.63 min (Z-isomer) and 18.93 min (E-isomer).
ret
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Preparation of methyl (S)-2-bromo-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-
-carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (10). 10 was
3
obtained from 1 (100 mg) and 38 (130 mg) following similar procedure with 4. The crude
product was purified by silica gel column chromatography (MeOH:DCM = 1:60 v/v) to
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0
afford the pure product as a light yellow solid 10 (95 mg, yield 73%). [훼] = −18.21 (c =
퐷
1
0
.19, DCM). H NMR (CDCl -d) δ 7.1 (dd, J = 9.0, 4.7 Hz, 2H), 6.9 (dt, J = 17.2, 9.5 Hz,
3
2H), 6.8 (d, J = 5.6 Hz, 1H), 6.3 (d, J = 24.8 Hz, 1H), 5.2 – 4.5 (m, 2H), 3.9 – 3.7 (m, 3H),
3.3 (d, J = 8.2 Hz, 2H), 3.1 (p, J = 11.6, 10.9 Hz, 2H), 2.5 – 2.4 (m, 3H), 2.2 (s, 1H), 2.2 –
13
2
.0 (m, 2H), 1.9 (d, J = 13.7 Hz, 1H), 1.7 (s, 1H), 1.5 (p, J = 13.1, 12.0 Hz, 2H). C NMR
CDCl -d) δ 174.9, 171.2, 170.4, 162.5, 162.0 (d, J = 245.1 Hz), 158.9, 158.4, 146.2,
(
3
132.1 (d, J = 3.2 Hz), 131.2, 131.1, 115.5, 115.3, 111.7, 101.4, 54.4, 53.5, 50.1, 42.4,
+
38.4, 38.1, 34.0, 26.9, 21.5, 12.4. HRMS (ESMS): C H BrFN NaO (M + Na) , calcd.
25
28
4
6
6
01.1068, found 601.1070. Purity: 97.3%, HPLC t = 18.52 min (Z-isomer) and 20.05 (E-
ret
isomer).
Preparation of ethyl (S)-2-chloro-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-3-
carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (11). 11 was
obtained from 1 (100 mg) and 39 (116 mg) following similar procedure with 4. The crude
product was purified by silica gel column chromatography (MeOH:DCM = 1:70 v/v) to
2
0
afford the pure product as a white solid 11 (100 mg, yield 81%). [훼] = −9.79 (c = 0.14,
퐷
1
DCM). H NMR (CDCl -d) δ 7.1 (q, J = 6.0, 5.5 Hz, 2H), 6.9 (q, J = 9.4 Hz, 2H), 6.6 (d, J
3
=
8.0 Hz, 1H), 6.3 (s, 1H), 5.3 – 4.7 (m, 2H), 4.3 (q, J = 7.2 Hz, 2H), 3.4 – 3.2 (m, 2H), 3.1
pd, J = 13.2, 11.3, 6.7 Hz, 2H), 2.4 (s, 3H), 2.2 (p, J = 5.9 Hz, 1H), 2.1 (h, J = 8.1, 6.0
Hz, 2H), 1.9 (d, J = 13.3 Hz, 1H), 1.7 (d, J = 10.8 Hz, 1H), 1.6 – 1.4 (m, 2H), 1.3 (q, J =
(
1
3
7
1
5
.1 Hz, 3H). C NMR (CDCl -d) δ 175.1, 171.3, 170.4, 162.1, 162.0 (d, J = 245.1 Hz),
3
58.9, 158.4, 142.1, 132.1 (d, J = 3.4 Hz), 131.2, 131.1, 123.1, 115.5, 115.3, 101.4, 62.6,
4.4, 47.8, 42.4, 38.5, 38.3, 34.2, 27.1, 21.5, 14.2, 12.4. HRMS (ESMS):
+
C H ClFN NaO (M + Na) , calcd. 571.1730, found 571.1735. Purity > 99%, HPLC t
26
30
4
6
ret
=
21.03 min (Z-isomer) and 22.53 min (E-isomer).
26
ACS Paragon Plus Environment

Page 27 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Preparation of ethyl (S)-2-bromo-4-((S)-3-(4-fluorophenyl)-2-(5-methylisoxazole-
-carboxamido)propanamido)-5-((S)-2-oxopiperidin-3-yl)pent-2-enoate (12). 12 was
3
obtained from 1 (100 mg) and 40 (136 mg) following similar procedure with 4. The crude
product was purified by silica gel column chromatography (MeOH:DCM = 1:65 v/v) to
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
0
afford the pure product as a yellow solid 12 (95 mg, 71%). [훼] = −5.56 (c = 0.11, DCM).
