Synthesis and evaluation of the HIF-1α inhibitory activity of 3(5)-substituted-4-(quinolin-4-yl)- and 4-(2-phenylpyridin-4-yl)pyrazoles as inhibitors of ALK5

Article abstract of DOI:10.1016/j.bmcl.2019.126822

The transcription factor hypoxia-inducible factor-1α (HIF-1α) plays an important role in apoptosis, metastasis, and proliferation and is recognized as an important potential therapeutic target for cancer. Six series of 3(5)-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)pyrazoles (11a–d, 12a–d, and 18a–d) and 3(5)-(6-methylpyridin-2-yl)-4-(2-phenyl-pyridin-4-yl)pyrazoles (19a–d, 20a–d, and 21a–d) were synthesized and evaluated for activin receptor-like kinase 5 (ALK5) and HIF-1α inhibitory activity at the enzyme and cell levels. The effect of the lead compound 20d (J-1012) on HIF-1α activation in HCT116 cells was investigated. J-1012 markedly decreased the hypoxia-induced or TNF-induced accumulation of HIF-1α protein dose-dependently. Analysis revealed that J-1012 inhibited HIF-1α protein synthesis, without affecting the degradation of HIF-1α protein. Furthermore, by inhibiting the activation of HIF-1α, J-1012 suppressed the metastasis and proliferation and promoted apoptosis of HCT116 cells. These results suggest that J-1012 may be a potential therapeutic agent against human colon cancer.

Full text of DOI:10.1016/j.bmcl.2019.126822

Journal Pre-proofs
Synthesis and evaluation of the HIF-1α inhibitory activity of 3(5)-substituted-4-
(quinolin-4-yl)- and 4-(2-phenylpyridin-4-yl)pyrazoles as inhibitors of ALK5
Yan-Wei Li, Xiang-Yu Li, Shanji Li, Li-Min Zhao, Juan Ma, Hu-Ri Piao, Zhe
Jiang, Cheng Hua Jin, Xuejun Jin
PII:
S0960-894X(19)30791-7
DOI:
https://doi.org/10.1016/j.bmcl.2019.126822
BMCL 126822
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
4 September 2019
3 October 2019
10 November 2019
Please cite this article as: Li, Y-W., Li, X-Y., Li, S., Zhao, L-M., Ma, J., Piao, H-R., Jiang, Z., Jin, C.H., Jin, X.,
Synthesis and evaluation of the HIF-1α inhibitory activity of 3(5)-substituted-4-(quinolin-4-yl)- and 4-(2-
phenylpyridin-4-yl)pyrazoles as inhibitors of ALK5, Bioorganic & Medicinal Chemistry Letters (2019), doi: https://
doi.org/10.1016/j.bmcl.2019.126822
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier Ltd.

Synthesis and evaluation of the HIF-1α inhibitory
activity of 3(5)-substituted-4-(quinolin-4-yl)- and
4-(2-phenylpyridin-4-yl)pyrazoles as inhibitors of
ALK5
Yan-Wei Lia,c,1, Xiang-Yu Lia,1, Shanji Lib,1, Li-Min Zhaoa, Juan Maa, Hu-Ri
Piaoa, Zhe Jiangc,, Cheng Hua Jina,, Xuejun Jina,
a
Key Laboratory of Natural Resources of Changbai Mountain and Functional Molecules, Ministry of
Education, Molecular Medicine Research Center, College of Pharmacy, Yanbian University, Yanji
133002, PR China.
b The School of Public Health, Jilin Medical University, Jilin, 132013, PR China.
c Pharmacy Intravenous Admixture Services, Yanbian University Hospital, Yanji, 133000, PR China
 Corresponding authors.
E-mail addresses: jinchenghua@ybu.edu.cn (C.H. Jin), xjjin@ybu.edu.cn (X. Jin),
jiangzhe95@163.com (Z. Jiang).
1 These authors contributed equally to this work.

Abstract
The transcription factor hypoxia-inducible factor-1α (HIF-1α) plays an important
role in apoptosis, metastasis, and proliferation and is recognized as an important
potential therapeutic target for cancer. Six series of
3(5)-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)pyrazoles (11a–d, 12a–d, and 18a–d)
and 3(5)-(6-methylpyridin-2-yl)-4-(2-phenyl-pyridin-4-yl)pyrazoles (19a–d, 20a–d,
and 21a–d) were synthesized and evaluated for activin receptor-like kinase 5 (ALK5)
and HIF-1α inhibitory activity at the enzyme and cell levels. The effect of the lead
compound 20d (J-1012) on HIF-1α activation in HCT116 cells was investigated.
J-1012 markedly decreased the hypoxia-induced or TNF-induced accumulation of
HIF-1α protein dose-dependently. Analysis revealed that J-1012 inhibited HIF-1α
protein synthesis, without affecting the degradation of HIF-1α protein. Furthermore,
by inhibiting the activation of HIF-1α, J-1012 suppressed the metastasis and
proliferation and promoted apoptosis of HCT116 cells. These results suggest that
J-1012 may be a potential therapeutic agent against human colon cancer.
Keywords: TGF-β; ALK5 inhibitor; HIF-1α; Apoptosis; Docking

Cancer is one of the most serious clinical problems in the world. Statistics from the
World Health Organization indicate that cancer accounted for approximately 9.6
million deaths globally in 2018, 1 and the incidence of cancer is increasing every year.
As well as surgery and radiation treatment, chemotherapy has been proven to be very
useful in cancer therapy.2,3 Patients treated with conventional cancer chemotherapies
usually suffer from side effects of the drugs because of the non-selective action of
chemotherapeutic drugs.4 Thus, the development of potent and effective novel
antineoplastic drugs is one of the most intensely persuaded goals of world.
Hypoxia-inducible factor-1 (HIF-1) is composed of α and β subunits. HIF-1α is the
most widely expressed isoform, and is regulated mainly post-translationally. In
normoxia conditions, the hydroxylation of the HIF-1α protein by prolylhydroxylases
causes its degradation by the proteasome,5 while hydroxylation by the factor
inhibiting HIF inhibits HIF-1α binding to CBP/p300.6 In hypoxia conditions, the
proteasomal degradation of HIF-1α is inhibited. Stabilized HIF-1α then dimerizes
with HIF-1β and translocates to the nucleus,7 where it induces the activation of genes
containing hypoxia-responsive elements in their promoters.8 Activated HIF-1 plays an
important role in the adaptive responses of tumor cells through the transcriptional
activation of over 100 downstream genes that regulate vital biological processes
required for tumor survival and progression, including genes involved in cell
proliferation and angiogenesis.9 Although many studies have indicated that HIF-1 is
an important regulator of the genetic response to hypoxia, HIF-1α can also be
activated by inflammatory cytokines, such as tumor necrosis factor (TNF),
interleukin-1β (IL-1β), prostaglandin E2 (PGE2) and lipopolysaccharide (LPS), which
suggests that HIF is implicated in the inflammatory process.10,11 It has recently been
shown that certain pro-inflammatory cytokines can stabilize HIF-1α protein
expression and activate HIF-1.12,13 These results suggest that HIF-1 may be involved
in the regulation of gene expression during inflammation.
Transforming growth factor-β (TGF-β) is a multifunctional growth factor, which is
widely found in the normal tissue cells and transformed cells of animals, and can be
produced by almost all cells in the human body.14 TGF-β family members include

