Discovery of a novel selective water-soluble SMAD3 inhibitor as an antitumor agent

Article abstract of DOI:10.1016/j.bmcl.2020.127396

Targeting the SMAD3 protein is an attractive therapeutic strategy for treating cancer, as it avoids the potential toxicities due to targeting the TGF-β signaling pathway upstream. Compound SIS3 was the first selective SMAD3 inhibitor developed that had acceptable activity, but its poor water solubility limited its development. Here, a series of SIS3 analogs was created to investigate the structure–activity relationship for inhibiting the activation of SMAD3. On the basis of this SAR, further optimization generated a water-soluble compound, 16d, which was capable of effectively blocking SMAD3 activation in vitro and had similar NK cell-mediated anticancer effects in vivo to its parent SIS3. This study not only provided a preferable lead compound, 16d, for further drug discovery or a potential tool to study SMAD3 biology, but also proved the effectiveness of our strategy for water-solubility driven optimization.

Full text of DOI:10.1016/j.bmcl.2020.127396

Journal Pre-proofs
Discovery of a Novel Selective Water-Soluble SMAD3 Inhibitor as an Antitu‐
mor Agent
Nannan Wu, Guangyu Lian, Jingyi Sheng, Dan Wu, Xiyong Yu, Huiyao Lan,
Wenhui Hu, Zhongjin Yang
PII:
S0960-894X(20)30507-2
DOI:
Reference:
https://doi.org/10.1016/j.bmcl.2020.127396
BMCL 127396
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
22 May 2020
2 July 2020
6 July 2020
Please cite this article as: Wu, N., Lian, G., Sheng, J., Wu, D., Yu, X., Lan, H., Hu, W., Yang, Z., Discovery of a
Novel Selective Water-Soluble SMAD3 Inhibitor as an Antitumor Agent, Bioorganic & Medicinal Chemistry
Letters (2020), doi: https://doi.org/10.1016/j.bmcl.2020.127396
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.
© 2020 Elsevier Ltd. All rights reserved.

Discovery of a Novel Selective Water-Soluble SMAD3 Inhibitor as an Antitumor
Agent
Nannan Wu, †, # Guangyu Lian, ‡,# Jingyi Sheng, ‡,# Dan Wu†, Xiyong Yu, † Huiyao
Lan,‡,* Wenhui Hu, †,* and Zhongjin Yang †,*
†Key Laboratory of Molecular Target & Clinical Pharmacology and State Key
Laboratory of Respiratory Disease, School of Pharmaceutical Sciences & the Fifth
Affiliated Hospital, Guangzhou Medical University, Guangzhou, Guangdong, 511436,
China.
‡ Department of Anatomical and Cellular Pathology, The Chinese University of Hong
Kong, Hong Kong SAR 999077, China.
# These authors contributed equally to this work.
* Corresponding authors:
For Zhongjin Yang: phone: +86 20 31100920; E-mail: yzj23@163.com.
For Wenhui Hu: phone: +86 20 31100920; E-mail: huwenhui@gzhmu.edu.cn.
For Huiyao Lan: E-mail: email: hylan@cuhk.edu.hk.
1

Abstract
Targeting the SMAD3 protein is an attractive therapeutic strategy for treating
cancer, as it avoids the potential toxicities due to targeting the TGF-β signaling pathway
upstream. Compound SIS3 was the first selective SMAD3 inhibitor developed that had
acceptable activity, but its poor water solubility limited its development. Here, a series
of SIS3 analogs was created to investigate the structure−activity relationship for
inhibiting the activation of SMAD3. On the basis of this SAR, further optimization
generated a water-soluble compound, 16d, which was capable of effectively blocking
SMAD3 activation in vitro and had similar NK cell-mediated anticancer effects in vivo
to its parent SIS3. This study not only provided a preferable lead compound, 16d, for
further drug discovery or a potential tool to study SMAD3 biology, but also proved the
effectiveness of our strategy for water-solubility driven optimization.
KEYWORDS: SMAD3 inhibitor, tumor, NK cell, tumor microenvironment.
2

