Structure-Based Discovery of SD-36 as a Potent, Selective, and Efficacious PROTAC Degrader of STAT3 Protein

Article abstract of DOI:10.1021/acs.jmedchem.9b01530

Signal transducer and activator of transcription 3 (STAT3) is a transcription factor and an attractive therapeutic target for cancer and other human diseases. Despite 20 years of persistent research efforts, targeting STAT3 has been very challenging. We report herein the structure-based discovery of potent small-molecule STAT3 degraders based upon the proteolysis targeting chimera (PROTAC) concept. We first designed SI-109 as a potent, small-molecule inhibitor of the STAT3 SH2 domain. Employing ligands for cereblon/cullin 4A E3 ligase and SI-109, we obtained a series of potent PROTAC STAT3 degraders, exemplified by SD-36. SD-36 induces rapid STAT3 degradation at low nanomolar concentrations in cells and fails to degrade other STAT proteins. SD-36 achieves nanomolar cell growth inhibitory activity in leukemia and lymphoma cell lines with high levels of phosphorylated STAT3. A single dose of SD-36 results in complete STAT3 protein degradation in xenograft tumor tissue and normal mouse tissues. SD-36 achieves complete and long-lasting tumor regression in the Molm-16 xenograft tumor model at well-tolerated dose-schedules. SD-36 is a potent, selective, and efficacious STAT3 degrader.

Full text of DOI:10.1021/acs.jmedchem.9b01530

Subscriber access provided by Karolinska Institutet, University Library
Article
Structure-Based Discovery of SD-36 as a Potent, Selective
and Efficacious PROTAC Degrader of STAT3 Protein
Haibin Zhou, Longchuan Bai, Renqi Xu, Yujun Zhao, Jianyong Chen, Donna McEachern, Krishnapriya
Chinnaswamy, Bo Wen, Lipeng Dai, Praveen Kumar, Chao-Yie Yang, Zhaomin Liu, Mi Wang,
Liu Liu, Jennifer L. Meagher, Han Yi, Duxin Sun, Jeanne A. Stuckey, and Shaomeng Wang
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b01530 • Publication Date (Web): 20 Nov 2019
Downloaded from pubs.acs.org on November 20, 2019
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 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 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.
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 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Structure-Based Discovery of SD-36 as a Potent, Selective and
Efficacious PROTAC Degrader of STAT3 Protein
Haibin Zhou,†+± Longchuan Bai,†+± Renqi Xu , Yujun Zhao , Jianyong Chen , Donna
†+±
†+±
†+±
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†+
#
▽
▽
▽
McEachern , Krishnapriya Chinnaswamy , Bo Wen , Lipeng Dai , Praveen Kumar , Chao-Yie
†
+
†+
†+
†+
#
†+
▽
Yang , Zhaomin Liu , Mi Wang , Liu Liu , Jennifer L. Meagher , Han Yi , Duxin Sun ,
Jeanne Stuckey , and Shaomeng Wang†+¶§*
#
†
+
¶
Rogel Cancer Center, Departments of Internal Medicine and Pharmacology, Medical School,
§
#
▽
Medicinal Chemistry, College of Pharmacy, Life Sciences Institute, Department of
Pharmaceutical Sciences, College of Pharmacy, University of Michigan, Ann Arbor, Michigan
48109, United States
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT:
Signal transducer and activator of transcription 3 (STAT3) is a transcription factor and an
attractive therapeutic target for cancer and other human diseases. Despite twenty-years of
persistent research efforts, targeting STAT3 has been very challenging. We report herein the
structure-based discovery of potent small-molecule STAT3 degraders based upon the proteolysis
targeting chimera (PROTAC) concept. We first designed SI-109 as a potent, small-molecule
inhibitor of the STAT3 SH2 domain. Employing ligands for cereblon/Cullin 4A E3 ligase and SI-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
09, we obtained a series of potent PROTAC STAT3 degraders, exemplified by SD-36. SD-36
induces rapid STAT3 degradation at low nanomolar concentrations in cells and fails to degrader
other STAT proteins. SD-36 achieves nanomolar cell growth inhibitory activity in leukemia and
lymphoma cell lines with high levels of phosphorylated STAT3. A single dose of SD-36 results in
complete STAT3 protein degradation in xenograft tumor tissue and normal mouse tissues. SD-36
achieves complete and long-lasting tumor regression in the Molm-16 xenograft tumor model at
well-tolerated dose-schedules. SD-36 is a potent, selective and efficacious STAT3 degrader.
2
ACS Paragon Plus Environment