TGF-βs, activins, inhibins, growth and differentiation factors (GDFs), nodal, and
anti-Müllerian hormone and bone morphogenetic proteins.15 TGF-β have three major
isoforms such as TGF-β1, TGF-β2, and TGF-β3, which are expressed in mammals,
and each isoform is encoded by a unique gene and expressed in a tissue-specific
manner. TGF-β is known to regulate cell proliferation, cell migration, extracellular
matrix synthesis, and inflammatory processes. Furthermore, TGF-β plays a critical
role in the initiation and progression of fibrosis in various organ systems, including
the kidneys,16 heart,17 lung,18 and liver.19 TGF-β transduces signaling through TGF-β
type I (activin receptor-like kinase 5, ALK5) and type II receptors.20 ALK5 is
phosphorylated by the combination of TGF-β and the type II receptor, which adjusts
its kinase activity.21 The activated ALK5 phosphorylated Smad2/Smad3 proteins,
form a heteromeric complex with Smad4, which is delivered into the nuclei, where it
regulate the expression of several hundred genes. Overexpression of TGF-β signaling
leaded to various human diseases such as cancer,22 pancreatic diseases,23 and
hematological
malignancies.24
Hence
the
regarding
TGF-β-mediated
ALK5-dependent signaling pathway at the receptor level has been highlighted as a
potential method of inhibiting TGF-β signaling. It has been reported that several small
molecule ATP-competitive ALK5 inhibitors such as compounds 1 (SB-505124),25 2
(SD-208),26 3 (A-83-01),27 4 (LY-2157299),28 and 5 (EW-7197),29 (Fig. 1) inhibited
the autophosphorylation of ALK5 and the TGF-β-induced transcription of matrix
genes in reporter assays at sub-micromolar concentrations. Compound 2 effectively
retarded the progress of fibrosis in the kidney, liver and lung, and compounds 2 and 4
strongly inhibited the growth and invasiveness of cancer cells in animal models.
Compound 4 has entered phase II clinical trials against glioma, hepatocellular
carcinoma,30 glioblastoma, and pancreatic cancer.31–33 In addition, another compound
5, also entered phase II clinical trials, inhibited TGF-β-induced tumor cell migration
and invasion in breast cancer cells, and inhibited hepatic, renal and pulmonary fibrosis
by blocking TGF-β/Smad and reactive oxygen species signaling.34

Fig. 1. ALK5 inhibitors under development
A class of ALK5 inhibitors that have a thioamide linkage between a pyrazole ring
and a phenyl ring has been reported, and among them, A-83-01 exhibited significant
inhibition of the transcriptional activity induced by ALK5.27 However, the thioamide
linkage was rather unstable and was slowly cleaved, to release the pyrazole ring
during long-term storage.
We have previously shown that a series of compounds based on compound 6
containing a quinolin-6-yl moiety, except for the methoxy-substituted analogues,
showed significant ALK5 inhibition in an enzymatic assay.35,36 These compounds
showed high selectivity for ALK5 relative to p38α mitogen activated protein (MAP)
kinase. Previous studies have shown that ALK5 inhibitors with high selectivity for
p38α MAP kinase also show high selectivity for other kinases.20 The most active
compound of the series inhibited ALK5 phosphorylation with an IC50 value of 0.022
μM and a selectivity index of >45 against p38α MAP kinase. In this study, we also
found that insertion of a methoxy group into the quinoline ring did not improve the
ALK5 inhibitory activity of the derivatives, which may be because of steric hindrance
from the resulting rigid structures. Therefore, we hypothesized that introducing a
methylene group between the pyrazole ring and the thioamide, to give a structure
without the rigidity and with less steric hindrance at the 4-positoin of the central

pyrazole ring in compound 3, may enhance the ALK5 inhibitory activity and the
stability. To test this hypothesis, we inserted a methylene group between the pyrazole
ring and thioamide, that is, we used a 2-phenylpyridin-4-yl group instead of
quinolin-4-yl moiety in compound 3.
Based on previous research, we replaced the thioamide linkage with a chemically
stable thioamidomethylene linkage and, thus, designed compounds 20a–d and 21a–d
(Fig. 2). To compare the effect of the thioamidomethylene linkage in 20a–d and
21a–d on the ALK5 inhibitory activity, the corresponding derivatives 11a–d, 12a–d,
18a–d and 19a–d with an amidomethylene linkage were also designed. The target
compounds 11b–d, 12b–d, 18b–d, 19b–d, 20b–d, and 21b–d each possess a
substituent, either o-F, m-F or m-CN on phenyl ring because these substituents were
previously found to be beneficial for ALK5 inhibitory activity and selectivity.29
Fig. 2. Design strategy based on compounds 4 and 5.
In the present study, we explored the antitumor effects of J-1012 on colon cancer

