Inhibiting aberrant signal transducer and activator of transcription protein activation with tetrapodal, small molecule Src homology 2 domain binders: Promising agents against multiple myeloma

Article abstract of DOI:10.1021/jm3017255

The signal transducer and activator of transcription (STAT) proteins represent a family of cytoplasmic transcription factors that regulate a pleiotropic range of biological processes. In particular, Stat3 protein has attracted attention as it regulates the expression of genes involved in a variety of malignant processes, including proliferation, survival, migration, and drug resistance. Multiple myeloma (MM) is an incurable hematologic malignancy that often exhibits abnormally high levels of Stat3 activity. Although current treatment strategies can improve the clinical management of MM, it remains uniformly incurable with a dismal median survival time post-treatment of 3-4 years. Thus, novel targeted therapeutics are critically needed to improve MM patient outcomes. We herein report the development of a series of small molecule Stat3 inhibitors with potent anti-MM activity in vitro. These compounds showed high-affinity binding to Stat3's SH2 domain, inhibited intracellular Stat3 phosphorylation, and induced apoptosis in MM cell lines at low micromolar concentrations.

Full text of DOI:10.1021/jm3017255

Subscriber access provided by RMIT University Library
Article
Inhibiting Aberrant Signal Transducer and Activator of Transcription
(STAT) Protein Activation with Tetrapodal, Small Molecule Src Homology
2 domain binders: Promising Agents Against Multiple Myeloma
Brent D. G. Page, Danielle C Croucher, Zhihua Li, Yoong Lim Wong, Sina Haftchenary,
Victor H Jimenez-Zepeda, Jennifer Atkinson, Paul A. Spagnuolo, Robert Colagouri,
andrew martin lewis, Aaron D. Schimmer, Suzanne Trudel, and Patrick Thomas Gunning
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm3017255 • Publication Date (Web): 22 Aug 2013
Downloaded from http://pubs.acs.org on September 11, 2013
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Inhibiting Aberrant Signal Transducer and Activator of
Transcription (STAT) Protein Activation with Tetrapodal,
Small Molecule Src Homology 2 domain binders: Promising
Agents Against Multiple Myeloma
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Brent D. G. Page,1 Danielle C. Croucher,2 Zhi Hua Li,2 Yoong Lim Wong,1 Sina
Haftchenary, 1 Victor H. Jimenez-Zepeda,2 Jennifer Atkinson,2 Paul A. Spagnuolo,2
Robert Colagouri,1 Andrew M. Lewis,1 Aaron D. Schimmer,2 Suzanne Trudel,2*
Patrick T. Gunning1*
1Department of Chemical and Physical Sciences, University of Toronto Mississauga, 3359
Mississauga Road, Mississauga ON Canada, L5L 1C6
2Ontario Cancer Institute, Princess Margaret Hospital, 620 University Avenue, Toronto, ON,
Canada, M5G 2C1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 35
1
2
3
4
5
6
ABSTRACT
7
8
9
The signal transducer and activator of transcription (STAT) proteins represent a family of
cytoplasmic transcription factors that regulate a pleiotropic range of biological processes. In
particular, Stat3 protein has received particular attention as it regulates the expression of genes
involved in a variety of malignant processes including proliferation, survival, migration and drug
resistance. Multiple myeloma (MM) is an incurable haematologic malignancy that often exhibits
abnormally high levels of Stat3 activity. Although current treatment strategies can improve the
clinical management of MM, it remains uniformly incurable with a dismal median survival time
post-treatment of 3-4 years. Thus, novel targeted therapeutics are critically needed to improve
MM patient outcomes. We herein report the development of a series of small molecule Stat3
inhibitors with potent anti-MM activity in vitro. These compounds showed high-affinity binding
to Stat3’s SH2 domain, inhibited intracellular Stat3 phosphorylation, and induced apoptosis in
MM cell lines at low micromolar concentrations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
INTRODUCTION
7
8
9
As a master regulator of cell signaling and tumourigenesis, signal transducer and activator of
transcription (Stat) 3 has emerged at the forefront of cancer drug discovery programs.1-3
Aberrant activation of Stat3 has been observed in a number of solid and haematologic
malignancies and is correlated with a variety of hallmark oncogenic processes including cell
growth and survival, angiogenesis, inflammation, and metastatic potential.1-7 Multiple myeloma
(MM) is the second most common haematologic malignancy and is responsible for
approximately 13 % of blood cancers and 1 % of all cancers.8 Although MM is generally
regarded as incurable, traditional high-dose chemotherapeutics and currently available targeted
therapies can improve the prognosis of MM patients when used as part of an aggressive
treatment regimen.8-11 Despite this, the median survival time after conventional treatment remains
disappointingly low (3-4 years).8 In the search for novel molecular targets in MM, Stat3 has
emerged as a driving force behind the maintenance and progression of the disease and it is
anticipated that Stat3 inhibitors will provide a novel and effective weapon in the fight against
MM.12-14
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The Stat3 signaling cascade is initiated by binding of extracellular ligands such as cytokines and
growth factors to their respective cell surface receptors.1, 4, 15 A subsequent conformational
change in the receptor results in intrinsic or kinase-mediated receptor activation via the
phosphorylation of key tyrosine residue within the receptor’s cytoplasmic domain. These
phosphorylated tyrosine residues (pTyr) serve as docking sites for a number of Src-homology 2
(SH2)-domain containing proteins such as Stat3. Thus, the phosphorylated receptor complex
recruits unphosphorylated Stat3 via its SH2 domain and in turn Stat3 is activated by
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 35
1
2
3
4
5
6
7
8
9
phosphorylation of a C-terminal tyrosine residue (Tyr705). Following dissociation from the
receptor complex, phosphorylated Stat3 (pStat3) proteins form transcriptionally active pStat3–
pStat3 homo-dimers via reciprocal interactions between the SH2 domain of one activated Stat3
monomer and the pTyr705 of another. Activated pStat3 dimers subsequently translocate to the
nucleus, where they bind specific DNA response elements to induce target gene expression. In
contrast to the transient nature of Stat3 activation in normal cells, constitutive Stat3 activity, such
as that observed in MM, drives the expression of target genes with known oncogenic roles,
contributing to the maintenance and progression of tumorigenic processes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The SH2 domain of Stat3 plays a critical role in the Stat3 signaling cascade, facilitating
recruitment of Stat3 to activating cell surface receptors, and playing a key role in the formation
of Stat3 homo-dimers.5, 6, 16 Thus, there has been significant effort to silence aberrant Stat3
signaling by targeting the SH2 domain.1-4 As a therapeutic target, the SH2 domain of Stat3, like
many protein–protein interaction interfaces, is devoid of a classical binding pocket and is
relatively flat and hydrophobic.17, 18 However, the SH2 domain possesses a hydrophilic region
(containing residues Lys591, Arg609 and Glu612) responsible for binding to pTyr residues.18
Many groups, including our own, have targeted this sub-pocket for developing Stat3 inhibitors,
employing polar groups such as, phosphate esters,17, 19-23 catechols,24 carboxylates25 and salicylic
acids15, 26-29 that mimic the native pTyr705 substrate. To design Stat3 selective SH2 domain
binders, two proximal sub-pockets (one hydrophobic and one amphipathic) have been
simultaneously targeted using tripodal Stat3 inhibitors.25
Our research group has approached the development of Stat3 inhibitors by performing extensive
structure activity relationship (SAR) analysis of a known Stat3-SH2 domain binder, S3I-201 (1).
This compound was identified through an in silico high-throughput screen,29 and led to the
ACS Paragon Plus Environment