퐷
1
H NMR (CDCl -d) δ 7.1 (dt, J = 8.7, 4.8 Hz, 2H), 7.0 – 6.9 (m, 2H), 6.6 (d, J = 8.6 Hz,
3
1H), 6.3 (s, 1H), 5.2 – 4.6 (m, 2H), 4.3 – 4.2 (m, 2H), 3.3 (s, 2H), 3.1 – 3.0 (m, 2H), 2.4 (s,
3H), 2.2 (s, 1H), 2.1 (d, J = 15.3 Hz, 2H), 1.9 (d, J = 14.1 Hz, 1H), 1.7 (d, J = 9.1 Hz, 1H),
1
3
1
1
1
.5 (dt, J = 22.9, 11.0 Hz, 2H), 1.3 (q, J = 5.9 Hz, 3H). C NMR (CDCl -d) δ 174.9, 171.2,
3
70.4, 162.0, 162.0 (d, J = 244.6 Hz), 158.9, 158.4, 145.7, 132.1 (d, J = 3.2 Hz), 131.2,
31.1, 116.2, 115.5, 115.3, 101.4, 62.8, 54.4, 50.4, 42.4, 38.5, 38.3, 34.0, 27.2, 21.6, 14.3,
+
12.4. HRMS (ESMS): C H BrFN NaO (M + Na) , calcd. 615.1225, found 615.1230.
26
30
4
6
Purity: 95.9%, HPLC t = 21.35 min (Z-isomer) and 23.06 (E-isomer).
ret
Preparation of N-((S)-1-(((S,Z)-4-cyano-1-((S)-2-oxopiperidin-3-yl)-4-phenylbut-3-
en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-carboxamide
(
13). 13 was obtained from 1 (100 mg, 0.23 mmol, 1.0 equiv) and 41 (115 mg, 0.46 mmol,
.0 equiv) following similar procedure with 4. The crude product was purified via silica gel
column chromatography (MeOH:DCM = 1:65 v/v) to afford the pure product as a white
2
2
0
1
solid 13 (92 mg, yield 75%). [훼] = −4.85 (c = 0.29, DCM). H NMR (CDCl -d) δ 7.5 – 7.4
퐷
3
(
m, 2H), 7.4 – 7.3 (m, 3H), 7.1 (dd, J = 8.4, 5.4 Hz, 2H), 6.8 (td, J = 8.6, 1.8 Hz, 2H), 6.3
s, 1H), 6.2 (d, J = 8.6 Hz, 1H), 5.0 – 4.9 (m, 1H), 4.9 (dddd, J = 12.5, 10.2, 6.1, 3.7 Hz,
(
1H), 3.3 (dq, J = 8.1, 4.7, 3.7 Hz, 2H), 3.1 (qd, J = 13.6, 6.9 Hz, 2H), 2.4 (s, 3H), 2.3 – 2.2
(m, 2H), 2.1 (ddt, J = 11.6, 6.1, 2.8 Hz, 1H), 1.9 (dt, J = 13.1, 4.2 Hz, 1H), 1.7 (dddd, J =
13
1
9.2, 14.2, 8.8, 3.1 Hz, 1H), 1.6 (dq, J = 10.5, 4.6, 3.2 Hz, 2H). C NMR (CDCl -d) δ
3
175.0, 171.3, 170.4, 161.9 (d, J = 245.1 Hz), 158.9, 158.4, 146.1, 132.4, 132.1 (d, J = 3.1
Hz), 131.2, 131.1, 129.5, 129.1, 125.9, 116.0, 115.8, 115.4, 115.2, 101.4, 54.2, 49.7, 42.4,
+
38.3, 35.7, 27.0, 21.6, 12.4. HRMS (ESMS): C H FN NaO (M + Na) , calcd. 566.2174,
30
30
5
4
found 566.2178. Purity: 95.5%, HPLC t = 8.46 min.
ret
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 59
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Preparation of N-((S)-1-(((S,Z)-4-cyano-4-(naphthalen-2-yl)-1-((S)-2-oxopiperidin-
3
-yl)but-3-en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-
carboxamide (14). 14 was obtained from 1 (100 mg) and 42 (115 mg) following similar
procedure with 4. The crude product was purified by silica gel column chromatography
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
MeOH:DCM = 1:65 v/v) to afford the pure product as a white solid 14 (92 mg, yield 75%).