cells in vitro. Based on the combined study results, the following conclusions were
reached: (1) J-1012 suppressed HIF-1 activation. (2) J-1012 markedly decreased the
hypoxia-induced or TNF-induced accumulation of HIF-1α protein dose-dependently.
(3) J-1012 inhibited HIF-1α protein synthesis, without affecting the degradation of
HIF-1α protein. (4) J-1012 by inhibiting the activation of HIF-1α, suppressed
metastasis and proliferation and promoted apoptosis. Our findings indicated that
J-1012 could be a valuable preclinical candidate for the development of a cancer
therapeutic.
The 3-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)pyrazoles 11a–d was synthesized as
shown in Scheme 1. The quinoline-4-carbaldehyde (7) was coupled with diphenyl
(6-methylpyridin-2-yl)(phenylamino)methylphosphonate37 in a mixture of THF and
i-PrOH (4:1) in the presence of Cs2CO3 at room temperature, followed by treatment
with 1 N HCl to afford the corresponding ketone 8 in 88% yield. Treatment of 8 with
N,N-dimethylformamide dimethyl acetal (DMF•DMA) in DMF at 80oC, followed by
cyclization with hydrazine monohydrate in absolute EtOH, produced the pyrazole 9 in
70% yield.38 The pyrazole 9 was alkylated with substituted phenylacetamides 10a–d39
in the presence of NaH in anhydrous DMF to yield the target compounds 11a–d and
their positional isomer 12a–d in 32%–35% and 7%–15% yields, respectively. The
positional isomers were separated by column chromatography and their structures
were confirmed by nuclear overhauser enhancement (NOE) experiments. In NOE
experiments, irradiation of the methylene protons of compound 11b at δ 5.17 ppm
gave an enhancement of the proton H-5 in the pyrazole ring at δ 7.78 ppm, while
irradiation of the methylene protons of compound 12b at δ 5.28 ppm gave no
enhancement of the proton H-5 in the pyrazole ring at δ 7.86 ppm, confirming the
respective alkylation positions.

Scheme
1.
Reagents
and
conditions:
(a)
(i)
diphenyl(6-methylpyridin-2-yl)(phenylamino)methylphosphonate, Cs2CO3, THF: i-PrOH (4:1), rt, 16 h;
(ii) 1 N HCl, rt, 1 h; (b) (i) DMF•DMA, DMF, 80oC, 4 h; (ii) N2H4•H2O, EtOH, reflux, 4 h; (c) NaI
(cat.), NaH, anhydrous DMF, rt, 2 h.
The 3-(6-methylpyridin-2-yl)-4-(2-phenylpyridin-4-yl)pyrazoles 18a–d was
synthesized as shown in Scheme 2. The 2-phenylisonicotinaldehyde (15) was
synthesized from commercially available phenylboronic acid (13) and
2-bromoisonicotinaldehyde (14) under Suzuki-Miyaura cross-coupling reaction
condition.40 The target compounds 18a–d and their positional isomers 19a–d were
synthesized from compound 15 via three steps in the same reaction condition as
described in Scheme 1. And these positional isomers were also separated by column
chromatography and their structures were confirmed by NOE experiments.

Scheme 2. Reagents and conditions: (a) Na2CO3, Pd(PPh3)4, THF: H2O (10:1), reflux, 4 h; (b) (i)
diphenyl(6-methylpyridin-2-yl)(phenylamino)methylphosphonate, Cs2CO3, THF: i-PrOH (4:1), rt, 4 h;
(ii) 1 N HCl, rt, 1 h; (c) (i) DMF•DMA, DMF, 90oC, 4 h; (ii) N2H4·H2O, EtOH, reflux, 4 h; (d) NaI
(cat.), NaH, anhydrous DMF, rt.
Thionation of compounds 11a–d and 18a–d with Lawesson´s reagent in anhydrous
DME at 85oC produced the thioamides 20a–d in 33%–97% yields and 21a–d in
34%–97% yields (Scheme 3). As we expected, all synthesized target compounds were
quite stable during long-term storage at room temperature.

Scheme 3. Reagents and conditions: (a) Lawesson's reagent, 85oC, 4–12 h.
To investigate whether compounds 11a–d, 12a–d, 18a–d, 19a–d, 20a–d, and
21a–d would inhibit ALK5, a kinase assay for preliminary screening was performed
using purified human ALK5 kinase domain produced in Sf9 insect cells, and test
compounds at 10 μM (Table 1). Compound 4 (LY-2157299) was used as a positive
control and DMSO was used as a negative control. All the compounds possessing a
quinolin-4-yl moiety (11a–d, 12a–d, and 20a–d) displayed more potent ALK5
inhibition (21%–100%) than those possessing a 2-phenylpyridin-4-yl moiety (18a–d,
19a–d and 21a–d, 0%–97%). The amides 11a–d and 18a–d showed more potent
ALK5 inhibition (98%–100% and 80%–97%, respectively) than their respective
positional isomers, 12a–d (21%–83%) and 19a–d (0%–7%). Among the compounds
containing a quinolin-4-yl moiety, the thioamides 20a–d showed similar ALK5
inhibition (97%–98%) to their corresponding amides 11a–d at 10 μM. Among the
compounds containing a 2-phenylpyridin-4-yl moiety, the thioamides 21a–d showed
slightly more potent ALK5 inhibition (88%–94%) than the amides 18a–d, at 10 μM.
We selected p38α MAP kinase to survey the selectivity profile of this series of
compounds because its kinase domain is one of the most homologous domains to
ALK5.41 All the tested compounds, except compounds 19a–d (0%–34% activity),
showed good activity (49%–97%) against p38α MAP kinase at 10 μM. Furthermore,

these two series of compounds showed a similar activity pattern to that found for the
ALK5 residual assay.
Table 1. Residual activities of 3(5)-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)- and
3(5)-(6-methylpyridin-2-yl)-4-(2-phenylpyridin-4-yl)pyrazoles 11a–d, 12a–d, 18a–d, 19a–d,
20a–d, and 21a–d on ALK5 and p38α MAP kinase
Compound
X
R
Residual activitya,d
p38αb
100±0.1
10±0
ALK5c
DMSO
11a
100±0.1
2±0
O
O
O
O
O
O
O
O
S
H
o-F
11b
6±0
0±0
11c
m-F
m-CN
H
4±0
0±0
11d
6±0
1±0
12a
24±0
71±1.0
79±2.5
39±4.0
17±1.0
3±0.5
2±0.5
2±0.5
2±0
12b
o-F
50±3.5
31±0.5
51±3.5
3±0.5
4±0.5
3±0.5
3±0
12c
m-F
m-CN
H
12d
20a
20b
S
o-F
20c
S
m-F
m-CN
H
20d
S
18a
O
O
O
O
O
O
O
O
S
11±1.0
20±1.5
14±0
12±0.5
20±1.0
3±0.5
18±0
18b
o-F
18c
m-F
m-CN
H
18d
24±0
19a
66±6.0
96±3.5
81±3.5
135±1.5
10±1.5
4±1.0
9±2.5
15±0.5
4±0.5
100±0.5
102±2.5
93±6.5
105±6.5
11±1.0
11±1.0
6±0.5
12±0.5
1±0
19b
o-F
19c
m-F
m-CN
H
19d
21a
21b
S
o-F
21c
S
m-F
m-CN
21d
S
4 (LY-2157299)
a
Residual kinase activities were measured for each compound at 10 μM, in duplicate, in reactions
b
containing p38α and ALK5 protein kinases. p38α MAP kinase was expressed in E. coli as the
c
untagged human recombinant protein. ALK5 was expressed in Sf9 insect cells as the human
recombinant GST-fusion protein using the vaculovirus expression system. d Activity is given as the

