Design, Synthesis, and Pharmacological Evaluation of Novel Multisubstituted Pyridin-3-amine Derivatives as Multitargeted Protein Kinase Inhibitors for the Treatment of Non-Small Cell Lung Cancer

Article abstract of DOI:10.1021/acs.jmedchem.7b00076

A novel series of pyridin-3-amine derivatives were designed, synthesized, and evaluated as multitargeted protein kinase inhibitors for the treatment of non-small cell lung cancer (NSCLC). Hit 1 was first disclosed by in silico screening against fibroblast

Full text of DOI:10.1021/acs.jmedchem.7b00076

Subscriber access provided by EAST TENNESSEE STATE UNIV
Article
Design, Synthesis, and Pharmacological Evaluation of Novel Multi-
substituted Pyridin-3-amine Derivatives as Multi-Targeted Protein
Kinase Inhibitors for the Treatment of Non-Small Cell Lung Cancer
Wei Zhu, Hui Chen, Yulan Wang, Jiang Wang, Xia Peng, Xian-Jie Chen, Yinglei
Gao, Chunpu Li, Yulong He, Jing Ai, Mei-Yu Geng, Mingyue Zheng, and Hong Liu
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 22 Jun 2017
Downloaded from http://pubs.acs.org on June 22, 2017
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis, and Pharmacological Evaluation of Novel
Multiꢀsubstituted
Pyridinꢀ3ꢀamine
Derivatives
as
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MultiꢀTargeted Protein Kinase Inhibitors for the Treatment of
NonꢀSmall Cell Lung Cancer
†
,‡,§,
║
†,§,
║
†,§,
║
†,‡,§
†,§
Wei Zhu
, Hui Chen , Yulan Wang , Jiang Wang , Xia Peng , Xianjie
†
,‡,§
†,§
†,‡,§
†,‡,§
*,†,§
†,§
Chen , Yinglei Gao , Chunpu Li , Yulong He , Jing Ai , Meiyu Geng ,
*,†,§
*,†,‡,§
Mingyue Zheng
and Hong Liu
†
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica,
Chinese Academy of Sciences, 555 Zu Chong Zhi Road, Shanghai, 201203, China.
‡
CAS Key Laboratory of Receptor Research, Shanghai Institute of Materia Medica,
Chinese Academy of Sciences, 555 Zu Chong Zhi Road, Shanghai 201203, P. R.
China.
§
University of Chinese Academy of Sciences, No.19A Yuquan Road, Beijing 100049,
China.
║
These authors contributed equally to this study.
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 70
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:
A novel series of pyridinꢀ3ꢀamine derivatives were designed, synthesized and
evaluated as multiꢀtargeted protein kinase inhibitors for the treatment of nonꢀsmall
cell lung cancer (NSCLC). The hit 1 was firstly disclosed by in silico screening
against fibroblast growth factor receptors (FGFR), which was subsequently validated
by in vitro experiments. The structureꢀactivity relationship (SAR) of its analogs was
then explored to afford novel FGFR inhibitors 2aꢀ2p and 3aꢀ3q. Among them, 3m
showed potent inhibition against FGFR1, 2, and 3. Interestingly, compound 3m not
only inhibited various phosphorylation and downstream signaling across different
oncogenic forms in FGFR overactivated cancer cells, but also showed nanomolar
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
,
level inhibition against several other NSCLCꢀrelated oncogenes kinases including
RET, EGFR, EGFR/T790M/L858R, DDR2, and ALK. Finally, in vivo pharmacology
evaluations of 3m showed significant antitumor activity (TGI = 66.1%) in
NCIꢀH1581 NSCLC xenografts with good PK profiles.
Key words: tyrosine kinase, FGFR inhibitors, virtual screening, antitumor.
2
ACS Paragon Plus Environment