Page 5 of 35
Journal of Medicinal Chemistry
1
2
3
4
generation of two lead compounds, 7 and (BP-1-102) 9, that are now entering advanced pre-
5
6
clinical trials as cancer therapeutic agents (Table 1).15, 26, 30
7
8
9
27
Table 1: Preliminary SAR leading to the development of lead compounds 7 and 9.15,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound activity was assessed using an electrophoretic mobility shift assay (EMSA) as
previously reported.26, 30
In previous work, we discovered several tripodal inhibitors of Stat3 that have demonstrated
promising anti-cancer activity.26, 30 Using the parent compound 1, we generated two lead
compounds 7 and 9 that have demonstrated more potent Stat3 SH2 domain binding, improved
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 35
1
2
3
4
5
6
7
8
9
inhibition of cellular Stat3 activity, and promising activity against cancer cell lines and xenograft
models.15, 27, 30, 31 Interestingly, substitution of the N-methyl group with an oxygen atom, NH or
N-Boc resulted in significant changes in inhibitor activity. This prompted further investigation
into the sulphonamide nitrogen substituent.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MATERIALS AND METHODS
GOLD Docking Simulations
Inhibitors were docked using GOLD docking software to Stat3 crystal structure, pdb 1BG1.
Compounds were first optimized into a low energy geometry. The compound binding site was
set to an area with a 12 Å radius surrounding Ser636. Best solutions were visualized using
Pymol, which was utilized to create the images shown in Figures 1, 2 and 4C and in
supplementary Figure S.1.
Fluorescence Polarization Assay
The fluorescence polarization assay was performed as previously reported. Briefly, fluorescently
labelled peptide probe (5-FAM-GpYLPQTV-NH2, CanPeptide, Pointe-Claire (Montreal),
Quebec, Canada) was incubated with Stat3 protein (SignalChem, Richmond, British Columbia,
Canada), and inhibitor for 30 minutes then analyzed on a Tecan M1000 fluorimeter (Tecan,
Mannedorf, Switzerland). Polarized fluorescence was plotted against concentration of inhibitor
and IC50 values were determined by fitting to a dose response curve. Representative curves of
the top compounds are shown in the supplementary information.
Cell Viability Assays. The MTS and MTT assays were used to measure cellular metabolic
activity, which reflects the number of viable cells. Cells were seeded in 96 well plates at 1-3 x
ACS Paragon Plus Environment