mean ± SD of two independent experiments run in duplicate relative to control incubations with
DMSO vehicle.
In previous studies, we found that sulfonamide compounds were more active than
the corresponding amide compounds.21,42 To evaluate the ALK5 inhibitory activity
and selectivity of the compounds possessing quinolin-4-yl or 2-phenylpyridin-4-yl
moieties, the thioamides 20a–d and 21a–d were selected based on residual activity
and previous data, and their half maximal inhibitory concentration (IC50) values were
measured. All compounds with a quinolin-4-yl moiety (20a–d) showed potent ALK5
inhibition (IC50 = 0.043–0.069 μM), whereas those with a 2-phenylpyridin-4-yl
moiety (21a–d) showed no appreciable ALK5 inhibitory activity (IC50 = 1.56–2.20
μM). (Table 2) Compound 20c showed the most potent ALK5 inhibitory activity of
these two series of compounds, with an IC50 value of 0.043 μM, which was 2.8-fold
more potent than compound 4 (IC50 = 0.119 μM). It is noteworthy that these two
series of compounds showed good p38α MAPK inhibitory activity (IC50 = 0.063–1.15
μM). This result was different to previous data.21,35 In particular, all compounds with
a quinolin-4-yl moiety (20a–d) showed potent p38α MAPK inhibitory activity (IC50 =
0.063–0.104 μM). For all compounds with a 2-phenylpyridin-4-yl moiety (21a–d), the
inhibitory activity towards p38α MAPK was better than towards ALK5. The most
potent compound 21b was 4.2-fold more selective for p38α MAPK (IC50 = 0.487 μM)
than for ALK5 kinase. (IC50 = 2.05 μM). The above result indicated that displacement
of the nitrogen atom and insertion of a less rigid structure for the quinolin-4-yl moiety
of compound 6 did not appear to improve the selectivity for ALK5 kinase. However,
it has been reported that p38α MAPK plays an essential role in cancer and is highly
expressed in invasive breast cancers43 and p38α inhibitors have been shown to be
effective for the treatment of p53 mutant and estrogen receptor-negative breast
cancers in recent studies.44 Therefore, these two series of compounds may have
potential as anticancer drugs. In these two series of compounds, the effects of the
substituents (F and CN) were not obvious.
Table
2.
The
IC50
of
3(5)-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)-
and
3(5)-(6-methylpyridin-2-yl)-4-(2-phenylpyridin- 4-yl)pyrazoles 20a–d and 21a–d on ALK5
and p38α MAP kinase

Compound
R
IC50 (μM)
p38αa
ALK5b
20a
H
o-F
0.074±0.0002
0.068±0.0004
0.063±0.0005
0.104±0.0007
0.536±0.0006
0.487±0.0008
0.761±0.0005
1.150±0.0006
0.487±0.0005
0.057±0.0003
0.053±0.0004
0.043±0.0006
0.069±0.0003
1.689±0.0007
2.050±0.0006
1.560±0.0004
2.200±0.0003
20b
20c
m-F
m-CN
H
20d
21a
21b
o-F
21c
m-F
m-CN
21d
4 (LY-2157299)
0.119±0.0006
a
b
p38α MAP kinase was expressed in E. coli as untagged human recombinant protein. ALK5 was
expressed in Sf9 insect cells as a human recombinant GST-fusion protein using the vaculovirus
expression system. a,b Values are the average of three independent experiments run in triplicate.
To rationalize the structure-activity relationships (SAR) shown in Tables 1 and 2,
we examined the binding modes of two representative ligands, 20b and 21b, using the
semi-flexible molecular docking programs DS CDOCKER.45 Docking analyses were
performed using the recently reported X-ray structure of ALK5 in complex with a
pyrazole ALK5 inhibitor (PDB: 1RWB).29 As shown in Figure 3, the carbonyl group
of Lys337 formed hydrogen bond network with the methylene group, the NH group
and the fluoride atom of the side chain of compound 20b. The quinoline ring in 20b
formed pi-anion interaction and pi-alkyl bond with Lys232. The pyridine ring and
pyrazole ring in 20b formed pi-alkyl bond with Leu340. The phenyl ring of
2-phenylpyridin-4-yl moiety in 21b formed pi-alkyl bond with Lys232. The pyridine
ring of the 2-phenylpyridin-4-yl moiety in 21b formed pi-alkyl bond with Leu340.
The quinolin-4-yl moiety of 20b and the 2-phenylpyridin-4-yl moiety of 21b were

stretched to the back of the hydrophobic pocket. Compound 20b showed higher
ALK5 inhibitory activity than compound 21b, which may be related to the following
two reasons. Compound 20b showed more binding with previously reported key
amino acids (Lys337, Glu245, Lys232 and Leu340) than did compound 21b (Leu340
and Lys232). Not only did the side chain of 21b not form any bonds with key amino
acids, but it was also stretched to the outside of the pocket. Therefore, compound 20b
appeared to be more favorably accommodated in the binding pocket of ALK5 than
compound 21b (Figs. 3B and 3D).
Fig. 3. Docking conformation of compounds 20b and 21b in the active site of ALK5 (PDB: 1RWB).
(A) 2D binding model of compound 20b. (B) Proposed conformation of compound 20b in the binding
pocket of ALK5. (C) 2D binding model of compound 21b. (D). Proposed conformation of compound
21b in the binding pocket of ALK5. The ligands are shown in yellow.
Both compounds 20b and 21b did not form any bonds with Ser280. As bonding