Page 3 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
Lung cancer is one of the most common cancers and nonꢀsmall cell lung cancer
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(NSCLC) accounts for 80–85% of all lung cancers. An epidermal growth factor
receptor (EGFR) inhibitor has recently been developed and has been shown to be
effective against NSCLC as more than 60% of NSCLCs express genetic mutations of
2
EGFR. However, the emergence of drugꢀresistant variants of NSCLC has greatly
reduced the clinical efficacy of EGFR inhibitors such as gefitinib. Multiꢀtargeted
tyrosine kinase inhibitors (TKIs), such as sunitinib, motesanib, sorafenib, lapatinib,
and vandetanib, have therefore been designed based on these drugꢀresistant variants.
For EGFRꢀTKIsꢀresistant patients with unknown driving mechanism, the concept of
"
cancer driver gene" may help to increase the efficacy of multiꢀtargeted drugs for
3
NSCLC. In recent years, various driver oncogenic genes in NSCLC have been
characterized, including EGFR mutation (15ꢀ20%), ALK rearrangements (5%), MET
amplification, ErbB2 mutations (1ꢀ3%), PI3KCA mutation (5ꢀ10%), NF1 mutations
(5%), KRAS mutations (25ꢀ30%), FGFR1 amplifications (15ꢀ20%), FGFR 2/3/4
mutations/rearrangements (5ꢀ10%), DDR2 mutations (4%), ROS1 rearrangements
4
(1ꢀ2%), and RET rearrangements (1ꢀ2%), and so on.
FGFRs are highly conserved transmembrane tyrosine kinase receptors. In human,
there are four FGFRs that are typical tyrosine kinase receptors (FGFR1ꢀ4), and they
are involved in several basic biologic processes, including tissue development,
5,6
angiogenesis, and tissue regeneration. An initial screen of 155 squamous cell lung
cancer specimens using single nucleotide polymorphism arrays identified focal
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 70
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
7
amplifications of the FGFR1 gene. Survival of FGFR1ꢀamplified lung cancer cell
lines was additionally shown to be dependent on overexpression of the FGFR1
8
kinase. As shown in Figure 1, there are many FGFR inhibitors in development for
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
9
NSCLC, including both selective FGFR inhibitors (LY2874455 ) and multiꢀtargeted
TKIs with FGFR activity. Currently, several multiꢀtargeted FGFR inhibitors have
1
0
11
been approved for NSCLC treatment, including vandetanib,
lenvatinib,
1
2
13
nintedanib, ponatinib, and several others are under clinical trial studies
14
15
16
17
18
19
(
anlotinib, cediranib, lucitanib, brivanib, danusertib, CPꢀ547632, and
2
0
ATꢀ9283 ). For example, dovitinib has been investigated as a nonspecific FGFRs
inhibitor, and it also targets vascular endothelial growth factor receptor 1/2/3
(VEGFR1/2/3), plateletꢀderived growth factor receptor α/β (PDGFRꢀα/β), stem cell
factor receptor (cꢀKit), FMSꢀlike tyrosine kinase 3 (FLT3), and colonyꢀstimulating
factor 1 receptor (CSF1R). In phase II trial, dovitinib exhibited efficacy in patients
21
with advanced squamous NSCLC with FGFR1 amplification. Another example is
crizotinib, a dual inhibitor of ALK and ROS1 signaling, showed marked antitumor
22,23
activity in patients with ROS1 rearranged or ALKꢀrearranged advanced NSCLC.
In 2013 and 2016, FDA approved crizotinib for the treatments of patients with
advanced NSCLC whose tumors have ROS1 rearrangements or ALK rearrangements,
respectively. Since it has now been accepted that NSCLC is not a singular entity but is
in fact multiple pathologies with unique molecular signatures, there has been a
renewed interest in developing multiꢀtargeted TKIs targeting different signaling
pathways and oncogenic drivers for enhancing efficacy and response rates. Discovery
4
ACS Paragon Plus Environment

Page 5 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and development of multiꢀtargeted TKIs with FGFR activity is therefore a promising
direction for the treatment of NSCLC.
In this study, molecular dockingꢀbased virtual screening was used to discover novel
FGFR inhibitors, based on the clarification of FGFR kinase structures. Fiftyꢀseven
compounds were selected for FGFR1 inhibitor bioactivity testing, and a compound
with the pyridinꢀ3ꢀamine scaffold 1 showed high inhibition ratio at a concentration of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
0 µM. Further bioactivity testing on compound 1 revealed that this compound
possesses a low micromolar level inhibitory activity with IC50 value of 3.8 ± 0.5 µM
against FGFR1. To improve the inhibition potency of compound 1 against FGFR1, we
analyzed its binding mode with FGFR1 and performed a structureꢀbased structural
optimization. The most potent compound 3m displayed nanomolar IC50 against FGFR
and FGFRꢀdepended cancer cell line. To better understand the pharmacological
profile of 3m, a kinase panel screening on 3m was performed, the effects of 3m on
protein kinase phosphorylation and downstream signaling were studied, and the
antiproliferative effects of 3m against several cancer cell lines that harboring the
oncogenic kinases targeted by 3m were detected. It was shown that compound 3m
targets several NSCLC related oncogenes kinases with nanomolar IC50, including
against FGFRs, RET, EGFR, EGFR/T790M/L858R, DDR2, and ALK. Meanwhile,
compound 3m could inhibit cancer cell lines proliferation harboring frequently
occurring oncogenic forms of FGFRs, DDR2, EGFR, and RET. Finally, in vivo
studies shown that compound 3m has good pharmacokinetic profiles and in vivo
activities against NCIꢀH1581 xenografts for the treatment of NSCLC.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 70
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
Multi-targeted FGFR inhibitors
H
H
F
Br
H
N
H
N
N
N
HN
O
O
N
O
N
O
O
O
N
Cl
O
O
O
O
F
O
O
F
N
N
H2N
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
H2N
N
N
N
Vandetanib, Approved 2011
Lenvatinib, Approved 2015
Anlotinib, Phase II/III
Cediranib, Phase III
O
H
N
N
N
F
O
N
F
F
H
N
N
N
N
2HN
O
N
NH
O
HN
O
N
N
O
O
O
F
O
O
N
O
N
N
H2N
O
N
H
O
Lucitanib, Phase II
Nintedanib, Approved 2014
Ponatinib, Approved 2012
Brivanib, Phase III
HN
N
F
H
O
N
N
N
N
N
O
NH2
F
HN
H
O
N
S
O
Br
N
N
O
N
N
N
O
O
N
N
H
F
H
H
N
NH
N
HN
N
O
N
H
NH2
O
Danusertib, Phase II
CP-547632, Phase II
Dovitinib, Phase III
AT-9283 Phase II
Selective FGFR inhibitors
OH
N
O
N
O
N
Cl
N
Cl
O
N
N
N
O
HN
O
O
HN
N
O
N
N
N
N
NH
N
H
H
Cl
Cl
N
H
Ly2874455, Phase I
15 (BGJ-398), Phase II
16 (AZD4547), Phase II
Figure 1. The representative structures of multiꢀtargeted and selective inhibitors with
FGFR inhibitory activities for the treatment of NSCLC.
6
ACS Paragon Plus Environment