TGF-β /SMAD signaling plays an important role in cancer progression, switching
from tumor suppression to tumor promotion.1,2 In fact, TGF-β signaling is often
overactivated in many advanced tumor types.3-9 This can cause cancer cells to lose their
epithelial characteristics, such as cell-cell adhesion and cell polarity, and acquire
migratory and invasive properties.2 Moreover, TGF-β/SMAD signaling can fuel
heterogeneity in cancer stem cells and drug resistance,10, 11 and is associated with poor
prognosis in patients.3 It is noted that TGF-β receptors not only activate SMAD-
independent canonical pathways, but also allow SMAD-independent signaling
responses and crosstalk with different signaling pathways (Figure 1).12
Figure 1. Overview of TGF‑β/SMAD signaling.
Due to TGF-β’s pleiotropic roles and complex functions, its inhibition can lead to
side effects.13, 14 SMAD3, but not SMAD2 and SMAD4, is a key downstream mediator
of TGFβ signaling, because it contains DNA binding domains that allow for it to bind
directly to the promoters in its target genes and thereby modulate transcription.15,16
Notably, SMAD3 has attracted an extraordinary amount of research interest due to its
ability to modulate the expression of a key protein that is a target for the
3

immunotherapeutic treatment of cancer. More specifically, SMAD3 can upregulate the
expression of programmed death-1 (PD-1) on T cells, which is a coinhibitory receptor
that inhibits the activity of tumor-infiltrating lymphocytes (TILs) in cancer. It should
be noted that SMAD2 cannot upregulate the expression of PD-1.17 In addition, it has
been shown that silencing of SMAD3 can suppress cancer cell growth and metastasis
by enhancing the cancer-killing activity of NK cells.18,19 Thus, the selective inhibition
of the SMAD3 protein with a potent, low toxicity, drug could provide a promising
anticancer treatment.
SIS3, a selective SMAD3-phosphorylation inhibitor (Figure 1)20, contains a
Michael acceptor, but our experiment showed it had no addition reactivity with
nucleophilic groups from glutathione. Therefore, SIS3 cannot covalently bind to the
targets, and not belong to Pan Assay INterference compoundS (PAINS). It has been
proven to effectively enhance the anticancer activities of NK cells via an E4BP4-
dependent mechanism and significantly suppress tumor growth and invasion.18
However, SIS3 suffers from insolubility in water, actually dissolved in a mixed solution
of 2% DMSO, 2% Tween-80 and 96% H2O for in-vivo animal assays. The poor water
solubility can also cause unstable results of biological assay, poor pharmacokinetic
properties,21,22 and lead to real problems in developing acceptable formulations in the
later stages of development.23
These potential risks directed us to improve the water solubility of SIS3. As a drug-
like parameter reflecting water solubility, the calculated lipophilicity (cLogP) of SIS3
(4.7) is high. Generally, a lower cLogP, as well as a lower molecular weight (MW), are
favored to achieve good pharmacokinetic properties.24 However, both cLogP and MW
tend to rise during lead compound optimization.25 From this perspective, it is preferable
to develop a lead compound that has both a low cLogP and a low MW.
4

Scheme 1. Synthesis of SIS3 and its derivatives 5b~j, 9, 12, 13a~h and 14a~ba
O
O
N
O
X =
R4
OH
O
11
10
CHO
R1
OH
R2
R1
N
R2
R1
Y =
e or f
a, b, c or d
OH
O
B
B
N
B
N
A
A
N
A
Z =
O
O
N
4
3
5
O
A = N; B = C
: R1 = Ph, R2 = H, R4 = X, 10,11
: R1 = Ph, R2 = H, R4 = X
3a
3b
3c
4a
4b
: R1 = Ph
: R1 = Ph, R2 = H
SIS3
g
: R1 = H
5b
5c
: R1 = Ph, R2 = F
: R1 = Ph, R2 = H, R4 = Y, 10,11
: R1 = Cyclopropyl
4c
: R1 = Ph, R2 = CN
: R1 = H, R2 = H
: R1 = Ph, R2 = H, R4 = Z, 10,11
: R1 = Ph, R2 = F, R4 = X, 10,11
: R1 = Ph, R2 = CN, R4 = X, 10,11
5d
5e
5f
4d
4e:
R1 = Cyclopropyl, R2 = H
: R1 = H, R2 = H, R4 = X, 10,11
5g
: R1 = Cyclopropyl, R2 = H, R4 = X, 10,11
5h
A = C; B = C
3d
: R1 = Ph, R2 = H, R4 = X, 10,11
4f
: R1 = Ph
A = C; B = N
: R1 = Ph, R2 = H
5i
O
: R1 = Ph, R2 = H, R4 = X, 10,11
3e
4g
: R1 = Ph
: R1 = Ph, R2 = H
O
O
5j
OH
O
O
O
O
N
O
h
i
e
O
N
N
N
N
N
N
N
6
7
N
8
9
O
O
OH
N
O
O
H
j
O
e
N
N
N
N
N
N
N
N
N
10
11
12
*
13a
13b
: S, R1 = H, R2 = OMe
O
H
*

