Identification of Highly Selective Lipoprotein-Associated Phospholipase A2 (Lp-PLA2) Inhibitors by a Covalent Fragment-Based Approach

Article abstract of DOI:10.1021/acs.jmedchem.0c00372

Covalent ligands are of great interest as therapeutic drugs or biochemical tools. Here, we reported the discovery of highly selective and irreversible inhibitors of lipoprotein-associated phospholipase A2 (Lp-PLA2) using a covalent fragment-based approach. The crystal structure of Lp-PLA2 in complex with a covalent fragment not only reveals the covalent reaction mechanism but also provides a good starting point to design compound 8, which has a more than 130,000-fold and 3900-fold increase in potency and selectivity, respectively, compared to those of the covalent fragment. Furthermore, fluorescent probes with high selectivity and sensitivity are developed to characterize Lp-PLA2 and its enzymatic activity in vitro or even in living cells in a way more convenient than immunoblotting tests or immunofluorescence imaging. Overall, we provide a paradigm for application of the covalent fragment-based strategy in covalent ligand discovery and the advantage of enol-cyclocarbamate as a new warhead in designing covalent inhibitors of serine hydrolases.

Full text of DOI:10.1021/acs.jmedchem.0c00372

Subscriber access provided by NATIONAL UNIV OF SINGAPORE
Article
Identification of Highly Selective Lipoprotein-Associated Phospholipase
A2 (Lp-PLA2) Inhibitors by a Covalent-Fragment-Based Approach
Fubao Huang, Hangchen Hu, Kai Wang, Chengyuan Peng, Wenwei Xu, Yu Zhang, Jing
Gao, Yishen Liu, Hu Zhou, Ruimin Huang, Minjun Li, Jianhua Shen, and Yechun Xu
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00372 • Publication Date (Web): 27 May 2020
Downloaded from pubs.acs.org on May 27, 2020
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 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Identification of Highly Selective Lipoprotein-
Associated Phospholipase A2 (Lp-PLA2)
Inhibitors by a Covalent-Fragment-Based
Approach
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
,
a,#
,a,b,#
a
a
a
Fubao Huang , Hangchen Hu
, Kai Wang , Chengyuan Peng , Wenwei Xu , Yu
a
b
c
a
a
d
a
Zhang , Jing Gao , Yishen Liu , Hu Zhou , Ruimin Huang , Minjun Li , Jianhua Shen *
and Yechun Xua,b*
aState Key Laboratory of Drug Research, CAS Key Laboratory of Receptor Research,
Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai
201203, China.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 82
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
bUniversity of Chinese Academy of Sciences, Beijing 100049, China.
cNanchang University, Nanchang 330031, China.
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
dShanghai Synchrotron Radiation Facility, Shanghai Advanced Research Institute,
Chinese Academy of Science, Shanghai 201203, China
ACS Paragon Plus Environment
2

Page 3 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
ABSTRACT
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
Covalent ligands are of great interest as therapeutic drugs or biochemical tools. Here,
we reported the discovery of highly selective and irreversible inhibitors of lipoprotein-
associated phospholipase A2 (Lp-PLA2) using a covalent-fragment-based approach. The
crystal structure of Lp-PLA2 in complex with a covalent fragment not only reveals the
covalent reaction mechanism but also provides a good starting point to design compound
8, which has a more than 130,000-fold and 3,900-fold increase in potency and selectivity,
respectively, compared to those of the covalent fragment. Furthermore, fluorescent
probes with high selectivity and sensitivity are developed to characterize Lp-PLA2 and its
enzymatic activity in vitro or even in living cells in a way more convenient than
immunoblotting test or immunofluorescence imaging. Overall, we provide a paradigm for
application of the covalent-fragment-based strategy in covalent ligands discovery and the
advantage of the enol-cyclocarbamate as a new warhead in designing covalent inhibitors
of serine hydrolases.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 82
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
INTRODUCTION
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
Traditionally, a covalent modification strategy has been regarded as a liability in
pharmaceutical industry, mainly due to the safety concern, such as the possibility of non-
1
specific binding, haptenization and possible idiosyncratic drug reactions. Nevertheless,
considering the therapeutic benefits of extended duration of action, increased potency or
binding efficiency and binding to otherwise “intractable” targets, covalent ligands are of
2
interest as therapeutic drugs. In addition, covalent ligands are frequently used as
biochemical tools for selective covalent modification of a target protein, in particular using
activity-based protein profiling (ABPP),3, 4 providing a powerful and applicable way to
characterise proteins and their functions.6, 7
5
To date, a number of covalent ligands (e.g. aspirin, penicillin, omeprazole, clopidogrel,
neratinib etc.) have been approved as treatments for diverse clinical indications, making
8
a significant impact on human health. However, these drugs identified in natural
9
2
products were unexpectedly resulted from high-throughput screening campaigns.
ACS Paragon Plus Environment
4

