Development of new N-arylbenzamides as STAT3 dimerization inhibitors

Article abstract of DOI:10.1039/c3md20323a

The O-tosylsalicylamide S3I-201 (10) was used as a starting point for design and synthesis of novel STAT-3 dimerization inhibitors with improved drug-like qualities. The phosphonic acid 12d and salicylic acids 13f, 13g with a shorter amide linker lacking

Full text of DOI:10.1039/c3md20323a

View Article Online
View Journal
MedChemComm
Accepted Manuscript
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Accepted Manuscripts are published online shortly after acceptance, which is prior
www.rsc.org/medchemcomm
Volume 1 | Number 2 | August 2010 | Pages 105–172
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
ISSN 2040-2503
REVIEW
Elisabeth-Maria Bissinger, Ralf Heinke,
Wolfgang Sippl and Manfred Jung
Targeting epigenetic
modifiers: Inhibitors of histone
methyltransferases
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
REVIEW
Kenneth K. Y. Wong and Xuelai Liu
Silver nanoparticles—the real
“silver bullet”in clinical medicine?
2040-2503(2010)1:2;1-
X
www.rsc.org/medchemcomm
Registered Charity Number 207890

Page 1 of 19
Medicinal Chemistry Communications
View Article Online
Development of new N-Arylbenzamides as STAT3 Dimerization Inhibitors
Murali K. Urlam,†¶ Roberta Pireddu,†,¶ Yiyu Ge,‡,¶ Xiaolei Zhang,† Ying Sun,† Harshani R. Lawrence,†,‡,§,
Wayne C. Guida,†,‡,║ Saïd M. Sebti,†,§,┴ Nicholas J. Lawrence.†,§,*
†
‡
Departments of Drug Discovery, and the Chemical Biology Core, Moffitt Cancer Center, 12902 Magnolia
Drive, Tampa, FL 33612, USA
Departments of §Oncologic Sciences, ║Chemistry and ┴Molecular Medicine, University of South Florida,
Tampa, FL 33620, USA
Corresponding Author: Nicholas J. Lawrence. Eꢀmail: Nicholas.Lawrence@moffitt.org; Phone: +813 745
6037 (NJL), Fax: +813 745 6748
¶
These authors contributed equally.
Abstract
The Oꢀtosylsalicylamide S3I-201 (10) was used as a starting point for design and synthesis of novel STATꢀ3
dimerization inhibitors with improved drugꢀlike qualities. The phosphonic acid 12d and salicylic acids 13f, 13g
with a shorter amide linker lacking the Oꢀtosyl group had improved STATꢀ3 inhibitory activity. The equivalent
potencies observed by the replacement of phosphonic acid moiety of 12d with 5ꢀaminoꢀ2ꢀhydroxybenzoic acid
group as in 13f further validates 5ꢀaminoꢀ2ꢀhydroxybenzoic acid as a phosphotyrosine mimic. The salicylic
acid 13f displayed improved whole cell activity. The focused library of salicylic acids 13 with benzamide
linker indicated that hydrophobic heptyl and cyclohexyl are the best tolerated R groups and a biphenyl ether (as
the Ar group) significantly contributes to STAT3 inhibitory activity. Our docking studies indicated that the
acidic groups of 12d, 13f and 13g interact in the pꢀTyrꢀ705 binding site in a broadly similar manner, while the
phenoxybenzoyl group and the cyclohexylbenzyl group occupying pY+1 and pYꢀX hydrophobic pockets
respectively. The in vitro and cell based potency of 13f warrants further development of this scaffold as STAT3
inhibitors.
1

Medicinal Chemistry Communications
Page 2 of 19
View Article Online
Introduction
STAT3 is a signal transducer and activator of transcription that transmits signals from cell surface receptors to
the nucleus. STAT3 is frequently hyperactivated in many human cancers. Under normal conditions, STAT3
activation is transient and tightly regulated. Upon cellular stimulation by ligands such as growth factors or
cytokines, STAT3 is phosphorylated on a critical tyrosine residue (Tyr705). This pTyrꢀSTAT3 dimerizes
through two reciprocal phosphotyrosine (pTyr)ꢀSrcꢀhomology 2 (SH2) interactions. The STAT3 dimers then
translocate to the nucleus and bind to specific DNAꢀresponse elements in the promoters of target genes thereby
activating transcription.1 STAT3 is found to be constitutively activated in many tumor cell types and
contributes to tumor progression through the modulation of target of antiapoptotic genes such as BclꢀxL, Bclꢀ2,
Mclꢀ1 and survivin along with genes driving cell cycle progression such as cꢀMyc and cyclinꢀD1.1, 2 The
association of aberrant STAT3 activation with many types of human malignancies and solid tumors3 has made
STAT3 an attractive molecular target for the development of novel cancer therapeutics.4, 5
The design of compounds that target STAT3 has been the subject of several excellent reviews.6, 7 These
include a survey of patents8 and a broader review of inhibitors of dimeric transcription factors.9 The direct
targeting of STAT3 is a particularly attractive way to inhibit its function. Several approaches have been taken
to inhibit the dimerization of phosphorylated STAT3 by blocking the SH2 domain binding site of the
phosphorylated STAT3 tyrosineꢀ705 residue. The first inhibitors of STAT3 dimerization were peptides and
phosphopeptides.10, 11 Significant advances have been made by the groups of McMurray12ꢀ14 and Wang15 by
using structureꢀbased approaches resulting in potent peptideꢀlike inhibitors incorporating a phosphotyrosine
residue such as the pYLPQ mimics 1 and 2 respectively. These potent cell permeable STAT3 dimerization
inhibitors have considerable ADME liabilities since the high affinity SH2 domain binding derives, at least in
part, from the necessary presence of a hydrolyzable phosphate group or phosphonate prodrug. Indeed, to date,
there are no reports of the activities of these compounds in animal models. As an alternative approach
considerable attention has been paid to the discovery of nonꢀpeptidic small molecule drugꢀlike inhibitors of
STAT3 dimerization seeking to avoid some of the ADME challenges inherent in the development of peptideꢀ
2