Page 7 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Result and Discussions
Identification and Validation of Multiꢀsubstituted Pyridinꢀ3ꢀamine Scaffold by
Structureꢀbased Virtual Screening
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To discover novel chemotypes of FGFR inhibitors, a dockingꢀbased virtual
2
4
screening was performed on an inꢀhouse compound library that contains more than
25ꢀ33
6
,000 compounds with structural diversity.
The crystal structure of FGFR1 in
3
4
complex with potency inhibitor 15 (PDB code 3TT0) was used as the receptor for
docking by the Schrӧdinger Suite 2015 software package. The ligand database was
prepared with LigPrep module to generate energy minimized threeꢀdimensional (3D)
coordinates, in which the protonation states of molecules were calculated with Epik.
The crystal structure of receptor was converted into allꢀatom and refined with the
Protein Preparation Wizard in Schrӧdinger, where the force field used was OPLS3.
Finally, the prepared ligands database was docked into prepared receptor with extra
35
precision (XP) mode in Glide module. The topꢀranked 1,120 compounds with XP
scores lower than ꢀ7.0 kcal/mol were kept and the duplicated compounds were
removed. Then, the resulting compounds were clustered based on structure similarity
using Pipeline Pilot 7.5 software. For each cluster, only one or two compounds with
higher docking scores were retained. In the end, by considering both the Glide XP
scores and structural diversity, 57 compounds were selected for FGFR1 inhibition
assays to validate them in vitro activity. With the ELISA kinase assay at 50 µM,
compound 1 showed inhibitory potency with inhibition ratio of 76.5% (Figure 2A).
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 70
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
Further evaluation demonstrated that compound 1 possesses a low micromolar level
activity with IC50 value of 3.8 ± 0.5 µM against FGFR1.
Binding Mode Analysis of Compound 1
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
To gain deeper insights into the molecular basis of the inhibitory activity, we analyzed
the putative binding mode of compound 1 obtained by molecular docking. As shown
in Figure 2B, the active site of FGFR can be generally segmented into three sections:
a hinge region, a hydrophobic pocket, and a Pꢀloop region. As depicted in Figure 2B,
compound 1 binds in the active site of FGFR1. The pyrazolyl moiety establishes two
hydrogen bonds with the backbone amino and carbonyl of Ala564 and Glu562 in the
hinge region, and the furyl group is located in the hydrophobic pocket of FGFR
protein. The hydroxyl of compound 1 could form a hydrogen bond with the backbone
carbonyl of Leu484 in the Pꢀloop, and the amino group may form a weak hydrogen
bond with the backbone carbonyl of Glu486. Comparing the putative binding modes
36
of compound 1 with the crystal structures of reported FGFR inhibitors 15 and 16,
we may find that the 3,5ꢀdimethoxyꢀphenyls of the reported two inhibitors occupy the
same hydrophobic pocket of the furyl of compound 1 (Figure 2C). However, the furyl
ring doesn’t fit the hydrophobic pocket well, which is more spacious and may
accommodate a larger group. Intuitively, modifying the furyl group to better fill the
hydrophobic pocket is the first round of scheme for the chemical optimization of
compound 1. Cheng et al. reported a series of 3,5ꢀdisubstutiedꢀ pyridine derivatives as
37
CDK inhibitors for the treatment of cancers.
Compared with
8
ACS Paragon Plus Environment

Page 9 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
,5ꢀdisubstutiedꢀpyridine derivatives, the distinct substitutes at the 6ꢀposition of the
pyridine ring may have great influence on the inhibitory activities against FGFR.
Subsequently, the structure−activity relationship (SAR) at Pꢀloop interaction region,
hydrophobic pocket, and hinge interaction region of the pyridine core will be further
explored (Scheme 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Identification of compound 1 as a hit against FGFR. (A) Inhibitory activity
of the 57 selected compounds at 50 µM. Compound 16 (black column) was used as a
positive control and the inhibition ratio of compound 1 (red column) is 76.5%. (B)
The putative binding mode of compound 1 with FGFR1, where hydrogen bonds are
depicted as dashed red lines and hydrophobic pocket of FGFR1 is marked as an
orange circle. (C) Superimposition of the binding mode of compound 1 (green) with
the crystal structures of compounds 15 (slate) and 16 (orange), where the hydrophobic
pocket of FGFR1 is marked as an orange circle.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 70
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
Scheme 1. SAR exploration of pyridinꢀ3ꢀamine derivatives as potent FGFR
inhibitors.
Chemistry
Two synthetic routes were exploited to access the desired compounds 2 and 3
(
Schemes 2 and 3). With the treatment of Grignard reagent MeMgBr, aldehydes 4
were converted to ethanols, which yielded corresponding ketones 5 upon oxidation of
MnO (Scheme 2). Pyrazoles 6 were then prepared by following refluxing with
2
hydrazine hydrate and the following protection with a Boc group. The boronic esters 7
were provided via a Miyaura boylation reaction. Pyridinꢀ3ꢀamine 8 underwent
2
Sandmeyer reaction in the presence of NaI, NaNO , and hydrochloric acid to afford 9,
which was coupled with ( )ꢀ2ꢀaminoꢀ2ꢀphenylethanol with CuI as catalyst to give key
R
intermediate 10. The Suzuki coupling of 7 and 10 provided chloropyridine 11, which
underwent another coupling with commercial available boronic acids to yield the
desired compounds 1, 2aꢀ2p, 3a, and 3kꢀ3q. The FGFR inhibitors 3bꢀ3e could be
obtained by the coupling of intermediate 12 with different hetero boronic acids and
5ꢀchloroꢀ2ꢀhydroxyphenylboronic
acid.
As
shown
in
Scheme
3,
3
ꢀbromoꢀ2ꢀchloropyridines 13 were prepared by copper catalyzed coupling of
intermediate 9 with amino alcohols. The following twoꢀstep Suzuki coupling of 13
10
ACS Paragon Plus Environment