Page 5 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Recently, several strategies have been developed to design covalent ligands. For
example, the choice of a highly reactive cysteine rather than lysine, serine or tyrosine of
targeted proteins as the covalent site could efficiently form covalent bonds to compounds
with otherwise low reactivity, resulting in significantly decreased risk of promiscuous
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
protein labeling. Moreover, mimicking the intermediate state in the chemical reaction (i.e.
mechanism-based covalent inhibitors, Figure 1A) or directly adding reactive groups, often
named covalent warheads, to highly selective reversible ligands (Figure 1B) rather than
starting from a covalent fragment (Figure 1C) is generally used to derive potency and
selectivity of resulting covalent ligands.2, 10-13 With these strategies, several purposely
designed inhibitors covalently attaching to a specific cysteine of protein kinases have
been approved as drugs in recent years,14 highlighting the utility and feasibility of
irreversible ligand design. Other strategies, such as DNA-encoded libraries featuring
electrophilic ligands, can also be used to develop covalent ligands, although examples of
such applications are sparse.1
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 82
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 1. The strategies of covalent ligand design.
Although successful application has been achieved in targeting cysteine in protein
kinases, these approaches are not suitable in every case of covalent ligands design.
Covalent inhibitors design targeting lipoprotein-associated phospholipase A2 (Lp-PLA2,
UniProt accession code: Q13093), a serine phospholipase belonging to group VII of PLA2
super-family, is exactly the case. This enzyme degrades certain oxidized phospholipids
(
oxPLs) which are resulted from the oxidative attack on phospholipid components of
cellular membranes and lipoproteins into proinflammatory products, such as
15
lysophosphatidylcholine (lysoPC) and oxidized nonesterified fatty acids (oxNEFA). Lp-
PLA2 is considered to be closely related to vascular inflammation and the FDA had
ACS Paragon Plus Environment
6

Page 7 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
approved a reagent for measuring the Lp-PLA2 activity to identify individuals at higher
risk of cardiovascular disease events.16 Furthermore, Lp-PLA2 inhibitors have been
studied for treating inflammation-related diseases, such as coronary heart diseases,15
Alzheimer’s disease17, 18 and diabetic macular edema.19-21 However, darapladib (1), an
Lp-PLA2 inhibitor, was unable to meet its primary end points in phase III clinical trials on
coronary heart diseases.22, 23 In this context, designing covalent Lp-PLA2 inhibitors with
higher potency and selectivity may be beneficial to take full advantages of the superiority
of covalent drugs. In addition, the covalent inhibitor also could be used to develop specific
probes as biochemical tools for elucidating functional versatility of the enzyme.
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
Given that serine and cysteine are the two most common covalent targeting residues,
the catalytic serine (S273), which is the only serine within the substrate binding pocket of
Lp-PLA2, could serve as an ideal covalent site. However, S273 is deeply located inside
the pocket (Figure S1A) and most of the potent reversible inhibitors completely occupied
the area around S273 (Figure S1B-F), leading to little space left for placement of an
additional warhead. It is thus inappropriate to form a covalent bond with S273 by directly
adding a warhead to known reversible inhibitors. Alternatively, replacing part of the non-
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 82
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
covalent inhibitors with various warheads might be an option, whilst it is not easy to keep
the binding pose of the inhibitors before and after the replacement, and to ensure that the
introduced warhead is able to form a covalent bond with S273. Therefore, the
aforementioned strategies are not suitable to design the covalent inhibitors of Lp-PLA2
and another strategy is required. In addition, although a series of reversible Lp-PLA2
inhibitors have been disclosed,15, 24 covalent inhibitors with high selectivity for Lp-PLA2
over PLA2VIIB have not been reported. PLA2VIIB, a homologous protein of Lp-PLA2,
has a sequence identity of 41% and a conserved catalytic site to Lp-PLA2, resulting in a
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
15
very similar substrate specificity of the two enzymes. Therefore, the high selectivity of a
covalent inhibitor towards Lp-PLA2 over PLA2VIIB should be considered.
Fragment-based lead discovery (FBLD) is a powerful approach for efficient discovery
25
of novel ligands that regulate protein functions. With the wide and successful application
26
of FBLD strategy in non-covalent ligand development, this strategy was also explored
to design covalent ligands. Some covalent fragment screening efforts have been made
recently27-32 and a successful example is to design ligands covalently binding to the non-
33
catalytic cysteine in protein kinases. However, to the best of our knowledge, application
ACS Paragon Plus Environment
8

