Discovery of 3-(5′-Substituted)-Benzimidazole-5-(1-(3,5-dichloropyridin-4-yl)ethoxy)-1H-indazoles as Potent Fibroblast Growth Factor Receptor Inhibitors: Design, Synthesis, and Biological Evaluation

Article abstract of DOI:10.1021/acs.jmedchem.6b00056

Fibroblast growth factor receptor (FGFR) represents an attractive oncology target for cancer therapy in view of its critical role in promoting cancer formation and progression, as well as causing resistance to approved therapies. In this article, we describe the identification of the potent pan-FGFR inhibitor (R)-21c (FGFR1-4 IC₅₀ values of 0.9, 2.0, 2.0, and 6.1 nM, respectively). Compound (R)-21c exhibited excellent in vitro inhibitory activity against a panel of FGFR-amplified cell lines. Western blot analysis demonstrated that (R)-21c suppressed FGF/FGFR and downstream signaling pathways at nanomolar concentrations. Moreover, (R)-21c provided nearly complete inhibition of tumor growth (96.9% TGI) in NCI-H1581 (FGFR1-amplified) xenograft mice model at the dose of 10 mg/kg/qd via oral administration.

Full text of DOI:10.1021/acs.jmedchem.6b00056

Subscriber access provided by - Access paid by the | UCSB Libraries
Article
Discovery of 3-(5'-Substituted)-benzimidazole-5-(1-(3,5-dichloropyridin-4-
yl)ethoxy)-1H-indazoles as Potent Fibroblast Growth Factor
Receptor Inhibitors: Design, Synthesis and Biological Evaluation
Wei Yan, Xinyi Wang, Yang Dai, Bin Zhao, Xinying Yang, Jun Fan, Yinglei Gao,
Fanwang Meng, Yuming Wang, Cheng Luo, Jing Ai, Mei-Yu Geng, and Wenhu Duan
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00056 • Publication Date (Web): 27 Jun 2016
Downloaded from http://pubs.acs.org on June 27, 2016
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 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 71
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
Discovery of 3ꢀ(5'ꢀSubstituted)ꢀbenzimidazoleꢀ5ꢀ(1ꢀ(3,5ꢀ
dichloropyridinꢀ4ꢀyl)ethoxy)ꢀ1Hꢀindazoles as Potent
Fibroblast Growth Factor Receptor Inhibitors: Design,
Synthesis and Biological Evaluation
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
,#
2,6,#
2
3
2
3
2
Wei Yan, Xinyi Wang,
Yang Dai, Bin Zhao, Xinying Yang, Jun Fan, Yinglei Gao, Fanwang
4,5
3
5
2,*
2,*
1,3,*
Meng, Yuming Wang, Cheng Luo, Jing Ai, Meiyu Geng, and Wenhu Duan
1
School of Pharmacy, East China University of Science & Technology, 130 Mei Long Road, Shanghai
00237, P.R. China
2
2
Division of Antiꢀtumor Pharmacology, State Key Laboratory of Drug Research, Shanghai Institute of
Materia Medica, Chinese Academy of Sciences, 555 Zu Chong Zhi Road, Shanghai 201203, P.R. China
3
Department of Medicinal Chemistry, Shanghai Institute of Materia Medica, Chinese Academy of
Sciences, 555 Zu Chong Zhi Road, Shanghai 201203, P.R. China
4
Department of Chemistry, College of Sciences, Shanghai University, 99 Shang Da Road, Shanghai
200444, P.R. China
5
Drug Discovery & Design Center, State Key Laboratory of Drug Research, Shanghai Institute of
Materia Medica, Chinese Academy of Sciences, 555 Zu Chong Zhi Road, Shanghai 201203, P.R. China
1
ACS Paragon Plus Environment

Page 3 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
6
University of Chinese Academy of Sciences, Beijing, 100049, P.R. China
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
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
Fibroblast growth factor receptor (FGFR) represents an attractive oncology target for cancer therapy
in view of its critical role in promoting cancer formation and progression, as well as causing resistance
to approved therapies. In this paper, we describe the identification of the potent panꢀFGFR inhibitor (R)-
21c (FGFR1–4 IC values of 0.9, 2.0, 2.0, and 6.1 nM, respectively). Compound (R)-21c exhibited
5
0
excellent in vitro inhibitory activity against a panel of FGFRꢀamplified cell lines. Western blot analysis
demonstrated that (R)-21c suppressed FGF/FGFR and downstream signaling pathways at nanomolar
concentrations. Moreover, (R)-21c provided nearly complete inhibition of tumor growth (96.9% TGI) in
NCIꢀH1581 (FGFR1ꢀamplified) xenograft mice model at the dose of 10 mg/kg/qd via oral
administration.
INTRODUCTION
Fibroblast growth factor receptors (FGFRs) are a family of four receptor tyrosine kinases (RTKs), and
each of the four FGFRs (FGFR1–4) consists of three extracellular immunoglobulin domains, a single
1
transmembrane domain, and a cytoplasmic tyrosine kinase domain. Upon binding to fibroblast growth
factors, the receptor undergoes dimerization and autophosphorylation of tyrosine residues on
cytoplasmic tail, resulting in activation of downstream signaling including MEKꢀERK, PLCγ, and PI3Kꢀ
Akt. Physiologically, FGFR pathway plays important roles in embryogenesis, tissue repair and wound
2
healing. FGFR is an oncogenic driver of a broad panel of malignant solid tumors: FGFR1 amplification
is identified in about 20% of squamous nonꢀsmall cell lung carcinoma and about 10% of hormone
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 71
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
receptorꢀpositive breast cancers; FGFR2 is also amplified in about 10% of gastric cancer and 4% of
tripleꢀnegative breast cancers; mutations of FGFR3 are found in about half of bladder cancers which are
strongly related to lowꢀgrade tumors; activating mutations of FGFR4 (affecting the kinase domain) are
found in 6% to 8% of patients with rhabdomyosarcoma, and overexpression of FGFR4 was found in
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
ꢀ8
50% of hepatocellular carcinomas. Moreover, aberrant FGFR activation has been associated with poor
prognosis and metastatic progression. In addition, overꢀactivation of FGFR signaling is responsible for
9
ꢀ11
de novo and acquired resistance to anticancer therapy.
Therefore, the blockade of FGF/FGFR
signaling is considered as an attractive approach for the treatment of cancers.
Figure 1. Structures of representative FGFR inhibitors.
The early examples of FGFR inhibitors were nonꢀselective. Owing to the high degree of homology of
FGFRs with VEGFRs (vascular endothelia growth factor receptors), many compounds developed as
12
FGFR inhibitors also inhibit VEGFR family members at similar potencies, such as 1 (TKI258),
2
3
ACS Paragon Plus Environment