Page 11 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with intermediate 7a and 2ꢀhydroxyphenylboronic acids afforded the desired
compound 3fꢀ3j.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
Scheme 2
a
o
Reagents and conditions: (a) MeMgBr, THF, 0 C; then, MnO
(b) N (75 % in H O), reflux; (c) Boc O, Et
,4,4',4‘,5,5,5',5'ꢀoctamethylꢀ2,2'ꢀbi(1,3,2ꢀdioxaborolane),
2
, 1,4ꢀdioxane, reflux;
2
H
4
2
2
3
N, DCM, r.t.; then,
Pd(dppf) Cl ·CH Cl ,
4
2
2
2
2
o
o
KOAc, 1,4ꢀdioxane, 80 C; (d) NaI, HCl (6 M), NaNO
2
, H
2
O, 0 C – r.t.; (e)
i
(R
)ꢀ2ꢀaminoꢀ2ꢀphenylꢀethanol, CuI,
Pd(dppf) Cl ·CH Cl , Na CO , 1,4ꢀdioxane, H
acids, Pd(dppf) Cl ·CH Cl , Na CO
K
3
PO
4
,
ethylene glycol, PrOH; (f)
o
2
2
2
2
2
3
2
O, 90 C; then TFA, DCM; (g) boronic
o
2
2
2
2
2
3
, 1,4ꢀdioxane, H
2
O, 120 C, MW. (h) boronic
o
acids, Pd(dppf) Cl ·CH Cl , K CO , 1,4ꢀdioxane, H O, 90
C; (i)
2
2
2
2
2
3
2
5
ꢀchloroꢀ2ꢀhydroxyphenylboronic acid, Pd(dppf)
2
Cl
2
·CH
2
Cl
2
, Na
2
CO
3
, 1,4ꢀdioxane,
o
H
2
O, 120 C, MW.
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 70
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
a
Scheme 3
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
i
Reagents and conditions: (a) amines, CuI, K
Cl ·CH Cl , Na CO , 1,4ꢀdioxane, H
acids, Pd(dppf) Cl ·CH Cl , Na CO
3
PO
4
, ethylene glycol, PrOH; (c) 7a,
o
Pd(dppf)
2
2
2
2
2
3
2
O, 90 C; then TFA, DCM; (d) boronic
o
2
2
2
2
2
3
, 1,4ꢀdioxane, H
2
O, 120 C, MW.
Structure−Activity Relationships of MultiꢀSubstituted Pyridineꢀ3ꢀamine
Derivatives
Based on the modeled complex structure of hit compound 1 with FGFR1, we have
designed and synthesized dozens of pyridineꢀ3ꢀamine derivatives to explore their
structure and activity (SAR) relationship. As shown in Table 1, the preliminary SAR
at the hydrophobic pocket was firstly investigated. Replacement of 3ꢀfuran fragment
of compound 1 with various heteroaryl moieties or a benzene ring afforded
compounds 2aꢀ2g. Compared with furan and thiophene analogs 2aꢀ2c, benzene
substituted derivative 2d displayed improved activity. Incorporation of
N
ꢀcontainingꢀheteroaryl substituents to yield compounds 2eꢀ2g resulted in decrease of
inhibitory activities. Encouraged by these results, analogs containing substituted
12
ACS Paragon Plus Environment

Page 13 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
phenyl groups at the 6ꢀposition of pyridine nucleus were prepared for further activity
optimization (2hꢀ2p). Introduction of fluorine (2h) had only marginal influence on
FGFR inhibition, while trifluoromethyl analog 2i showed complete loss of activity.
Interestingly, compounds 2jꢀ2l, bearing Hꢀbond donor fragment such as amide or
hydroxymethyl group, maintained or slightly enhanced inhibitory activity against
FGFR. To our delight, 2ꢀOH substituted derivative 2m demonstrated potent FGFR
inhibition, with an enzymatic IC50 of 14.9, 3.0, and 8.0 nM against FGFR1, 2, and 3,
respectively. Blocking of the phenolic hydroxyl group (2n) led to sharply reduced
potency, which indicated the critical role of this potential hydrogenꢀbond interaction
with FGFR. Such pseudoꢀcyclic motifs have been reported for kinase inhibitor
hingeꢀbinders with an intramolecular hydrogen bond. However, based on putative
binding mode, it is unlikely to form an internal Hꢀbond in compound 2m in our case
(For details, see Figures S1 and S2 in supporting information). Compared with
compound 2m, 3ꢀOH substituted analog 2o exhibited 5ꢀfold decreased potency
against FGFR1. With the incorporation of chlorine at the 4 position of the phenol
moiety in 2m, compound 2p also showed potent FGFR inhibition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figures 3A and 3B displayed the putative binding modes of compounds 1, 2d, and 2p
with FGFR1. Compared with compounds 1 and 2d, the replacement of furyl to phenyl
could gain more hydrophobic shape complementarity with the hydrophobic pocket of
FGFR, which is consistent with their inhibitory potency difference in the enzymatic
assays. In addition, the phenyl group of 2p occupies the hydrophobic pocket like 2d,
and the hydroxyl substituent on the phenyl of 2p could form two hydrogen bonds with
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 70
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 backbone amino and side chain carboxyl of Asp641. Accordingly, the inhibitory
activity of 2p was improved as compared with 2d.
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
14
ACS Paragon Plus Environment