Page 9 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the covalent-fragment-based approach to discover potent covalent ligands targeting
serine has not been exemplified. This phenomenon is mainly caused by the concern that
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
covalent fragments must have relatively high reactivity to gain a reactivity-driven affinity
rather than a reversible binding affinity, resulting in subsequent bad selectivity.34, 35 Here,
we utilize a covalent-fragment-based approach (Figure 1C) to design covalent Lp-PLA2
inhibitors and select an applicable covalent fragment binding to S273 of Lp-PLA2 at the
beginning. With the aid of multiple crystal structures, we quickly fill the remaining space
of the substrate binding pocket and grow the fragment into a covalent Lp-PLA2 inhibitor
with encouraging potency and selectivity. A further modification on this covalent inhibitor
generates Lp-PLA2-specific fluorescent probes with high selectivity and sensitivity, which
are useful to evaluate the binding efficiency of competitive ligands and label Lp-PLA2 in
vitro or in living cells. These probes could also help to elucidate unexplored functions of
the enzyme in future studies.
RESULTS
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 82
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
The choice of a covalent fragment/warhead. Given a relatively weak reactivity of
2
serine compared to cysteine in the enzymes, we initially sought a covalent warhead from
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
known covalent ligands to ensure the formation of a covalent bond to S273. Among the
published Lp-PLA2 inhibitors, SB-222657 (2),36 JMN4 (3)37 and SB-253514 (4)38 were
suggested to covalently binding to Lp-PLA2 (Figure 2A). However, the detailed covalent
binding modes of these ligands with the enzyme are not known yet. In contrast to the
extensive studies of β-lactam (a core of 2) and carbamate (a core of 3) covalent moieties
in antibiotics and probes, respectively,39, 40 the novel enol-cyclocarbamate covalent
moiety (core of 4) is rarely investigated. 4 is a natural product isolated from Pseudomonas
fluorescens (DSM 11579) and was studied for antibiotics discovery with a focus on its Z-
41
isomer. It shows an inhibitory activity against Lp-PLA2 with an IC of 51 nM and also
50
shows good selectivity for Lp-PLA2 over a range of other serine proteases, such as
porcine elastase, trypsin, chymotrypsin, thermolysin, fungal aspartic protease, bacterial
type IX metalloprotease, and the herpes protease enzymes.38 Furthermore, the
enantiomer of the chiral warhead of 4 has been successfully synthesized and confirmed
to be hydrolytically stable.42, 43 However, several issues of the enol-cyclocarbamate
ACS Paragon Plus Environment
10