Page 5 of 71
Journal of Medicinal Chemistry
1
2
3
13
14
(
BIBFꢀ1120), and 3 (AZD2171) (Figure 1). Although therapies based on inhibiting VEGF/VEGFR
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
5
signaling pathway have been approved for the treatment of cancers, e.g. 10 (sunitinib), 11
1
6
17
18
(
sorafenib), 12 (pazopanib), and 13 (vandetanib) (Figure 2), undesired side effects such as
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
19ꢀ21
hypertension and bleeding have been reported.
Recently, medicinal chemists tend to develop more
selective FGFR inhibitors, and several orally available FGFR selective inhibitors have been reported to
22
23
24
enter clinical trials, for example, 4 (NVPꢀBGJ398), 5 (AZD4547), 6 (CH5183284), and 7
2
5
(
LY2874455) (Figure 1). The 3,5ꢀdimethoxyꢀphenyl moiety of 4 and 5 occupies the same hydrophobic
pocket and forms a hydrogen bond with Asp641, which is identified by 8 (PD173074) to contribute to
26ꢀ27
24
their selectivity for FGFR1;
similar interactions are also observed between 6 and FGFR1.
Therefore, it is possible that a common single mutation would confer resistance to these inhibitors,
which is classified as a category of genetic alteration contributed to the resistance of FGFR inhibitors
besides other categories of compensatory pathway activation and epithelialꢀtoꢀmesenchymal transition,
2
4, 28ꢀ30
as revealed by limited studies.
2
5
As illustrated in the binding model of 7 in the ATP binding site of FGFR3, the indazole core binds in
the ATP binding pocket and forms two hydrogen bonds with the carbonyl of Glu230 and the NH of
Ala232; the vinyl pyrazole fragment extends toward solvent, making important hydrophobic
interactions.
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 71
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. Structures of representative approved VEGFR2 inhibitors.
Additionally, a hydrogen bond between the pyridyl nitrogen and Asn236 in hydrophobic pocket is
observed; this additional hydrogen bond increases FGFR binding affinity. We envisioned that the
interaction between the pyridine fragment and the hydrophobic pocket might overcome the resistance of
the former three FGFR selective inhibitors (4, 5, and 6) caused by the single mutation as discussed
above.
3
1
McBride et al. reported compound 9 as a dual FGFR1/VEGFR2 inhibitor (Figure 1); this compound
showed moderate in vitro and in vivo antitumor efficiency. To understand the ligandꢀprotein
interactions, we developed a binding model of compound 9 in the ATP binding site of FGFR1 with a
2
4
known crystal structure of the FGFR1 kinase domain (PDB ID: 3WJ6). As shown in Figure 3,
compound 9 binds to the hinge region of the kinase and forms two hydrogen bonds with the carbonyl of
Glu562 and the NH of Ala564; an additional hydrogen bond occurs between the carbonyl of Ala564 and
N(1)ꢀH of benzimidazole moiety; the benzimidazole moiety extends from the indazole 3ꢀposition toward
solvent exposed region. In view of the high degree of homology of FGFR1 with FGFR3, it was
reasonable to assume that 7 and 9 had similar binding patterns with FGFR in the hinge region and the
5
ACS Paragon Plus Environment

Page 7 of 71
Journal of Medicinal Chemistry
1
2
3
solvent exposed region. In view of the importance of the interactions between the pyridine fragment and
the surrounded hydrophobic pocket, we envisioned that the introduction of a pyridylalkoxy moiety to the
indazole benzimidazole scaffold might form a novel series of FGFR inhibitors with improved potency
and selectivity. To test our hypothesis, we designed and synthesized a series of novel 3ꢀbenzimidazoleꢀ
5ꢀpyridylalkoxyꢀ1Hꢀindazole derivatives (Figure 4), and these new derivatives displayed excellent in
vitro and in vivo antitumor activities. Herein, we report the detailed synthesis and biological evaluation
of these compounds.
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 3. Proposed binding mode of compound 9 to FGFR1 kinase domain. The atoms of compound 9
are colored as follows: carbon orange, oxygen red, nitrogen blue, fluorine light blue, and hydrogen
white. The docked protein structure of FGFR1 is shown in the cartoon and key residues are colored as
follows: carbon green, oxygen red, nitrogen blue, and hydrogen white. The hydrogen bonds are colored
in black.
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 71
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. Design of our compounds.
RESULTS AND DISCUSSION
Lead Optimization and Identification of Compound (R)-21c.
To evaluate the aforementioned hypothesis, we prepared compound 14a as the proofꢀofꢀconcept
compound, which possessed moderate enzymatic activity (Table 1). In order to get a putative binding
mode, a molecular docking simulation was performed. As depicted in Figure 5, 14a maintained all the
critical interactions observed in compounds 7 and 9: the indazole core was anchored to the hinge region
of kinase domain via two highly conserved hydrogen bonds with the carbonyl of Glu562 (1.8 Å) and the
amino group of Ala564 (2.1 Å); an additional hydrogen bond occurred between the carbonyl of Ala564
and N(1)ꢀH of benzimidazole moiety (2.3 Å); a hydrogen bond between pyridyl nitrogen atom and
amino group of Asn568 (2.3 Å) was formed as expected, which reinforced this binding conformation.
7
ACS Paragon Plus Environment

