Discovery, synthesis,andevaluation of small-molecule signal transducer and activator of transcription 3 inhibitors

Article abstract of DOI:10.1248/cpb.c12-00745

The signal transducer and activator of transcription 3 (STAT3) oncogene is a promising molecular target and its inhibitors have great potential as anticancer drugs. To identify novel and STAT3-selective inhibitors, a virtual screening based on Specs and M

Full text of DOI:10.1248/cpb.c12-00745

1574
Regular Article
Chem. Pharm. Bull. 60(12) 1574–1580 (2012)
Vol. 60, No. 12
Discovery, Synthesis, and Evaluation of Small-Molecule Signal
Transducer and Activator of Transcription 3 Inhibitors
Zhi-Bing Shi,a Dan Zhao,a Yan-Yan Huang,b Yun Du,a Xiang-Rong Cao,c Zhu-Nan Gong,c
,a
Rui Zhao,b and Jian-Xin Li*
a State Key Laboratory of Analytical Chemistry for Life Science, School of Chemistry and Chemical Engineering,
b
Nanjing University; Nanjing 210093, P.R. China: Beijing National Laboratory for Molecular Sciences, Laboratory
of Analytical Chemistry for Life Science, Institute of Chemistry, Chinese Academy of Sciences; Beijing 100080, P.R.
c
China: and Jiangsu Key Laboratory for Molecular and Medical Biotechnology, Nanjing Normal University; Nanjing
210097, P.R. China. Received August 22, 2012; accepted September 15, 2012
The signal transducer and activator of transcription 3 (STAT3) oncogene is a promising molecular target
and its inhibitors have great potential as anticancer drugs. To identify novel and STAT3-selective inhibi-
tors, a virtual screening based on Specs and Maybridge databases was conducted and a 6,6′-bibenzoxazole
type small molecule, compound 3a with a inhibition constant Ki value of 494.32nM to STAT3 was explored.
Further, a novel series of derivatives originally derived from 3a was synthesized and evaluated through cell-
based assays using human breast cancer cell lines, MDA-MB-468 and MCF-7 with or without constitutive
expression of STAT3, respectively. In the series, 3a, 3c, 3d and 4e showed a better inhibitory activity with a
good selectivity. Among them, 3a and 3c significantly inhibited STAT3 protein level and also displayed bind-
ing affinity for STAT3 that detected with flow injection analysis-quartz crystal microbalance (FIA-QCM)
analysis system. The results provided a new lead for future design and development of potent STAT3 inhibi-
tors.
Key words signal transducer and activator of transcription 3; inhibitor; cancer; bibenzoxazole
The signal transducers and activators of transcription application prospect of such compounds. The second one
(STATs) are a class of transcription factor proteins with seven is non-peptidic small molecules, most of them, such as
members including 1, 2, 3, 4, 5a, 5b and 6 that regulate cell S3I-M2001,13) STA-21,14) S3I-201,15) more recent STX-0119,16)
growth and survival by modulating the expression of specific and others17) were identified through structure-based virtual
target genes.1,2) STAT3, a member of STATs family, was iden- screening of chemical libraries (Fig. 1). Those compounds are
tified by homology screening of cDNA libraries.3) It is found generally cell permeable and have great potential for further
to be activated by epidermal growth factor (EGF), interleu- research and development, however, most of them have weak
kin-6 (IL-6), and several other growth factors and cytokines binding affinities or a lower selectivity to STAT3. Therefore,
that share the gpl30 receptor subunit and its homologs.4) In there is still a great need to develop novel molecules with
normal cells, the activation of STAT3 is rapid and transient. better bioactivity and selectivity to STAT3.
Constitutive activation of STAT3 is frequently detected in the
In the present study, as an effort to develop novel small
specimens from cancer patients with advanced diseases and a molecule STAT3 inhibitors, we report the discovery of a
number of tumor cell lines, but not in normal epithelial cells, small molecule STAT3 inhibitor, 2,2′-bis(arylsulphonylamino)-
and also identified as an essential step during the transition of 6,6′-bibenzoxazoles (3a) through a virtual screening of 2 data-
normal cells to neoplastic cells.5,6) The structure of STAT3 is bases, and synthesis of the derivatives originally derived from
composed of an amino-terminal domain, a coiled-coil domain, 3a. The bioactivity and the selectivity of the compounds were
a DNA-binding domain, a linker, a Src Homology 2 (SH2) evaluated using two human breast cancer cell lines, MDA-
domain, and a transcriptional activation domain.7) The STAT3 MB-468 and MCF-7 with or without constitutive expression
signal pathway is composed of several distinct steps,8) among of STAT3, respectively. Furthermore, inhibitory activity of
them, the dimerization of two STAT3 monomers through SH2 the compounds with potent bioactivity on STAT3 protein level
domains is a decisive event for its activation and transcrip- and the binding affinity to STAT3 were also assayed.
tional activity.2) Blocking dimerization through small modules
provides a promising approach to the development of molecu-
larly targeted therapies for the treatment of cancers. Since the
Results and Discussion
As mentioned above, to be activated, STAT3 must firstly
X-ray structure of STAT3 dimer complexing with DNA was form a dimer via its SH2 domain and then translocate to the
reported by Becker,9) the design of SH2 inhibitors has been nucleus to targeted genes.9) Therefore, blocking of the dimer-
actively performed for anticancer drug discovery.
ization of STAT3 monomers is directly associated STAT3
Currently, two main approaches are being performed in inhibitors. In order to discover basic skeleton of the inhibitors,
the design of small molecule inhibitors that suppress STAT3 we conducted a virtual screening against the STAT3 SH2
dimerization. The first one is peptidomimetics, such as domain using the available databases including Specs with
ISS610,10) ISS840 and others.7,11,12) The peptide-based ligands 321374 compounds and Maybridge with 79299 ones. The 3D
usually achieve quite high binding affinities to STAT3, but structures of those compounds were download from ZINC
they usually have poor or no cellular activity that limits the database18) and used during the docking studies. The crystal
structure of dimerized STAT3 at 2.25Å was obtained from
RCSB protein databank (PDB code: 1BG1). Autodock 4.219)
The authors declare no conflict of interest.