Page 11 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
scaffold remain to be addressed. First, the accurate covalent binding mechanism of the
enol-cyclocarbamate with Lp-PLA2 is still unrevealed. The cyclocarbamate (C3) and the
α,β-unsaturated carbonyl groups (C1), are possible electrophiles for S273 (Figure 2A).
Second, the enol-cyclocarbamate derivatives have poor selectivity for Lp-PLA2 over
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
38
PLA2VIIB. The IC of 4 for Lp-PLA2 is a 9-fold of the IC for PLA2VIIB. Given the good
50
50
selectivity of 4 for Lp-PLA2 over other serine hydrolases, unsolved problems with the
enol-cyclocarbamate series as Lp-PLA2 inhibitors and the novelty of this warhead derived
from a natural product, we decided to focus on the enol-cyclocarbamate warhead. In
parallel, we tested the β-lactam and carbamate warheads using compounds SB-222657
(2) and JMN4 (3) as representatives but were unable to successfully determine their co-
crystal structures with Lp-PLA2. Such a failure might be due to a low solubility of these
two compounds in aqueous solution (2, cLogP: 5.5; 3, cLogP: 6.6).
Characterization of the enol-cyclocarbamate warhead reacting with Lp-PLA2. It has
been reported that the stereochemistry of the chiral carbon at the ring junction is not
critical to the Lp-PLA2 inhibition.42 Following the reported synthesis route,42 we first
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 82
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
synthesized fragment 5, an enantiomer of the chiral warhead in 4. The enzymatic assay
shows that this fragment has a millimolar inhibitory activity against both Lp-PLA2 and
PLA2VIIB with no selectivity (Figure 2A). We also determined a co-crystal structure of Lp-
PLA2 bound with 5 (PDB code: 6M06) to provide structural insight into the reaction
mechanism of this type of warhead (Figure 2B). Contiguous electron density was clearly
shown between S273 and 5, strongly implying the formation of a covalent bond, and
allowed to exactly place a covalent bond between the hydroxyl oxygen of S273 and the
C3 but not C1 atom of fragment 5.
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
Based on this crystal structure, a presumed mechanism of 5 reacted with Lp-PLA2 is
shown in Scheme 1. In brief, an acylation reaction proceeds through the nucleophilic
addition of the hydroxyl oxygen of S273 to the carbonyl carbon (C3) of fragment 5,
simultaneously facilitated by proton transfer from S273 to H351, which serves as a base,
resulting in a tetrahedral intermediate. Through H-bonding interactions, the major role of
the oxyanion hole formed by NH groups of F274 and L153 is to stabilize the negative-
charged tetrahedral intermediate. From the intermediate to products, a proton transfers
from H351 to the oxygen linked to C1 of 5 and the breakage of the scissile bond happens
ACS Paragon Plus Environment
12

Page 13 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
spontaneously. In addition, the terminal carboxyl group of the covalently bound fragment
H-bonds to Q352, and the pyrrolidine moiety forms hydrophobic interactions with W298,
F322 and H351 (Figure 2B).
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. Covalent binding of the enol-cyclocarbamates to Lp-PLA2. (A) Structures of
three suspected covalent inhibitors of Lp-PLA2 and fragment 5. The inhibition of fragment
5
against Lp-PLA2 and PLA2VIIB was shown under the structure. (B) A co-crystal
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 82
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
structure of fragment 5 covalently bound to Lp-PLA2 (PDB code: 6M06). The direction for
fragment 5 to grow into a more potent ligand is shown by a blue arrow. A close-up view
of the binding pocket, in which fragment 5 (yellow) together with surrounding residues
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
(cyan) are shown as sticks and H-bonds are represented by dashed lines. The mesh
depicts the (2F -F ) difference electron-density maps of the fragment contoured at 1.2 σ.
o
c
Scheme 1. A presumed mechanism for the reaction of fragment 5 with catalytic residues
of Lp-PLA2 based on the crystal structure of Lp-PLA2 in complex with 5.
A low reactivity of the enol-cyclocarbamate towards cysteine. Since fragment 5 can
covalently modify the catalytic serine of Lp-PLA2, it is also possible to react with cysteine,
thereby resulting in unexpected covalent modification. It is thus important to explore the
electrophilicity and lability of fragment 5 targeting cysteine. We measured the half-life for
the reaction of 5 with glutathione (GSH t ), which is widely used to mimic proteinaceous
1/2
ACS Paragon Plus Environment
14