Page 3 of 19
Medicinal Chemistry Communications
View Article Online
like inhibitors. Representative small molecule16 STAT3 dimerization inhibitors are shown in Figure 1 include
the catechol 3;17 the 1,3,4ꢀoxadiaxole 4;18 the benzothiophene dioxide 5 (Stattic);19 the anthraquinone 6 (STAꢀ
21);20 the purine 7;21 the oxazole 8;22 and the Celecoxibꢀlike pyrazole 9.23
CH2Ph
NH
O
HN
O
O
H2N
N
N
H
N
H
N
O
O
N
O
O
R
O
N
N
H
O
P
N
H
O
P
HN
HO
HO
RO
RO
O
1, R = CH2OtBu
2, R = COC15H31
F
F
H
N
O
HO
HO
O
O
O
O
N
O
N
S
O
O2N
O
OH
O
O
N
OH
3
4
5 (Stattic)
6, (STA-21)
F
Br
O
H
N
N
N
O
N
H
N
N
N
O
S
N
H
O
O
N
9
O
P
HO
N
NH2
HO
HO
O
7
8
Figure 1. Chemical structures of representative STAT3 dimerization inhibitors
We identified the STAT3 inhibitor S3Iꢀ201 (10, NSCꢀ74859)24 (Figure 2) from the NCI chemical collection by
using structureꢀbased virtual screening with a model based on the Xꢀray crystal structure25 of the STAT3β
homodimer (pdb code 1BG1). S3Iꢀ201 (10) inhibited STAT3:STAT3 complex formation and STAT3 DNAꢀ
binding and transcriptional activities. Furthermore, S3Iꢀ201 has been shown to exert antitumor effects against
human breast24 and liver26 cancer xenografts in mouse models via mechanisms that are consistent with
inhibition of STAT3 dimerization.
3

Medicinal Chemistry Communications
Page 4 of 19
View Article Online
Figure 2. Chemical structures of S3Iꢀ201 (10, NSCꢀ74859), scaffolds 12 and 13 and reported S3Iꢀ201
derivatives 14-16.
In this communication, we describe our efforts toward improvement of S3Iꢀ201 (10). We focused on the
replacement of the potentially reactive Oꢀsulfonylglycine portion of 10, and investigated alternative
phosphotyrosine mimicking groups. We show that the phosphotyrosine mimicking 4ꢀaminoꢀ2ꢀhydroxybenzoic
acid of 10 can be replaced by its isomeric partner 5ꢀaminoꢀ2ꢀhydroxybenzoic acid in the sulfonylglycine series
11. We also show that the sulfonylglycine linker separating the two aryl groups of 11 by 5 atoms can be
replaced by a shorter linker as in phosphonic acids 12 and salicylic acids 13. Gunning and Turkson have
developed a library 14, based on 10, in which the Oꢀsulfonyl group has been replaced its Nꢀsulfonyl
counterpart.27, 28 The major findings of their work show that a large hydrophobic group as R1 is beneficial,
especially a paraꢀcyclohexylbenzyl group. The analogs had an R2 group as H, Me or BOC (tertꢀ
butyloxycarbonyl). The methyl containing compounds were the most potent. The sulfonamide 15 (SF-1-066)
was shown to be the most potent in the fluorescence polarization assay (IC50 15ꢀ20 ꢁM). Subsequently the
pentafluorophenylsulfonamide 16 has been shown to be orally bioavailable and impressively inhibits the growth
of human breast and lung tumor xenografts.29
4