Page 15 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Table 1. SAR at the 6ꢀposition of the Pyridine Scaffold
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
, 2aꢀ2p
FGFR1
(IC50, nM)
3800 ± 500
> 1000
FGFR2
(IC50, nM)
> 1000
FGFR3
(IC50, nM)
> 1000
Compd.
R
1
furanꢀ3ꢀyl
furanꢀ2ꢀyl
thiophenꢀ3ꢀyl
thiophenꢀ2ꢀyl
Phꢀ
2a
100ꢀ1000
56.9 ± 9.0
100ꢀ1000
26.3 ± 6.2
> 1000
> 1000
2
b
327.4 ± 35.1
> 1000
969.7 ± 236.8
> 1000
2c
2
d
127.9 ± 13.3
> 1000
279.9 ± 30.5
> 1000
2e
pyridinꢀ3ꢀyl
quinolinꢀ8ꢀyl
indolꢀ7ꢀyl
2ꢀFꢀPhꢀ
2
f
> 1000
> 1000
> 1000
2g
456.0 ± 122.4
437.3 ± 113.3
> 1000
56.4 ± 5.2
44.6 ± 12.9
> 1000
560.7 ± 35.1
430.5 ± 68.1
> 1000
2
h
2i
2ꢀCF
2ꢀNHAcꢀPhꢀ
2ꢀCH OHꢀPhꢀ
2ꢀCONH ꢀPhꢀ
3
ꢀPhꢀ
2
j
418.4 ± 64.6
167.3 ± 49.7
48.8 ± 10.5
14.9 ± 2.9
> 1000
250.2 ± 49.1
50.5 ± 0.6
72.9 ± 10.3
3.0 ± 1.0
100ꢀ1000
3653.5 ± 283.9
250.5 ± 35.6
388.0 ± 73.8
8.0 ± 1.6
2k
2
2
l
2
2m
2ꢀOHꢀPhꢀ
2
n
2ꢀOMeꢀPhꢀ
> 1000
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 70
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
2
o
3ꢀOHꢀPhꢀ
66.5 ± 5.2
55.8 ± 3.9
0.7 ± 0.2
21.4 ± 4.0
433.8 ± 32.3
73.1 ± 4.3
5.5 ± 0.4
2
p
2ꢀOH, 5ꢀClꢀPhꢀ
8.6 ± 2.0
1.1 ± 0.1
1
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
The IC50 values are shown as the mean ± SD (nM) from two separate experiments.
As shown in Table 2, the SARs at the moiety interacting with the hinge region were
also explored, with the privileged 4ꢀchloroꢀphenol fragment retained. As mentioned
above, the pyrazole group forms two hydrogen bonds with the hinge region. To
investigate the importance of the interaction, we synthesized derivatives 3aꢀ3e, of
which the pyrazole group was replaced with other azaꢀheterocyclic or noncyclic
groups (Table 2). Among these derivatives, compound 3a, isoquinoline analog 3b, and
indolone analog 3c exhibited similar or slightly increased potency compared with
compound 2p, whereas the ringꢀopening derivative (3d) and indole derivative (3e)
displayed significantly reduced activity, demonstrating the pivotal role of the
hydrogen bond donating group at the hinge region of FGFR. To account for the
chemical basis of this result, we analyzed the putative binding modes of 3b and 3c. As
shown in Figure 3C, 3b could form one hydrogen bond with the hinge region through
quinolyl nitrogen atom to Ala564, and the hydroxyl forms a hydrogen bond with
Glu484. Compound 3c could maintain two hydrogen bonds with the hinge region,
while the hydroxyl doesn’t form any hydrogen bond interaction, which should be
responsible for the decreased inhibitory potency of 3c. In addition, the noncyclic
derivative 3d and indoleꢀsubstituted derivative 3e cannot form stable anchoring
16
ACS Paragon Plus Environment

Page 17 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
interaction with hinge region, which may lead to its reduced inhibitory activity.
The SAR of the 1ꢀhydroxymethylꢀ1ꢀphenylꢀmethylamnio moiety of 2p was also
explored, which interacts with Pꢀloop through two hydrogen bonds based on the
model. Either altering the chirality (3f) or removing the terminal hydroxyl group (3g)
resulted in significant loss of potency, highlighting the importance of Pꢀloop region
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
interaction. As shown in Figure 3D, (
R
)ꢀisomer 2p could form two hydrogen bonds
with Leu484 and Glu486 in the Pꢀloop based on the docking, while (
S
)ꢀisomer 3f only
could form one hydrogen bond through the amino to Glu486, thus showing a
decreased activity.
Finally, further modification at the phenol segment at the 6ꢀposition of the pyridine
nucleus led to the preparation of compounds 3kꢀ3q. The replacement of the chlorine
substituent with methyl group (3k) retained the potent inhibition against FGFR,
however, the bulky isopropyl group (3l) led to loss of inhibitory activity. Interestingly,
the following Fꢀscan at the phenol moiety afforded a handful of FGFR inhibitors
3mꢀ3q with significantly enhanced activities, which can be attributed to the feature of
relatively small and electronꢀwithdrawing fluorine atom (For details, see Figure S3 in
38,39
supporting information).
1
2
3
Table 2. SAR of Substituents R , R and R on the 2ꢀ(Pyridinꢀ2ꢀyl)Phenol
a
Scaffold
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 70
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
3aꢀ3q
FGFR1
FGFR2
FGFR3
1
2
3
Compd.
R
R
R
(IC50, nM) (IC50, nM) (IC50, nM)
3a
5ꢀCl
5ꢀCl
5ꢀCl
5ꢀCl
5ꢀCl
5ꢀCl
24.9 ± 3.1
16.1 ± 4.5
91.5 ± 8.5
>1000
6.6 ± 1.5
2.0 ± 0.1
93.9 ± 25.7
23.0 ± 8.2
3b
153.5 ±
591.5 ±
57.2
3c
11.9
3d
>1000
>1000
>1000
>1000
>1000
>1000
3
e
>1000
3f
>1000
18
ACS Paragon Plus Environment