Page 15 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
cysteine reactivity.30, 44 To determine the rates of reaction with GSH, the N-phenyl
acrylamide (a positive control) or fragment 5 was incubated with GSH, and the remaining
GSH at varying time after incubation was determined. The GSH t1/2 of the N-phenyl
acrylamide is 285 min which is close to the reported value and for fragment 5 it is 1888
min, which is longer than most available warheads targeting cysteine and some approved
covalent drugs (Figure S2).45 This result demonstrates that the enol-cyclocarbamate
warhead shows low reactivity towards cysteine, suggesting the low probability of non-
specific binding to cysteine.
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
A fusion of the covalent enol-cyclocarbamate and darapladib. As shown in Figure 2B
with a blue arrow, the complex structure also revealed that the terminal carboxyl group of
fragment 5 offers a perfect growth vector towards the unoccupied pocket which is exactly
the binding pocket for reversible inhibitors, providing a great opportunity for fragment
evolution in the next step. Subsequently, to accelerate the growth of fragment 5, we
overlaid the crystal structures of Lp-PLA2 in complex with fragment 5 as well as previously
reported non-covalent inhibitors to dig out a suitable scaffold fitting well into the
unoccupied pocket. To our delight, we found that the biphenyl ethanediamine moiety
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 82
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
(
fragment 6) of darapladib, the blue part shown in Figure 3A, could be a good candidate,
as the amide group of darapladib is proximal to the carboxyl group of 5 bound in Lp-PLA2
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
(a red circle shown in Figure 3A). Given that darapladib possesses superb Lp-PLA2
46
inhibitory potency (IC : 0.7 nM) and decent selectivity (1,200-fold) against PLA2VIIB,
50
we designed and synthesized compound 7 in which fragment 5 is fused into darapladib
based on the superimposition of two complex structures. Surprisingly, compound 7 not
only gains a high inhibitory activity against Lp-PLA2 with an IC of 25 nM, ~68,000-fold
50
improvement compared to 5, but also exhibits more than 1,000-fold selectivity for Lp-
PLA2 over PLA2VIIB (Figure 3A). In addition, compound 7 showed a time-dependent
inhibition on Lp-PLA2, indicating the irreversible binding feature as expected. A plot of the
observed rates of inhibition (k ) of 7 against Lp-PLA2 versus concentration revealed a
obs
-1
-1
kinact/K value of 3140 M ·s , which is a typical sign for an irreversible inhibitor (Figure
I
S3A and S3B). A protein mass spectrometry analysis confirmed the covalent adduct of
Lp-PLA2 with a mass matching that of the wild-type enzyme modified by a condensation
with one molecule of compound 7 (Figure S4). We further solved the co-crystal structure
of compound 7 bound to Lp-PLA2 (PDB code: 6M08), and found that only one molecule
ACS Paragon Plus Environment
16

Page 17 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of 7 covalently binds to S273 and no extra unfilled electronic density left behind. This is
consistent with the results from the enzymatic activity assay and the mass spectrometry
analysis, showing that compound 7 reacted with Lp-PLA2 with 1:1 stoichiometry.
The crystal structure of the Lp-PLA2-7 complex revealed that the enol-
cyclocarbamate warhead of 7 adopts a highly conserved binding pose of fragment 5 and
the covalent bond is exactly maintained (Figure 3B). Moreover, the biphenyl group of 7
inserts into a hydrophobic sub-pocket formed by L371, L369, F357, A355, Q352, G154,
L121, L111, and F110, while the ethanediamine moiety extends to the solvent region. The
crystal structures of Lp-PLA2 in complex with compound 7 superimposed on darapladib
demonstrate that the protein-ligand interaction patterns of the biphenyl ethanediamine
moiety in two complexes are highly conserved (Figure 3C). Therefore, in the Lp-PLA2-7
complex, both the warhead which is almost identical to fragment 5 and the non-covalent
interacting portion borrowed from darapladib bind to the pocket as we desired (Figure
3D), highlighting the power and advantage of the structure-based drug design strategy.
Notably, although compound 7 selectively inhibited Lp-PLA2 at a nanomolar
concentration, both the warhead (fragment 5) and the biphenyl ethanediamine moiety
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
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 82
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
(
fragment 6) showed extremely weak potency and low selectivity. Inhibition of fragment 6
on Lp-PLA2 is only 43% at a concentration of 200 µM, and it is 44% for PLA2VIIB at a
concentration of 250 µM (Figure 3A). It is thus suggested that contribution of the enol-
cyclocarbamate warhead and the reversible fragment 6 has a significantly synergetic
effect on the potency and selectivity improvement.
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 3. The discovery of compound 7 by a fusion of the covalent enol-cyclocarbamate
and darapladib. (A) Schematic description for the design of compound 7 based on
ACS Paragon Plus Environment
18