Page 5 of 19
Medicinal Chemistry Communications
View Article Online
Results and Discussion
The routes to the analogs of 10 are shown in Schemes 1 and 2. First, the reductive amination of paraꢀ
substituted arylaldehydes 18 with methyl 5ꢀaminosalicylate (17) provided the corresponding Nꢀ
arylmethylaminosalicylates 19 in good to excellent yields. The coupling reaction of the series of substituted
aminosalicylates 19 with Nꢀsulfonylglycine derivatives 20 was conveniently achieved following a modified
procedure with dichlorotriphenylphosphorane (PPh3Cl2) to furnish the tertiary amides 21 in very good yields.30
Subsequent hydrolysis of the methyl esters under basic conditions (NaOHꢀTHFꢀH2O) resulted in the formation
of the desired salicylic acids 11. The Nꢀsulfonylglycine derivatives 20 were prepared from either Nꢀbenzyl or
Nꢀmethyl glycine (22 and 23 respectively) via Nꢀsulfonylation with a range of substituted arylsulfonyl chlorides
to first provide the sulfonamides 24 which were then hydrolyzed.
A library of related Nꢀbenzamides lacking the potentially reactive methylenoxysulfonyl group of 10,
incorporating an arylphosphonate (examples 12a and 12b) or benzylphosphonate (examples 12c and 12d) as a
phosphotyrosine mimetic,31 was prepared as shown in Scheme 2. The amino diethyl phosphonate esters
25a,b,d were prepared by methods shown in the supporting information (Scheme S1), whilst phosphonate 25c
was commercially available. The phosphonateꢀcontaining amides 26a-d were prepared by coupling the amines
25a-d with 4ꢀphenoxybenzoic acid in the presence of EDC and catalytic amount of DMAP in DCM. This was
followed by Nꢀalkylation by treatment with sodium hydride and 1ꢀ(bromomethyl)ꢀ4ꢀcyclohexylbenzene (27, see
Scheme S1) in THF to provide the tertiary amides 28a-d. The final phosphonic acids 12a-d were obtained by
treatment of 28a-d with bromotrimethylsilane (TMSBr) (10 eq.) in dichloromethane. Similarly hydrolysis of
the intermediate amides 26a-c provided the phosphonic acidꢀcontaining amides 29a-c. The library of Nꢀ
benzamides 13 in which the phosphonic acid of 12 is replaced by a salicylic acid was prepared as shown in
Scheme 3. The amide library 30, incorporating a methyl salicylate, was prepared from the amine library 19
5

Medicinal Chemistry Communications
Page 6 of 19
View Article Online
(Scheme 1) via reaction with either a carboxylic acid and dichlorotriphenylphosphorane or directly with an acyl
chloride. Finally, hydrolysis of the methyl salicylate ester of library 30, performed in a sealed tube on a heating
block, provided the desired library of salicylic acids 30 in good yields.
6

Page 7 of 19
Medicinal Chemistry Communications
View Article Online
Scheme 1. Synthetic route to library 11
Scheme 2. Synthetic route to phosphonate containing benzamide analogs 12a-d
O
O
HO
H
N
NH2
O
EDC/DCM/DMAP (cat.)
R1
25a-d
R1
O
, X = Et
26a (42%), 26b (90%)
26c (96%), 26d (98%), X = Et
a, R1 = p-PO(OX)2
b, R1 = m-PO(OX)2
c, R1 = p-CH2PO(OX)2
d, R1 = m-CH2PO(OX)2
Br
27
NaH/THF
O
TMSBr/DCM
0 ºC to rt, 4 h
O
N
N
O
R1
R1
O
12a (56%), 12b (55%)
12c (51%), 12d (81%), X = H
28a (25%), 28b (35%)
28c (69%), 28d (72%), X = Et
O
O
H
N
H
N
TMSBr/DCM
0 ºC to r.t.
R1
R1
O
O
26a-c, X = Et
29a (76%), 29b (81%)
29c (100%), X = H
Scheme 3. Synthetic route to Nꢀbenzamide library 13
7