: R, R1 = H, R2 = OMe
N
13c: *
S, R1 = OH, R2 = OMe
: S, R1 = H, R2 = OH
: *R, R1 = H, R2 = OH
: *S, R1 = OH, R2 = OH
: *R, R1 = Cl, R2 = OMe
R1
COR2
*
13d
13e
13f
e
N
N
13g
13h
13a h
~
4a
N
: *R, R1 = H, R2=
N
e
R3
H
O
N
N
N
14a
14b
: R3
=
=
N
N
: R3
N
N
14a~b
aReagents and conditions: (a) methyl diethylphosphonoacetate, NaH, DMF, 0oC to r.t., 8 h; then add NaOH,
MeOH, and H2O; (b) malonic acid, piperidine, pyridine, 110oC, 6 h (c) Ethyl fluoroacetate, TiCl4, TEA, CH2Cl2, r.t.,
4 h; then add LiOH, THF, H2O, 3 h; (d) t-butyl cyanoacetate; piperidine, AcOH, toluene, 100oC, 15 h; then add TFA,
CH2Cl2, r.t., 12 h; (e) HATU, TEA, corresponding amines, DMF, 2 h. (f) oxalyl chloride, DMF, THF; then add
DIPEA and amine X. (g) H2, Pd/C, MeOH, r.t. overnight. (h) methyl chlorooxoacetate, AlCl3, CH2Cl2, 0oC to r.t.,
5

overnight; (i) LiOH, MeOH, H2O, r.t. overnight. (j) t-butyl diethylphosphonoacetate, then add TFA, CH2Cl2, r.t., 12
h.
To achieve this objective, we firstly designed and synthesized SIS3 analogs to
study its structure−activity relationships (SAR) to determine the impact of molecular
fragment on anti-SMAD3-activation. As shown in Scheme 1, the synthesis started with
the known aldehydes 3a~e, which were readily obtained from indoles, according to
similar reported procedures.26 A Horner−Wadsworth−Emmons (HWE) reaction of
compounds 3a, 3b, or 3d and methyl diethylphosphonoacetate was followed by ester
hydrolysis to generate the corresponding acids, 4a, 4d, or 4f. Alternatively, compounds
4e or 4g were obtained by the Knoevenagel reaction of 3c or 3e and malonic acid.27
Compound 3a and ethyl fluoroacetate were treated with TiCl4 and TEA in CH2Cl2 to
afford the desired 4b.28 Similar to the synthetic method for 4e, treatment of 3a with t-
butyl cyanoacetate in the presence of piperidine, followed by removal of the t-butyl
group with TFA provided 4c. Treatment of the acids 4a~d with oxalyl chloride or
HATU, and then condensation with the corresponding secondary amines gave amides
SIS3 and 5c~j.29 Similarly, using aldehyde 10 as the substrate, amide 12 was obtained
in two steps.30 Reduction of the double bond in SIS3 with H2 and Pd/C in MeOH yielded
5b.31 Finally, compound 9 was prepared in three steps as follows: 7-azaindole 6 was
treated with AlCl3 in CH2Cl2 at room temperature followed by the addition of methyl
oxalyl chloride to yield compound 7;32 Next, compound 7 was hydrolyzed to the acid,
and then condensed with 6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline to obtain the
desired compound 12. With 4a in hand, compounds 13a-h and 14a-b were easily
obtained by a condensation reaction with the corresponding amines in the presence of
HATU.
6

Table 1. Inhibition of SMAD3-phosphorylation by SIS3, 4a, 5b~h, 9, and 12.
B
O
19
O
13
A
18
O
N
11
16
10
3
N
N
1
7
C
D
Region
Inhibition %a
Cmpd
A
B
C
D
5 μM
10 μM
SIS3
5b
–
–
–
–
–
51.8±1.4
79.3±0.2
O
–
–
10.2±13.0
12.3±4.9
O
9
–
–
–
–
–
–
–
-25.2±4.2 -40.7±1.5
O
O
O
5e
56.4±5.3
-13.5±6.7
95.1±1.2
-2.6±9.7
F
5f
4a
5c
–
–
–
–
–
–
CN
–
OH
-9.2±0.3
59.8±5.8
2.6±7.9
OH
OH
N
–
–
–
–
–
–
–
91.1±5.2
OMe
O
O
N
2
5d
5i
–
–
–
–
–
44.6±5.9
45.7±7.1
30.8±5.7
19.1±7.8
29.7±12.9
18.5±14.3
78.5±3.2
54.0±11.6
53.1±2.9
52.1±6.3
48.7±6.7
53.7±5.2
2
OMe
–
–
–
–
–
N
N
12
5j
N
N
N
N
5g
5h
–
H
–
a The % inhibition values are the means of three experiments.
7