Page 19 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fragment 5 and darapladib. A red circle indicates the proximity between the amide of
darapladib and the carboxyl of 5 by superimposing the co-crystal structures of Lp-PLA2
in complex with 5 (yellow) and darapladib (blue). (B) The Lp-PLA2-7 complex structure
reveals interactions of 7 (green) with residues (blue) in the substrate binding pocket (PDB
code: 6M08). H-bonds are represented by dashed lines. (C, D) Structure superimposition
of Lp-PLA2-7 on Lp-PLA2-darapladib (C) or Lp-PLA2-5 (D).
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
Introduction of a substituent filling into an unoccupied sub-pocket. After a thorough
investigation on the complex structure of Lp-PLA2-7, we considered an unoccupied sub-
pocket formed by L107, A155, L159, A355, and F357 near the trifluoromethyl substituted
benzene ring of the biphenyl group. Substituents at the meta-position of the
trifluoromethyl on this benzene ring might occupy this hydrophobic sub-pocket (Figure
4A-B). Accordingly, we introduced a linear and a cyclic lipophilic substituent in
compounds 8 and 9, respectively, in order to fill the sub-pocket (Figure 4 and Figure S5).
The enzymatic activity assay demonstrated that a linear 1,1,1-trifluoroethoxyl group
introduced in compound 8 enhanced ~2-fold inhibitory potency to IC = 13 nM. The
50
-1
-1
kinact/K value of compound 8 (6835 M ·s ) also showed ~1-fold increase compared to 7
I
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 82
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
(
Figure S3C and S3D). Most importantly, the selectivity of 8 for Lp-PLA2 over PLA2VIIB
reaches ~3,900-fold (Figure 4A), which is almost four times that of 7 (~1040-fold).
Compound 8 is by far the most selective inhibitor of Lp-PLA2 to the best of our knowledge.
The determined crystal structure of compound 8 in complex with Lp-PLA2 (PDB code:
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
6
M07) revealed the conserved protein-ligand interactions as those presented in the co-
crystal structure of Lp-PLA2-7 except that the hydrophobic sub-pocket mentioned above
was perfectly occupied by the substituted 1,1,1-trifluoroethoxyl group at the terminal
benzene ring of 8 (Figure 4B). By comparison, compound 9, introducing a benzyloxy
group at the same position, showed a similar potency compared to 8, while its selectivity
for Lp-PLA2 over PLA2VIIB dramatically declines to ~360-fold (Figure S5). The reason
for such a drop in selectivity from 8 to 9 is that 9 is a more potent inhibitor of PLA2VIIB,
but the mechanism of action of the substituted benzyloxy group on PLA2VIIB remains
unknown.
ACS Paragon Plus Environment
20

Page 21 of 82
Journal of Medicinal Chemistry
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. The discovery of compound 8 with high potency as well as selectivity for Lp-
PLA2 over PLA2VIIB. (A) Schematic description for structure-based optimization of
compound 7 to yield compound 8. (B) The co-crystal structures of Lp-PLA2-7 and Lp-
PLA2-8 (PDB code: 6M07) complexes, in which a sub-pocket marked with a gray arrow
was perfectly occupied by the introduced trifluoro-ethoxyl group.
High selectivity of compound 8 for Lp-PLA2 over other serine hydrolases. Because of
the concerns of nonspecific or off-target adduct formation, selectivity is one of the key
issues in design of safe and effective covalent ligands. It is especially important for
inhibitors of serine hydrolases like Lp-PLA2, due to the existence of a huge number of
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 82
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
4
7
these enzymes in cells. With these considerations in mind, we applied the activity-based
protein profiling (ABPP) method to investigate the selectivity of compound 8 in HEK293T
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
37
cells. ABPP mainly utilizes the active-site directed covalent probes to target specific one
or subsets (families) of proteins in complex proteomes so as to provide a quantitative
4
readout of the functional state of the targeting protein(s). The gel-based competitive
ABPP analysis was performed using FP-TAMRA which is a non-selective covalent
fluorescent probe of serine hydrolases. We first incubated the HEK293T lysate with
compound 8 at different concentrations and then added FP-TAMRA to the lysates. An in-
gel fluorescence scanning was performed on samples, showing that compound 8 did not
cross react with other serine hydrolases at concentrations up to 50 μM (Figure 5).
Presumably overlapped by other enzymes, a small amount of the basal Lp-PLA2 in the
HEK293T cell was not clearly visualized in the gel. To highlight the band of Lp-PLA2, we
repeated the same experiment with the HEK293T lysate added with the recombined
human Lp-PLA2 which was also used in other experiments involved in this study,
distinctly showing the changes of the Lp-PLA2 band before and after adding 8 at different
ACS Paragon Plus Environment
22