20
1
[
훼] = −42.77 (c = 0.11, DCM). H NMR (CDCl -d) δ 8.0 (d, J = 1.8 Hz, 1H), 7.9 – 7.8 (m,
퐷
3
3
H), 7.5 (ddt, J = 8.5, 6.0, 3.1 Hz, 3H), 7.1 (dd, J = 8.5, 5.4 Hz, 2H), 6.8 (td, J = 8.7, 2.2
Hz, 2H), 6.4 (d, J = 8.6 Hz, 1H), 6.3 (s, 1H), 5.0 (q, J = 7.4 Hz, 1H), 4.9 (qd, J = 8.6, 7.1,
3
.1 Hz, 1H), 3.3 (dp, J = 11.4, 3.7, 3.1 Hz, 2H), 3.1 (qd, J = 13.7, 6.9 Hz, 2H), 2.4 (d, J =
.5 Hz, 3H), 2.3 (ddt, J = 17.3, 11.6, 5.4 Hz, 2H), 2.2 – 2.1 (m, 1H), 1.9 (tt, J = 7.4, 3.8
3
13
Hz, 1H), 1.7 – 1.5 (m, 3H). C NMR (CDCl -d) δ 175.0, 171.3, 170.4, 161.8 (d, J = 245.0
3
Hz), 158.9, 158.4, 146.1, 133.6, 133.2, 132.2 (d, J = 2.9 Hz), 131.2, 131.1, 129.6, 128.9,
1
28.5, 127.8, 127.2, 127.0, 126.2, 122.3, 116.1, 115.9, 115.4, 115.2, 101.4, 54.3, 49.8,
+
4
2.4, 38.3, 38.3, 35.8, 27.0, 21.6, 12.4. HRMS (ESMS): C H FN NaO (M + Na) , calcd.
34
32
5
4
616.2331, found 616.2334. Purity: 95.8%, HPLC t = 10.16 min.
ret
Preparation
of
N-((S)-1-(((S,Z)-4-cyano-1-((S)-2-oxopiperidin-3-yl)-4-(4-
(
trifluoromethyl)phenyl)but-3-en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-
methylisoxazole-3-carboxamide (15). 15 was obtained from 1 (100 mg) and 43 (135 mg)
following similar procedure with 4. The crude product was purified by silica gel column
chromatography (MeOH:DCM = 1:60 v/v) to afford the pure product as a yellow solid 15
2
0
1
1
(
93 mg, yield 70%). [훼] = −12.77 (c = 0.094, DCM). H NMR (CDCl -d) H NMR (CDCl -
퐷
3
3
d) δ 7.7 (d, J = 8.3 Hz, 2H), 7.6 (d, J = 8.3 Hz, 2H), 7.2 (dd, J = 8.5, 5.4 Hz, 2H), 6.9 (t, J
= 8.6 Hz, 2H), 6.4 (s, 2H), 5.0 – 4.9 (m, 1H), 4.9 – 4.7 (m, 1H), 3.3 (dd, J = 10.8, 4.1 Hz,
2
H), 3.2 – 3.1 (m, 2H), 2.4 (d, J = 1.4 Hz, 3H), 2.3 (dq, J = 13.1, 5.6 Hz, 1H), 2.1 (s, 2H),
13
1
.9 – 1.8 (m, 1H), 1.7 (dd, J = 14.5, 7.2 Hz, 1H), 1.6 (ddd, J = 14.1, 5.6, 2.8 Hz, 2H). C
NMR (CDCl -d) δ 175.0, 171.2, 170.4, 161.8 (d, J = 245.3 Hz), 158.9, 158.3, 148.6, 135.8,
3
132.0 (d, J = 3.2 Hz), 131.4 (d, J = 32.9 Hz), 131.1, 131.0, 129.2, 126.2, 126.0, 126.0,
1
23.7 (q, J = 272.2 Hz), 115.6, 115.4, 115.2, 114.9, 101.3, 54.2, 50.3, 42.3, 38.5, 37.9,
28
ACS Paragon Plus Environment

Page 29 of 59
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
3
5.4, 27.4, 21.5, 12.3. HRMS (ESMS): C H F N NaO (M + Na) , calcd. 634.2048,
31 29
4
5
4
found 634.2050. Purity: 97.9%, HPLC t = 10.06 min.