To evaluate the ability of the compounds to inhibit SMAD3 activation, all the
synthesized compounds were tested for their ability to inhibit SMAD3 phosphorylation
at two different concentrations (5 μM and 10 μM) using a CAGA-based luciferase
reporter assay.33 As shown in Table 1, the parent compound SIS3 inhibited SMAD3
phosphorylation by 51.8% at 5 μM and 79.3% at 10 μM. To confirm the tolerability of
substitution in the A-region, various functional groups were introduced into this region.
Among these, compound 5b or 9, derived from a reduction or carbonyl substitution of
the 10,11-double bond in SIS3, exhibited diminished or lost inhibitory activity. A
higher reactive Michael acceptor α-cyanoacrylamide (5f) resulted in a complete loss of
activity, which indicated they were reversible small molecule inhibitors. In stark
contrast, fluorination of the 11-position in SIS3 (5e, 95% inhibition at 10 μM) slightly
improved activity, possibly due to the fact that fluorine is more lipophilic than hydrogen
and significantly more lipophilic than C=O or C≡N substituents. In the B-region, a
complete loss of activity was observed following the removal of the
tetrahydroisoquinoline moiety (4a), while elimination or substitution of the two methyl
groups on the phenyl ring (5c or 5d) of SIS3 had either a slightly beneficial effect or no
effect on activity, respectively. For the C-region and D-region modified compounds, a
moderate decrease was observed in their inhibition of SMAD3 phosphorylation. For
example, 5i, 5j, and 12 all showed a reduced inhibition of SMAD3 phosphorylation
(approximately 50% at 10 μM for all three compounds). A similar result was seen upon
replacement of the phenyl group with a hydrogen atom (5g, 48.7% inhibition at 10 μM)
or a cyclopropyl group (5h, 53.7% inhibition at 10 μM). These findings indicate that
the 10,11-double bond is essential for inhibitory activity towards SMAD3
phosphorylation, and increasing lipophilicity in the A-region can have a beneficial
effect on activity. Notably, the C-18 or C-19 positions in the tetrahydroisoquinoline
moiety can tolerate certain types of groups, such as hydroxyl and ether groups.
8

Table 2. Inhibition of SMAD3 phosphorylation by compounds 13a~h and 14a~b
O
H
N

R1
COR2
O
N
O
N
13a h
~
O
N
R3
O
N
15
N
N
N
SIS3
N
N
14a~b
Group
R1
Inhibition, %a
Cmpd
Chirality
S
R2
R3
5 μM
10 μM
13a
13b
13c
13d
13e
13f
H
H
OMe
n.a. b
32.4±3.0
47.6±8.0
41.9±10.4
1.8±29.3
19.6±12.3
< 0%
67.1±6.9
R
S
OMe
OMe
OH
n.a. b
n.a. b
n.a. b
n.a. b
n.a. b
n.a. b
n.a. b
51.4±3.3
62.6±9.5
6.6±9.6
31.2±10.5
< 0%
OH
H
S
R
H
OH
S
OH
Cl
OH
13g
13h
14a
R
OMe
44.8±3.1
n.d.c
55.3±8.7
< 0%
N
H
R
H
N
N
N
n.a. b
n.a. b
n.a. b
26.4±17.0
37.3±0.6
9