Page 3 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Signal transducer and activator of transcription 3 (SATT3) is a member of the STAT family
1
of transcription factors that transmits signals from cell surface receptors to the nucleus. Persistent
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
activation of STAT3 is often associated with poor prognosis of human cancers since activated
STAT3 signaling not only provides advantages of growth, survival and metastasis to tumor cells,
but also suppresses the antitumor immune response.2-3 STAT3 is an attractive therapeutic target
for human cancer and other human diseases.
In the canonical mechanism, STAT3 is recruited from the cytosol onto the cell membrane
where it interacts with different cytokine receptors through its Src homology 2 (SH2) domain. This
results in phosphorylation of its Tyr705 residue (Y705), and its subsequent dimerization and
1
translocation into the nucleus for gene transcription. It has been proposed that dimerization of
STAT3 is required for its translocation from cytosol into the nucleus and for its transcriptional
activity, and pharmacological inhibition of STAT3 SH2 domain could effectively block the
1
dimerization and transcriptional activity of STAT3. Consequently, considerable efforts have been
made in the last two decades to develop small molecular inhibitors of the STAT3 SH2 domain.3-4
Because STAT3 homodimerization is mediated by a phosphotyrosine-containing peptide
from one monomer and a well-defined binding pocket in the SH2 domain of the second monomer,
peptidomimetic inhibitors have been designed based upon STAT3 phosphotyrosine-containing
peptides. Although previous peptidomimetic inhibitors have been reported to achieve high binding
affinities to STAT3 in biochemical binding assays, they are generally not cell-permeable due to
3
-4
the presence of their negatively charged phosphate group and their peptidic characteristics. Non-
peptide small-molecule inhibitors of the STAT3 SH2 domain have been reported, but they often
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
have weak to moderate binding affinities to STAT3, making it difficult to establish firmly that
STAT3 is the primary cellular target.3-4 Furthermore, because STAT3 shares a high sequence
homology of its SH2 domain with other STAT members, it has been challenging to design SH2
domain inhibitors for STAT3 that are highly selective over other STAT proteins. Significantly, in
contrast to initial hypotheses, it has now become clear that the monomeric STAT3 protein can
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
translocate from cytosol into the nucleus and is transcriptionally active. Consequently, even if a
potent, cell-permeable and highly selective inhibitor of the STAT3 SH2 domain is ultimately
discovered, it can be predicted that it would only be able to partially block the gene transcriptional
activity of STAT3. Accordingly, there is a clear need to develop a new therapeutic approach to
targeting STAT3.
Therefore, in a therapeutic setting, specific down-regulation or knock-down of STAT3 may
6
be much more effective than inhibition of its dimerization. To support this idea, AZD9150, which
was designed as a STAT3 antisense oligonucleotide (ASO), has shown promising preclinical and
3
, 7
clinical activity, despite some major limitations associated with ASO molecules .
Recently, the Proteolysis Targeting Chimeras (PROTAC) strategy has gained momentum
for its promise in the discovery and development of an entirely new type of therapeutics for the
treatment of human diseases.8-11 A PROTAC molecule is a bifunctional small-molecule designed
8
to recruit cellular degradation machinery to induce targeted protein degradation. We therefore
proposed to design PROTAC degraders of STAT3 as a new strategy to effectively target STAT3.
In the present study, we report our structure-based discovery of SD-36 as a potent, exceptionally
selective, and highly efficacious PROTAC degrader of the STAT3 protein.
RESULTS AND DISCUSSION
4
ACS Paragon Plus Environment