*To whom correspondence should be addressed. e-mail: lijxnju@nju.edu.cn
© 2012 The Pharmaceutical Society of Japan

December 2012
1575
compounds 2a–l. Then, the heterocyclizations were conducted
by reaction of 2a–l with 3,3′-dihydroxybenzidine in the pres-
ence of potassium hydroxide (KOH), the products were treated
with water and recrystallized in methanol to afford the target
compounds 3a–l with satisfactory yields (see Supplementary
data).
There were about three methods to synthesize the skeleton
of compounds 4a–g.23—25) Based on the synthetic method of
3a–l, we designed a novel and simple synthetic route of 4a–g.
Compounds 2a–l were reacted with O-aminophenol with dif-
ferent substitutions, respectively, to directly provide analogues
4a–g with good yields.
With the above synthesized compounds in hand, the anti-
proliferative activity of the compounds against breast cancer
cell line (MDA-MB-468) with constitutive expression of
STAT3 was firstly evaluated using a 3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The cells
were treated with each compound at the concentrations of
1µM, 10µM and 100µM for 48h, and the results are summa-
rized in Table 1.
As can been seen from Table 1, the same as the virtual
screening result, parent compound 3a showed a inhibitory
activity on MDA-MB-468 cells at 100µM concentration, while
no activity at 10µM. The derivatives 3c, 3d and 4e displayed a
better activity compared with 3a even at 10µM. Unfortunately,
Fig. 1. Typical Small-Molecule STAT3 Inhibitors
and PyRx 0.5 programs were employed to the virtual screen- compounds 3b, 3e–l and the monomers 4a–g except for 4e
ing. The AutoDockTools was used to prepare the screening exhibited almost no activity on the cells at 10µM. Above data
target, chain B of protein 1BG1 and water were removed from clearly revealed that the substitution at moiety of arylsulfon-
the structure and only monomer A was used as the docking amide in 3a–l gave a great impact on the activity. Meta and
target. All hydrogens were added then non polar hydrogens para-methyl substitutions elevated the activity, while ortho-
were merged. The Grid center was set SH2 domain and the methyl group damaged the activity. The halogen substitutions
GridBox was set big enough to cover the entire dimeriza- at any positions of arylsulfonamide moiety deactivated the
tion interface. Lamarckian Genetic Algorithm was used as a derivatives. The activity of the monomers (4a–g) was weaker
searching method, the number of GA runs was set at 10.
than that of 3a–l, suggesting that to get better activity, link-
The docking result was rearranged by the binding energy age of two basic skeletons was essential, which was almost in
(see Supplementary data), compound 3a with bis-oxazole skel- agreement with docking results. Compared with reported non-
eton and the binding energy of −8.6kcal/mol and inhibition peptidic small-molecule STAT3 inhibitors such as S3I-M2001
constant Ki value of 494.32nM was selected from the above and STX-0119, the present compounds showed comparative
rows for further study. The predicted binding model (Fig. activity on MDA-MB-468, while, the compounds were easier
2) showed that 3a has significant interactions with Lys591, to synthesize from sulfonamides compared with preparations
Arg636, Lys626, and Gln635. Structure of two analogues of S3I-M2001 and STX-0119 or other small-molecule STAT3
3d and 3i were also found in ZINC database (Chart 1), the inhibitors.13–17)
docking model showed similar interactions with STAT3 SH2
domain.
In order to understand the inhibitory selectivity of those
compounds with better activity, another breast cancer cell
Bis-oxazole, as an important structure motif exist in many line, MCF-7 cells with no constitutive expression of STAT3
natural products, its derivatives such as hennoxazole and was involved. Four compounds 3a, 3c, 3d and 4e at 100µM
diazonamide A possess significant bioactivities including anti- concentration were tested. As shown in Fig. 3, four com-
fungal, cytotoxic, anthelmintic properties and so on.20,21)
pounds displayed no or very weak suppressive activity toward
To confirm the docking results and further get preliminary MCF-7, suggesting those compounds had a good inhibitory
structure–activity relationship (SAR), compound 3a and its selectivity on STAT3.
derivatives 3b–l were designed and synthesized, and the
In order to verify whether the observed inhibitory activity
preparations of derivatives 4a–g (monomers) of the half struc- on MDA-MB-468 cells is related to the STAT3 protein, the
ture of the 3a were also performed, respectively. The synthetic STAT3 protein level of the cells cultured with compounds
routes of 3a–l22) and 4a–g analogues were outlined in Chart 3a, 3c, 3d and 4e was assayed with a STAT3 ELISA kit. As
1. Firstly, the commercially available starting materials, aryl- shown in Fig. 4, 3a and 3c at 100µM inhibited more than 70%
sulfonamine 1a–l were reacted with carbon disulfide (CS2) in STAT3 protein, while 3d and 4e did not display suppressive
an alkaline medium using N,N-dimethylformamide (DMF) as activity. Because activation of STAT3 leads to cell-cycle pro-
solvent to form corresponding intermediates. The intermedi- gression, anti-apoptotic effects, and tumor invasion and me-
ates were successively reacted with dimethylsulfate (Me2SO4) tastasis,26) the current results revealed that 3a and 3c inhibited
as a methylation agent for 2h, and the reaction products were the proliferation of MDA-MB-468 cells at least via a suppres-
treated with water without any isolation processes to provide sion of STAT3 protein content.