Page 23 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
concentrations (Figure 5). It is thus concluded that compound 8 has an excellent
selectivity for Lp-PLA2 over other serine hydrolases in the cells.
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 5. Selectivity evaluation of compound 8 in cells by the gel-based competitive ABPP
probed with FP-TAMRA which is a non-selective covalent fluorescent probe of serine
hydrolases. By comparison, the recombined human Lp-PLA2, the HEK293T lysate and
the mixture of these two were used in the first, second and third four samples,
respectively, and an in-gel fluorescence scanning was performed on all the samples.
Fluorescent probes design based on compound 8. Inspired by the excellent selectivity
of compound 8 in HEK293T cells, we decided to label compound 8 with different
fluorescent groups so as to provide an ideal fluorescent probe for Lp-PLA2. According to
the co-crystal structure of Lp-PLA2-8, the ethanediamine moiety is exposed to the solvent
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 82
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
accessible region and is thereby an ideal site to attach fluorophores without interfering
with the inhibitory activity. Utilizing difluoroboron dipyrromethene and lissamine
rhodamine B as the luminophore, we designed and synthesized two fluorescent
compounds, 8-BODIPY and 8-RH, respectively (Figure 6A), based on compound 8. The
enzymatic activity assay showed that two fluorescent compounds maintain the nanomolar
potency against Lp-PLA2 (8-BODIPY: IC = 14 nM; 8-RH: IC = 25 nM). The gel-based
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
50
50
fluorescent labelling analysis with 8-BODIPY or 8-RH shown in Figure S6 revealed that
these two fluorescent compounds hardly label any other proteins except Lp-PLA2 and its
degradation at concentrations up to 5 μM, which is ~385-fold and 200-fold of the in-vitro
IC s of 8-BODIPY and 8-RH, respectively. This again manifests the excellent selectivity
50
of the two fluorescent probes and their parent compound (8). An immunoblotting test on
the same gel used for fluorescence scanning resulted in the bands identical to the
fluorescent bands, demonstrating that the selectivity and sensitivity of our fluorescent
probes are comparable to the antibody of Lp-PLA2 (Figure S6).
ACS Paragon Plus Environment
24

Page 25 of 82
Journal of Medicinal Chemistry
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 6. (A) Chemical structures of 8-BODIPY and 8-RH. (B, C) An in-gel fluorescence
scanning using 8-BODIPY (B) or 8-RH (C) as a fluorescent probe with the HEK293T
lysate added with different concentrations of recombinant human Lp-PLA2. An
immunoblotting test was performed on the same gel used for the fluorescence scanning.
Fluorescent labeling of Lp-PLA2 by 8-BODIPY and 8-RH. With the high potency and
selectivity, 8-BODIPY and 8-RH are capable of specifically modifying Lp-PLA2 or
qualitative and even semi-quantitative characterization of the enzyme. We incubated the
two probes with the HEK293T lysate added with different concentrations of recombinant
human Lp-PLA2, an in-gel fluorescence scanning was then performed on samples, and
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 82
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
eventually an immunoblotting test was carried out on the same gel used for the
fluorescence scanning (Figure 6B and 6C). With a more convenient way, the bands,
which qualitatively and semi-quantitatively label Lp-PLA2 in the lysate, resulted from the
in-gel fluorescence scanning are highly consistent with those from the immunoblotting
test.
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
In addition, 8-BODIPY or 8-RH can also be used as an activity-based probe for
profiling Lp-PLA2 inhibited by a competitive binding ligand. The gel-based competitive
ABPP analysis of compound 8 with the use of 8-BODIPY and 8-RH as probes confirmed
the inhibition of Lp-PLA2 by 8, which is not accessible by the immunoblotting test (Figure
S7).
Then the probes were tested in HEK293T cells which were transfected with the Lp-
PLA2 expression vector. Due to a poor solubility and membrane permeability of 8-RH,
only 8-BODIPY was tested. HEK293T cells were transfected with Lp-PLA2 and cultured
for 48 h followed by the incubation with 1 μM 8-BODIPY for 10 min. Then the cells were
collected for SDS-PAGE followed by the in-gel fluorescent scanning which showed a
distinct band of Lp-PLA2 (Figure S8). An immunoblotting test on the same gel resulted in
ACS Paragon Plus Environment
26