Page 7 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
104 cells/well in 90 µL of fresh culture medium. Prior to the addition of cell suspensions, 10 µL
of test compound (or vehicle control) was added to wells in triplicate. Cultures were then
incubated for 72 hours at 37 °C, 5% CO2. Following treatment, cell viability was assessed by
MTS assay (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-
tetrazolium for AML2, DU145 and MDA468 cell lines, or MTT assay (3-(4,5-Dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide for MM cell lines.15 Relative cell viability (to DMSO
control) was determined colorometrically and EC50 values determined by fitting to a standard
dose response curve when applicable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Immunoblot Analysis. Following treatment, target cells were harvested and washed twice in ice
cold PBS. The resulting cell pellets were lysed in lysis buffer ((50mM Tris-HCl pH 7.4, 150 mM
NaCl, 1 mM ethylenediaminetetra acetic acid (EDTA), and 1% NP-40) supplemented with 1 mM
phenylmethanesulfonyl fluoride (PMSF), 2 mM sodium vanidate (Na3VO4) and protease
inhibitor cocktail (Roche Applied Science) for 30 minutes on ice. Protein lysates were collected
by centrifugation at 14,000 g for 15 minutes. Protein concentration was determined by Bradford
Assay (Thermo Scientific, Rockford, IL) and normalized with lysis buffer before the addition of
β-mercaptoethanol-supplemented Lamelli sample buffer (Bio-Rad Laboratories, Hercules, CA).
Proteins (10-30 µg) were resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis
(SDS-PAGE) using 7-10% gels, and transferred to polyvinylidene fluoride (PVDF) membranes
using wet transfer at 70 V for 1 hour. Membranes were rinsed in Tris-buffered Saline with 0.01%
Tween-20 (TBST) and blocked for 1 hour at room temperature in TBST containing 5% bovine
serum albumin (BSA) powder, followed by overnight incubation with primary antibodies at 4
°C. Primary antibodies against the indicated proteins were diluted in TBST with either 5% BSA
or 5% milk, as specified by manufacturer. Following three 15 minute washes in TBST,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 35
1
2
3
4
5
6
7
8
9
membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit or anti-
mouse secondary antibodies (Thermo Scientific Pierce, Rockford, IL) diluted 1:4000 in TBST
for 1 hour at room temperature. Membranes were developed using the enhanced
chemiluminescence kit (Perkin Elmer, Waitham, MA) according to the manufacturer’s
instructions and visualized by autoradiography. Resulting autoradiographs were analyzed by
densitometry using the Gel Doc XR station and Quantity One Software (Bio-Rad, Hercules, CA).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Luciferase Reporter Assay
Target cells were transduced with replication incompetent, VSV-g pseudotyped lentiviral
particles containing the Stat3-driven Firefly luciferase reporter constructs (pCignal Lenti-
Stat3TRE-FLuc). Transductions were performed with polybrene (8 µg/µl) in accordance with the
Cignal Lenti Reporter Assay Kit (SA Biosciences, Frederick, MD). The pCignal Lenti-Stat3TRE
-
FLuc reporter construct is under the control of a basal promoter element (TATA box) joined to
tandem repeats of a specific Stat3 transcriptional response element (TRE), and regulates the
expression of the mammalian codon-optimized, non-secreted form of the Firefly luciferase gene.
Stably transduced cells were selected using puromycin (2 µg/µl) for 2 weeks. As an internal
control, Stat3FLuc-expressing cells were stably transfected with the pCignal Lenti-CMV-RLuc
reporter construct, which contains a CMV immediately early enhancer/promoter that
constitutively drives Renilla luciferase expression. Transductions were performed as previously
described and stable cells selected with hygromycin (50 µg/µl) for 2 weeks. In vitro reporter
construct activity of drug treated cells was measured using the Stop & Glo® Dual Luciferase
Assay System (Promega, Madison, WI), with data presented as relative luciferase units (RLU =
Firefly luciferase/Renilla luciferase).
ACS Paragon Plus Environment