1576
Vol. 60, No. 12
Fig. 2. The Predicted Binding Model of Compound 3a with the STAT3-SH2 Domain Generated by AutoDock (v4.2)
(A) Only the residues that form hydrogen bonds with 3a are shown. (B) STAT3-SH2 domain is shown in surface model.
Reagents and conditions: (a) CS2, NaOH, DMF, 0°C; (b) Me2SO4, RT; (c) 3,3′-Dihydroxybenzidine, KOH, DMF, N2, reflux. (d) KOH, DMF, N2, reflux.
Chart 1. Synthesis of compounds 3a–l and 4a–g
In order to understand the interaction of the inhibitors and commercial available STAT3 solution contains glutathione,
STAT3, flow injection analysis-quartz crystal microbalance to eliminate its interference, a STAT3 modified QCM chip
(FIA-QCM) binding experiments with lead inhibitors 3a and (QCM-S) and only glutathione modified chip (QCM-G) were
3c were conducted with STAT3 protein immobilized QCM prepared. The binding curves of frequency versus time of 3a
chip. As an extremely sensitive surface mass sensor, quartz under the optimal conditions were recorded as graphs and
crystal microbalance (QCM) has been used for the measure- illustrated in Fig. 5. From the binding curves, the binding
ment of mass change in a variety biological studies and has equilibrium constants of 3a were evaluated to be 4.31×103 L/
become an ideal method to entirely monitor the association mol and 8.50×102 L/mol on QCM-S (Fig. 5A) and QCM-G
and disassociation processes of molecules in real time without (Fig. 5B) chips, respectively. The results clearly demonstrated
the use of labels.27) Combined with flow injection analysis that 3a strongly bound with immobilized STAT3. While, 3c
(FIA), QCM allows online monitoring of the analyte binding, displayed only a weak binding affinity with STAT3, because
and provides a much more convenient tool for the determina- the binding equilibrium constants were 6.89×102 L/mol on
tion of real-time study of analytes interactions.28) Because the QCM-S chip and 6.81×102 L/mol on QCM-G chip, which

December 2012
1577
Table 1. Inhibitory Effects of Synthesized Compounds on MDB-MA-468
Cells
Inhibition rate (%)
Compd.
1µM
10µM
100µM
3a
3b
3c
−8.42 1.53
−10.50 0.84
3.96 1.05
4.73 0.95
−3.19 1.83
16.63 0.81**
23.60 0.20**
0.19 1.10
26.94 1.04*
−6.24 0.10
45.41 0.76**
37.41 0.64**
−5.76 0.57
20.59 0.20**
−15.29 0.21
8.30 1.70
3d
3e
5.41 0.47
−18.61 3.23
−9.25 0.76
−15.38 0.53
−10.81 0.67
−15.70 0.17
−16.74 0.25
−14.14 0.69
−10.08 0.70
−9.77 0.50
−7.17 0.40
−9.98 0.57
−6.86 0.40
−2.60 0.53
−12.37 0.93
−10.19 0.46
3f
4.44 1.15
3g
−3.19 1.14
1.88 0.17
3h
3i
2.61 0.99
19.65 0.31*
−16.12 1.44
20.00 0.55**
0.24 1.21
3j
1.83 0.44
3k
3l
3.28 0.50
−0.19 0.17
−2.99 0.15
−2.70 0.81
−0.48 0.69
0.87 0.32
Fig. 3. Inhibitory Effects of the Selective Compounds on MDA-
MB-468 and MCF-7
4a
9.76 0.81
4b
4c
17.88 0.93*
7.76 0.29*
Human breast cancer cells MDA-MB-468 and MCF-7 were treated with 100µM
selected compounds for 48h and assayed for viability using MTT method. Control
group without any treatment was pegged as 0.00%, while other data were calculat-
ed relative to it. Each value represents the mean S.D., n=3. Significant differences
compared with control group. *p<0.05, **p<0.01.
4d
4e
25.18 0.10*
22.82 0.49**
−0.94 0.78
−1.29 0.50
7.63 0.52*
1.64 0.67
4f
4g
0.29 0.96
Control
0.00 4.01
Human breast cancer cell line MDA-MB-468 with constitutive activation of
STAT3 was used for the assay. The inhibition rate cells of control group without any
treatment was pegged as 0.00%, while other data were calculated relative to it. Data
are expressed as mean S.D., n=4. Significant differences compared with control
group, *p<0.05, **p<0.01. “−” a minus sign of the data means enhancement effect
on cell viability.
suggested that the main binding effect was caused from glu-
tathione. As 3c exhibited inhibitory effect on STAT3 level,
it is possible that 3c might inhibit both STAT3 and upstream
kinase, to clarify it, further research work is essential.