Page 5 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of small molecule STAT3 SH2 domain inhibitors containing a novel phosphotyrosine
mimetic
Our laboratory has previously reported the design of peptidomimetic compounds with high
12
binding affinities to the STAT3 SH2 domain, such as CJ-887 (Figure 1), which binds to the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
human recombinant STAT3 SH2 protein with K = 47 nM in a fluorescence polarization (FP) assay
i
(Figure S1 and Table S1). Despite this high binding affinity to STAT3, CJ-887 is ineffective in
inhibition of STAT3 Y705 phosphorylation in cells, suggesting its poor cell permeability. We
hypothesized that the poor cell-permeability of CJ-877 could be largely attributed to its doubly
ionized natural phosphate group.13-14 A second major issue associated with CJ-877 is its expected
susceptibility to degradation by protein tyrosine phosphatases (PTPs).13-14 To overcome these
major limitations, we have performed extensive optimization of CJ-887, with the key steps
outlined in Figure 1.
Figure 1. Design of potent and cell-permeable small molecule inhibitors of the STAT3 SH2
domain. Chemical structures and binding affinities of our previously reported compound CJ-887
and newly designed and synthesized inhibitors.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Cyclization of the amino group with the phenyl group in CJ-887 forms an indole (2), which
converts CJ-887, a phosphotyrosine into a conformationally constrained, non-amino acid
phosphate. The resulting compound 2 binds to STAT3 with a K
i
value of 25 nM, and is thus twice
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
as potent as CJ-887 (Figure 1 and Table S1). To address the potential susceptibility of the
phosphate ester bond in 2 to hydrolysis by PTPs, we replaced the -OPO
3
2
H group with a non-
hydrolyzable -CH
2
-PO
3
H
2
group13-14, yielding compound 3. Compound 3 binds to STAT3 with a
K
i
value of 761 nM, and is thus 30 times weaker than 2. We reasoned that the reduced binding
affinity of 3 is partially attributable to the weaker acidity of the -CH PO group compared to
2
3
H
2
that of the -OPO
3
H
2
group. To test this idea, we replaced the -CH
2
PO
3
H
2
group with -CF
2
PO
3
2
H ,
which has been used previously in the design of phosphotyrosine mimetics.13-17 Compound 4,
containing the -CF
100-times more potent than 3 and 7 times more potent than the initial peptidomimetic 1 (CJ-877).
Our predicted binding model of compound 4 in a complex with the STAT3 SH2 domain
Figure S2) suggested that part of the 8-membered ring in 4 is exposed to the solvent environment.
2
PO
3
H
2
group binds to STAT3 with a very high affinity (K = 7 nM), and is
i
>
(
To facilitate the synthesis of STAT3 degraders, we replaced one of the carbon atoms in the eight-
membered ring with a nitrogen atom, producing 5 with a methyl substituent, and 6 with an acetyl
substituent. Both 5 and 6 bind to STAT3 with the same high affinity, achieving K values of 14
i
nM. Next we installed a second phenyl group at the α position of the benzyl group in 5 and 6 to
restrict the conformation of the original phenyl group, yielding 7 and SI-109 (8), respectively.
Compounds 7 and SI-109 bind to STAT3 with K values of 12 nM and 14 nM, respectively.
i
To compare directly the phosphotyrosine mimetic used in compounds 4-8 with other
phosphotyrosine mimetics, we used SI-109 as the template molecule and synthesized 9, 10 and 11
(Figure 1). Compound 9 in which the -CF2-PO
3
2
H moiety in SI-109 is replaced with a phosphate (-
6
ACS Paragon Plus Environment