n.a. b
n.a. b
n.a.b
N
14b
35.4±8.0
69.0±5.2
a The % inhibition values are the means of three experiments. bn.a. means not applicable.
On the basis of the above analysis, we chose the well-tolerated
tetrahydroisoquinoline moiety as a region for further modification. By utilizing drug
design strategies including ring variation, scaffold hopping, and bioisosteres, two series
of compounds were designed, synthesized, and evaluated for their inhibitory activity
towards SMAD3 phosphorylation (Table 2). To begin with, a polar carboxyl group was
introduced at the C-15 position, with the six-membered ring being opened, to construct
commercially available phenylalanine fragments. As shown in Table 2, compound 13d
(1.8% inhibition at 5 μM), 13e (19.6% inhibition at 5 μM), and 13f (< 0% inhibition at
5 μM), with the introduction of a carboxyl group at the C-15 position, exhibited a
significant drop in inhibition compared to SIS3. Unsurprisingly, compound 13h with a
polar and hydrophilic fragment at the C-15 position lost inhibition, even at 10 μM.
When the carboxyl group was changed to a lipophilic ester group, such as in compounds
13a, 13b, and 13c, a recovery was observed in the inhibition of anti-SMAD3
phosphorylation (Table 2). Comparing the % inhibition of 13a (32.4%) with 13b
(47.6%), and 13d (1.8%) with 13e (19.6%), all at 5 μM, the R configuration at the C-
15 position appeared to have a weak advantage over the S configuration. Similar to the
data shown in Table 1, substitution of the phenyl ring with various groups (13a, 13b,
and 13g) had no obvious effects on inhibitory activity. In another series of compounds,
conversion of the benzene ring to an electron-deficient pyridine or pyrimidine ring, 14a
and 14b, resulted in reduced inhibitory activity at 5 μM.
10

Relatively wide
tolerance
O
O
Essential to activity
O
N
16
10
SIS3
15
Lipophilic group
favored
11
Ph > cyclopropyl, H
N
N
N atom at
positon 7 required
Figure 2. The preliminary structure−activity relationship of SIS3
Figure 2 summarizes the conclusions drawn from these studies on the effect of
structural changes within the four domains on the ability to inhibit SMAD3
phosphorylation. These SARs provide further guidance for the design of new optimized
SIS3 analogues.
Scheme 2. Synthesis of SIS3 derivatives 16a~da
O
16a
: n=1, R1 =
O
O
O
N
O
R1
O
O
O
O
NH
n
N
16b
16c
16d
O
: n=2, R1 =
: n=1, R1 =
: n=1, R1 =
n
n
b
a
N
N
N
N
N
N
N
N
N
3a
15a:
15b
n=1
: n=2
16
a Reagents and conditions: (a) LDA, THF, 0oC to 45oC, 48 h; (b) NaH, DMF, 0oC to r.t., overnight.
Based on the above SAR findings, we reasoned that in order to improve the water-
solubility properties, the introduction of polar or hydrophilic substituents around the
tolerant region A might improve solubility. Considering that increasing the lipophilicity
at the C-11 and C-15 positions had beneficial effects, the 15, 16- ethylene group was
transferred to the C-11 position to form new skeleton compounds 16a-d. Their synthesis
route is shown in Scheme 2: the 3-alkylidenelactams 15a and 15b were prepared by
11

aldol condensation of aldehyde 3a and lactams (acetamide protecting group),34 and then
treatment of 15a or 15b with the corresponding alkyl halides or methanesulfonate in
the present of NaH resulted in the desired products 16a~16d.35
Table 3. Inhibition of SMAD3-phosphorylation by compounds 16a~d
O
R3
O
O
N
O
N
n
15
N
N
N
N
SIS3
16a~d
Group
Inhibition %a
Cmpd
clogPb
n
R3
5 μM
10 μM
SIS3
16a
16b
16c
n.a.c
n.a.c
4.7
4.5
5.0
4.7
3.7
51.8±1.4
79.3±0.2
91.2±1.1
86.7±2.1
3.8±10.3
62.6±3.4
O
1
2
1
1
54.2±1.9
57.2±5.9
-10.2±3.4
44.4±7.6
O
O
O
O
N
N
16d
N
a The % inhibition values are the means of three experiments. bCalculated logarithm of the octanol/water distribution
coefficient using ChemBioDraw Ultra 14.0. c n.a. means not applicable.
12