Conclusion
In summary, we have identified a series of novel STAT3
targeting small molecule inhibitors through virtual screening,
preliminary structure optimization and in vitro assays. Among
the compounds, 3a and 3c exhibited a potent anti-proliferation
activity on STAT3-depend human breast cancer cells with a
good selectivity, significantly inhibited STAT3 protein level
of MDA-MB-468 cells, and 3a displayed a strong binding ac-
tivity on STAT3. The current results provided a new lead for
further design and optimization of more potent and selective
STAT3 inhibitors as a new class of anticancer drugs.
Fig. 4. Effects of the Selective Compounds on STAT3 Protein Content
in MDA-MB-468 Cells
Human breast cancer cells MDA-MB-468 were treated with 10µM and 100µM
of selected compounds and the content of STAT3 was assayed with STAT3 ELISA
kit. Each value represents the mean S.D., n=3. Significant differences compared
with control group (0.00%, the STAT3 concentration was 587.36pg/mL), *p<0.05,
**p<0.01.
Chemical Synthesis. General Procedure of Synthesis of
N-Bis(methylthio)methylene Arylsulfonamide Type Com-
pounds (2a–l) To a solution of arylsulfonamide (1mmol)
Experimental
General 1H- and 13C-NMR spectra were measured on a in dried DMF (2mL) was added NaOH (80mg) and carbon
Bruker Advance II 300 spectrometer using tetramethylsilane disulfide (1mmol) at 0°C over a few minutes. The reaction
as internal standard. Chemical shift (δ) are reported in parts solution was stirred for 30min at 0°C, and Me2SO4 (2mmol)
per million (ppm) and coupling constants J are reported in was added dropwisely over several minutes. After warming up
hertz (Hz), 13C-NMR spectra were fully decoupled, and the to room temperature, ample water was added and the resulting
following abbreviations are used: singlet (s), doublet (d), and crystalline precipitate was collected via filtration. The filter
multiplet (m).
MDA-MB-468 and MCF-7 cells were obtained from the title compounds.
Shanghai Cell Bank of the Chinese Academy of Sciences General
cake was washed with water, methanol and dried to afford the
Procedure
of
Synthesis
of
(Shanghai, China). STAT3 was purchased from PeproTech 2,2′-Bis(arylsulfonamino)-6,6′-bibenzoxazole (3a–l) To a
Inc., U.S.A. Human signal transducer and activator of tran- solution of 3,3′-dihydroxybenzidine (1mmol) in dried
scription 3 (STAT3) Kit was from Rapidbio, RB.
DMF (8mL) was added 5M NaOH solution (0.2mL), and

1578
Vol. 60, No. 12
Fig. 5. The Binding Processes of 3a on STAT3 (A, QCM-S) and Glutathione (B, QCM-G) Immobilized Chips
stirred at room temperature for 30min. Then, solution of 156.85 (s), 145.54 (s), 144.79 (d, J=6.6Hz), 135.57 (s), 131.84
N-bis(methylthio)methylene arylsulfonamide (2, 2mmol) in (d, J=7.8Hz), 130.02 (s), 124.39 (s), 122.75 (s), 119.83 (d,
2mL DMF was added in dropwise, refluxed under N2 for 8h. J=21.1Hz), 113.60 (d, J=24.3Hz), 112.61 (s), 109.11 (s);
After cooling, acetic acid (2mL) was added. The percipitate positive ESI-MS m/z 583 [M+1]+; HR-MS (ESI): Calcd for
obtained after ample water was added, and then washed with C26H17F2N4O6S2 [M+H]+ 583.0558, Found 583.0535.
1
water, and recrystallized using methanol to afford the target
Compound 3f (436mg, 75%): H-NMR (300MHz, DMSO-
d6) δ: 12.49 (s, 2H), 8.07–7.94 (m, 3H), 7.91–7.79 (m, 3H),
compounds.
Compound 3a (436mg, 80%): H-NMR (300MHz, DMSO- 7.61 (d, J=8.4Hz, 2H), 7.47–7.30 (m, 6H); 13C-NMR (75MHz,
d6) δ: 7.94 (m, 4H), 7.87 (s, 1H), 7.60 (m, 8H), 7.37 (d, DMSO-d6) δ: 164.44 (d, J=243.9Hz), 156.58 (s), 145.44 (s),
J=7.0Hz, 2H); 13C-NMR (75MHz, DMSO-d6) δ: 156.52, 139.10 (s), 135.55 (s), 129.83 (s), 129.50 (d, J=9.5Hz), 126.94
145.34, 142.63, 135.53, 132.79, 129.75, 129.45, 126.41, 124.37, (s), 125.58 (s), 124.39 (s), 116.54 (d, J=22.5Hz), 112.52 (s),
112.48, 109.10; positive electrospray ionization (ESI)-MS 109.11 (s); positive ESI-MS m/z 583 [M+1]+; HR-MS (ESI):
1
m/z 547 [M+1]+; high resolution (HR)-MS (ESI): Calcd for Calcd for C26H17F2N4O6S2 [M+H]+ 583.0558, Found 583.0541.
1
C26H19N4O6S2 [M+H]+ 547.0746, Found 546.0742.