with Ser280 is known to be critical for the selectivity of ligands towards ALK5 over
p38α MAP kinase, this may be the reason for the low selectivity of the two
compounds. Indeed, the results suggested that the two compounds had a good affinity
for p38α related protein. To investigate the binding affinity in the p38α active site, we
selected the same compounds (20b and 21b) and examined the binding using DS
CDOCKER. The docking performance and accuracy of the DS2017 program was
assessed by the semi-flexible docking of the native co-crystalized ligand (SB203580)
into p38α MAP-kinase (PDB: 1A9U).46 As shown in Figure 4, the carboxylic acid
group of Asp168 formed a hydrogen bond network with the NH group and the
fluoride atom of the side chain of compound 20b, formed a salt bridge with the NH
group of the side chain and a pi-anion bond with the pyrazole ring of compound 20b.
The carboxylic acid group of Glu71 formed a pi-anion bond with the quinoline and
pyridine rings of compound 20b. The two rings of the quinoline in 20b formed
pi-anion bonds with Lys53 and Glu71 and pi-alkyl bonds with Ala51 and Ile84. The
pyridine ring in compound 20b formed a pi-anion bond with Lys53 and a pi-cation
bond with Glu71. The carboxylic acid group of Glu71 formed a pi-cation bond with
the phenyl ring of the 2-phenylpyridin-4-yl moiety of compound 21b. The pyridine
ring of the 2-phenylpyridin-4-yl moiety in compound 21b formed a pi-alkyl bond with
Lys53. The carboxylic acid of Asp168 formed a hydrogen bond with the NH group of
the side chain of compound 21b and a pi-cation bond with the pyrazole ring.
Compound 20b showed higher p38α inhibitory activity than compound 21b, which
may be related to the following two reasons. Compound 20b showed more bonding
with previously reported key amino acids (Asp168, Ala51, Lys53, Ile84, and Glu71)
than did compound 21b (Asp168, Lys53, and Glu71).46,47 Moreover, compound 21b
was more stretched out to the edge of the p38α binding pocket than compound 20b
(Figs. 4C and 4D). Hence, compound 20b appeared to be more favorably
accommodated in the binding pocket of p38α than compound 21b.

Fig. 4. Docking conformation of compounds 20b and 21b in the active site of p38α (PDB: 1A9U). (A)
2D binding model of compound 20b. (B) Proposed conformation of compound 20b in the binding
pocket of p38α. (C) 2D binding model of compound 21b. (D). Proposed conformation of compound
21b in the binding pocket of p38α. The ligands are shown in yellow.
The HIF-1α inhibitory activity of all compounds (20a–d and 21a–d) was evaluated
and the pretest results showed that compound 20d (J-1012) had the best activity. To
explore the interaction between J-1012 and HIF-1α, the binding mode was
investigated through a molecular docking study. It was shown that the pyridine
moiety and side chain of J-1012 were stretched into the back of the binding pocket
consisting of Arg342, Thr301, Cys308, Phe263, His307, Ser262, and Gln320. The
nitrogen atom on the side chain of Gln320 formed hydrogen bonds with the pyridine
and central pyrazole ring of J-1012. The carboxylic acid group of Phe263 formed a

hydrogen bond with the NH group of the side chain of J-1012 (Figs. 5A and 5B).
Subsequently, to investigate the effect of J-1012 on the transcriptional activity of
HIF-1, we transfected HCT116 cells with a luciferase reporter gene driven by six
specific HREs. A significant increase in luciferase activity was observed in cells
cultured under hypoxic conditions, while J-1012 dose-dependently inhibited
hypoxia-induced luciferase activity (Fig. 5C). To evaluate the effect of J-1012 on cell
viability in HCT116 cells, a MTT assay was performed. J-1012 did not display
significant cellular toxicity towards HCT116 cells at concentrations up to 50 μM (Fig.
5D).
Fig. 5. A) J-1012 was docked into the binding site of HIF-1α (3D binding mode). The ligand is shown
yellow. B) J-1012 was docked into the binding site of HIF-1α (2D binding mode). C) HCT116 cells
were transiently co-transfected with a pGL3-HRE-Luciferase and pRL-CMV vectors. Following 24 h
incubation, the cells were incubated under hypoxia in the absence or presence indicated concentrations

of J-1012. Luciferase activities were determined as described in “Materials and Methods”. Data
represented as mean ± standard deviation of three independent experiments. **p < 0.01, ***p < 0.001,
significant with respect to control. D) MTT assay to measure cell viability in HCT116 cell after 24 h
treatment with the indicated concentrations of J-1012 concentrations.
To explore the effects of J-1012 on HIF-1α expression under hypoxic conditions,
HIF-1α protein levels were determined in HCT116 cells treated with J-1012.
Following 12 h treatment, J-1012 exhibited dose-dependent inhibition of the HIF-1α
protein levels induced by hypoxia in HCT116 cells (Fig. 6A). Several studies have
demonstrated that some inflammatory cytokines, such as TNF, IL-1β, and LPS, can
induce the transcriptional activity of HIF-1α.48 Thus, we exposed HCT116 cells to the
inflammatory factor TNF (10 ng/mL) for 12h. TNF dramatically promoted HIF-1α
protein expression and J-1012 inhibited this TNF-induced HIF-1α protein expression
(Fig. 6B); LY-2157299, the positive control group, had a similar effect to that of
J-1012. Generally, the accumulation of HIF-1α protein is dependent on the balance
between HIF-1α protein synthesis and degradation.49 To further investigate whether
the mechanism of inhibition was the result of increased degradation, or decreased
synthesis, of HIF-1α protein, the proteasome inhibitor MG-132 was used to prevent
HIF-1α protein degradation. After treatment with MG-132, the level of HIF-1α
protein was significantly increased under hypoxic conditions (lane 2 compared with
lane 4 in Fig. 6C). In contrast, J-1012 inhibited HIF-1α accumulation even under
co-treatment with MG-132 (lanes 3 and 5 in Fig. 6C). These results indicated that
HIF-1α protein synthesis in HCT116 cells is markedly impaired in the presence of
J-1012. Next, to investigate the effect of J-1012 on the stability of the HIF-1α protein,
the protein translation inhibitor CHX was used to prevent HIF-1α protein synthesis.
We first induced HIF-1α accumulation by exposing the cells to 1% O2 for 4 h, and
then added CHX alone or in conjunction with J-1012. Under these conditions, the
HIF-1α protein levels reflected the rate of HIF-1α degradation. In the presence of
CHX, the HIF-1α levels declined rapidly as expected, however, J-1012 did not alter
the degradation rate of HIF-1α (Fig. 6D).