Medicinal Chemistry Communications
Page 8 of 19
View Article Online
The ability of the libraries 11, 12 and 13 to inhibit STAT3 dimerization was evaluated by a competitive,
fluorescenceꢀpolarization (FP)ꢀbased assay, as developed by Schust and Berg,19, 32 and described in the
Supporting Information, using full length STAT3 (Nꢀterminal GST, SignalChem, Richmond, BC, Canada) and
the fluorescent probe peptide 5ꢀFAM–G(pTyr)LPQTV–CONH2 (GenScript, Piscataway, NJ, USA).32 This
peptide, derived from the gp130 IL6 receptor binds to the STAT3 SH2 domain; blocking its binding provides a
measure of inhibition of STAT3 dimerization which binds through the sequence (pTyr)LKTKF.33 Table 1
shows the STAT3 inhibitory activities of the library of sulfonamides 11 bearing a 5ꢀamidoꢀ2ꢀhydroxybenzoic
acid group. Compound 11a, which bears a tosyl group, equivalent to the sulfonyl Bꢀring of S3Iꢀ201, and a
benzyl group on each nitrogen atom, is weakly active (11a, IC50 = 201.3 ± 1.5 ꢂM). When the tosyl group was
replaced with a biphenylsulfonyl group, a 9ꢀfold increase in the activity was observed (11b, IC50 = 22 ± 9.1 ꢂM)
indicating that further analogs should be pursued. Although, the biphenylsulfonyl group appeared to be
superior to the tosyl group, both series were prepared to further explore SAR relationship among these analogs.
The paraꢀchlorobenzyl derivative 11c is 4ꢀfold more potent (IC50 = 50 ± 3.8 ꢂM) than the unsubstituted analog
11a (IC50 = 201.3 ± 1.5 ꢂM). Consistent with the above observations, the biphenylsulfonyl derivate (11d, IC50
= 15 ± 1.2 ꢂM) showed better activity than its tosyl analog 11c. Other groups such methoxy, cyclohexyl, nꢀ
heptyl, and isoꢀbutyl groups at the paraꢀposition of the amido Nꢀbenzylated derivatives (11e, 11f, 11g and 11h
respectively) resulted in activities similar to that of 11b and 11d (IC50 values 17ꢀ23 ꢂM).
The effect of a methyl group as the R1 group was next assessed. Thus the library of Nꢀmethylsulfonamides 11i-
r was prepared as shown in Scheme 1 from commercially available sarcosine methyl ester (23). In most cases
the Nꢀmethylsulfonamides, are less active than the corresponding Nꢀbenzylsulfonamide (R1 = Bn) counterparts.
For example the methyl group in the biphenylsulfonyl example 11j (IC50 = 57 ± 13 ꢂM) and 11k (IC50 = 61 ± 6
ꢂM), reduces activity compared to their respective benzyl analog 11b (IC50 = 22 ± 9.1 ꢂM) and 11d (IC50 = 15
± 1.2 ꢂM). The presence of an alkyl benzyl group (R = cyclohexyl and nꢀheptyl, Table 1) in the Nꢀmethyl
sulfonamides 11m (IC50 = 32 ± 12 ꢂM) and 11n (IC50 = 22 ± 8 ꢂM) did not alter their inhibitory as compared to
the corresponding Nꢀbenzyl sulfonamides 11f (IC50 = 23.3 ± 2 ꢂM) and 11g (IC50 = 19 ± 3.5 ꢂM). Indeed the
8

Page 9 of 19
Medicinal Chemistry Communications
View Article Online
presence of the cyclohexyl group dramatically improves the activity of the equivalent in tosylsulfonamide 11l
(IC50 = 45 ± 12 ꢂM) (cf. unsubstituted benzyl derivative 11i, (IC50 >1000 ꢂM). This improvement was also
seen with the 2ꢀhydroxyꢀ4ꢀaminobenzoic acid series of Gunning and Turkson.27 The tosylsulfonamide 11l (IC50
= 45 ± 12 ꢂM) is the isomeric form of 15 (SF-1-066) (IC50 = 15 ± 5 ꢂM, FP; IC50 = 35 ± 9 ꢂM, EMSA). In our
hands the activity of 15 (SF-1-066) (IC50 = 50 ± 7 5 ꢂM, FP) is similar to that of 11l. We were unable to
determine an IC50 of 11i since it is so poorly active; the equivalent 2ꢀhydroxyꢀ4ꢀaminobenzoic acid from
Gunning and Turkson27 (compound 14 in reference 27; IC50 = 292 ± 35 ꢂM, EMSA) is also significantly less
active than its cyclohexyl substituted benzyl counterpart. Thus both isomers of the salicylic acid are of
equivalent potency and improved over the IC50 of 10 (S3Iꢀ201) in the EMSA assay (IC50 = 86 ± 33 ꢂM,
EMSA).27, 28 A chloro or fluoro biphenylsulfonamide group of the molecule (Ar) was tolerated (compounds
11o–11r) showing similar activities to their biphenylsulfonamide where made. A small number of heterocyclic
containing (as the R substituent) analogs 11sꢀ11w were prepared to reduce the overall lipophilicity of the
compounds and provide a basic site for salt formation. The sulfonamides 11w (IC50 = 35.3 ± 12.7 ꢂM) and 11t
(IC50 = 33.5 ± 1.8 ꢂM) possessing an Nꢀbenzylamide bearing a metaꢀ4ꢀpyridyl group had moderate inhibitory
activity. The Nꢀbenzylamides 11s (IC50 >300 ꢂM), 11u (IC50 = 94.7 ± 0.32 ꢂM) and 11v (14.4% inhibition at
50 ꢂM) with heterocycles in the paraꢀposition were significantly less active.
We next focused on reducing the size of the glycine linker of 10 to separate the two aryl groups by 2 atoms by
using a simple amide group as shown in 12 and 13 (Figure 1). We first made the Nꢀbenzamides 12a and 12b
which incorporate an arylphosphonic acid as their nonꢀhydrolyzable phosphotyrosine mimic, as shown in
Scheme 2.31 The paraꢀsubstituted isomer 12a (IC50 = 42.0 ± 0.8 ꢂM) was moderately active and better than the
meta isomer 12b (18% inhibition at 50 ꢂM) (Table 2). The two benzylphosphonic acids 12c (IC50 = 28.4 ± 2.9
ꢂM) and 12d (IC50 = 18.9 ± 1.1 ꢂM) showed improved activity compared to 12a. The Nꢀpꢀcyclohexylbenzyl
group clearly contributes to the activity of 12a,c,d since the unsubstituted amides 29a-c are all significantly less
active. The STAT3 inhibitory activity of 12d indicated that the benzamide scaffold merited further attention.
9