Page 9 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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 5. Proposed binding mode of compound 14a to FGFR1 kinase domain. The atoms of compound
1
4a are colored as follows: carbon orange, oxygen red, nitrogen blue, and hydrogen white. The docked
protein structure of FGFR1 is shown in the cartoon and key residues are colored as follows: carbon
green, oxygen red, nitrogen blue, and hydrogen white. The hydrogen bonds are colored in black.
For initial structureꢀactivity relationship exploration, 14a served as a benchmark to analyze both
enzymatic and cellular potency contributions from different pyridine ring substitution patterns, and
compound 5 was used as the positive control. Despite its moderate potency against FGFR1, 14a
displayed a weak inhibitory effect in NCIꢀH1581 proliferation assay (IC of 5016.9 nM). Incorporation
5
0
of a chlorine atom at the 3ꢀposition of the pyridine ring provided 14b, which displayed measurable
increases in potencies; when an additional chlorine was introduced to the 5ꢀposition, the 3,5ꢀdichloro
derivative 14c exhibited a 31ꢀfold increase compared to 14a in cellular potency with an IC of 158.7
5
0
nM; further introduction of an αꢀmethyl into 14c afforded 15a, which demonstrated a further elevenꢀfold
enhancement of cellular potency compared to 14c (IC50 of 14.1 nM) and 7ꢀ and 54ꢀfold increases
compared with
Table 1. Structureꢀactivity Relationship of Substitutions on Racemic 5ꢀPyridylalkoxy Group
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 71
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
FGFR1
NCIꢀH1581 Cell
Compound
Pyridyl
R
a
a
(
IC , nM)
(IC , nM)
5
0
50
1
4a
4b
H
H
20.8 ± 2.3
13.6 ± 1.7
5016.9 ± 450.0
3071.1 ± 252.2
1
14c
H
6.6 ± 0.8
158.7 ± 3.0
1
4d
H
H
1.7 ± 0.1
4.1 ± 0.8
1859.1 ± 156.4
1232.4 ± 121.7
1
4e
1
5a
5b
Me
Et
2.7 ± 0.8
5.7 ± 2.3
14.1 ± 2.1
1
380.1 ± 21.8
9
ACS Paragon Plus Environment

Page 11 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
15c
Me
22.4 ± 4.2
30.8 ± 4.2
9
7
5
–
–
–
–
–
–
20.3 ± 1.3
0.3 ± 0.1
0.6 ± 0.2
749.1 ± 47.2
< 0.5
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
92.1 ± 2.8
a
Values are the mean ± SD of two independent assays.
compound 9 in enzymatic and cellular potencies, respectively. The increases in potency of compounds
4b, 14c, and 15a might be explained by the increased lipophilicity by inclusion of the chlorine atom
1
3
2
and the αꢀmethyl. The ethyl analogue 15b retained enzymatic potency but turned out to be 27ꢀfold less
active than 15a in the NCIꢀH1581 proliferation assay. Next, we assessed the importance of the position
of the pyridine nitrogen by synthesizing compounds 14d and 14e. Interestingly, when pyridine nitrogen
was moved to the 2ꢀposition, the resulting compound 14d was more potent than the 1ꢀ or 3ꢀpyridyl
analogues (14a and 14e, respectively). These results implied that the 2ꢀpyridyl isomer interacted in a
more favorable manner with the protein; however, when 3,5ꢀdichloro and αꢀmethyl were introduced to
14d, the resulting compound 15c was eightꢀ and twoꢀfold less active than 15a in enzymatic and cellular
potencies, respectively. Therefore, 15a with the 3,5ꢀdichloro and αꢀmethyl substitution pattern was
chosen as the compound for further optimization.
Table 2. Structureꢀactivity Relationship of 3ꢀHeterocycle Substitutions on Racemic 5ꢀ(1ꢀ(3,5ꢀ
Dichloropyridinꢀ4ꢀyl)ethoxy)ꢀ1Hꢀindazole
Compound
R
FGFR1
NCIꢀH1581 Cell
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 71
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
a
(
IC , nM)
(IC , nM)
50
5
0
1
6a
1.8 ± 0.1
5.1 ± 0.6
402.7 ± 13.6
3035.7 ± 593.3
> 10000
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
16b
1
6c
228.2 ± 27.1
2.4 ± 0.6
17
817.2 ± 13.5
a
Values are the mean ± SD of two independent assays.
We next turned our attention to the benzimidazole moiety by replacing it with three other heterocyclic
bioisosteres. The substituent at the 5’ꢀposition was removed for rapid evaluation of the impact of this
replacement (Table 2), which led to compound 16a with modest cellular potency. Compared with 16a,
benzoxazole analogue 16b displayed threeꢀ and eightꢀfold decreases in enzymatic and cellular potencies,
respectively. The decrease in potency of 16b could be explained by the absence of the hydrogen bond
formed between benzimidazole NH and Ala564. Replacement of the benzimidazole moiety by
benzothiazole completely eroded the cellular activity (16c). Indole analogue 17 retained FGFR activity
but was twoꢀfold less active than 16a in NCIꢀH1581 proliferation assay. These results revealed that the
benzimidazole moiety was optimal for FGFR activity.
Table 3. Structureꢀactivity Relationship of Substitutions on Benzimidazole
11
ACS Paragon Plus Environment