Fig. 6. A) HCT116 cell were incubated under normoxia, or hypoxia conditions for 12h, in the indicated
concentration of J-1012, or LY-2157299 (50 µM). Nuclear extracts were then prepared and assayed for
HIF-1α protein expressions by western blotting. The same blot was reprobed with an anti-Topo-I
antibody as a loading control. B) HCT116 cell were exposed to 10 ng/mL TNF for 24 h and incubated
concentration of J-1012, or LY-2157299 (50 µM). Total protein of the HCT116 cells was extracted,
and the HIF-1α expression levels were then detected by western blotting. The same blot was reprobed
with an anti-Topo-I antibody as a loading control. C) HCT116 cells were pretreatment with proteasome
inhibitor MG-132 (10 µM) for 30 min and incubated with J-1012 (50 µM) in the presence of 1% O2.
After 12 h incubation, the nuclear extract for HIF-1α was analyzed by Western blotting. The same blot
was reprobed with an anti-Topo-I antibody as a loading control. D) Mix cycloheximide (CHX) (10 µM)
and J-1012 (50 µM) with medium and treated HCT116 cells. After 20, 40, 60 min incubation, the
nuclear extract for HIF-1α was analyzed by Western blotting. The same blot was reprobed with an
anti-Topo-I antibody as a loading control.
Recent studies have strongly suggested that HIF-1α acts as a potent inducer of cell
apoptosis.50 We wondered if J-1012 could enhance the apoptosis induced by TNF. To
test our hypothesis, we carried out Annexin V-FITC/PI double staining. As shown in
Figs 7A and 7B, treatment of HCT116 cells with vehicle only, TNF alone, J-1012
alone, LY-2157299 alone, TNF and LY-2157299 combined treatment, and TNF and
J-1012 combined treatment induced apoptosis by 1.2%, 4%, 6.3%, 5.3%, 6.5% and
8%, respectively. Because caspases are a group of cysteine proteases critical for the
apoptosis of eukaryotic cells, we investigated whether J-1012 affected the
TNF-induced activation of caspase-8 and caspase-3. TNF alone had only a minimal
effect, and J-1012 alone significantly affected the activation of caspase-8 and

caspase-3, whereas co-treatment with TNF and J-1012 potentiated activation of the
caspases, as indicated by the presence of cleaved caspases (Fig. 7C, top three panels).
In addition, the PARP cleavage assay was used to detect TNF-induced apoptosis.
J-1012 potentiated the effect of TNF-induced PARP cleavage, although J-1012 alone
also induced PARP cleavage (Fig. 7C, bottom second panel). These results indicate
that J-1012 (50 μM) could significantly enhance the TNF-induced apoptosis of
HCT116 cells compared with LY-2157299 (50 μM) (Fig. 7A).
Fig. 7. A) HCT116 cells were pretreated with 50 μM J-1012 or 50 μM LY-2157299 for 12 h and then
incubated with TNF (10 ng/mL) for 24 h, and stained with Annexin V-FITC and propidium iodide (PI),
followed by analysis using a flow cytometer. Early apoptotic cells (Annexin-V+ and PI-) are displayed
in the lower right quadrant and late apoptotic cells (Annexin-V+ and PI+) are shown in the upper right
quadrant. B) The percentage of apoptotic cells was counted. C) HCT116 cells were pretreated with 50
μM J-1012 or 50 μM LY-2157299 for 12 h and then incubated with TNF (10 ng/mL) for 24 h, Whole
cell extracts were analyzed by western blot analysis using indicated antibodies for cleaved caspase-8,

cleaved capase-3, cleaved PARP and tubulin.
Previous studies have demonstrated that the HIF-1α regulated expressions of
VEGF, MMP-9 and ICAM-1 is involved in tumor-cell proliferation, angiogenesis,
invasion and metastasis.51,52 We therefore examined whether J-1012 can suppress the
expression of these proteins. VEGF, MMP-9 and ICAM-1 protein levels were
measured by western blot analysis in HCT116 cells. The results indicated that TNF
induced the expression of ICAM-1, MMP-9 and VEGF, whereas J-1012 inhibited the
expression (Fig. 8A). These results were also reflected in the mRNA levels (Fig. 8B).
The migration of endothelial cells is a prerequisite for solid tumor angiogenesis.53 The
cell scratch assay was performed to examine the influence on the motile activity of
HUVECs. The results showed that J-1012 (50 μM) could significantly inhibit the
TNF-induced migration and invasive capacity of HCT116 cells compare with
LY-2157299 (50 μM).
Fig. 8. A) HCT116 cells were pretreated with 50 μM J-1012 or 50 μM LY-2157299 for 12 h and then
incubated with TNF (10 ng/mL) for 24 h, Whole cell extracts were analyzed by western blot analysis
using indicated antibodies for ICAM-1, MMP-9, VEGF and tubulin. B) HCT116 cells were
preincubated with indicated concentrations of J-1012 or 50 μM LY-2157299 for 12 h, and then treated
with 10 ng/mL TNF for an additional 12 h. The expression levels of ICAM-1, VEGF, MMP-9 were
verified by RT-PCR. GAPDH was used to show equal loading of total RNA. C) HUVEC cells were
pretreatment with indicated concentrations of J-1012 or 50 μM LY-2157299 in the presence or absence

of TNF (10 ng/mL). Photographs at time after 24ꢀh are shown. The area delimited with dashed line
represents the uncovered area.
To determine whether J-1012 could affect cell proliferation, the Edu incorporation
assay was performed. Following 12 h treatment, J-1012 (50 μM) inhibited the number
of Edu-positive cells induced by TNF in HCT116 cells. J-1012 (50 μM) significantly
inhibited the proliferation of HCT116 cells compared with LY-2157299 (50 μM)
(Figs. 9A and 9C). To determine the effect of the long-term antiproliferative activity
of J-1012, we used a colony formation assay. The results showed that the levels of
HCT116 cells were reduced in a dose-dependent manner (Figs. 9B and 9D) by
J-1012.
Fig. 9. A) HCT116 cells were incubated with J-1012 (50 μM) or LY-2157299 (50 μM) for 12 h and
treated with TNF (10 ng/ml) for 30 min. The EdU-labeled proliferations cells were examined under a
fluorescence microscope. B) HCT116 cells treated with J-1012 10, 30 50 μM or 50 μM LY-2157299