Medicinal Chemistry Communications
Page 10 of 19
View Article Online
Since the phosphonates would likely require a prodrug protection strategy12 to render them cell permeable, we
next investigated salicylic acid containing Nꢀbenzamides 13. The analog of 12d with a simple unsubstituted Nꢀ
benzyl substitutent was inactive (Table 3, 13a, 3% inhibition at 50 ꢁM). The 4ꢀmethoxybenzyl derivative 13b
was weakly active (IC50 118.1 ± 8.6 ꢁM) as was its 3,4ꢀdimethoxy analog 13c. Some improvement in activity
was observed when the R substituent was a halogen (13d, IC50 48 ± 9.4 ꢁM and 13e, IC50 52 ± 4.3 ꢁM). The
presence of an alkyl group (R = nꢀheptyl and cyclohexyl, Table 3) resulted in further improvement (13f, IC50
12.8 ± 0.5 ꢁM and 13g, IC50 15 ± 4.4 ꢁM respectively). This effect of the paraꢀalkyl group of the Nꢀ
benzylamide of 11f-h also resulted in analogs with similar activities (Table 1). Replacement of the
phenoxybenzoyl group of 13f (IC50 12.8 ± 0.5 ꢁM) by benzoyl (13h, IC50 32 ± 4 ꢁM) and 3ꢀmethoxybenzoyl
(13i, IC50 33.3 ± 1.6 ꢁM) resulted in a twoꢀfold reduction in activity in both cases. Substitution of the phenyl
group of 13h by a pyridyl group as in 13j, 13k and 13l resulted in significant reduction in activity. The
carboxylic acid 33, an isomer of 13g, possessing the isomeric salicylic acid group, was prepared from 4ꢀaminoꢀ
2ꢀhydroxybenzoic acid (Scheme S2, Supporting Information). This was essentially inactive in the FP assay
(12% inhibition at 50 ꢁM and 24% at 200 ꢁM). At least in this case the 5ꢀaminoꢀ2ꢀhydroxybenzoic acid of 13g
is critical for activity.
10

Page 11 of 19
Medicinal Chemistry Communications
View Article Online
Table 1: In vitro FP inhibitory activity, for disruption of the STAT3–pYLPQTV complex, of library 11
Compound
R
H
R1
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Bn
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Bn
Bn
Bn
Me
Me
Ar
IC50 (µM)a
201.3 ± 1.5
22.2 ± 9.1
50 ± 3.8
15 ± 1.2
22 ± 10
23.3 ± 2
19 ± 3.5
17 ± 3.6
> 1000
11a
11b
11c
11d
11e
11f
H
Cl
Cl
OCH3
n-C7H13
11g
11h
11i
isoꢀButyl
H
H
57 ± 13
61 ± 6
11j
Cl
11k
11l
45 ± 12
32 ± 12
22 ± 8
11m
11n
11o
11p
11q
11r
11s
11t
n-C7H13
isoꢀButyl
Cl
20 ± 7.1
43 ± 6
OCH3
50 ± 9.1
23 ± 0.3
>300
n-C7H13
4ꢀPiperidyl
3ꢀ(4ꢀPyridyl)
4ꢀMorpholinyl
4ꢀPiperidyl
3ꢀ(4ꢀPyridyl)
33.5 ± 1.8
94.7 ± 0.32
11u
11v
11w
14.4 ± 0.6%
inhibition at 50 ꢁM
35.3 ± 12.7
a: The IC50 is defined as the concentration that gives an FP signal half that of the difference between the bound and free states of the
STAT3–pYLPQTV complex (see supporting information).
11

Medicinal Chemistry Communications
Page 12 of 19
View Article Online
Table 2: In vitro FP inhibitory activity, for disruption of the STAT3–pYLPQTV complex, of libraries 12 and 29.
Compound
12a
Structure
IC50 (µM)a
42.0 ± 0.8
12b
12c
17.9 ± 0.5% inhibition at 50 µM
28.4 ± 2.9
18.9 ± 1.1
12d
29a
7.0 ± 2.9% inhibition at 50 µM
29b
29c
9.0 ± 1.7% inhibition at 50 µM
32.8 ± 6.3% inhibition at 50 µM
a: The IC50 is defined as the concentration that gives an FP signal half that of the difference between the bound and free states of the
STAT3–pYLPQTV complex (see supporting information).
12