Compound 3g (442mg, 72%): H-NMR (300MHz, DMSO-
1
Compound 3b (430mg, 75%): H-NMR (300MHz, DMSO- d6) δ: 8.14 (m, 2H), 7.97–7.76 (m, 2H), 7.71–7.49 (m, 8H), 7.42
d6) δ: 8.13–7.97 (m, 2H), 7.84 (d, J=1.2Hz, 2H), 7.64–7.45 (d, J=7.3Hz, 2H); 13C-NMR (75MHz, DMSO-d6) δ: 156.95,
(m, 4H), 7.36 (m, J=10.0, 6.0Hz, 6H), 2.61 (s, 6H); 13C-NMR 145.24, 139.85, 135.75, 134.65, 134.25, 132.08, 131.31, 129.85,
(75MHz, DMSO-d6) δ: 156.45, 145.29, 140.77, 136.95, 135.52, 127.94, 124.54, 112.71, 109.21; positive ESI-MS m/z 615
132.76, 132.61, 129.81, 127.65, 126.38, 124.36, 112.45, 109.06, [M+1]+; HR-MS (ESI): Calcd for C26H17Cl2N4O6S2 [M+H]+
20.20; positive ESI-MS m/z 575 [M+1]+; HR-MS (ESI): Calcd 614.9967, Found 614.9959.
for C28H23N4O6S2 [M+H]+ 575.1059, Found 575.1033.
Compound 3h (461mg, 75%): 1H-NMR (300MHz,
1
Compound 3c (459mg, 80%): H-NMR (300MHz, DMSO- DMSO-d6) δ: 11.72 (s, 2H), 8.00–7.80 (m, 6H), 7.70 (d, J=
d6) δ: 7.84–7.88 (m, 2H), 7.68–7.80 (m, J=8.7Hz, 4H), 8.8Hz, 2H), 7.60 (m, 4H), 7.38 (d, J=7.3Hz, 2H); 13C-NMR
7.56–7.64 (m, J=7.9Hz, 2H), 7.51–7.30 (m, 6H), 2.38 (s, 6H). (75MHz, DMSO-d6) δ 156.74, 145.47, 144.52, 135.62, 133.97,
13C-NMR (75MHz, DMSO-d6) δ: 156.52, 145.34, 142.57, 132.72, 131.57, 129.78, 126.15, 125.20, 124.45, 112.59, 109.16;
138.33, 135.54, 133.39, 129.82, 129.30, 126.69, 124.58, 123.55, positive ESI-MS m/z 615 [M+1]+; HR-MS (ESI): Calcd for
112.49, 109.10, 21.18; positive ESI-MS m/z 575 [M+1]+; C26H17Cl2N4O6S2 [M+H]+ 614.9967, Found 614.9944.
HR-MS (ESI): Calcd for C28H23N4O6S2 [M+H]+ 575.1059,
Compound 3i (462mg, 75%): H-NMR (300MHz, DMSO-
d6) δ: 7.94 (m, 4H), 7.88 (d, J=1.1Hz, 2H), 7.62 (m, 6H), 7.37
1
Found 575.1042.
1
Compound 3d (402mg, 70%): H-NMR (300MHz, DMSO- (d, J=8.3Hz, 2H); 13C-NMR (75MHz, DMSO-d6) δ: 156.80,
d6) δ: 7.93–7.74 (m, 6H), 7.61 (d, J=8.4Hz, 2H), 7.36 (d, m, 145.51, 144.68, 135.59, 131.78, 129.93, 128.92, 125.55, 124.42,
6H), 2.35 (s, 6H); 13C-NMR (75MHz, DMSO-d6) δ: 156.59, 122.26, 112.62, 109.13; positive ESI-MS m/z 616 [M+1]+;
145.41, 143.00, 139.92, 135.50, 129.96, 129.84, 126.50, 124.33, HR-MS (ESI): Calcd for C26H17Cl2N4O6S2 [M+H]+ 614.9967,
112.50, 109.05, 21.32; positive ESI-MS m/z 575 [M+1]+; Found 614.9961.
HR-MS (ESI): Calcd for C28H23N4O6S2 [M+H]+ 575.1059,
Compound 3j (528mg, 75%): H-NMR (300MHz, DMSO-
d6) δ: 8.17 (dd, J=7.7, 1.6Hz, 2H), 7.92–7.78 (m, 4H),
1
Found 575.1051.
Compound 3e (436mg, 75%): 1H-NMR (300MHz, 7.70–7.48 (m, 6H), 7.41 (d, J=8.3Hz, 2H); 13C-NMR (75MHz,
DMSO-d6) δ: 7.88 (d, J=1.1Hz, 2H), 7.82–7.69 (m, 4H), DMSO-d6) δ: 157.01, 152.87, 145.28, 141.58, 135.69, 135.53,
7.56–7.67 (m, 4H), 7.54–7.42 (m, 2H), 7.38 (d, J=8.3Hz, 2H). 134.19, 130.01, 128.42 (s), 124.51, 120.00, 112.70, 109.20;
13C-NMR (75MHz, DMSO-d6) δ: 161.99 (d, J=248.1Hz), positive ESI-MS m/z 704 [M+1]+; HR-MS (ESI): Calcd for

December 2012
1579
C26H17Br2N4O6S2 [M+H]+ 704.8936, Found 704.8920.
111.20, 107.65; positive ESI-MS m/z 320 [M+1]+; HR-MS
1
Compound 3k (528mg, 75%): H-NMR (300MHz, DMSO- (ESI): Calcd for C13H10N3O5S [M+H]+ 320.0341, Found
d6) δ: 8.07 (m, 2H), 7.94 (d, J=7.8Hz, 2H), 7.88 (d, J=1.2Hz, 320.0317.