Page 9 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
RESULTS
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. GOLD32, 33 docking images of compounds bound to Stat3’s SH2 domain (Stat3 pdb
1BG118). A: Compound 1; B: Compound 7; C: Compound 9; D: Compound 8.
Docking simulations were utilized to explore potential binding interactions of compounds 8
which possessed the hydrophobic Boc group appended to the sulfonamide nitrogen. Comparing
8 to parent compounds 1, 7 and 9, we observed interesting differences. Compounds 1, 7 and 9
were found to dock to the SH2 domain of Stat3 with similar conformations as previously
reported.25 We presumed that the salicylic acid group mimics the pTyr motif and facilitated
docking with the polar phosphate binding region. The N-cyclohexylbenzyl substituent, common
to both 7 and 9, was found to interact in silico with the hydrophobic residues, Val637 and
Trp623. The N-methyltoluenesulphonamide group interacted with the amphipathic region which
contained Ile634 and Glu594 as well as the hydrophobic side-chain of Lys591.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 35
1
2
3
4
5
6
7
8
9
When the N-Boc derivative (8) was docked in silico, it was found to orient similarly to 7 and 9.
However, the bulky, hydrophobic t-butyl group was found to disfavorably orientate away from
the protein surface (Figure 2, A and B). However, this seemingly unfavorable docking position
was not reflected in the in vitro EMSA disruption assay, with only a slight decrease in potency,26
and an improved binding affinity observed in the fluorescence polarization (FP) binding assay
(Compound 8 IC50 = 15.8 ± 0.2 µM c.f. compound 7 IC50 = 31.0 ± 9.4 µM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Further docking studies revealed an alternative binding mode for 8 where the Boc group
contributed to protein surface binding (Figure 2, C and D). However, unlike previous studies,
the N-cyclohexylbenzyl moiety was positioned within the amphipathic binding pocket containing
residues Ile634, Glu594 and the side chain of Lys591 (Figure 2). As a result, the substituted
sulphonamide group projected into the hydrophobic cleft composed of residues including,
Trp623 and Val637. The substituent on the sulphonamide nitrogen interacted with Trp623 and
Phe716 and placed the sulfonamide S-substituent in closer proximity to Cys712. This orientation,
facilitated by the adoption of the more unfavourable cis amide, allowed for improved interaction
between the protein and larger hydrophobic substituent appended to the sulphonamide nitrogen.
ACS Paragon Plus Environment

Page 11 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Popular binding modalities of 8 bound to Stat3’s SH2 domain (pdb 1BG118). Images A
and B show the previously predicted docking poses of 8. Images C and D show an possible
alternative binding mode, where the N-cyclohexylbenzyl and sulfonamide substituents are
orientated differently.
To further explore the in vitro Stat3 binding potency and whole cell biological effects of N-
alkylated analogs of both 7 and 9, we herein report the synthesis of a novel library of salicylic
acid-based inhibitors. Inhibitors were functionalized with select substituents stemming from the
sulfonamide nitrogen to furnish a series of tolyl-N-alkyl and perfluorobenzene-N-alkyl
derivatives of 7 and 9, respectively.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 35
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1: Synthesis of the tolyl-N-alkyl derivatives. a: i. AcOH, ii. NaCNBH3, MeOH, 6 h, 45
°C; b: TsCl, DIPEA, MeCN, 16 h, 0 °C to RT; c: Boc2O, DIPEA, CH2Cl2, 1 h, RT; d:
LiOH.H2O, THF:H2O (3:1), 16 h, RT; e: PPh3Cl2, CHCl3, 0.5 h, 110 °C, microwave; f: RBr or
RCl, Cs2CO3, DMF, 1 h, RT; g: H2, Pd/C, THF:MeOH (1:1), 6 h, RT or i. LiOH.H2O, H2O:THF,
16 h, RT; ii. TFA:toluene (1:1), 1 h, RT.
The tolyl-N-alkyl derivatives were prepared from doubly O-benzyl protected 4-aminosalicylic
acid (10) which was coupled to 4-cyclohexylbenzaldehyde under standard reductive amination
conditions using NaCNBH3. The resultant secondary aniline, 11, was then coupled to
functionalized carboxylic acid, 13. Carboxylate 13 was prepared from glycine methyl ester
hydrochloride which was tosylated using tosyl chloride and then Boc protected using Boc-
anhydride. Saponification of the methyl ester gave the acid (13) which was coupled to the
aniline using triphenylphosphine dichloride. The peptide coupling also cleaved the Boc-
protecting group as two equivalents of HCl are generated during this reaction. The
sulphonamide nitrogen was functionalized with a variety of alkyl bromides or alkyl chlorides and
then deprotected using hydrogenolysis or a step-wise saponification of the benzyl ester followed
by treatment with TFA to cleave the benzyl ether. This protocol was used to produce a library of
45 compounds in a divergent fashion. All compounds were subjected to analysis in a
ACS Paragon Plus Environment