Page 13 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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
FGFR1
NCIꢀH1581 Cell
Compound
4’
5’
a
a
(
IC , nM)
(IC , nM)
5
0
50
1
6a
H
H
H
1.8 ± 0.1
402.7 ± 13.6
> 10000
18a
18b
19a
Me
16.2 ± 2.4
> 1000
b
H
NT
H
H
H
F
2.1 ± 0.4
2.0 ± 0.1
2.8 ± 0.6
582.3 ± 66.4
922.3 ± 94.6
315.3 ± 11.7
1
9b
Me
19c
OMe
1
9d
H
H
2.1 ± 0.2
1.0 ± 0.1
5.5 ± 2.4
1.6 ± 1.2
20
2
2
1a
H
0.8 ± 0.1
10.3 ± 0.65
1b
H
H
1.4 ± 0.1
1.0 ± 0.2
1.9 ± 0.5
1.0 ± 0.2
21c
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 71
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
1d
H
3.6 ± 0.9
5.9 ± 2.9
21e
H
H
H
H
0.2 ± 0.0
7.6 ± 1.4
0.4 ± 0.2
0.3 ± 0.1
2.2 ± 0.5
8.1 ± 1.1
0.8 ± 0.1
1.1 ± 0.2
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
2
1f
1
9e
2
1g
1h
2
H
H
0.6 ± 0.1
1.5 ± 0.1
1.2 ± 0.6
3.3 ± 1.1
2
2
a
b
Values are the mean ± SD of two independent assays.
Not tested.
Finally, systematic structural optimization of benzimidazole of 16a revealed that substituents on the
benzimidazole dramatically impacted FGFR inhibition. As summarized in Table 3, the 4’ꢀmethyl
analogue 18a showed a nineꢀfold drop in enzymatic potency compared with 16a and had a cellular IC50
greater than 10000 nM; moreover, the attachment of a 4’ꢀpiperidyl group to the benzimidazole moiety
(
18b) resulted in a significant loss of activity (FGFR1 kinase IC greater than 1000 nM). These results
50
revealed that substitution at 4’ꢀposition of the benzimidazole moiety was not tolerable. On the contrary,
inclusion of substituents at the 5’ꢀposition of the benzimidazole had favorable impacts: derivatives
bearing hydrophobic groups (such as fluoro, methyl, and methoxyl) retained enzymatic and cellular
potencies (compounds 19a–c); moreover, when the hydrophilic methylamino was attached to the 5’ꢀ
position of benzimidazole, the resulting derivative 19d exhibited a cellular potency with an IC less
5
0
13
ACS Paragon Plus Environment

Page 15 of 71
Journal of Medicinal Chemistry
1
2
3
than 10 nM in the NCIꢀH1581 proliferation assay. Therefore, we introduced a number of amines and
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
alcohol to the 5’ꢀposition of the benzimidazole moiety. In general, sixꢀmembered ring analogues (20 and
2
1b–e) had comparable potencies; but 1ꢀethylpiperazine analogue 21d was nearly fourꢀ and sixꢀfold less
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
active than methyl analogue 21c in enzymatic and cellular potencies, respectively; the
piperidinylpiperidine analogue 21a had a FGFR1 IC of 0.8 nM, but suffered a twelveꢀfold enzymeꢀtoꢀ
5
0
cell shift, resulting in a cellular IC of 10 nM; the homopiperazine analogue 21f was eightꢀfold less
50
active compared with 21c. The relatively low potency of 21a, 21d, and 21f indicated that the larger ring
substitution was less tolerable. Compounds bearing chain substituents (19e, 21g–h, and 22) also turned
out to be very potent, and they had comparable potencies to the cyclic amines. As discussed above, a
range of functional groups were tolerated at the 5’ꢀposition of the benzimidazole moiety, potentially
owing to this region extending out of the kinase domain of FGFR1 into a solvent exposed region.
Because of the undesired sideꢀeffects caused by inhibiting VEGFR2 signaling pathway, five
compounds with excellent enzymatic and cellular potencies were selected to be evaluated for their
VEGFR2 kinase inhibition. As listed in Table 4, all tested compounds displayed sixꢀ to eightꢀfold
selectivity for FGFR1 over VEGFR2, while compound 9 was nearly equipotent against FGFR1 and
VEGFR2. This result
Table 4. FGFR Selectivity of Selected Compounds
FGFR1
VEGFR2
VEGFR2/FGFR1
Compound
a
a
(
IC , nM)
(IC , nM)
ratio
7.8
5
0
50
2
0
1.0 ± 0.1
1.4 ± 0.1
1.0 ± 0.2
0.4 ± 0.2
7.8 ± 1.3
9.0 ± 1.0
8.0 ± 0.5
2.5 ± 0.5
21b
6.4
21c
19e
8.0
6.2
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 71
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
2
1.5 ± 0.1
20.3 ± 1.3
0.3 ± 0.1
0.6 ± 0.2
12.6 ± 0.9
16.4 ± 0.9
5.8 ± 0.4
8.4
0.8
9
7
5
14.5
45.0
27.0 ± 3.4
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
Values are the mean ± SD of two independent assays.
revealed that incorporation of 5ꢀ(1ꢀ(3,5ꢀdichloropyridinꢀ4ꢀyl)ethoxyl fragment at the 5ꢀposition of the
indazole enhanced FGFR inhibitory activity and selectivity over VEGFR2. On the basis of their
enzymatic and cellular inhibitory activities, compounds 21c and 22 were chosen for further evaluation.
Compounds 21c and 22 were racemates; the enantiomers of 21c and 22 were resolved to evaluate the
relative FGFR activity. As illustrated in Figure 6, the R enantiomers were significantly more potent; the
result was also consistent with the specific binding revealed in a docking model of compound (R)-21c to
FGFR1 (Figure 9, vide infra).
Figure 6. Structures and activities of the enantiomers of 21c and 22.
Pharmacokinetic Profiles of Compounds (R)-21c and (R)-22.
15
ACS Paragon Plus Environment