1
2H), 7.83 (dd, J=8.0, 0.9Hz, 2H), 7.62 (dd, J=8.3, 1.4Hz,
Compound 4g (266mg, 80%): H-NMR (300MHz, DMSO-
2H), 7.53 (t, J=7.9Hz, 2H), 7.38 (d, J=8.3Hz, 2H); 13C-NMR d6) δ: 8.13 (dd, J=8.9, 2.3Hz, 1H), 8.02 (d, J=2.3Hz, 1H),
(75MHz, DMSO-d6) δ: 156.80, 145.51, 144.68, 135.59, 131.78, 7.82 (d, J=8.2Hz, 2H), 7.71 (d, J=8.9Hz, 1H), 7.37 (d, J=
130.92, 129.93, 128.92, 125.55, 124.42, 122.26, 112.62, 109.13; 8.2Hz, 2H), 2.36 (s, 3H); 13C-NMR (75MHz, DMSO-d6) δ:
positive ESI-MS m/z 704 [M+1]+; HR-MS (ESI): Calcd for 157.18, 148.64, 144.99, 143.33, 139.41, 131.99, 129.89, 126.54,
C26H17Br2N4O6S2 [M+H]+ 704.8936, Found 704.8931.
120.18, 111.16, 107.64, 21.33; positive ESI-MS m/z 334 [M+1]+;
1
Compound 3l (492mg, 70%): H-NMR (300MHz, DMSO- HR-MS (ESI): Calcd for C14H12N3O5S [M+H]+ 334.0498,
d6) δ: 7.92–7.83 (m, 6H), 7.77 (d, J=8.6Hz, 4H), 7.61 (d, Found 334.0477.
J=8.4Hz, 2H), 7.37 (d, J=8.2Hz, 2H); 13C-NMR (75MHz,
Biological Assays. Cell Culture Human breast cancer cell
DMSO-d6) δ: 156.57, 145.47, 141.95, 135.59, 132.50, 129.71, lines (MDA-MB-468 and MCF-7) were grown in RPMI-1640
128.58, 126.94, 126.50, 125.59, 124.45, 112.53, 109.16; medium with 15% fetal bovine serum (FBS) for 24h, and
positive ESI-MS m/z 704 [M+1]+; HR-MS (ESI): Calcd for then the cells were harvested and counted. Solutions of tested
C26H17Br2N4O6S2 [M+H]+ 704.8936, Found 704.8908.
compounds (100µL) at the desired concentrations and MCF-7
General Procedure of Synthesis of 4a–g The synthetic cell solution (2.5×104 cells/mL, 100µL) were added incubated.
route was similar to that procedure for compound 3a, but O- Then, the cell viability was assayed by MTT method.
aminophenols with different substitutions were used instead of
3,3′-dihydroxybenzidine and the equivalent of O-aminophe- (DMSO) followed by dilution with the medium to desired
nols was equal to compound 2a. concentrations, and DMSO final concentration was 0.1% in
Each tested compound was dissolved in dimethyl sulfoxide
1
Compound 4a (244mg, 85%): H-NMR (300MHz, DMSO- all of the cultures. DMSO at 0.1% was added control groups
d6) δ: 12.69 (s, 1H), 8.01 (m, 1H), 7.52–7.15 (m, 7H), 2.61 (s, and showed no effects on cells. All the cultures were kept in
3H); 13C-NMR (75MHz, DMSO-d6) δ: 156.17, 144.42, 140.78, a CO2 incubator under moist condition of 5% CO2 in air at
136.95, 132.73, 132.59, 130.03, 127.65, 126.35, 125.63, 123.93, 37°C.
112.25, 110.71, 20.19; positive ESI-MS m/z 289 [M+1]+;
Cell Viability Assay Cells were seeded in 96-well plates
HR-MS (ESI): Calcd for C14H13N2O3S [M+H]+ 289.0647, (5×104 cells/well) in triplicate and treated with compounds
Found 289.0633. 3a–l, 4a–g in 1µM, 10µM, 100µM. After a 48h incubation pe-
1
Compound 4b (230mg, 80%): H-NMR (300MHz, DMSO- riod, 20µL MTT reagent (5mg/mL) was added, and the cells
d6) δ: 12.70 (s, 1H), 7.81 (d, J=7.8Hz, 2H), 7.49 (d, J=7.8Hz, were incubated for 4h. The supernatants were aspirated, and
1H), 7.40–7.16 (m, 5H), 2.35 (s, 3H); 13C-NMR (75MHz, the formazan crystals in each well were dissolved in 200µL of
DMSO-d6) δ: 156.22, 144.45, 143.00, 139.87, 130.07, 129.82, dimethyl sulfoxide for 30min at 37°C. The absorbance value
126.47, 125.62, 123.92, 112.26, 110.72, 21.30; positive ESI- was monitored by microplate reader at 490nm and used to
MS m/z 289 [M+1]+; HR-MS (ESI): Calcd for C14H13N2O3S calculate cell viability.
[M+H]+ 289.0647, Found 289.0619.