Page 13 of 19
Medicinal Chemistry Communications
View Article Online
Table 3: In vitro FP inhibitory activity, for disruption of the STAT3–pYLPQTV complex, of library 13.
IC50 (µM)a unless otherwise
Compound
13a
R
Ar
specified
H
3% inhibition at 50 µM
OMe
118.1 ± 8.6
13b
13c
3,4ꢀDiꢀOMeb
13 ± 8% inhibition at 50 µM
Cl
Br
48 ± 9.4
52 ± 4.3
12.8 ± 0.5
15 ± 4.4
13d
13e
13f
13g
Heptyl
Cyclohexyl
Heptyl
Heptyl
32 ± 4
13h
13i
33.3 ± 1.6
Heptyl
Heptyl
Heptyl
13j
13k
13l
22 ± 5% inhibition at 50 µM
21 ± 3% inhibition at 50 µM
31 ± 6% inhibition at 50 µM
Cyclohexyl
33
12 ± 0.2% inhibition at 50 µM
a: The IC50 is defined as the concentration that gives an FP signal half that of the difference between the bound and free states of the
STAT3–pYLPQTV complex (see supporting information). b: indicates 3,4ꢀdisubstitution of the phenyl group bearing the R
substituent.
13

Medicinal Chemistry Communications
Page 14 of 19
View Article Online
The phosphonic acid 12d, and salicylic acids 13f and 13g were docked to the STAT3 SH2 using GLIDE, using
methods described previously.24 The lowꢀenergy docking poses are shown in Figure 3 position the acidic
groups in the pTyrꢀ705 binding site and are broadly similar. The superimposition of all three inhibitors 12d
(green), 13f (yellow) and 13g (blue) is shown in Figure 3a. The metaꢀphosphonic acid group of 12d occupies
the phenylphosphate binding pocket (Figure 3b). Indeed, the phosphorus atom is only 0.2 Å from the position
of the STAT3 pTyrꢀ705 residue Xꢀray coordinates (see Figure S1, supporting information). Hydrogen bonds
are formed between the three phosphate oxygen atoms and with residues Lysꢀ591, Argꢀ609, Gluꢀ612 and Serꢀ
611 (Figure 3c). Charged interactions are evident between the phosphonate and the Lysꢀ591 and Argꢀ609
residues. The phenoxybenzoyl group is positioned with the terminal phenyl group occupying the pY+1
hydrophobic pocket. The pꢀcyclohexylbenzyl group places the cyclohexyl group deep within the hydrophobic
pYꢀX pocket.34
Figure 3. Docking of phosphonic acid 12d and salicylic acids 13f and 13g to the STAT3 SH2 domain. a: overlay of the docked
poses of phosphonate 12d (carbon atoms shown green) and salicylic acids 13f (carbon atoms shown yellow) and 13g (carbon atoms
shown blue) with hydrogen atoms omitted. b: surface rendering of 12d docked to the STAT3 pꢀTyr binding site (carbon and hydrogen
atoms shown gray; oxygen atoms shown red and nitrogen atoms shown blue). c: Schematic binding mode of 12d to the STAT3 SH2
domain showing the hydrogen bonds (red) and hydrophobic interactions (green).
The salicylic acids 13f and 13g dock with their hydroxycarboxylic acid groups deep within the pꢀTyr binding
site. The docking pose of 13g, discussed in detail elsewhere,35 positions its pꢀcyclohexylbenzyl group in the
pYꢀX pocket and the phenoxyphenyl group in a region close to the pY+1 pocket. A similar pose is also
obtained for salicylic acid 13f (shown in yellow in Figure 3a) with the pꢀheptylbenzyl group occupying the pYꢀ
14

Page 15 of 19
Medicinal Chemistry Communications
View Article Online
X pocket. The terminal phenyl group of the phenoxybenzoyl group is located in the pY+1 pocket (see Figure
S2a, supporting information). Overall the docking shows that all three molecules are able to adopt reasonable
conformations that results in polar interactions of the acid group with the pꢀTyr binding pocket and hydrophobic
interaction of each of the sides chains in both the pYꢀX pocket and the area close to or in pY+1 pocket.
Figure 4. Effects of selected compounds upon pSTAT3 levels in MDAꢀMBꢀ468 human breast cancer cell lines after 4 h treatment by
immunoblotting. A: treatment at single dose of 200 ꢁM for all compounds. B: Dose response for selected compounds at 50, 100 and
200 ꢁM.
A selection of the most potent compounds in the FP assay from the libraries 11, 12 and 13 were assessed in a
cellular assay. Their potency derives from their ability to displace fluoresceinꢀGpYLPQTV suggesting that they
bind at the phosphotyrosineꢀ705 binding site in the SH2 domain of STAT3. STAT3 is known to be
phosphorylated on Y705 by the EGF receptor (EGFR) tyrosine kinase. This phosphorylation of Y705 is
preceded by recruitment of STAT3 to EGFR by binding of the STAT3ꢀSH2 domain to pY1068 or pY1086 of
EGFR.36 A compound binding to the SH2 domain of STAT3 is therefore expected to block the binding of
STAT to EGFR and subsequent phosphorylation of Y705 of STAT3. Human breast cancer MDAꢀMBꢀ468 cells
15