ret
Preparation of N-((S)-1-(((S,E)-4-cyano-1-((S)-2-oxopiperidin-3-yl)-4-(pyridin-2-
yl)but-3-en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-
carboxamide (16). 16 was obtained from 1 (100 mg) and 44 (115 mg) following similar
procedure with 4. The crude product was purified by silica gel column chromatography
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
MeOH:DCM = 1:65 v/v) to afford the pure product as a light yellow solid 16 (20 mg, yield
2
0
1
20%). [훼] = −28.71 (c = 0.058, DCM). H NMR (CDCl -d) δ 8.5 (d, J = 4.8 Hz, 1H), 7.8
퐷
3
–
7.6 (m, 2H), 7.2 – 7.2 (m, 1H), 7.1 (dd, J = 8.4, 5.3 Hz, 2H), 6.9 (t, J = 8.5 Hz, 2H), 6.2
s, 1H), 6.1 (d, J = 8.5 Hz, 1H), 5.5 (q, J = 7.5 Hz, 1H), 4.9 (q, J = 7.5 Hz, 1H), 3.2 (s, 2H),
.1 – 2.9 (m, 2H), 2.4 (s, 3H), 2.2 (qd, J = 9.5, 6.2, 5.5 Hz, 1H), 2.1 – 2.0 (m, 2H), 1.9 –
(
3
13
1.7 (m, 2H), 1.6 (d, J = 11.2 Hz, 2H). C NMR (CDCl -d) δ 174.1, 170.3, 169.6, 161.0 (d,
3
J = 245.5 Hz), 157.8, 157.4, 152.4, 150.1, 148.3, 136.3, 131.1 (d, J = 3.1 Hz), 130.1,
1
30.1, 122.7, 122.4, 117.6, 114.6, 114.4, 113.3, 100.4, 53.4, 46.6, 41.5, 37.6, 37.5, 33.2,
+
25.8, 20.4, 11.4. HRMS (ESMS): C H FN NaO (M + Na) , calcd. 567.2127, found
29
29
6
4
567.2130. Purity: 95.4%, HPLC t = 6.97 min.
ret
Preparation of N-((S)-1-(((S,Z)-4-cyano-1-((S)-2-oxopiperidin-3-yl)-4-(pyridin-2-
yl)but-3-en-2-yl)amino)-3-(4-fluorophenyl)-1-oxopropan-2-yl)-5-methylisoxazole-3-
carboxamide (17). 17 was obtained from 1 (100 mg) and 44 (115 mg) following similar
procedure with 4. The crude product was purified by silica gel column chromatography
(
MeOH:DCM = 1:65 v/v) to afford the pure product as a light yellow solid 17 (60 mg, yield
20
1
6
0%). [훼] = −20.01 (c = 0.05, DCM). H NMR (CDCl -d) δ 8.6 (dd, J = 9.7, 5.5 Hz, 1H),
퐷
3
7.7 (td, J = 7.8, 1.8 Hz, 1H), 7.5 (dd, J = 11.9, 8.2 Hz, 2H), 7.3 – 7.2 (m, 1H), 7.2 – 7.0 (m,
H), 6.8 (t, J = 8.4 Hz, 1H), 6.3 (s, 1H), 4.9 (q, J = 7.3 Hz, 1H), 4.9 – 4.7 (m, 1H), 3.2 (dt,
3
J = 15.8, 7.6 Hz, 2H), 3.1 – 2.9 (m, 2H), 2.4 (s, 3H), 2.2 (td, J = 11.2, 6.4 Hz, 2H), 2.1 –
13
2
.0 (m, 1H), 1.8 (dt, J = 13.6, 4.2 Hz, 1H), 1.7 – 1.6 (m, 1H), 1.6 – 1.4 (m, 2H). C NMR
(CDCl -d) 174.9, 171.2, 170.4, 161.9 (d, J = 245.0 Hz), 158.9, 158.4, 150.1, 149.8, 137.4,
3
1
32.0 (d, J = 2.9 Hz), 131.2, 131.1, 124.0, 121.2, 115.5, 115.5, 115.3, 101.4, 54.3, 49.7,
29
ACS Paragon Plus Environment