Page 27 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the bands identical to the fluorescent bands, demonstrating the labelling of endogenous
Lp-PLA2 by 8-BODIPY (Figure S9).
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
Florescence polarization-based competitive experiment between 8-BODIPY and
darapladib. To investigate the competitive binding of our covalent inhibitors versus non-
covalent inhibitors to Lp-PLA2, an experiment of competition between 8-BODIPY and
darapladib was performed with a fluorescence polarization-based assay, as the
fluorescent probe, 8-BODIPY, is derived from the covalent inhibitor 8. Darapladib is the
most potent non-covalent inhibitor of Lp-PLA2 with an IC of 0.7 nM and thus was used
50
for this experiment. Recombinant Lp-PLA2 was first incubated with gradiently diluted
darapladib followed by an addition of 10 nM 8-BODIPY, and the signal of fluorescence
polarization decreased with the increased concentration of darapladib (Figure S10). This
result reveals that the half maximal competitive concentration of darapladib is 11.65 nM
in the presence of 10 nM 8-BODIPY, demonstrating the competition between 8-BODIPY
and darapladib. In contrast, the signal of fluorescence polarization only slightly changed
with the increased concentration of darapladib (Figure S10) when recombinant Lp-PLA2
was first incubated with 10 nM 8-BODIPY followed by the addition of gradiently diluted
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 82
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
darapladib. These results show an irreversible binding of 8-BODIPY to Lp-PLA2 and
reveal that the higher binding ability of the covalent inhibitors allows them to compete for
potent non-covalent inhibitors.
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
Fluorescence imaging of 8-BODIPY in living cell. Next, 8-BODIPY was tested for live-
cell imaging. Living PC-3 cells, in which the endogenous Lp-PLA2 is expressed, were
treated with 8-BODIPY, ER-Tracker Red and Hoechst 33342 which characterizes Lp-
PLA2, endoplasmic reticulum (ER) and nucleus, respectively, for a confocal laser
scanning. The resulting imaging shows colocalization of fluorescence of 8-BODIPY and
ER-Tracker Red (Figure S11), suggesting Lp-PLA2 tends to co-localize with ER. Such a
result is consistent with the characteristic of secreted proteins like Lp-PLA248 and
matches an immunofluorescence imaging of Lp-PLA2 in a previous work.49
To further confirm the binding of 8-BODIPY with Lp-PLA2 in the living cells, an
experiment of competitive binding with Lp-PL2 between 8-BODIPY and darapladib was
performed. The result showed that the florescent signal of 8-BODIPY within cells
decreased after the competitive binding of darapladib with Lp-PLA2 (Figure S12),
demonstrating the binding of 8-BODIPY with Lp-PLA2 in the living cells.
ACS Paragon Plus Environment
28

Page 29 of 82
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
DISCUSSION AND CONCLUSIONS
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
In summary, to accomplish the evolution of a covalent fragment/warhead into a potent
and selective ligand, we efficiently conducted two rounds of optimization and synthesised
compounds with the aid of the co-crystal structures. The resulting compound 8 has a
more than 130,000-fold increase in the inhibitory activity on Lp-PLA2 and a 3,900-fold
increase in the selectivity for Lp-PLA2 over PLA2VIIB. It is by far the most selective
inhibitor of Lp-PLA2. Furthermore, it shows an excellent selectivity over a wide range of
other serine hydrolases in the cells, providing a qualified covalent lead for further
development. Future studies are warranted to evaluate the potency of compound 8
against Lp-PLA2 in vivo as well as the efficacy of the compound in the animal models for
Lp-PLA2-related diseases. Additionally, based on this compound, we designed selective
and sensitive Lp-PLA2 probes to characterize Lp-PLA2 in vitro or even in living cells in a
more convenient way, and also to be used as probes of ABPP tailored for Lp-PLA2. Based
on these results, these probes could be considered as a useful chemical biology probe
ACS Paragon Plus Environment
29