Medicinal Chemistry Communications
Page 16 of 19
View Article Online
were treated with the compounds (200 ꢁM) and the levels of pSTAT3 assessed by immunoblotting after 4 hours
(Figure 4A). Most of the compounds reduced the amount of pSTAT3. It was not surprising that the phosphonic
acid 12d did not affect the pSTAT3 levels, since phosphonic acids are not usually cell permeable. Five of the
best compounds, 11o, 11r, 11n, 11h, and 13f were then tested at 50, 100 and 200 ꢁM (Figure 4B) and the
results as shown in Figure 4B indicate that inhibition of pSTAT3 was dose dependent. At 100 ꢁM 11o appears
to increase pSTAT3 and decrease total STAT3 and therefore at this concentration may have activated STAT3.
Nevertheless at 200 ꢁM it is clear that for all compounds shown in Figure 4B that STAT3 is significantly
deactivated. These results are consistent with the ability of the compounds to block the binding of the
fluorescent labeled peptide to the STAT3 SH2 domain. One compound, the benzamide 13g, was selected for
further biological characterization. Its ability to inhibit STAT3 dimerization in vitro and in intact cells and to
suppress malignant transformation in human cancer cells that depend on STAT3 is reported by us in detail
elsewhere.35
In conclusion, we have shown that the phosphotyrosine mimicking 5ꢀaminoꢀ2ꢀhydroxybenzoic acid can be
incorporated into analogs of 1 and provides significantly active STAT3 dimerization inhibitors. We have also
developed a series of Nꢀbenzylbenzamides 13 by removing the reactive sulfonyloxymethyl moeity of the linking
group of 1, as STAT3 dimerization inhibitors with improved potency. Importantly, the equivalent potencies of
13g and 12d further validates 5ꢀaminoꢀ2ꢀhydroxybenzoic acid as a phosphotyrosine mimic.
Acknowledgments
We thank the Moffitt Chemical Biology Core for their outstanding technical assistance and maintenance of the
NMR, mass spectrometry and HPLC instruments. This work was partially supported by NCI grant CA140681
(SMS and NJL).
16

Page 17 of 19
Medicinal Chemistry Communications
View Article Online
Supporting Information: Experimental details of the compound synthesis and characterization, the FP assay
and molecular modeling. Figures S1 and S2 and Schemes S1 and S2.
17

Medicinal Chemistry Communications
Page 18 of 19
View Article Online
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
J. E. Darnell, Jr., Science, 1997, 277, 1630ꢀ1635.
J. E. Darnell, Nat. Med., 2005, 11, 595ꢀ596.
H. Yu and R. Jove, Nat. Rev. Cancer, 2004, 4, 97ꢀ105.
P. Yue and J. Turkson, Expert Opin. Inv. Drug, 2009, 18, 45ꢀ56.
D. Masciocchi, A. Gelain, S. Villa, F. Meneghetti and D. Barlocco, Future Med. Chem., 2011, 3, 567ꢀ597.
A. K. Mankan and F. R. Greten, Expert Opin. Inv. Drug, 2011, 20, 1263ꢀ1275.
A. Lavecchia, C. Di Giovanni and E. Novellino, Curr. Med. Chem., 2011, 18, 2359ꢀ2375.
B. D. Page, D. P. Ball and P. T. Gunning, Expert Opin. Ther. Pat., 2011, 21, 65ꢀ83.
J. L. Yap, J. Chauhan, K.ꢀY. Jung, L. Chen, E. V. Prochownik and S. Fletcher, MedChemComm, 2012, 3, 541ꢀ
551.
10.
11.
12.
13.
J. Turkson, D. Ryan, J. S. Kim, Y. Zhang, Z. Chen, E. Haura, A. Laudano, S. Sebti, A. D. Hamilton and R. Jove,
J. Biol. Chem., 2001, 276, 45443ꢀ45455.
D. R. Coleman, Z. Ren, P. K. Mandal, A. G. Cameron, G. A. Dyer, S. Muranjan, M. Campbell, X. Chen and J. S.
McMurray, J. Med. Chem., 2005, 48, 6661ꢀ6670.
P. K. Mandal, F. Gao, Z. Lu, Z. Ren, R. Ramesh, J. S. Birtwistle, K. K. Kaluarachchi, X. Chen, R. C. Bast, W. S.
Liao and J. S. McMurray, J. Med. Chem., 2011, 54, 3549ꢀ3563.
P. K. Mandal, D. Limbrick, D. R. Coleman, G. A. Dyer, Z. Ren, J. S. Birtwistle, C. Xiong, X. Chen, J. M. Briggs
and J. S. McMurray, J. Med. Chem., 2009, 52, 2429ꢀ2442.
14.
15.
P. K. Mandal, Z. Ren, X. Chen, C. Xiong and J. S. McMurray, J. Med. Chem., 2009, 52, 6126ꢀ6141.
J. Chen, L. Bai, D. Bernard, Z. NikolovskaꢀColeska, C. Gomez, J. Zhang, H. Yi and S. Wang, ACS Med. Chem.
Lett., 2010, 1, 85ꢀ89.
16.
17.
M. Zhao, B. Jiang and F. H. Gao, Curr. Med. Chem., 2011, 18, 4012ꢀ4018.
W. Hao, Y. Hu, C. Niu, X. Huang, C.ꢀP. B. Chang, J. Gibbons and J. Xu, Bioorg. Med. Chem. Lett., 2008, 18,
4988ꢀ4992.
18.
K. Matsuno, Y. Masuda, Y. Uehara, H. Sato, A. Muroya, O. Takahashi, T. Yokotagawa, T. Furuya, T. Okawara,
M. Otsuka, N. Ogo, T. Ashizawa, C. Oshita, S. Tai, H. Ishii, Y. Akiyama and A. Asai, ACS Med. Chem. Lett.,
2010, 1, 371ꢀ375.
19.
20.
21.
J. Schust, B. Sperl, A. Hollis, T. U. Mayer and T. Berg, Chem. Biol., 2006, 13, 1235ꢀ1242.
H. Song, R. X. Wang, S. M. Wang and J. Lin, P. Natl. Acad. Sci. USA, 2005, 102, 4700ꢀ4705.
V. M. Shahani, P. Yue, S. Haftchenary, W. Zhao, J. L. Lukkarila, X. Zhang, D. Ball, C. Nona, P. T. Gunning and
J. Turkson, ACS Med. Chem. Lett., 2011, 2, 79ꢀ84.
22.
23.
K. A. Z. Siddiquee, P. T. Gunning, M. Glenn, W. P. Katt, S. Zhang, C. Schroeck, S. M. Sebti, R. Jove, A. D.
Hamilton and J. Turkson, ACS Chem. Biol., 2007, 2, 787ꢀ798.
H. Li, A. Liu, Z. Zhao, Y. Xu, J. Lin, D. Jou and C. Li, J. Med. Chem., 2011, 54, 5592ꢀ5596.
18