Page 17 of 71
Journal of Medicinal Chemistry
1
2
3
Compound (R)-21c or (R)ꢀ22 was administrated to male SpragueꢀDawley (SD) rats: either as a 1
mg/kg intravenous or orally by gavage at a 5 mg/kg dose and the relevant pharmacokinetic parameters
were summarized in Table 5. Both compounds exhibited moderate and similar plasma clearance (2.0 and
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
1
.7 L/h/kg, respectively) and oral exposed (798.6 and 663.9 ng·h/mL, respectively). Compound (R)ꢀ22
had a long halfꢀtime (7.3 h) but also had a very large volume of distribution (17.6 L/kg) and relatively
low oral bioavailability (22%); while (R)-21c had a moderate halfꢀtime (2.0 h), volume of distribution
(5.8 L/kg), and oral bioavailability (33%). Taking all their data into consideration, we selected
compound (R)-21c to further evaluate its in vitro and in vivo antitumor activities.
a
Table 5. Pharmacokinetic Profile of (R)-21c and (R)ꢀ22 in Rats
Dose
Cmax
Tmax T1/2
AUC0→∞
(ng·h/mL)
485.3
CL
Vss
F
Compound
(
mg/kg) (ng/mL)
(h)
–
(h)
2.0
–
(L/h/kg) (L/kg) (%)
1
(iv)
300.3
97.2
2.0
–
5.8
–
–
33
–
(R)-21c
5 (po)
1 (iv)
5 (po)
4.0
–
798.6
735.4
195.3
7.3
–
582.3
1.7
–
17.6
–
(R)ꢀ22
0.8
663.9
22
a
Cmax, maximum concentration; Tmax, time of maximum concentration; T , elimination halfꢀlife;
1/2
AUC0→∞, area under the plasma concentration time curve; CL, clearance; V , volume of distribution; F,
ss
oral bioavailability.
Kinase Inhibition Profile of (R)-21c.
By use of an ELISAꢀbased enzymatic assay, (R)-21c was distinguished for its high potency against
FGFR1 kinase activity. Then we were prompted to investigate its activity against a panel of tyrosine
kinases. Compound (R)-21c showed almost equal potency against FGFR2 and FGFR3, a little weaker
against FGFR4, with IC values of 2.0, 2.0, and 6.1 nM, respectively, indicating (R)-21c was a panꢀ
50
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 71
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
FGFR inhibitor (Table 6). Compared with compound 5 (FGFR4 IC of 165 nM, literature reported
5
0
2
3
data ), (R)-21c was significantly more potent in inhibiting FGFR4, indicating its potential use for the
8
, 33
treatment of FGFR4ꢀdriven cancer types, for example, hepatocellular carcinoma (HCC).
Furthermore,
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
(R)-21c was extended to be profiled against a panel of 54 human kinases, including FGFR homologous
PDGFR kinase family (PDGFRα, PDGFRβ). Compared to its high potency against FGFR, (R)-21c
exhibited much weaker potency against these tested kinases (IC >100 nM, or 1000 nM) (Table 6 and
5
0
Table S1), indicating that (R)-21c was a selective FGFR inhibitor.
Table 6. Kinase Inhibition Profile of (R)ꢀ21c
a
b
a
b
Tyrosine kinase
FGFR1
IC (nM)
Tyrosine kinase
PDGFRβ
cꢀMet
IC (nM)
50
50
0.9 ± 0.2
2.0 ± 0.5
2.0 ± 0.3
6.1 ± 0.4
7.5 ± 0.8
> 1000
> 1000
> 100
> 100
> 100
> 100
> 100
> 100
> 100
FGFR2
FGFR3
Flt1
FGFR4
Flt3
VEGFR2
EGFR
RET
cꢀSrc
ErbB2
> 1000
BcrꢀAbl
EPHA2
PDGFRα
781.8 ± 7.6
a
EGFR, epidermal growth factor receptor; PDGFR, platelet derived growth factor receptor; cꢀMet,
mesenchymal epithelial transition growth factor; Flt, fmsꢀlike tyrosine kinase; RET, rearranged during
transfection; cꢀSrc, cellular Src kinase; BcrꢀAbl, breakpoint cluster regionꢀabelson; EPHA,
erythropoietinꢀproducing hepatocellular carcinoma receptor.
b
Values are the mean ± SD of three independent assays.
Inhibition on FGFR Signaling of Compound (R)-21c
17
ACS Paragon Plus Environment

Page 19 of 71
Journal of Medicinal Chemistry
1
2
3
To further evaluate the cellular activity of (R)-21c targeting FGFR kinase, we analyzed its effects on
the phosphorylation of FGFR and its major downstream signaling molecules, PLCγ and Erk. Two
representative cancer cell lines with different mechanisms of FGFR activation, FGFR1ꢀfusion protein
driven human myeloid leukemia cancer cell line KGꢀ1 and FGFR2ꢀamplified dependent human gastric
cancer cell line SNUꢀ16, were chosen. As shown in Figure 7, (R)-21c significantly inhibited the
phosphorylation of FGFR1 in a dose dependent manner in KG1 cells. In line with suppressed FGFR1
phosphorylation, the phosphorylation of PLCγ and Erk, the main downstream molecules of FGFR, was
also strongly inhibited. Similar results were observed in (R)-21c treated SNUꢀ16 cells. These results
suggested that (R)-21c exhibited an effective blockage of FGFRs signaling.
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 7. Compound (R)-21c effectively inhibits the phosphorylation of FGFR and the downstream
effectors PLCγ and Erk in KGꢀ1 and SNUꢀ16 cells. KG1 (A) or SNUꢀ16 (B) cells treated with (R)-21c
for 2 h at the indicated concentrations were lysed and subjected to western blot analysis.
Inhibition on FGFR Signaling-driven cancer cell proliferation of (R)-21c
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 71
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
Increased FGFR signaling promotes cancer cell proliferation. To elucidate the impact of (R)-21c on
FGFR mediated cancer cell proliferation, four FGFRꢀdriven cancer cell lines harboring the frequently
occurring oncogenic forms of FGFRs were chosen: FGFR1 translocated KG1 cells, FGFR1ꢀamplified
NCIꢀH1581 cells, FGFR2ꢀamplified SNU16 cells, FGFR3ꢀamplified RT112 cells. As shown in Table 7,
compound (R)-21c strongly inhibited cell proliferation of KG1, NCIꢀH1581, SNU16, RT112 cells, with
most of the IC values of less than 0.1 nM, which was much more potent than that of compound 5, 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
5
0
1
most advanced panꢀFGFR inhibitor in clinical studies. These findings suggested that (R)ꢀ21c strongly
inhibited FGFRꢀdriven cancer cell proliferation via blocking FGFR signaling.
Table 7. Antiproliferative Activity of (R)-21c on FGFR Addicted Cell Lines
a
IC (nM)
50
Cell lines
(R)-21c
< 0.1
5
KGꢀ1
NCIꢀH1581
SNU16
6.4 ± 1.1
58.8 ± 1.4
13.4 ± 0.3
88.4 ± 10.5
4.2 ± 2.5
< 0.1
RT112
< 0.1
a
Values are the mean ± SD of three independent assays.
In Vivo Antitumor Activity of Compound (R)-21c
To assess the in vivo antitumor efficacy of (R)-21c, NCIꢀH1581 xenograft mouse model, which is
driven by amplification of FGFR1, was used. Compound (R)-21c was orally administrated once daily at
doses of 2 or 10 mg/kg for 21 consecutive days. The results showed that (R)-21c could suppress tumor
19
ACS Paragon Plus Environment