As shown in Table 3, 16a did show a slight increase in inhibitory activity (91.2%)
at 10 μM while maintaining activity at 5 μM (54.2%). However, compound 16b, which
contained three carbon atoms in the new linker region, did not show increases in
inhibitory activity at either 5 μM or 10 μM. While replacing the dimethoxybenzyl in
16a with a polar bioisostere appeared to be a good strategy to improve water-solubility,
unfortunately, when the dimethoxybenzyl group was replaced with 4-methoxy-3,5-
dimethylpyridyl in compound 16c, the % inhibition dropped to 3.8% at 10 μM. As
another bioisosteric replacement, the introduction of 2,3,5-trimethylpyrazyl in place of
the dimethoxybenzyl group (16d), resulted in slight reductions in the % inhibition
values, being 44.4% at 5 μM and 62.6% at 10 μM. In spite of this, its cLogP value was
remarkably improved from 4.7 to 3.7.
Table 4. Drug-like properties comparison of SIS3 and 16d
b,c
b,d
Solubility
ED50
TD50
Cmpd MW clogPa
μM
μM
in water (HCl)
in PBS (pH 7.4)
SIS3 453
16d 437
4.7
3.7
4.53
18.66
0.006 mg/mL
0.141 mg/mL
> 94 mg/mL
6.63
22.04
0.014 mg/mL
a Calculated logarithm of the octanol/water distribution coefficient using ChemBioDraw Ultra 14.0. bThe inhibition
c
concentration are the means of three experiments. Concentration required to inhibit SMAD3 phosphorylation by
d
50% relative to controls. Concentration required to inhibit A549 cell growth by 50% relative to controls as
determined by MTT assay.
Compared with SIS3, our lead compound 16d showed similar inhibitory activity
towards SMAD3 phosphorylation and had similar cytotoxicity towards A549 cells as
determined by an MTT assay (Table 4). Furthermore, 16d had a decreased MW and a
reduced cLogP value of 3.7, which could better meet Lipinski’s “rule of five”. The
solubilities of compound 16d or its HCl salt were determined in both pure water and
phosphate buffer (pH 7.4) at room temperature by HPLC (Table 4)36. It was found that
the solubility of 16d as a free base (14 μg/mL) was 2-fold greater than SIS3 in PBS (6
13

μg/mL), while the solubility of 16d as the HCl salt (> 94 mg/mL) was 666-fold greater
than SIS3 (0.141 mg/mL). Obviously, the improvement in aqueous solubility is
attributable to the pyrazyl ring, which is more water soluble than the
tetrahydroisoquinoline moiety in SIS3.
Figure 3. Compounds 16a and 16d reduce the TGF-β-dependent increase in
phosphorylation of SMAD3. A549 cells were stimulated with 1 ng/mL TGF-β for 30
minutes with or without 2 hours pre-treatment with 16a or 16d at both 5 and 10 M.
Cells were lysed and the levels of phospho-SMAD3, SMAD3, phospho-SMAD2, and
SMAD2 were determined by western blot. GAPDH levels were used to ensure equal
protein loading.
In order to accurately assess selectivity, we chose two compounds, namely 16a
and 16d, to examine their effect on the TGF-β-induced phosphorylation of SMAD3 and
SMAD2. As shown by the results of a western blot analysis (Figure 3), the
phosphorylation of SMAD3, but not of SMAD2, was inhibited by both 16a and 16d.
14

These data suggest that compound 16d and its analog 16a are specific inhibitors of
SMAD3 phosphorylation.
Figure 4. Compound 16d suppresses cancer progression by enhancing NK cell
accumulation in the syngeneic Lewis lung carcinoma (LLC) model mice. (A)Tumor
volumes were measured every five days, and tumor weights were measured at the end
of the experiment. Data are expressed as the mean ± SD for groups of five mice. ns, no
significant difference, *p < 0.05, ***p < 0.001. (B) Representative images showing
immunofluorescent staining of NK1.1 (green) are shown. The nuclei were
counterstained with DAPI (blue). The percentage of NK cells in the tumor
microenvironment are shown in the right panel. Each bar represents the mean ± SD for
groups of three mice. *** p < 0.001 vs 16d – 0 μg/g, ### p < 0.001 vs 16d –1.25 μg/g.
15

The anticancer activity of 16d in vivo was evaluated in a syngeneic Lewis lung
carcinoma (LLC) model in C57BL/6 mice. Tumor bearing mice were randomized and
placed into different treatment groups and then treated intraperitoneally with SIS3 (a
solution of 2% DMSO, 2% Tween-80 and 96% H2O) or 16d (a solution of 2% Tween-
80 and 98% H2O) at the dosage of 0, 1.25, 2.5 or 5.0 μg·g-1·day-1 (i.p.) for 20
consecutive days (Figure 4A). In vivo treatment with 16d significantly reduced the
volume and weight of the LLC tumors in mice in a dose-dependent manner. Moreover,
there was no significant difference in the tumor suppressive effect between SIS3 and
16d in terms of the size and weight of the lung tumors in mice. Next, we investigated
the mechanisms through which the SMAD3 signaling inhibitor 16d attenuates LLC
progression. As previously reported, deletion or inactivation of Smad3 largely promotes
NK cell mediated immunity against tumors.17,37 Therefore, we examined the impact of
16d on NK cell accumulation in the tumor microenvironment. As shown in Figure 4B,
treatment with 16d increased the number of tumor-infiltrating NK cells in a dose-
dependent manner. Compared with the control mice, treatment with 16d at 2.5 or 5.0
μg/g significantly increased NK cell accumulation, which indicated the presence of an
enhanced antitumor immune response. These results suggest that compound 16d may
markedly suppress cancer progression by enhancing NK cell-mediated anticancer
immunity in LLC model mice.
In conclusion, modifications based on SIS3 resulted in the discovery of a library
of anti-SMAD3-phosphorylation agents. According to these results, the following
SARs were found: (1) the 10,11-double bond is essential for anti-SMAD3
phosphorylation activity; (2) diverse substitutions on C-18 and C-19 can be tolerated;
(3) variations in the 7-azaindole ring and the 2-phenyl group can affect the inhibitory
activity. Compared with SIS3, the novel lead compound 16d has higher water solubility,
and demonstrates a similar potency for the inhibition of SMAD3-phosphorylation. In
addition, 16d also had a high anticancer effect by enhancing NK cell-mediated
immunity in vivo. The work reported herein provides a potential tool to study SMAD3
biology or a desirable lead compound for further drug discovery.
16