Page 7 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
OPO
3
H
2
) is 2-times less potent than SI-109. Compound 10 in which the -CF
2
PO
3
H moiety in SI-
2
1
09 is replaced with –CH
2
PO
3
2
H is 14-times less potent than SI-109. Compound 11 containing 4-
1
7
phosphonodifluoromethylcinnamate , binds to STAT3 with K
i
value of 164 nM and is >10 times
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
less potent than SI-109. These binding data show that the combination of the bicyclic indole with
the difluoromethylphosphate moiety has yielded an excellent phosphotyrosine mimetic, which is
used in compounds 4-8.
Hence, optimization of 1 (CJ-887) has yielded a series of high-affinity, small-molecule
ligands for the STAT3 SH2 domain, including SI-109 (8), which was employed in our subsequent
design of PROTAC STAT3 degraders.
Determination of the co-crystal structure of SI-109 in a complex with STAT3
To understand the structural basis for the high binding affinity of our designed STAT3
inhibitors, we determined the co-crystal structure of SI-109 in a complex with the STAT3 SH2
domain at a resolution of 3.15 Å (PDB access code: 6NUQ, Figure 2 and SI, Table S2).
Comparison of the co-crystal structure of SI-109 in a complex with STAT3 and the STAT3
SH2 dimeric crystal structure showed that SI-109 captures all the critical interactions with STAT3
that are observed in the STAT3 dimeric complex. The difluoromethylphosphate in SI-109 forms
extensive hydrogen bonding interactions with the side chains of Arg609, Ser611 and Ser613 and
the backbone amide nitrogen atoms of Glu612 and Ser613. An additional hydrogen bond is formed
between the pro-(S)-F atom of the difluoromethyl group with Arg609. The indole moiety in SI-
1
09 sits in a shallow cleft formed between Pro639 and Ser636 of STAT3. The glutamine group in
SI-109 forms two hydrogen bonds with the side chain of Gln644. The backbone carbonyl moiety
of the glutamine group in SI-109 forms two water-mediated hydrogen bonds with the hydroxyl
group of Tyr647 and the amino group of Lys658. The pro-(S)-phenyl group of the diphenylmethyl
group in SI-109 interacts with Leu666, Met660 and Ile659, whereas a pro-(R)-phenyl group has
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
no specific interactions with any residue in STAT3 and is fully exposed to solvent environment.
Although no specific interactions are observed between the [8,5]-bicyclic ring system in SI-109
and the STAT3 protein, optimal hydrogen bonds are formed between the amide groups in the
proximity of the [8,5] bicyclic ring and the protein. Specifically, the amino group tethered to the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
-membered ring forms an optimal hydrogen bond with the backbone carbonyl group of Ser636
and the carbonyl group attached to the 5-membered ring forms a hydrogen bond with the hydroxyl
group of Tyr647 and additionally, a water-mediated hydrogen bond with the backbone carbonyl
group of Lys658. Hence, the bicyclic ring system functions as a scaffold which ensures that the
difluoromethylphosphonic acid and the glutamine moieties in SI-109 are ideally positioned and
oriented to achieve optimal interactions with STAT3.
Figure 2. Co-crystal structure of SI-109 with STAT3 (PDB code: 6NUQ). (A) The Fo-Fc electron
density generated from an omit map contoured at 3σ shows placement of atoms of SI-109; (B)
Detailed interactions of STAT3 with SI-109. SI-109 (green) and side chains of interacting residues
of STAT3 with SI-109 are shown in stick form. Hydrogen bonds are depicted as dashed lines.
This co-crystal structure of SI-109 in a complex with STAT3 provides the structural basis
for its high binding affinity to STAT3, and facilitated our design of STAT3 degraders.
8
ACS Paragon Plus Environment

Page 9 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of PROTAC STAT3 degraders
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Design of putative STAT3 PROTAC degraders based upon cereblon ligands and our
STAT3 SH2 domain inhibitor SI-109.
VHL/cullin 2 and cereblon(CRBN)/cullin 4A E3 ligase complexes have been successfully
employed in the design of PROTAC degraders for different proteins.8-11 For the VHL/cullin 2 E3
ligase complex, high-affinity peptidomimetic ligands have been discovered18-21 and have been
successfully used for the design of highly potent PROTAC degraders for different proteins.8-11 For
the cereblon(CRBN)/cullin 4A E3 ligase complex, ligands for cereblon, such as lenalidomide and
pomalidomide, are small-molecule drugs with molecular weights between 250 and 260, which are
ideal for the design of PROTAC degraders with excellent physiochemical properties. Accordingly,
we have decided to employ lenalidomide and pomalidomide (Figure 3) for the design of STAT3
degraders.
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Our co-crystal structure of SI-109 in a complex with the STAT3 SH2 domain shows that
the amino group of the eight-membered ring in SI-109 is exposed to the solvent (Figure 2) and is
a suitable site at which to tether a ligand for an E3 ligase complex in the design of PROTAC
STAT3 degraders. We have designed and synthesized a series of potential STAT3 degraders
linking this site in SI-109 with a cereblon ligand through a linker (Figure 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Determination of the optimal linker length
Our previous studies demonstrated that the length of the linker plays a key role in protein
degradation potencies in our BET and MDM2 PROTAC degraders.22-25 We synthesized and
evaluated a series of putative STAT3 degraders with different linker lengths (Table 1).
An acute myeloid leukemia (AML) cell line, Molm-16 expresses high levels of
phosphorylated STAT3 at the Tyr705 residue (pSTAT3Y705) protein and total (phosphorylated
2
6
and unphosphorylated) STAT3 protein. We thus used the Molm-16 cell line to evaluate our
putative STAT3 degraders for their ability to degrade both the total STAT3 protein and
pSTAT3Y705 protein. Because PROTAC degraders can induce protein degradation rapidly, we
treated the Molm-16 cells for 4 h with our putative STAT3 degraders and obtained the data
shown in Table 1 and Figure S3.
Table 1. STAT3 degraders with various linker lengths.
10
ACS Paragon Plus Environment