Page 19 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3
g
5ꢀCl
>1000
>1000
>1000
>1000
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3h
5ꢀF
5ꢀF
>1000
>1000
>1000
>1000
3
i
>1000
168.3 ±
3
j
5ꢀCl
71.9 ± 26.7 401.5 ± 1.1
51.7
3k
5ꢀMe
32.3 ± 10.8 16.1 ± 6.0
22.8 ± 4.9
i
3
l
5ꢀ Pr
>1000
>1000
>1000
3
m
5ꢀF
4ꢀF
3ꢀF
18.0 ± 4.9
25.2 ± 8.1
27.9 ± 0.4
1.6 ± 0.3
4.7 ± 0.7
2.5 ± 0.8
27.5 ± 9.4
15.9 ± 0.9
35.3 ± 4.1
3n
3o
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 70
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
3
p
4,5ꢀdiꢀF 10.6 ± 1.1
2.8 ± 0.8
27.6 ± 1.1
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
3q
3,5ꢀdiꢀF 63.5 ± 0.2
3.4 ± 0.3
1.1 ± 0.1
58.4 ± 14.0
5.5 ± 0.4
1
6
0.7 ± 0.2
a
The IC50 values are shown as the mean ± SD (nM) from two separate experiments.
Figure 3. The putative binding modes of pyridineꢀ3ꢀamine derivatives by docking in
FGFR1 protein (PDB code: 3TT0). (A) The binding modes of 1 (green), 2d (yellow)
and 2p (pink) in the hydrophobic pocket of FGFR1, of which the front side of pocket
20
ACS Paragon Plus Environment

Page 21 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
surface is clipped away for clarity and the hydrophobic pocket is depicted as a dashed
orange circle. (B) The binding modes of 2d (yellow) and 2p (pink) with FGFR1. (C)
Superimposition of the binding modes of 2p (pink), 3c (blue), and 3b (limon). (D)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The binding modes of 2p ((
R)ꢀisomer) and its enantiomer 3f ((S)ꢀisomer). Hydrogen
bonds are depicted as dashed red lines.
Cancer Cell Inhibitory Activities of the Selected Compounds:
Based on the enzymatic inhibitory potency against FGFR, compounds 2d, 2m, 2n, 2p,
3a, 3b, and 3mꢀ3q were selected to further evaluate their inhibitory activities against
FGFR2 amplification cell line SNU16 (Table 3). The panꢀFGFR inhibitor 16 was
tested as positive control. As shown in Table 3, the selected compounds exhibited
moderate to potent inhibition. Overall, the cellꢀbased antiꢀproliferative results of the
selected compounds generally agree with the FGFR inhibition results at the molecular
level. Among them, compound 3m demonstrated best inhibitory activity against cell
assay SNU16 with IC50 of 24.8 nM, which thus was selected for further investigation.
Table 3. Antiꢀproliferative Activities of Derivatives against FGFRꢀExpressing
a
Cancer Cell SNU16
Compd.
SNU16 (IC50, nM)
Compd.
SNU16 (IC50, nM)
2
d
677.7 ± 30.5
57.5 ± 3.7
3m
3n
24.8 ± 0.6
2
m
313.3 ± 25.5
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 70
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
2
o
36.2 ± 2.9
3o
3p
3q
16
293.5 ± 27.8
288.6 ± 84.2
280.1 ± 43.5
25.4 ± 1.2
2
p
24.8 ± 3.3
2281.3 ± 20.0
249.0 ± 44.8
3
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
3
b
a
The IC values are shown as the mean ± SD (nM) from two separate experiments.
50
Compounds 2p, 3b, 3m, and 3n as multiꢀtarget kinase inhibitors
Compounds 2p, 3b, 3m, and 3n were distinguished for their remarkable potency
against recombinant human FGFR kinase. Accordingly, we were prompted to
investigate whether this potency was specifically against FGFR. As shown in Table 4,
compounds 2p, 3b, 3m, and 3n also strongly inhibited the kinase activity of KDR,
VEGFRꢀ1, Ret, EGFR, DDR2, ABL, ErbB4, and cꢀSrc, with IC50 value at nanomole
or subꢀnanomole level. Compounds 2p, 3b, 3m, and 3n are, therefore,
multiꢀtargeted protein kinase inhibitors.
Given the relative high inhibitory potency of compounds 2p, 3b, 3m, and 3n against
FGFR, RET, DDR2, EGFR and VEGFR2, together with the fact that these kinases are
important targetable oncogenes in NSCLC and the most verified targets in cancer
therapy, we took these kinases as representatives to evaluate the inhibitory effect of
2
p, 3b, 3m, and 3n on protein kinasesꢀassociated events. Interestingly, small
structural changes may significantly alter the activity profile of the compounds. For
example, compound 3b is much more potent than 3m on FGFR4 and EGFR inhibition,
22
ACS Paragon Plus Environment