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (NSFC)
(NO. 81803364 to Z. Yang and NO. 81872743 to W. Hu).
Appendix A. Supplementary data
The following are the Supplementary data to this article. Supporting data to this article
can be found online.
References
1. Christofori, G. Nature 2006, 441, 444.
2. Massague, J. Cell 2008, 134, 215.
3. Bruna, A.; Darken, R. S.; Rojo, F.; Ocana, A.; Penuelas, S.; Arias, A.; Paris, R.; Tortosa, A.; Mora,
J.; Baselga, J.; Seoane, J. Cancer cell 2007, 11, 147.
4. Watanabe, T.; Wu, T. T.; Catalano, P. J.; Ueki, T.; Satriano, R.; Haller, D. G.; Benson, A. R.;
Hamilton, S. R. N Engl J Med 2001, 344, 1196.
5. Drabsch, Y.; Ten, D. P. J Mammary Gland Biol Neoplasia 2011, 16, 97.
6. Wiercinska, E.; Naber, H. P. H.; Pardali, E.; Pluijm, G.; Dam, H.; Dijke, P. Breast Cancer Res.
Treat 2011, 128, 657.
7. Arteaga, C. L. Curr Opin Genet Dev 2006, 16, 30.
8. Mundy, G. R. Nat Rev Cancer 2002, 2, 584.
9. Guise, T. A.; Chirgwin, J. M. Clin Orthop Relat Res 2003, S32.
10. Oshimori, N.; Oristian, D.; Fuchs, E. Cell 2015, 160, 963.
11. Tripathi, V.; Sixt, K. M.; Gao, S.; Xu, X.; Huang, J.; Weigert, R.; Zhou, M.; Zhang, Y. E. Mol cell
2016, 64, 549.
12. Akhurst, R. J.; Hata, A. Nat Rev Drug Discov 2012, 11, 790.
13. Shull, M. M.; Ormsby, I.; Kier, A. B.; Pawlowski, S.; Diebold, R. J.; Yin, M.; Allen, R.; Sidman,
C.; Proetzel, G.; Calvin, D.; Annunziata, N.; Doetschman, T. Nature 1992, 359, 693.
14. Bierie, B.; Chung, C. H.; Parker, J. S.; Stover, D. G.; Cheng, N.; Chytil, A.; Aakre, M.; Shyr, Y.;
Moses, H. L. J Clin Invest 2009, 119, 1571.
15. Meng, X. M.; Nikolic-Paterson, D. J.; Lan, H. Y. Nat Rev Nephrol 2016, 12, 325.
16. Bae, E.; Sato, M.; Kim, R. J.; Kwak, M. K.; Naka, K.; Gim, J.; Kadota, M.; Tang, B.; Flanders, K.
C.; Kim, T. A.; Leem, S. H.; Park, T.; Liu, F.; Wakefield, L. M.; Kim, S. J.; Ooshima, A. Cancer
Res 2014, 74, 6139.
17