Page 21 of 71
Journal of Medicinal Chemistry
1
2
3
growth in a doseꢀdependent manner, with the tumor growth inhibition rate (TGI) of 44.5% and 96.9% at
the doses of 2 mg/kg and 10 mg/kg, respectively, without significant body loss (Figure 8), indicating
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
(R)-21c elicited a robust antitumor efficacy in a FGFRꢀdependent tumor model. Therefore, (R)-21c was
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 potential FGFR inhibitor candidate for further drug development.
Figure 8. The inhibitory effect of (R)-21c on tumor growth in NCIꢀH1581 xenograft model. Compound
(R)ꢀ21c was administered orally once daily for 21 days. (A) The percentage of tumor growth inhibition
values (TGI) was measured on the final day of the study for the drugꢀtreated mice compared with the
vehicle control mice: (∗) P< 0.05, (∗) P< 0.01, (∗) P< 0.001, versus vehicle group, using Student’s t test.
Results are expressed as the mean ± SEM (n = 6 for inhibitorꢀtreated group, n = 12 for vehicle control
group). (B) Body weight measurements during the treatment.
Molecular Docking Study of Compound (R)-21c to FGFR1
As illustrated in Figure 9, (R)-21c had a similar binding mode with 14a: the indazole core was
anchored to the hinge region of FGFR1 kinase domain via two highly conserved hydrogen bonds with
Glu562 and Ala564; a hydrogen bond between pyridyl nitrogen atom and Asn568 was formed as
expected. The αꢀmethyl in the linkage between pyridine ring and the indazole core had a favored steric
effect: as indicated in Figure 9C, the Rꢀmethyl fit in a pocket surrounded by Asp641, Val492, and
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 71
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
Glu485, thus compelling the pyridine ring to form a strong hydrogen bond with Asn568; on the contrary,
an Sꢀmethyl would have an unfavored steric hindrance effect with Val492 which would destabilize the
active conformation. The benzimidazole moiety was located in a narrow hydrophobic cleft that was
formed by five residues (namely, Leu484, Tyr563, Ala564, Ser565, and Gly567): an additional
hydrogen bond occurred between the carbonyl of Ala564 and N(1)ꢀH of benzimidazole moiety that
reinforced this binding conformation; the 4’ꢀposition of benzimidazole was surrounded by Leu484,
Ser565, and Gly567; while the 5’ꢀsubstitution extended out of this pocket into the solvent exposed
region; this binding information explained why 4’ꢀsubstitution of benzimidazole was not tolerable, but a
range of functional groups were tolerated at the 5’ꢀposition.
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 9. The proposed binding mode of (R)-21c with FGFR1. (A) The atoms of (R)-21c are colored as
follows: carbon orange, oxygen red, nitrogen blue, chlorine green, and hydrogen white. The docked
protein structure of FGFR1 is shown in the cartoon and key residues are colored as follows: carbon
green, oxygen red, nitrogen blue, and hydrogen white. The hydrogen bonds are colored in black. (B) The
docked protein structure of FGFR1 is shown in surface. The surfaces of Gly485 and Glu486 are omitted
for clarity. (C) The Rꢀmethyl fit in a pocket surrounded by three residues (Asp641, Val492, and Glu485).
CHEMISTRY
21
ACS Paragon Plus Environment

Page 23 of 71
Journal of Medicinal Chemistry
1
2
3
a
Scheme 1
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
a
Reagents and conditions: (a) 5ꢀhydroxyindole, Cs CO , DMF, 70 °C; (b) NaNO , HCl, H O/dioxane,
2
3
2
2
0 °C to rt; (c) Et N, piperidine, NMP, 120 °C; (d) H , 10% Pd/C, MeOH, rt; (e) S , DMF, 90 °C.
3
2
8
As outlined in Schemes 1–2, benzyl chlorides 23 or methane sulfonates 31 were reacted with 5ꢀ
hydroxyꢀ1Hꢀindole to provide ethers 24 or 32, which were transformed to 3ꢀcarbaldehydeꢀ1Hꢀindazoles
3
4
2
5 or 33 by treatment of dilute nitrous acid. Nucleophilic displacement of the Cꢀ5 fluorine of 26a with
piperidine afforded 27. Reduction of 27 with hydrogen afforded diamine 28 which should be protected
in argon atmosphere during workup and reacted immediately after workup with aldehydes 25 or 33 to
afford compounds 14 and 15.
Compounds 16a and 16c were prepared by analogous methods (Scheme 3); while the stronger oxidant
DDQ rather than sulfur was used in the final cycloꢀcondensation step in the preparation of 16b. The
indoleꢀindazole derivative 17 was synthesized via palladiumꢀcatalyzed crossꢀcoupling reaction of iodide
3
5ꢀ36
3
5 and boronic acid 36; compounds 35 and 36 were prepared according to literature procedures.
a
Scheme 2
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 71
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
a
Reagents and conditions: (a) for 30a and 30c, LDA, acetaldehyde, THF, −78 °C; for 30b, LDA,
propionaldehyde, THF, −78 °C; (b) MsCl, Et N, CH Cl , 0 °C to rt; (c) 5ꢀhydroxyindole, Cs CO , DMF,
3
2
2
2
3
7
0 °C; (d) NaNO , HCl, H O/dioxane, 0 °C to rt; (e) 28, S , DMF, 90 °C.
2 2 8
a
Scheme 3
a
Reagents and conditions: (a) for 16a and 16c, S , DMF, 90 °C; for 16b, (i) DMF, 90 °C; (ii) DDQ,
8
CHCl , reflux; (b) (i) Cs CO , Pd(dppf) Cl , dioxane, 90 °C; (ii) 4 N HCl in MeOH.
3
2
3
2
2
Compounds 18–22 were prepared via a similar route (Scheme 4). An array of 2ꢀnitroaniline
intermediates (40) were prepared starting from 26a by displacement of fluorine with commercial
available amines (39). Acylation of amine 19d with bromoacetyl chloride followed by displacement of
23
ACS Paragon Plus Environment