STAT3 Protein Level Assay After treatment with 3a,
1
Compound 4c (231mg, 80%): H-NMR (300MHz, DMSO- 3c, 3d and 4e 10 µM, 100µM or DMSO for 24h, the samples
d6) δ: 12.65 (s, 1H), 7.98–7.86 (m, 2H), 7.67–7.48 (m, 3H), 7.32 were analyzed for STAT3 using the human signal transducer
(s, 1H), 7.20 (d, J=8.0Hz, 1H), 7.09 (d, J=8.0Hz, 1H), 2.33 (s, and activator of transcription 3 (STAT3) Kit (Rapidbio, RB)
3H); 13C-NMR (75MHz, DMSO-d6) δ: 156.28, 144.62, 142.73, with an enzyme-linked immunosorbent assay (ELISA)-based
133.87, 132.71, 129.58, 127.63, 126.38, 126.13, 111.81, 110.99, method following the manufacturer’s protocol. The result was
21.24; positive ESI-MS m/z 289 [M+1]+; HR-MS (ESI): Calcd used to calculate the concentration of STAT3 in the samples
for C14H13N2O3S [M+H]+ 289.0647, Found 289.0624.
and the inhibition rate.
STAT3 Binding Assay with FIA-QCM Immobilization
1
Compound 4d (241mg, 82%): H-NMR (300MHz, DMSO-
d6) δ: 12.61 (s, 1H), 7.80 (d, J=8.2Hz, 2H), 7.35 (d, J=8.2Hz, procedure of STAT3: The immobilization of STAT3 molecules
3H), 7.12 (s, 1H), 7.01 (d, J=8.3Hz, 1H), 2.34 (s, 3H), 2.33 (s, on the surface of quartz crystal microbalance (QCM) chip
3H); 13C-NMR (75MHz, DMSO-d6) δ: 156.43, 142.76, 141.63, was carried out as reported method26) with slight modification.
137.49, 135.39, 129.97, 129.51, 128.44, 124.50, 112.39, 110.32 Briefly, the chip surface was first cleaned with Piranha solu-
(s), 21.28; positive ESI-MS m/z 303 [M+1]+; HR-MS (ESI): tion (H2SO4–30% H2O2=3:1, v/v), and rinsed with water and
Calcd for C15H15N2O3S [M+H]+ 303.0803, Found 303.0801.
ethanol, then air-dried. The cleaned surfaces were immersed
1
Compound 4e (257mg, 80%): H-NMR (300MHz, DMSO- into a 20mM cysteamine hydrochloride solution for 12h in
d6) δ: 12.78 (s, 1H), 7.98–7.87 (m, 2H), 7.62 (d, J=8.6Hz, 2H), the dark at room temperature, and then the excess cysteamine
7.37 (d, J=8.3Hz, 1H), 7.13 (s, 1H), 7.02 (d, J=8.4Hz, 1H), was removed with ethanol and water washings. Glutaric dial-
2.34 (s, 3H); 13C-NMR (75MHz, DMSO-d6) δ: 156.34, 142.94, dehyde was used as activating reagent to introduce aldehyde
142.62, 139.94, 135.27, 130.06, 129.80, 126.45, 124.38, 112.33, groups. The quartz crystal was immersed into a 2.5% (v/v)
110.25; positive ESI-MS m/z 323 [M+1]+; HR-MS (ESI): Calcd glutaric dialdehyde solution (pH 7.4) at 40°C for 4h in an in-
for C14H12ClN2O3S [M+H]+ 323.0257, Found 323.0229.
cubator shaker, and stopped by washing with large amount of
1
Compound 4f (264mg, 83%): H-NMR (300MHz, DMSO- water. Then, the chip was exposed to 30µL of 1.6µM STAT3
d6) δ: 8.13 (dd, J=8.9, 2.4Hz, 1H), 8.03 (d, J=2.3Hz, 1H), solution at room temperature for 12h. In order to eliminate
7.99–7.88 (m, 2H), 7.72 (d, J=8.9Hz, 1H), 7.60 (ddd, J=16.0, the interference of glutathione contained in the STAT3 solu-
7.7, 2.1Hz, 3H); 13C-NMR (75MHz, DMSO-d6) δ: 157.23, tion, a reference chip (QCM-G) was also prepared follow-
148.61, 145.01, 142.20, 133.03, 131.84, 129.50, 126.47, 120.23, ing the same procedure with only the same concentration of

1580
Vol. 60, No. 12
1159–1168 (2008).
glutathione solution.
9) Becker S., Groner B., Müller C. W., Nature, 394, 145–151 (1998).
10) Turkson J., Kim J. S., Zhang S., Yuan J., Huang M., Glenn M.,
Haura E., Sebti S., Hamilton A. D., Jove R., Mol. Cancer Ther., 3,
261–269 (2004).
11) Gunning P. T., Katt W. P., Glenn M., Siddiquee K., Kim J. S., Jove
R., Sebti S. M., Turkson J., Hamilton A. D., Bioorg. Med. Chem.
Lett., 17, 1875–1878 (2007).
12) Shahani V. M., Yue P. B., Fletcher S., Sharmeen S., Sukhai M. A.,
Luu D. P., Zhang X. L., Sun H., Zhao W., Schimmer A. D., Turkson
J., Gunning P. T., Bioorg. Med. Chem., 19, 1823–1838 (2011).
13) Siddiquee K. A. Z., Gunning P. T., Glenn M., Katt W. P., Zhang S.,
Schroeck C., Sebti S. M., Jove R., Hamilton A. D., Turkson J., ACS
Chem. Biol., 2, 787–798 (2007).