Page 13 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
fluorescence polarization (FP) assay and MTS proliferation assay against DU145, MDA-468 and
AML2 cells.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2: Synthesis of perfluorobenzene-N-alkyl derivatives. a: PPh3Cl2, Fmoc-Gly-OH,
CHCl3, 0.5 h, 110 °C, microwave; b: DMF:piperidine (9:1), 0.5 h, RT; c: C6F5SO2Cl, K2CO3,
MeCN, 4 Å MS, 4 h, 0 °C - RT; d: RBr, Cs2CO3, DMF, 1 h, RT; e: H2, Pd/C, THF:MeOH (1:1),
6h, RT.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 35
1
2
3
4
5
6
7
8
9
To make the perfluorobenzene-N-alkyl derivatives, Fmoc-glycine was coupled to secondary
aniline, 11, using triphenylphosphine dichloride. The Fmoc group was removed using piperidine
in DMF to afford the free amine. Sulfonamide 19 was prepared by treating amine 18 with
pentafluorobenzenesulfonyl chloride. A variety of different alkyl bromides were then used to
furnish the sulphonamide nitrogen then treatment with hydrogen and 10% Pd/C gave the
deprotected final molecules. Of note, a step-wise deprotection procedure could not be used for
the synthesis of the perfluorobenzene-N-alkyl derivatives as treatment with LiOH.H2O led to
nucleophilic aromatic substitution of the perfluorobenzene ring, placing a hydroxyl group para-
to the sulphonamide substituent. Again, all compounds were subjected to analysis in an FP assay
and MTS cell proliferation assay against DU145, MDA-468 and AML2 cells. Selected
compounds were further evaluated via Western blot analysis for pStat3 and in MM cell lines for
cytotoxicity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2: Tolyl-N-alkyl Derivatives IC50 reported for FP assay.
ACS Paragon Plus Environment

Page 15 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 35
1
2
3
4
Table 3: Perfluorobenzene-N-alkyl derivatives.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An FP assay for measuring phosphopeptide:Stat3-SH2 domain disruption was performed as
previously reported.34 Relative to parent compounds (7 and 9), many of the N-alkyl derivatives
showed improved activity. Moderate improvements were observed through addition of simple
alkyl groups. However, it appeared that substituted benzyl substituents provided the greatest
enhancement of inhibitory activity. Substitution with the polar pyridine or aminobenzyl
appendages led to a marked loss in protein binding affinity. Compounds that incorporated an N-
(arylsulfonyl)amide exhibited lower binding affinities, with the exception of the N-Boc analogs.
As a preliminary screen, inhibitors were subjected to an MTS assay to assess the in vitro anti-
tumour activity of compounds. Most promisingly, a number of these compounds potently
inhibited the viability of a range of human cancer cell lines including prostate cancer (DU145),
breast cancer (MDA-468) and leukemia (AML2). EC50 data and corresponding structures for the
ACS Paragon Plus Environment

Page 17 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
leading five compounds from each family and their corresponding parent structures are
summarized in Figure 3. Dose response curves are shown in supplementary information.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. IC50 values for the top five tolyl-N-alkyl and perfluoro-N-alkyl compounds as
calculated by MTS assay.15, 26, 27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 35
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. A: Western blot analysis of basal pStat3 activation in a panel of human MM cell lines.
Quantitative analysis by densitometry shown reveals relative levels of pStat3 to total Stat3
protein, and relative total Stat3 protein to GAPDH. B: MTT assay with top eight N-alkylated
Stat3 inhibitors against 8226 and H929 MM cell lines. C: GOLD docking images of lead
compounds 21h and 16i. Compounds were found to optimally bind in a similar conformation to
compound 8, suggesting that the N-alkyl group improves binding potency due to a more
complete occupation of the Stat3 SH2 domain. D: MTT cell viability assay with 16i (upper) and
21h (lower) against panel of MM cell lines.
ACS Paragon Plus Environment