Page 23 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
while 3m is more active than 3b on IR and IGF1R. Such variance in kinase activity
profile may provide opportunity for future fineꢀtuning the efficacy of the compounds,
suggesting that these series of compounds are usefully multiꢀtargeted, instead of
nonspecific promiscuous kinase inhibitors. While, compounds 2p, 3b, 3m, and 3n
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
displayed no obvious inhibitory activity against CDK1, CDK4, and CDK6 (IC50
>
1
000 nM, Table 4), compared with 3,5ꢀdisubstitutedꢀpyridine analogs reported by
25
Cheng et al.
a
Table 4. The kinase panel screening data of compounds 2p, 3b, 3m, and 3n.
IC50 (nM)
Kinase
2
p
3b
3m
3n
FGFR1
FGFR2
FGFR3
FGFR4
KDR
10.1 ± 0.2
8.2 ± 1.3
15.9 ± 3.8
61.7 ± 6.4
3.7 ± 0.2
0.7 ± 0.1
3.1 ± 1.4
8.6 ± 2.5
ꢀ
12.7 ± 2.6
1.0 ± 0.1
6.0 ± 0.3
7.4 ± 3.2
0.8 ± 0.0
0.04 ± 0.01
4.2 ± 1.9
1.9 ± 0.8
ꢀ
8.7 ± 1.1
2.4 ± 1.3
7.1 ± 1.7
299.4 ± 34.5
1.7 ± 0.0
0.2 ± 0.1
4.7 ± 0.1
6.0 ± 5.9
0.8 ± 0.0
2.9 ± 1.8
6.2 ± 3.7
5.7 ± 2.8
24.4 ± 13.8
5.8 ± 0.5
32.9 ± 0.1
93.3 ± 19.3
2.6 ± 0.1
0.4 ± 0.0
8.1 ± 6.6
1.3 ± 0.1
ꢀ
Ret
EGFR
EGFR/T790M/L858R
DDR2
ABL
6.6 ± 1.8
5.3 ± 4.4
3.9 ± 1.8
4.3 ± 0.5
5.0 ± 4.9
6.1 ± 2.5
2.4 ± 0.1
6.0 ± 3.7
ErbB4
cꢀSrc
6.4 ± 4.3
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 70
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
VEGFRꢀ1
IR
>10
>10
1ꢀ10
>10
887.6 ±
163.5
>1000
>100
4
31.2 ±
91.6 ± 21.2
417.5 ± 112.6
183.2
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
IGF1R
PDGFRꢀα
PDGFRꢀβ
Mer
>100
>100
>1000
>100
>1000
>100
>100
1ꢀ10
>10
1ꢀ10
30.9 ± 13.7
1ꢀ10
4.4 ± 1.8
1ꢀ10
36.9 ± 20.8
>10
7.1 ± 2.5
1ꢀ10
ErbB2
ALK
231.1 ± 14.3
93.9 ± 27.7
101.1 ± 29.3 155.5 ± 10.4
2
99.5 ±
LTK
45.3 ± 15.3
122.9 ± 52.4
>1000
68.5 ± 11.3
>1000
123.4
2
72.9 ±
ROS1
168.5 ± 87.2
101.3
EPHꢀA2
c ꢀKit
Erk2
>10
>100
>10
>10
>10
>100
>10
>10
5
74.7 ±
120.1
960.2 ± 87.4
>1000
>1000
>1000
>1000
>1000
>1000
>1000
196.1 ± 43.7
>1000
>1000
>1000
>1000
>1000
24
>1000
>1000
>1000
>1000
>1000
>1000
>1000
cꢀMet
AXL
>1000
>1000
>1000
>1000
>1000
>1000
Tyro3
CDK1
CDK4
CDK6
ACS Paragon Plus Environment

Page 25 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
The IC50 values are shown as the mean ± SD (nM) from two separate experiments.
Compound 3m blocks protein kinase phosphorylation and downstream signaling
in cells.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We then investigated the intracellular targeting activity of compound 3m. Both
naturally and genetically kinase expressing cell lines were tested (Figure 4). Firstly,
the effects of compound 3m on the FGFR phosphorylation and downstream signaling
was investigated in representative FGFRꢀaddicted cancer cells. Compound 3m
inhibited the phosphorylation of FGFRs and FGFR substrate 2α (FRS2α), the key
adaptor protein largely specific to FGFR, in a doseꢀdependent manner in the tested
cell lines. The phosphorylation of PLCγ and ERK, key downstream molecules of
40
FGFR, was also inhibited upon compound 3m treatment (Figures 4Aꢀ4C). Similar
results were observed in DDR2, RET, and KDRꢀdriven NCIꢀH2286,
Ba/F3/CCDC6ꢀRET Ba/F3/TELꢀKDR cells, respectively (Figures 4Dꢀ4F). Together,
compound 3m exhibited an effective inhibition of FGFR, DDR2, RET, and KDR
activation and their signaling.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 70
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. Compound 3m blocks tyrosine kinase phosphorylation and downstream
signaling in cells. Cells including NCIꢀH1581(A, left), KG1(A, right), KATOIII(B),
UMUC14(C), NCIꢀH2286(D), Ba/F3/CCDC6ꢀRET(E), Ba/F3/TELꢀKDR (F) treated
with 3m for 2 h at the indicated concentrations were lysed and subjected to Western
blot analysis.
Compound 3m Impairs Cancer Cell Proliferation
To further detect the antiꢀproliferative effects of compound 3m targeting the indicated
protein kinases, we extended to a panel of human cancer cell and mode cell lines that
harboring frequently occurring oncogenic forms of FGFRs, DDR2, EGFR, RET, and
KDR (Table 5). As expected, compound 3m remarkably inhibited cell proliferation of
FGFRꢀaddicted cancer cells, including NCIꢀH1581, KG1, KATOIII, SNU16,
26
ACS Paragon Plus Environment