17. Park, B. V.; Freeman, Z. T.; Ghasemzadeh, A.; Chattergoon, M. A.; Rutebemberwa, A.; Steigner,
J.; Winter, M. E.; Huynh, T. V.; Sebald, S. M.; Lee, S. J.; Pan, F.; Pardoll, D. M.; Cox, A. L. Cancer
Discov 2016, 6, 1366.
18. Tang, P. M.; Zhou, S.; Meng, X. M.; Wang, Q. M.; Li, C. J.; Lian, G. Y.; Huang, X. R.; Tang, Y.
J.; Guan, X. Y.; Yan, B. P.; To, K. F.; Lan, H. Y. Nat Commun 2017, 8, 14677.
19. Wang, Q. M.; Tang, P. M.; Lian, G. Y.; Li, C.; Li, J.; Huang, X. R.; To, K. F.; Lan, H. Y. Cancer
Immunol Res 2018, 6, 965.
20. Jinnin, M.; Ihn, H.; Tamaki, K. Mol Pharmacol 2006, 69, 597.
21. Nassar, A. E.; Kamel, A. M.; Clarimont, C. Drug Discov Today 2004, 9, 1020.
22. Stella, V. J.; Nti-Addae, K. W. Adv Drug Deliv Rev 2007, 59, 677.
23. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv Drug Deliv Rev 2001, 46, 3.
24. Gleeson, M. P. J Med Chem 2008, 51, 817.
25. Oprea, T. I.; Davis, A. M.; Teague, S. J.; Leeson, P. D. J. Chem. Inf. Comput. Sci 2001, 41, 1308.
26. Nyerges, M.; Pintér, Á.; Virányi, A.; Bitter, I.; Tőke, L. Tetrahedron Lett 2005, 46, 377.
27. Kaminska, K.; Ziemba, J.; Ner, J.; Schwed, J. S.; Lazewska, D.; Wiecek, M.; Karcz, T.; Olejarz, A.;
Latacz, G.; Kuder, K.; Kottke, T.; Zygmunt, M.; Sapa, J.; Karolak-Wojciechowska, J.; Stark, H.;
Kiec-Kononowicz, K. Eur J Med Chem 2015, 103, 238.
28. Fujii, Y.; Wasano, N.; Tamura, N.; Shindo, M.; Matsumoto, K., Cis-cinnamic Acid Analogs and
Gravitropism Modifiers Containing Them. J.P. Patent 2016160246, 2016.
29. Yang, Z.; Kuang, B.; Kang, N.; Ding, Y.; Ge, W.; Lian, L.; Gao, Y.; Wei, Y.; Chen, Y.; Zhang, Q.
Eur J Med Chem 2017, 127, 296.
30. Kamal, A.; Bharath Kumar, G.; Lakshma Nayak, V.; Reddy, V. S.; Shaik, A. B.; Rajender; Kashi
Reddy, M. D. Med. Chem. Comm 2015, 6, 606.
31. Ma, Y.; Yin, D.; Ye, J.; Wei, X.; Pei, Y.; Li, X.; Si, G.; Chen, X.; Chen, Z.; Dong, Y.; Zou, F.;
Shi, W.; Qiu, Q.; Qian, H.; Liu, G. J. Med. Chem 2020, 63, 10, 5458.
32. Colley, H. E.; Muthana, M.; Danson, S. J.; Jackson, L. V.; Brett, M. L.; Harrison, J.; Coole, S. F.;
Mason, D. P.; Jennings, L. R.; Wong, M.; Tulasi, V.; Norman, D.; Lockey, P. M.; Williams, L.;
Dossetter, A. G.; Griffen, E. J.; Thompson, M. J. J. Med. Chem 2015, 58, 9309.
33. Luwor, R. B.; Wang, B.; Nheu, T. V.; Iaria, J.; Tsantikos, E.; Hibbs, M. L.; Sieber, O. M.; Zhu, H.
J. Growth Factors 2011, 29, 211.
34. Andreani, A.; Rambaldi, M.; Locatelli, A.; Leoni, A.; Bossa, R.; Chiericozzi, M.; Galatulas, I.;
Salvatore, G. J. Med. Chem 1993, 28, 825.
35. Liu, X.; Han, Z.; Wang, Z.; Ding, K. Angew Chem Int Ed Engl 2014, 53, 1978.
36. Gross, G.; Tardio, J.; Kuhlmann, O. Int J Pharm 2012, 437, 103.
37. Lian, G. Y.; Wang, Q. M.; Tang, P. M.; Zhou, S.; Huang, X. R.; Lan, H. Y. MOL THER 2018, 26,
2255.
Graphical Abstract
18

Relatively wide
tolerance
O
N
O
Essential to activity
O
N
O
N
N
SIS3
16d
Lipophilic group
favored
Ph > cyclopropyl, H
N
N
N
N
SIS3
Compared with
:
N atom at
positon 7 required
1. Lower cLogP and MW; higher water-
solubility
2. Similar potency in selectively inhibiting
SMAD3-activation
3. NK cell-mediated anticancer effect
19