Page 11 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DC (STAT3, µM)
Molm-16, 4h
5
0
Cmpds
SI-109
Linker
>10
1
1
2
3
2.63
0.26
SD-36 (14)
0.06
0.09
0.15
1
1
5
6
In this study, we examined the degradation of both the total STAT3 protein and
pSTAT3Y705 protein for our designed STAT3 degraders and found that our STAT3 degraders
induce degradation of the total STAT3 protein and pSTAT3Y705 protein with very similar
potencies. Hence, we determined the DC50 value, the concentration required to reduce the protein
level by 50% of the total STAT3 protein for each compound and used this value to establish the
structure-activity relationships.
Western blotting data show that the STAT3 inhibitor SI-109 has no effect on the level of
total STAT3 protein and only modestly decreases the level of the pSTAT3Y705 protein at
concentrations up to 1 µM. Compound 12, with a total of 6 non-hydrogen atoms in its linker
between SI-109 and the cereblon ligand, has no effect on the level of total STAT3 protein up to 1
µM and reduces pSTAT3Y705 protein only minimally at 1 µM (Table 1 and Figure S3). However,
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
with higher concentrations, 12 dose-dependently reduces the levels of both the total STAT3 and
pSTAT3Y705 proteins and has an DC50 value of 2.63 µM for the total STAT3 protein (Table 1 and
Figure S3). Increasing the length of the linker in 12 by one carbon yielded 13, which effectively
reduces the levels of both the total STAT3 and pSTAT3Y705 proteins and with DC50 = 0.26 µM, is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0 times more potent than 12. Increasing the linker length in 13 by one additional methylene group
yielded SD-36 (14). SD-36 is very effective and potent in reducing the levels of the total STAT3
and pSTAT3Y705 proteins and achieves a DC50 of 0.06 µM.
Increasing the linker length by introduction of one more methylene group in SD-36 resulted
in 15, which has a DC50 of 0.09 µM and is thus similarly potent as SD-36. Increasing the linker
length in 15 by two methylenes yielded 16, which has DC50 = 0.15 µM and is thus 2-3 times less
potent than SD-36. Our data thus established that a linker length of 8-9 atoms, as in SD-36 or 15,
is optimal for these STAT3 degraders to reduce the levels of both the total STAT3 and pSTAT3Y705
proteins.
Investigation of the linker composition
Our previous studies showed that the linker composition has a major effect on the ability
of our MDM2 and BET PROTAC degraders to induce degradation of the target proteins.23-25
Accordingly, we investigated the linker composition using SD-36 as the template compound, and
obtained the data summarized in Table 2 and Figure S4.
Changing the amide group in SD-36 to an amino group yielded 17, which has DC50 = 0.65
µM and is >10 times less potent than SD-36 in reducing the levels of total STAT3 and pSTAT3Y705
proteins (Table 2 and Figure S4A). Conversion of the amide group into a carbamate or urea group
led to 18 and 19, respectively. While 18 is about 4 times less potent than SD-36 in inducing STAT3
degradation, 19 is as potent as SD-36. Replacing a CH
2
-CH linker moiety in SD-36 with a
2
12
ACS Paragon Plus Environment