Page 27 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
UMUC14, and RT112, with IC50 values of 391.2, 14.5, 155.7, 24.8, 128.1, and 122.5
nM, respectively. Moreover, compound 3m significantly inhibited the proliferation of
HCC827 and PCꢀ9 cells, which feature EGFRꢀdependent cell growth, while
compound 3m inhibited the proliferation of H1975 cells with IC value of 447.6 nM.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50
Similar observation was also obtained in DDR2, RET, and KDRꢀconstitutively
activated NCIꢀH2286, Ba/F3/CCDC6ꢀRET and Ba/F3/TELꢀKDR cells. Our results
indicated that 3m inhibits cancer cell proliferation by targeting multiꢀkinase.
a
Table 5. Effects of 3m on cancer cell proliferation.
Cell Line
NCIꢀH1581
KG1
gene alteration
FGFR1 amplification
FGFR1OP2ꢀFGFR1
FGFR2 amplification
3m (IC50, nM)
391.2 ± 14.5
14.5 ± 4.5
16 (IC50, nM)
62.6 ± 5.7
16.1 ± 0.8
KATOIII
155.7 ± 21.2
14.1 ± 2.5
SNU16
FGFR2 amplification
FGFR3 mutation
24.8 ± 0.6
10.5 ± 2.0
27.1 ± 0.1
UMUC14
128.1 ± 0.3
RT112
HCC827
PCꢀ9
FGFR3 amplification
EGFR Exon 19 del
EGFR L858R
122.5 ± 4.1
0.4 ± 0.0
27.0 ± 1.8
/
/
/
/
/
/
5.3 ± 0.4
H1975
EGFR T790M/L858R
DDR2 I638F mutation
447.6 ± 45.7
317.2 ± 22.1
315.7 ± 1.7
9.8 ± 1.3
NCIꢀH2286
Ba/F3/CCDC6ꢀRET CCDC6ꢀRET fusion
Ba/F3/TELꢀKDR TELꢀKDR fusion
a
The IC50 values are shown as the mean ± SD (nM) from two separate experiments.
PK Profiles of Compound 3m
Pharmacokinetic (PK) profile of the candidate 3m was assessed in Sprague−Dawley
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 70
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
(
SD) rats and CDꢀ1 mice (Table 6). Compound 3m proved to be suitable for oral
administration with high area under the curve (AUC0ꢀ∞ = 18705.5 ng/mL*h), long
halfꢀlife (t1/2 = 8.0 h), and good oral bioavailability (F = 60%) in SD rats. Compound
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
m
displayed a lower halfꢀlife of 2.72 h and AUC0ꢀt of 2402 ng·h/mL at 20 mg/kg oral
administration in CDꢀ1 mice than in SD rats. The binding ratio of plasma protein to
m was 99.56%. Accordingly, the free Cmax for 3m in the plasma at the dose of 20
3
mg/kg and 5 mg/kg were calculated as 10.2 and 1.1 nM, respectively.
Table 6. PK profiles of compound 3m in SD rats and CDꢀ1 mice.
Dose
mg/kg)
20
T
max
C
max
AUC0ꢀt
ng/mL*h
16629.1
13748.4
2402
AUC0ꢀ∞
ng/mL*h
18705.5
13756.9
2409
MRT
t
1/2
CLz
F
%
Species
Rats
Compd.
(
h
ng/mL
2077.4
6535.2
961
h
h
L/h/kg
3
m (p.o.)
1
6.72
3.41
3.59
3.74
8.0
/
60%
3m (i.v.)
10
0.25
0.42
1.50
2.39
2.72
2.50
0.727
3m (p.o.)
20
/
/
Mice
3
m (p.o.)
5
106
289
290
28
ACS Paragon Plus Environment

Page 29 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vivo Antitumor Efficacy
Classic FGFRꢀdependent H1581 lung cancer cell line derived xenograft was used for
the in vivo efficacy evaluation, nude mice bearing NCIꢀH1581 tumors were
randomized and treated with 3m at doses of 5 or 20 mg/kg for 14 consecutive days.
As shown in Figure 5, compound 3m could suppress tumor growth in a
doseꢀdependent manner, with the tumor growth inhibition ratio (TGI) of 37.8% and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
6.1% at the doses of 5 mg/kg and 20 mg/kg, respectively, indicating compound 3m
elicited a marked antitumor efficacy in NCIꢀH1581 tumor model. However, since
adverse reactions such as diarrhea, emaciation and body weight loss were observed in
mice dosed at 20 mg/kg, higher dose treatment was not further carried out in this
study.
29
ACS Paragon Plus Environment