14) Bhasin D., Cisek K., Pandharkar T., Regan N., Li C., Pandit B., Lin
J., Li P. K., Bioorg. Med. Chem. Lett., 18, 391–395 (2008).
15) Siddiquee K., Zhang S., Guida W. C., Blaskovich M. A., Greedy B.,
Lawrence H. R., Yip M. L. R., Jove R., McLaughlin M. M., Law-
rence N. J., Sebti S. M., Turkson J., Proc. Natl. Acad. Sci. U.S.A.,
104, 7391–7396 (2007).
16) Matsuno K., Masuda Y., Uehara Y., Sato H., Muroya A., Takahashi
O., Yokotagawa T., Furuya T., Okawara T., Otsuka M., Ogo N.,
Ashizawa T., Oshita C., Tai S., Ishii H., Akiyama Y., Asai A., ACS
Med. Chem. Lett., 1, 371–375 (2010).
17) Zhao M., Jiang B., Gao F. H., Curr. Med. Chem., 18, 4012–4018
(2011).
18) Irwin J. J., Shoichet B. K., J. Chem. Inf. Model., 45, 177–182 (2005).
19) Morris G. M., Huey R., Lindstrom W., Sanner M. F., Belew R.
K., Goodsell D. S., Olson A. J., J. Comput. Chem., 30, 2785–2791
(2009).
Inhibitors-STAT3 binding assay: At the beginning, 10%
DMSO Tris–HCl buffer (50mM, pH 7.4) was flowed into
STAT3 modified chip (QCM-S) via the HPLC injection value
(flow rate, 40µL/min). After a stable baseline was achieved,
200µL of 3a or 3c solutions (dissolved in 10% DMSO Tris–
HCl buffer solution, and the final concentration was 0.5mg/
mL) were injected into the fluid system via the HPLC injec-
tion value. The curves of frequency versus time were recorded
and the entire real-time association and dissociation processes
were displayed continuously on the computer. The same ex-
periment was conducted on the glutathione modified chip
(QCM-G).
Statistical Analysis Data are presented as means S.D. of
the indicated number of experiments. Data were analyzed by
one-way analysis of variance (ANOVA) followed by Dunnett’s
test or Student’s t-test when appropriate. Probability values of
0.05 or less were considered to be statistically significant. All
statistic analyses were performed with SPSS 13.0 (SPSS Inc.,
Chicago, IL, U.S.A.).
Acknowledgments This work was supported by the Natu-
ral Science Foundation of Jiangsu Province (BK2009240), the
National Natural Science Fund for Creative Research Groups
of China (21121091) and the fund of Jiangsu Key Laboratory
for Molecular and Medical Biotechnology (2011MMBKF02).
Supplementary Data Supplementary data (the docking
results, 1H- and 13C-NMR spectra of selected synthesized
compounds) associated with this article can be found, in the
online version.
20) Ichiba T., Yoshida W. Y., Scheuer P. J., Higa T., Gravalos D. G., J.
Am. Chem. Soc., 113, 3173–3174 (1991).
21) Davies J. R., Kane P. D., Moody C. J., J. Org. Chem., 70, 7305–7316
(2005).
22) Garin J., Melendez E., Merchan F. L., Merino P., Orduna J., Tejero
T., J. Heterocycl. Chem., 27, 321–326 (1990).
23) Meléndez E., Merchán F. L., Merino P., Orduna J., Urchegui R., J.
Heterocycl. Chem., 28, 653–656 (1991).
24) Lai C., Gum R. J., Daly M., Fry E. H., Hutchins C., Abad-Zapatero
C., von Geldern T. W., Bioorg. Med. Chem. Lett., 16, 1807–1810
(2006).
25) Sławiński J., Brzozowski Z., Eur. J. Med. Chem., 41, 1180–1189
(2006).
26) Lieblein J. C., Ball S., Hutzen B., Sasser A. K., Lin H. J., Huang T.
H. M., Hall B. M., Lin J., BMC Cancer, 8, 302–309 (2008).
27) Tuantranont A., Wisitsora-at A., Sritongkham P., Jaruwongrungsee
K., Anal. Chim. Acta, 687, 114–128 (2011).
28) Zhang Q., Huang Y., Zhao R., Liu G., Chen Y., Biosens. Bioelec-
tron., 24, 48–54 (2008).
References
1) Darnell J. E. Jr., Science, 277, 1630–1635 (1997).
2) Horvath C. M., Darnell J. E. Jr., Curr. Opin. Cell Biol., 9, 233–239
(1997).
3) Zhong Z., Wen Z., Darnell J. E. Jr., Science, 264, 95–98 (1994).
4) Stahl N., Farruggella T. J., Boulton T. G., Zhong Z., Darnell J. E.
Jr., Yancopoulos G. D., Science, 267, 1349–1353 (1995).
5) Yu H., Jove R., Nat. Rev. Cancer, 4, 97–105 (2004).
6) Masciocchi D., Gelain A., Villa S., Meneghetti F., Barlocco D., Fu-
ture Med. Chem., 3, 567–597 (2011).
7) Timofeeva O. A., Gaponenko V., Lockett S. J., Tarasov S. G., Jiang
S., Michejda C. J., Perantoni A. O., Tarasova N. I., ACS Chem.
Biol., 2, 799–809 (2007).
8) Fletcher S., Turkson J., Gunning P. T., ChemMedChem, 3,