Page 19 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Based on protein binding affinity and inhibition of cell proliferation in DU145, AML2 and
MDA-468 cell lines we selected the top eight compounds for further evaluation in the context of
MM. We reasoned that lead compounds should demonstrate greater activity against MM cell
lines that harbor high levels of pStat3 as they would presumably be more reliant on aberrant
Stat3 activity. Therefore, prior to testing lead compound potency, baseline Stat3 activation was
first examined in a panel of genetically heterogeneous MM cell lines using Tyr705
phosphorylation as a surrogate marker of Stat3 activation. Whole cell lysates prepared from MM
cell lines in logarithmic growth conditions were subject to immunoblot analysis and probed with
antibodies against pStat3 (Tyr705) and total Stat3 protein. Although Stat3 protein was expressed
in all MM cell lines, albeit to varying degrees, constitutively active pStat3 was detected in 6 of 8
MM cell lines, with two cell lines, MM1.S and SKMM2, lacking detectable pStat3 (Figure 4A).
Densitometric analysis performed on immunoblots to quantitate the ratio of pStat3 to total Stat3
protein confirmed variability in baseline pStat3. In the interests of examining the effects of our
Stat3 inhibitors, we selected MM cell lines possessing a variety of pStat3 levels for screening
lead compounds, predicting that cell lines with high pStat3 should be more sensitive to Stat3
inhibition.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As a preliminary evaluation of anti-MM activity, lead agents were subjected to an MTT assay
against two MM cell lines that possessed constitutive Stat3 activation. We selected these two
cell lines based on levels of constitutively activated Stat3 protein. A low pStat3 (H929) and a
high pStat3 (RPMI-8226) cell line were chosen. We reasoned that cells with lower levels of
constitutively active Stat3 would be less sensitive to Stat3 inhibition. Compounds, 16i and 21h
demonstrated favorable activity and selectivity within their respective libraries and were chosen
to undergo further analysis against a larger panel of human MM cell lines.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 35
1
2
3
4
5
6
7
8
9
In docking analyses of inhibitor binding, 16i and 21h were found to dock to the Stat3 SH2
domain in a similar orientation to 8, where the N-alkyl substituents contributed to protein surface
binding. Notably, the 2-trifluoromethylbenzyl group interacted with the hydrophobic cleft
consisting of residues Trp623 and Val637, while the sulfonamide appendage was projected into a
previously unoccupied site containing Cys712, Phe 716 and Glu638. We hypothesized that this
interaction may lead to more complete occupation of the Stat3 SH2 domain.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As shown in Figure 4D, both 16i and 21h demonstrated dose-dependent inhibition of MM cell
viability after 72 hours of treatment, as assessed by MTT assay. Compared with immunoblot
analysis of relative baseline pStat3 levels in tested MM cell lines, we noted that 21h, the
pentafluorophenyl-containing analog, showed activity against non-pStat3 containing MM cell
lines such as SKMM2 and MM1.s, whereas 16i exhibited lower biological activity against these
cell lines. Moreover, JJN3, which contains high constitutive Stat3 activation, was more resistant
to 21h than 16i. Notably, these compounds were approximately two-fold more potent than
parent compounds 7 and 9 (data not shown). Taken together, these findings suggested that the
potent anti-MM activity of 21h may be, at least in part, due to off-target effects, whereas 16i
delivered a more desirable activity profile and will likely provide a larger therapeutic window.
Alternatively, the broad activity of 21h against this panel of MM cell lines, regardless of baseline
Stat3 phosphorylation status, may reflect a universal dependence of MM tumour cells on non-
canonical Stat3 signaling pathways that are dependent on a functional SH2 domain, but not Stat3
phosphorylation.
Given the activity profiles of 16i and 21h in MM cell lines, we next evaluated the ability of these
compounds to inhibit Stat3 phosphorylation. Exposure to 16i and 21h for 6 hours lead to dose-
dependent inhibition of pStat3, and as expected, no inhibition of total Stat3 protein levels (Figure
ACS Paragon Plus Environment

Page 21 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5A). As Stat3 is a master transcriptional regulator, we also employed a Stat3-driven luciferase
reporter construct to evaluate Stat3 transcriptional activity. In agreement with inhibition of Stat3
phosphorylation, treatment with 16i and 21h potently inhibited the transcriptional activity of
Stat3 in 8226 and XG6 cell lines, with reductions in relative luciferase ranging from
approximately 50-80% after 6 hours (Figure 5B). For 16i, the observed inhibition of
transcriptional activity correlated well with the initial MTT results. Conversely, treatment with
7.5 µM of 21h had little effect on luciferase production, however, dosing 21h at this
concentration had potent biological effects in the MTT assay. These results suggested that while
both 16i and 21h inhibited Stat3 phosphorylation and transcriptional activity, the increased
cellular activity of 21h may be due to off-target effects.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To evaluate whether 16i/21h-mediated inhibition of Stat3 phosphorylation and transcriptional
activity was sufficient to abrogate downstream Stat3-induced gene expression, we evaluated the
effect of these compounds on a known Stat3 target gene, c-Myc. Since this particular protein is
known to have a very short half-life (20-30 minutes),35 we evaluated the resulting effects of drug
treatment on c-Myc protein expression after 6 hours using immunoblot analysis. Consistent with
the previously observed decreases in Stat3 transcriptional activity, both 16i and 21h dose-
dependently reduced c-Myc protein expression (Figure 5A). However, in a separate analysis,
negligible decreases were observed in other known Stat3 targets such as Bcl-xL and survivin
(data not shown), which we speculate to be a result of differences in protein-specific kinetics.
To further confirm the induction of apoptosis following treatment with 16i and 21h, whole cell
lysates of inhibitor treated JJN3 cells were collected and subjected to immunoblot analysis for
cleaved Poly ADP-ribose polymerase (cPARP), a marker of apoptosis. As shown in Figure 5A,
both 16i and 21h induced PARP cleavage.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 35
1
2
3
4
5
6
7
8
9
To further characterize the cellular activity of 16i and 21h and the mechanisms by which they
effect MM cell viability, we evaluated 16i and 21h induced apoptosis using flow cytometric
analysis of Annexin V and PI staining. In 8226 cells, both 16i and 21h induced apoptosis in a
dose- and time-dependent manner as represented by a shift of cells from the lower left quadrant
(viable cells), to the lower right quadrant (early apoptotic cells) at 24 hours, and migration to the
upper right quadrant (late apoptosis) at 48 hours (Figure 5C). Promisingly, analysis of apoptosis
in MM cell lines with varying degrees of sensitivity to 16i and 21h revealed similar results to
those observed in the MTT assay, with a greater induction of apoptosis in 8226 cells compared to
XG6 and JJN3 cells (Figure 5D).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. A: Western blot analysis of 16i- and 21h- mediated effects on pStat3 inhibition in
JJN3 tumour cells, revealing dose-dependent inhibition of Stat3 phosphorylation, inhibition of
Stat3 target gene expression as demonstrated by c-Myc and induction of apoptosis shown by
ACS Paragon Plus Environment