Page 19 of 19
Medicinal Chemistry Communications
View Article Online
24.
K. Siddiquee, S. Zhang, W. C. Guida, M. A. Blaskovich, B. Greddy, H. R. Lawrence, M. L. R. Yip, R. Jove, M.
M. McLaughlin, N. J. Lawrence, S. M. Sebti and J. Turkson, P. Natl. Acad. Sci. USA, 2007, 104, 7391ꢀ7396.
S. Becker, B. Groner and C. W. Muller, Nature, 1998, 394, 145ꢀ151.
25.
26.
L. Lin, R. Amin, G. I. Gallicano, E. Glasgow, W. Jogunoori, J. M. Jessup, M. Zasloff, J. L. Marshall, K. Shetty,
L. Johnson, L. Mishra and A. R. He, Oncogene, 2009, 28, 961ꢀ972.
27.
28.
29.
S. Fletcher, B. D. G. Page, X. Zhang, P. Yue, Z. H. Li, S. Sharmeen, J. Singh, W. Zhao, A. D. Schimmer, S.
Trudel, J. Turkson and P. T. Gunning, ChemMedChem, 2011, 6, 1459ꢀ1470.
S. Fletcher, J. Singh, X. Zhang, P. Yue, B. D. G. Page, S. Sharmeen, V. M. Shahani, W. Zhao, A. D. Schimmer, J.
Turkson and P. T. Gunning, ChemBioChem, 2009, 10, 1959ꢀ1964.
X. Zhang, P. Yue, B. D. G. Page, T. Li, W. Zhao, A. T. Namanja, D. Paladino, J. Zhao, Y. Chen, P. T. Gunning
and J. Turkson, P. Natl. Acad. Sci. USA, 2012, 109, 9623ꢀ9628.
30.
31.
I. Azumaya, T. Okamoto, F. Imabeppu and H. Takayanagi, Tetrahedron, 2003, 59, 2325ꢀ2331.
T. R. Burke, Jr., M. S. Smyth, A. Otaka, M. Nomizu, P. P. Roller, G. Wolf, R. Case and S. E. Shoelson,
Biochemistry, 1994, 33, 6490ꢀ6494.
32.
33.
34.
35.
J. Schust and T. Berg, Anal. Biochem., 2004, 330, 114ꢀ118.
Z. Ren, L. A. Cabell, T. S. Schaefer and J. S. McMurray, Bioorg. Med. Chem. Lett., 2003, 13, 633ꢀ636.
I.ꢀH. Park and C. Li, J. Mol. Recognit., 2011, 24, 254ꢀ265.
X. Zhang, Y. Sun, R. Pireddu, H. Yang, M. K. Urlam, H. R. Lawrence, W. C. Guida, N. J. Lawrence and S. M.
Sebti, Cancer Res., Published OnlineFirst January 15, 2013, 73, doi:10.1158/0008ꢀ5472.CANꢀ1112ꢀ3175
H. Shao, H. Y. Cheng, R. G. Cook and D. J. Tweardy, Cancer Res., 2003, 63, 3923ꢀ3930.
36.
19