for 12 h and then treated with TNF (10 ng/ml) for 30 min. Next, the medium was replaced with a new
one. Keep cell growing for 10 days. Finally, the cells were stained with crystal violet and photographed.
C), D) Data represented as mean ± standard deviation of three independent experiments. *p < .05, **p
< .01, ***p < .001.
HIF-1 is overexpressed in many human cancers, and the activity of HIF-1 in cells is
correlated with tumorigenicity and angiogenesis.54 Under normal oxygen tension,
HIF-1α binds to the von Hippel-Lindau protein and is degraded via the
ubiquitin-dependent pathway.55 During this process, specific proline residues undergo
posttranslational hydroxylation by prolyl hydroxylases.56 In hypoxic conditions,
HIF-1 is stabilized and complexes with the β subunit to form a functional
transcription factor. The complete HIF-1 molecule translocates to the nucleus and
activates the expression of downstream genes in response to hypoxia.57 In this way,
HIF-1 activates different genes involved in angiogenesis, migration and survival, such
as erythropoietin and VEGFs.52,58 The same genes are often found to be
overexpressed in tumor cells, suggesting that these pathways are also involved in
tumor progression, development, invasion, metastasis, apoptosis and
tumorigenesis.49,52,59 Therefore, HIF-1 plays an important role in tumorigenesis and
development. Inhibition of HIF-1 activity has been shown to have a considerable
effect on tumor growth in preclinical studies.60,61 The development of HIF-1 inhibitors
represents a major challenge in the field of cancer treatment. In this study, we
demonstrated that J-1012 inhibited HIF-1α levels by impairing HIF-1α protein
synthesis.
With the expression of HIF-1α, tumor cells express high levels of VEGF and
interact with the VEGF receptor 2 (VEGFR2) to induce the growth of vascular
endothelial cells, the proliferation of endothelial cells and the formation of new
vessels.62 Cadherin and inter-cellular adhesion molecule-1 (ICAM-1) are adhesion
molecules that affect the adhesion, invasion and metastasis of tumor cells.62 In the
present study, we have shown that J-1012 suppressed TNF-induced MMP-9, ICAM-1
and VEGF mRNA levels.
Apoptosis plays an important role in the regulation of tumor cell growth, metastasis,
and invasion. Our study demonstrated that J-1012 potentiated TNF-induced apoptosis

of HCT116 cells, which was confirmed from flow cytometry analysis. Apoptosis is
characterized by specific morphological and biochemical features in which caspase
activation plays a central role. In this study, we found that J-1012 affected the
TNF-induced activation of caspase-8 and caspase-3, and PARP cleavage, indicating
that the apoptotic effects of TNF are enhanced by cryptopleurine. J-1012 also
inhibited TNF-induced HIF-1α protein expression.
In this report, six series of 3(5)-(6-methylpyridin-2-yl)-4-(quinolin-4-yl)pyrazoles
(11a–d,
12a–d
and
18a–d)
and
3(5)-(6-methylpyridin-2-yl)-4-(2-phenyl-pyridin-4-yl)- pyrazoles (19a–d, 20a–d and
21a–d) were synthesized and evaluated for ALK5 inhibitory activity in enzymatic
assays. We found that the introduction of a quinolin-4-yl moiety at the 4-position of
the pyrazole ring improved the inhibitory activity towards ALK5, while the
introduction of a 2-phenylpyridin-4-yl moiety at the same position did not improve
the inhibitory activity. The most potent compound, 20c, inhibited ALK5
phosphorylation with an IC50 value of 0.043 μM, and exhibited 98% inhibition of
ALK5 phosphorylation, at a concentration of 10 μM in an enzymatic assay. Molecular
docking studies were performed to identify the mechanism of action of the test
compounds and good binding interactions were observed. The results of the docking
studies supported the conclusion that the ring size of the heterocycle at the 4-position
in 1-substituted-3-(6-methylpyridin-2-yl)pyrazoles may be important for good ALK5
inhibition. Furthermore, the present study showed that J-1012 can inhibit HIF-1α
activity in HCT116 cells. In addition, we elucidated an important mechanism of the
anticancer activity of J-1012, related to cell invasion, metastasis, proliferation and
apoptosis, which are essential factors for tumor progression. This mechanism may in
part explain the anticancer effect of J-1012, and provides a rationale for the
development of J-1012 as an anticancer drug. Further studies need to be performed in
colon cancer patients to determine the clinical utility of J-1012.
Acknowledgments

This work was partially supported by the National Science Foundation of China
(No. 81560557, 81760657 and 81660608) and the Education Department of Jilin
Province
Scientific
Research
Fund
Project
(No.
2016-283).
This work was partially supported by Jilin Province Science and Technology Develop
ment Outstanding Young Talents (20180520057JH) and Yanbian University Youth
Research Fund Project (2017.No. 31) and the 13th five-year program of science and
technology of the ministry of education of jilin province (JJKH20191152KJ).
We
thank Victoria Muir, PhD, from Liwen Bianji, Edanz Group China
(www.liwenbianji.cn/ac), edited the English text of a draft of this manuscript.
Supplementary data
Supplementary data associated with this article can be found, in the online version.