Page 25 of 71
Journal of Medicinal Chemistry
1
2
3
the bromine with dimethylamine furnished amide 19e. Intermediates 38 and 41 were prepared according
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
7ꢀ38
to literature procedures.
a
Scheme 4
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
Reagents and conditions: (a) (i) H , 10% Pd/C, MeOH, rt; (ii) 33a, S , DMF, 90 °C; (b) for 26d,
methylammonium chloride, DIPEA, NMP, 140 °C; for 40, DIPEA, NMP, 120 °C; (c) (i) 2ꢀbromoacetyl
chloride, CH Cl , rt; (ii) dimethylamine, THF, rt.
2
8
2
2
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 71
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 enantiomers of 21c and 22 were resolved to evaluate the relative FGFR activities. The racemic
alcohol 30a was reacted with chiral acid 42 via kinetic resolution reaction (twice) to give (S)ꢀ30a with
3
9
an enantiomeric excess (ee) of 98% (Scheme 5). Mitsunobu reaction of (S)ꢀ30a with acetic acid
followed by hydrolysis with sodium hydroxide afforded (R)ꢀ30a in nearly complete configuration
inversion (ee 97%). Compound (R)-21c was prepared starting from (S)ꢀ30a in a similar manner as
described for the racemic 21c (Scheme 6). It was noted that partial racemization was observed in the SN2
ether formation step, and the ee of (R)ꢀ32a was 89%; the corresponding aldehyde was suspended in
ethyl acetate and filtered to generate enantiomerically pure (R)ꢀ33a with an ee of over 99%.
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
Scheme 5
a
Reagents and conditions: (a) (i) DCC, 4ꢀDMAP, toluene, 0 °C; (ii) NaOH, H O, EtOH, rt; (b) (i) HOAc,
2
DIAD, PPh , THF, 0 °C to rt; (ii) NaOH, H O, EtOH, rt.
3
2
CONCLUSION
In summary, we have developed a series of 3ꢀ(5'ꢀsubstituted)ꢀbenzimidazoleꢀ5ꢀ(1ꢀ(3,5ꢀ
dichloropyridinꢀ4ꢀyl)ethoxy)ꢀ1Hꢀindazoles as potent FGFR inhibitors. Structural optimization of 14a led
to the identification of (R)-21c, which had panꢀFGFR inhibitory activities against FGFR1–4 (IC values
50
of 0.9, 2.0, 2.0, and 6.1 nM, respectively) and eightꢀfold selectivity over VEGFR2. Compared with other
FGFR selective inhibitors, (R)-21c potently inhibited FGFR4, indicating its potential use for the
treatment of FGFR4ꢀdriven cancer types. Western blot analysis revealed that (R)-21c suppressed
FGF/FGFR and downstream signaling pathways at nanomolar concentrations. Furthermore, (R)-21c
displayed potent antiproliferative effect against a number of FGFRꢀamplified cell lines including NCIꢀ
25
ACS Paragon Plus Environment