Page 23 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
increased PARP cleavage. B: Luciferase assay demonstrating that after 6 hours, both 16i and 21h
dose dependently inhibit Stat3-driven luciferase expression. C: Flow cytometric analysis of 16i-
and 21h- mediated apoptosis as measured by Annexin V and PI staining. Representative
scatterplots for 8226 cells showing increased population of cells in the lower right quadrant after
24 hours of treatment, which migrate to the upper right quadrant after 48 hours. D: Analysis of
apoptosis in 8226, XG6 and JJN3 human MM cell lines upon treatment with 16i and 21h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 35
1
2
3
4
5
6
7
8
9
Figure 6. A: Analysis of 16i and 21h selectivity for cytokine-induced STAT phosphorylation.
Serum-starved U937 cells were treated with 16i or 21h for 4 h prior to stimulation with indicated
cytokine for 15 minutes. Graphical comparison of 16i and 21h selectivity for inhibiting Stat1, 3,
and 5 phosphorylation as assessed by phospho-flow cytometry. B: Fluorescence polarization
assays with Stat1, Stat3 and Stat5 were conducted with lead inhibitors 16i and 21h to investigate
isoform selectivity.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To address the selectivity of these compounds for inhibiting Stat3 over other STAT isoforms in
cells, we performed phospho-flow cytometry to investigate the effects of these compounds on
cytokine-induced Stat1/3/5 phosphorylation. Although both 16i and 21h were shown to inhibit
IL-6-induced Stat3 phosphorylation in these experiments, similar levels of inhibition were also
observed for GM-CSF-induced Stat5 phosphorylation and IFNλ-induced Stat1 phosphorylation
(Figure 6A). Limited STAT isoform selectivity was also demonstrated using reported FP assays
which measure inhibition of native phosphopeptide binding to Stat1. In this assay we found that
both 16i and 21h show little selectivity for the Stat3 isoform over Stat1 (16i, Stat1 IC50 = 5.8 ±
0.6 µM cf. Stat3 IC50 = 2.8 ± 4.3 µM; 21h Stat1 IC50 = 10.9 ± 0.8 µM cf. Stat3 IC50 = 12.8 ± 2.8
µM, Figure 6 B). Thus, improving STAT isoform selectivity remains a goal for future compound
libraries.
To further probe selectivity of 16i, we conducted a KinomeScan at 5 µM, analogous to
approximate IC50 values determined in MM cells, to assess potential off-target effects in 132
different cellular kinases, including 37 that possessed SH2 domains. Most encouragingly, this
assay revealed only minor off-target effects against SH2 domain containing kinases at 5 µM,
ACS Paragon Plus Environment

Page 25 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
only 2 of the 37 SH2 domain-containing kinases displayed greater than 25 % inhibition of kinase
activity. Of note, known upstream regulators of STAT activity, such as JAK1, JAK2, JAK3,
TYK2 and SRC, were not inhibited by 16i (relative activity between 84 and 100 %). A full list
of kinases screened and summary of relative inhibition can be found in the supplementary
materials.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Given the activity profile of 16i, we assessed this compound in primary MM patient samples.
Encouragingly, 16i demonstrated activity against malignant plasma cells (CD138+) from
primary MM patient samples (Figure 7A), with 20 µM treatment reducing MM tumour cell
viability by over 50% in 3 patient samples. Furthermore, at doses exceeding 20 µM, 16i
demonstrated little activity against non-MM (CD138-) cells (Figure 7B). Furthermore, at doses
of 30 µM, 16i had little effect on haematopoietic progenitor colony formation, suggesting that
this compound does not inhibit the ability of normal haematopoietic progenitors to proliferate or
form distinct colonies (Figure 7C). Taken together, our analysis of 16i in the context of primary
MM patient samples revealed a therapeutic window. Furthermore, although data presented here
suggests that 21h may be a less selective inhibitor as compared to 16i, it remains an intriguing
anti-cancer compound, displaying potent in vitro cytotoxic effects in MM cell lines at low µM
concentrations.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 35
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7. Activity of 16i against primary MM patient samples. Mononuclear cells (MNC) from
MM patients were obtained by Ficoll-Paque separation of 6 patient derived bone marrow
aspirates. Samples were cultured and treated with 16i followed by staining with antibodies
against CD138 (MM cell surface marker) or Annexin V (apoptosis). Results are presented as the
decrease in CD138+ cell population, representing MM cells (A), and decrease in CD138- cell
population, representing non-MM cells (B left). Alternatively, isolated MNCs were cultured in
MethoCult (StemCell Technologies), and treated with 16i to evaluate the activity of this agent on
healthy hematopoietic progenitor colony formation (B right).
ACS Paragon Plus Environment