References
1. Wu J, Zhang ZH, Zhang LH, et al. Bioorg Med Chem Lett. 2019;29:853.
2. Reddy CN, Nayak VL, Mani GS, et al. Bioorg Med Chem Lett. 2015;25:4580.
3. Kamel MM, Megally Abdo NY. Eur. J. Med. Chem. 2014;86:75.
4. Muhamad N, Plengsuriyakarn T, Na-Bangchang K. Int J Nanomedicine. 2018;13:3921.
5. Maxwell PH, Dachs GU, Gleadle JM, et al. Proc Natl Acad Sci USA. 1997;94:8104.
6. Lando D, Peet DJ, Gorman JJ, et al. Genes Dev. 2002;16:1466.
7. Jiang BH, Zheng JZ, Leung SW, et al. J. Biol. Chem. 1997;272:19253.
8. Gray MJ, Zhang J, Ellis LM, et al. Oncogene. 2005;24:3110.
9. Masoud GN, Li W. Acta. Pharm. Sin B. 2015;5:378.
10. Blouin CC, Pagé EL, Soucy GM, et al. Blood. 2004;103:1124.
11. Haddad JJ, Land SC. FEBS Lett. 2001;505:269.
12. Patten DA, Lafleur VN, Robitaille GA, et al. Mol Biol Cell. 2010;21:3247.
13. Pagé EL, Robitaille GA, Pouysségur J, et al. J Biol Chem. 2002;277:48403.
14. Serita S, Tomokiyo A, Hasegawa D, et al. Arch Oral Biol. 2017;78:135.
15. Schmierer B, Hill CS. Nature Rev Mol Cell Biol. 2007;8:970.
16. Jeong BY, Park SR, Cho S, et al. J Antimicrob Chemoth. 2018;73:962.
17. Lim H, Zhu YZ. Cell Mol Life Sci. 2006;63:2584.
18. Gu L, Zhu YJ, Yang X, et al. Acta Pharmacol Sin. 2007;28:382.
19. Zhang D, Hao X, Xu L, et al. Oncol Lett. 2017; 14:4511.
20. Zhao LM, Guo Z, Xue YJ, et al. Molecules. 2018;23:3369/1.
21. Jin CH, Sreenu D, Krishnaiah M, et al. Eur J Med Chem. 2011;46:3917.
22. Hendrik U, Frank G, Roland K, et al. Int J Mol Sci. 2018;19:1568/1.
23. Tobias B, Benjamin R, Dirk R, et al. Mol Cancer. 2015;14:199/1.
24. Dong M, Blobe GC. Blood. 2006;107:4589.
25. Byfield SD, Major C, Laping NJ, et al. Mol Pharmacol. 2004;65:744.
26. Uhl M, Aulwurm S, Wischhusen J, et al. Cancer Res. 2004;64:7954.
27. Tojo M, Hamashima Y, Hanyu A, et al. Cancer Sci. 2005;96:791.
28. Bueno L, De Alwis DP, Pitou C, et al. Eur J Cancer. 2008;44:142.
29. Jin CH, Krishnaiah M, Sreenu D, et al. J Med Chem. 2014;57:4213.
30. Giannelli G, Villa E, Lahn M. Cancer Res. 2014;74:1890.
31. Mundla SR. World Patent WO 2007018818 A1. 2007.
32. Fujiwara Y, Nokihara H, Yamada Y, et al. Cancer Chemother Pharmacol. 2015;76:1143.
33. Brandes AA, Carpentier AF, Kesari S, et al. Neuro-Oncology. 2016;18:1146.
34. Park SA, Kim MJ, Park SY, et al. Cell Mol Life Sci. 2015;72:2023.
35. Jin CH, Krishnaiah M, Sreenu D, et al. Bioorg Med Chem. 2011;19:2633.

36. Zhu WJ, Cui BW, Wang HM, et al. Eur J Med Chem. 2019;180:15.
37. Journet M, Cai D, Larsen RD, et al. Tetrahedron Lett. 1998;39:1717.
38. Sun LH, Niu DQ, Lee WC, et al. World Patent WO 2009009059 A1. 2009.
39. Mridula S, Agarwal SK, Patnaik GK, et al. J Med Chem. 1990;33:2970.
40. Hamed A, Christoph B, Christine B, et al. World Patent WO 2012063207 A1. 2012.
41. Eyers PA, Craxton M, Morrice N, et al. Chem Biol. 1998;5:321.
42. Zhao LM, Guo Z, Xue YJ, et al. Molecules. 2018;23:3369/1.
43. Chen L, Mayer JA, Krisko TI, et al. Cancer Res. 2009;69:8853.
44. Dyrager C, MÖllers LN, Kjäll LK, et al. J Med Chem. 2011;54:7427.
45. Wang ZC, Qin YJ, Wang PF, et al. Eur J Med Chem. 2013;66:1.
46. Lopes RDO, Romeiro NC, Lima CDFD, et al. Eur J Med Chem. 2012;54:264.
47. Gill K, Singh AK, Kapoor V, et al. Biochim Biophys Acta. 2013;1830:2763.
48. Tang M, Tian Y, Li D, et al. PLoS One. 2014;9:102967/1.
49. Wang JY, Wang Z, Li M, et al. Chem Biol Interact. 2018;296:134.
50. Greijer AE, van der Wall E. J Clin Pathol. 2004;57:1009.
51. Zhou B, Zhan H, Tin L, et al. Cancer Lett. 2016;382:11.
52. Semenza GL. Nat Rev Cancer. 2003;3:721.
53. Zheng LC, Wang XQ, Lu K, et al. Oncotarget. 2017;8:41348.
54. Wilson WR, Hay MP. Nat Rev Cancer. 2011;11:393.
55. Dai Y, Bae K, Siemann DW. Int J Radiat Oncol Biol Phys. 2011;81:521.
56. Loboda A, Jozkowicz A, Dulak J. Mol Cells. 2010;29:435.
57. Wang G, Wang JJ, Fu XL, et al. Oncol Rep. 2017;37:657.
58. Semenza GL. J Clin Invest. 2013;123:3664.
59. Semenza GL, Roth PH, Fang HM, et al. J Biol Chem. 1994;269:23757.
60. Vaupel P. Oncologist. 2004;9:10.
61. Wouters A, Pauwels B, Lardon, F, et al. Oncologist. 2007;12:690.
62. Petersen LH, Dzialowski E, Huggett DB. Comp Biochem Physiol A Mol Integr Physiol.
2011;160:311.

Figure captions:
Table 1. Residual activities of compounds 11a–d, 12a–d, 18a–d, 19a–d, 20a–d, and
21a–d on ALK5 and p38α MAP kinase
Table 2. The IC50 of compounds 20a–d and 21a–d on ALK5 and p38α MAP kinase
Fig. 1. ALK5 inhibitors under development
Fig. 2. Design strategy based on compounds 4 and 5
Fig. 3. Docking of compounds 20b and 21b in the active site of ALK5
Fig. 4. Docking of compounds 20b and 21b in the active site of p38α
Fig. 5. Identification of J-1012 as a HIF-1 pathway inhibitor from a cell-based
screening assay
Fig. 6. Effect of J-1012 on HIF-1α protein synthesis
Fig. 7. Effect of J-1012 on the TNF-induced apoptosis in HCT116 cells
Fig. 8. Effect of J-1012 on the TNF-induced invasion
Fig. 9. J-1012 inhibits the proliferation of cells.
Scheme 1. Synthetic scheme for the synthesis of compounds 11a–d and 12a–d
Scheme 2. Synthetic scheme for the synthesis of compounds 18a–d and 19a–d
Scheme 3. Synthetic scheme for the synthesis of compounds 20a–d and 21a–d

Fig. 1. ALK5 inhibitors under development