Page 27 of 71
Journal of Medicinal Chemistry
1
2
3
H1581, KG1, SNU16, and RT112. After an oral administration of compound (R)-21c at a dose of 10
mg/kg/qd, almost complete tumor stasis was observed in nude mice bearing NCIꢀH1581 human lung
cancer xenograft after 21 days of initial administration, indicating that (R)-21c was a potential FGFR
inhibitor for further drug development.
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
a
Scheme 6
a
Reagents and conditions: (a) MsCl, Et N, CH Cl , 0 °C to rt; (b) 5ꢀhydroxyindole, Cs CO , DMF, 70
3
2
2
2
3
°
C; (c) NaNO , HCl, H O/dioxane, 0 °C to rt; (d) (i) 40c, H , 10% Pd/C, MeOH, rt; (ii) S , DMF, 90 °C.
2 2 2 8
EXPERIMENTAL SECTION
Chemistry.
All starting materials and reagents were either obtained from commercial suppliers or prepared
according to literature reported procedures. All purchased chemicals and solvents were used without
further purification unless otherwise noted. Flash chromatography was performed using silica gel
(300−400 mesh). All reactions were monitored by thinꢀlayer chromatography (TLC), using silica gel
1
plates with fluorescence F254 and visualized under UV light. H NMR spectral data were recorded on
1
3
Varian Mercury 300 or 400 NMR spectrometer, and C NMR was recorded on Varian Mercury 126 or
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 71
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
51 NMR spectrometer at ambient temperature. Chemicals shifts (δ) were reported in parts per million,
coupling constants (J) values were in hertz, and the splitting patterns were abbreviated as follows: s for
singlet; d for doublet; t for triplet; q for quartet and m for multiplet. The low or high resolution of EIꢀMS
was recorded on a Finnigan/MAT95 spectrometer. All tested compounds were purified to ≥95% purity
as determined by high performance liquid chromatography (HPLC).
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
General procedure A: Synthesis of target compounds 14–15, 16a and 16c. A mixture of the
aldehyde (25a–e or 33a–c) (0.3 mmol), aniline (28, 34a or 34c) (0.3 mmol), and sulfur (0.6 mmol) was
4
0
heated to 90 °C for 2 h under argon atmosphere . Ethyl acetate was added and the reaction mixture was
partitioned between ethyl acetate and water. The organic layer was washed with water and brine
successively, dried over anhydrous sodium sulfate, and concentrated in vacuum. The resulting residue
was purified by silica gel chromatography (dichloromethane/methanol + 2% of aqueous ammonia, v/v,
9
9:1 to 97:3) to give the desired product.
3
-(5-(Piperidin-1-yl)-1H-benzo[d]imidazol-2-yl)-5-(pyridin-4-ylmethoxy)-1H-indazole (14a). The
title compound was prepared from 25a and 28 following general procedure A. Yellow solid (56%), mp
1
232 °C dec. H NMR (300 MHz, CD OD) δ 8.54 (dd, J = 4.6, 1.6 Hz, 2H), 8.00 (d, J = 2.0 Hz, 1H),
3
7
2
.57 (dd, J = 19.5, 7.6 Hz, 4H), 7.25 (dd, J = 9.1, 2.4 Hz, 2H), 7.08 (dd, J = 8.8, 2.1 Hz, 1H), 5.31 (s,
13
H), 3.19–3.08 (m, 4H), 1.87–1.71 (m, 4H), 1.66–1.55 (m, 2H). C NMR (126 MHz, acetoneꢀd ) δ
6
155.1, 152.1, 150.8 (2C), 150.5, 147.3, 139.9, 138.7, 137.5, 136.0, 122.7 (2C), 122.7, 120.1, 115.5,
+
1
12.3, 104.3, 103.1, 69.4, 53.2 (2C), 27.0 (2C), 25.1. HRMS (EI ) m/z calcd for C H N O 424.2012;
2
5
24
6
found 424.2011.
5
-((3-Chloropyridin-4-yl)methoxy)-3-(5-(piperidin-1-yl)-1H-benzo[d]imidazol-2-yl)-1H-indazole
(14b). The title compound was prepared from 25b and 28 following general procedure A. Yellow solid
27
ACS Paragon Plus Environment

Page 29 of 71
Journal of Medicinal Chemistry
1
2
3
1
(
50%), mp 140–142 °C. H NMR (300 MHz, CD OD) δ 8.59 (s, 1H), 8.51 (d, J = 5.0 Hz, 1H), 8.03 (s,
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
H), 7.76 (d, J = 4.9 Hz, 1H), 7.58 (d, J = 3.6 Hz, 1H), 7.55 (d, J = 3.9 Hz, 1H), 7.30–7.23 (m, 2H), 7.10
13
(d, J = 8.8 Hz, 1H), 5.37 (s, 2H), 3.22–3.13 (m, 4H), 1.85–1.75 (m, 4H), 1.67–1.58 (m, 2H). C NMR
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
(151 MHz, acetoneꢀd ) δ 154.8, 149.8, 149.2, 147.6, 144.8, 139.9, 138.8, 138.6, 137.4, 130.8, 123.9,
6
+
122.6, 120.0 (2C), 116.2, 112.4, 104.5, 100.0, 67.1, 53.4 (2C), 26.8 (2C), 24.9. HRMS (EI ) m/z calcd
for C H ClN O 458.1622; found 458.1619.
2
5
23
6
5
-((3,5-Dichloropyridin-4-yl)methoxy)-3-(5-(piperidin-1-yl)-1H-benzo[d]imidazol-2-yl)-1H-
indazole (14c). The title compound was prepared from 25c and 28 following general procedure A.
1
Yellow solid (72%), mp 205 °C dec. H NMR (300 MHz, CD OD) δ 8.61 (s, 2H), 8.11 (d, J = 2.2 Hz,
3
1H), 7.64–7.50 (m, 2H), 7.30 (s, 1H), 7.19–7.09 (m, 2H), 5.45 (s, 2H), 3.26–3.14 (m, 4H), 1.89–1.75 (m,
13
4
H), 1.68–1.57 (m, 2H). C NMR (126 MHz, acetoneꢀd ) δ 155.3, 149.9 (2C), 148.9, 147.7, 141.2,
6
1
38.9, 137.3, 134.3 (2C), 122.6, 120.2, 116.3, 112.5, 104.3, 98.9, 65.5, 53.5 (2C), 26.7 (2C), 24.8.
+
HRMS (EI ) m/z calcd for C H Cl N O 492.1232; found 492.1232.
2
5
22
2
6
3
-(5-(Piperidin-1-yl)-1H-benzo[d]imidazol-2-yl)-5-(pyridin-3-ylmethoxy)-1H-indazole (14d). The
title compound was prepared from 25d and 28 following general procedure A. Yellow solid (54%), mp
1
1
7
52–154 °C. H NMR (300 MHz, CD OD) δ 8.72 (s, 1H), 8.51 (d, J = 5.0 Hz, 1H), 8.08–7.97 (m, 2H),
3
.63–7.45 (m, 3H), 7.28 (s, 1H), 7.22 (dd, J = 9.1, 2.2 Hz, 1H), 7.12 (dd, J = 8.8, 2.1 Hz, 1H), 5.30 (s,
13
2H), 3.24–3.15 (m, 4H), 1.86–1.74 (m, 4H), 1.69–1.57 (m, 2H). C NMR (126 MHz, acetoneꢀd ) δ
6
1
1
4
55.3, 150.5, 150.3, 150.2, 147.7, 138.6, 137.5, 136.4, 133.9, 124.3, 122.7, 120.3, 116.2, 112.3, 104.1,
+
02.5, 68.8, 53.2 (2C), 27.0 (2C), 25.1. HRMS (EI ) m/z calcd for C H N O 424.2012; found
25 24
6
24.2004.
28
ACS Paragon Plus Environment