Page 27 of 35
Journal of Medicinal Chemistry
1
2
3
4
DISCUSSION AND CONCLUSIONS
5
6
7
8
9
We have presented a novel library of salicylic acid-based small molecule Stat3 inhibitors that
offer promising Stat3-SH2 binding affinity and anti-MM activity. Lead compounds, 16i and 21h,
offer improved in vitro binding activity over precursors, 7 and 9, respectively, and improved
anti-cancer whole cell activity. In silico binding evidence suggested that these compounds bound
with greater efficiency to the Stat3 SH2 domain. While NMR and X-ray crystallographic
structural studies are ongoing to identify exact binding modes, we can deduce from the presented
SAR is that Stat3’s SH2 domain might be accommodating of larger, tetrapodal analogs of the 7
and 9 scaffolds. The caveat however, is that this modification appears to reduce STAT isoform
selectivity. Both 16i and 21h were shown to disrupt phopohopeptide:Stat3 protein complexes,
inhibit Stat3 phosphorylation as well as block target gene transcription. Moreover, both
compounds have significant anti-MM activity, potently decreasing MM cell viability and
promoting the induction of apoptosis. While compound 21h was the more potent inhibitor,
further studies to identify potential off-target effects must be undertaken. Although not as
potent, 16i was shown to have a more favorable toxicity profile, with no observed cytotoxicity in
healthy hematopoietic cells or in MM cell lines that harbor minimal pStat3. Furthermore, we
have demonstrated that lead agent 16i has minimal effects against a panel of cellular kinases in
vitro, suggesting that this Stat3 inhibitor is not likely interacting with kinases to inhibit Stat3
activation. Thus, while we cannot claim that the anti-MM activity of lead agents is a sole result
of Stat3 inhibition, evidence does suggest that anti-Stat3 activity does a play a role in the
observed biological results. Ongoing biological studies aim to delineate the exact target or
combination of targets to better explain the promising results presented herein. Further
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 35
1
2
3
4
experiments are underway to characterize the therapeutic utility of both 16i and 21h in in vivo
5
6
studies to identify a Stat3 inhibitor candidate suitable for advanced preclinical trials in MM.
7
8
9
ACKNOWLEDGEMENTS
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
This work was supported by the National Science Engineering Research Councils (NSERC)
(PTG), University of Toronto (PTG), and by an NSERC CGSD fellowship (BDGP).
Corresponding authors
Patrick T. Gunning, E-mail: patrick.gunning@utoronto.ca; Tel.: 905-828-5354; Fax: 905-569-
5425.
Suzanne Trudel, E-mail: strudel@uhnres.utoronto.ca; Tel.: 416-946-4501 (Ext 4566).
Supporting Information Available: Chemical methods, characterization, and additional
information for biological assays.
Abbreviations
SH2
Src Homology 2 domain
STAT3
cPARP
GM-CSF
MM
Signal Transducer and Activator of Transcription (3) Protein
cleaved Poly ADP-ribose polymerase
Granulocyte macrophage colony-stimulating factor
Multiple Myeloma
MTT
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, a yellow tetrazole
ACS Paragon Plus Environment

Page 29 of 35
Journal of Medicinal Chemistry
1
2
3
4
5
6
MTS
(3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-
tetrazolium)
7
8
9
siRNA
RNAi
PI
small interfering RNA
RNA interference
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Propidium iodide
FP
Fluorescence Polarization
Mononuclear cells
MNC
GOLD
Genetically Optimized Ligand Docking
REFERENCES
1. Haftchenary, S.; Avadisian, M.; Gunning, P. T. Inhibiting aberrant Stat3 function with
molecular therapeutics: A progress report. Anticancer Drugs 2011, 22, 115-127.
2. Page, B. D. G.; Ball, D. P.; Gunning, P. T. Signal transducer and activator of transcription 3
inhibitors: a patent review. Exp. Opin. Ther. Pat. 2011, 21, 65-83.
3. Fletcher, S.; Turkson, J.; Gunning, P. T. Molecular approaches towards the inhibition of the
Signal Transducer and Activator of Transcription 3 (Stat3) Protein. ChemMedChem, 2008, 3,
1159-1168.
4. Fletcher, S.; Drewry, J. A.; Shahani, V. M.; Page, B. D. G.; Gunning, P. T. Molecular
disruption of oncogenic signal transducer and activator of transcription 3 (STAT3) protein.
Biochem. Cell Biol. 2009, 87, 825-833.
ACS Paragon Plus Environment