Page 13 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pyrazole ring in either side of the linker resulted in 20 and 21, respectively, and both compounds
are 3-4 times less potent than SD-36. Replacement of a CH
2
-CH
2
-CH linker moiety in SD-36 with
2
a piperidine ring yielded 22 or 23. Compound 22 is 3 times less potent than SD-36, and 23 is 40
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
times less potent than SD-36 (Figure S4B).
Table 2. STAT3 degraders with various linker compositions.
DC (STAT3, µM)
Molm-16, 4h
5
0
Cmpds
SD-36
Linker
0.06
1
7
0.65
0.24
0.05
0.18
18
1
9
0
2
2
2
2
2
1
2
3
4
0.25
0.18
2.47
0.38
0.16
0.09
25
2
6
Because both 17 and 23 contain a linker with a positively charged group and are much less
potent than SD-36, a positive charge in the linker of these degraders would appear to be detrimental
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to their ability to degrade STAT3 protein. To test this idea further, we changed the amino group in
2
3 to an amide, affording compound 24. Although 24 is still 6 times less potent than SD-36, it is
more potent than 23 in reducing the levels of the total STAT3 and the pSTAT3Y705 proteins,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
supporting the hypothesis. Changing the ethynyl group in SD-36 to -CH
2
-CH
2
- or -NH-CH - led
2
to compounds 25 and 26, respectively, both of which are 2-3 times less potent than SD-36 in
reducing the levels of STAT3 and pSTAT3Y705
.
Taken together, these data show that both the length and the composition of the linker play
an important role in the ability of our designed STAT3 degraders to reduce the levels of STAT3
and pSTAT3Y705, and the optimal linkers are those used in SD-36 and compound 19.
Examination of the CRBN ligand portion in SD-36
We next performed modifications of the CRBN ligand portion in SD-36 (Table 3).
27
Thalidomide and lenalidomide bind to CRBN with similar modes and affinities. We
replaced the lenalidomide moiety in SD-36 with a thalidomide moiety, producing 27. Surprisingly,
27 has DC50 = 3.54 µM and is thus >50 times less potent than SD-36 (Table 3 and Figure S5).
The co-crystal structure of lenalidomide in a complex with CRBN suggests two additional
linking positions in the phenyl ring of lenalidomide and accordingly, we synthesized 28 and 29.
While these two compounds effectively degrade STAT3 and pSTAT3Y705, both are 5-10 times less
potent than SD-36 (Table 3 and Figure S5).
Table 3. STAT3 degraders with modifications of the Cereblon ligand portion.
14
ACS Paragon Plus Environment

Page 15 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DC (STAT3, µM)
Molm-16, 4h
5
0
Code Name
SD-36
R
0.06
2
7
3.54
0.41
0.36
28
2
9
Modifications of the STAT3 ligand portion of SD-36
The co-crystal structure of SI-109 in a complex with STAT3 shows that one of the phenyl
groups lacks specific interactions with STAT3 protein (Figure 2). We modified this part of the
STAT3 ligand in SD-36 to investigate the effect on STAT3 degradation (Table 4).
Removal of one phenyl group from SD-36 yielded compound 30, which is 4 times less
potent than SD-36 in reducing STAT3 protein (Table 4 and Figure S6). Removal of both the phenyl
groups from SD-36 resulted in 31, which achieves a DC50 = 0.64 µM and is thus 10-times less
potent than SD-36. These data indicate that the two phenyl groups in SD-36 are important for SD-
3
6 to achieve its high potency in inducing degradation of STAT3.
Table 4. STAT3 degraders with modified STAT3 ligand portion.
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DC (STAT3, µM)
Molm-16, 4h
5
0
Cmpds
SD-36
R
0.06
3
3
0
1
0.22
0.64
Evaluation of various phosphotyrosine mimetics
In SD-36, we employed 5-(difluoro(phosphoro)methyl)-1H-indole-2-carboxylic acid to
mimic the phosphotyrosine group. To investigate the importance of this phosphotyrosine mimic in
achieving potent and effective STAT3 degradation by SD-36, we synthesized 32, 33 and 34 (Table
5
) with three different phosphotyrosine mimetics15-17
.
Compounds 32 and 33 fail to degrade both the total STAT3 and pSTAT3Y705 proteins at
concentrations up to 10 µM (Figure S7). Compound 34 reduces the levels of both the total STAT3
and pSTAT3Y705 proteins with DC50 = 4.45 µM, and is >70-times less potent than SD-36 (Table 5
and Figure S7). Hence, the phosphotyrosine mimic in SD-36 is critically important for achieving
potent and effective degradation of the total STAT3 and pSTAT3Y705 proteins.
Table 5. STAT3 degraders with modifications of the STAT3 Ligand Portion.
16
ACS Paragon Plus Environment

Page 17 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
DC (STAT3, µM)
Molm-16, 4h
50
Cmpds
SD-36
R
0.06
3
2
>10
>10
4.45
33
3
4
Design and synthesis of a control compound for SD-36
To facilitate our investigation of the cellular mechanism of action of our potent STAT3
degrader SD-36, we synthesized SD-36Me (Figure 4), in which a methyl group was installed to
block the binding of SD-36 to cereblon.27-28
SD-36Me is completely ineffective in inducing STAT3 degradation in Molm-16 cells at
concentrations up to 10 µM (Figure S8), clearly suggesting that the degradation of STAT3 by SD-
36 is cereblon and cullin-4A ligase-dependent, consistent with the PROTAC mechanism. We
employed SD-36Me and the STAT3 inhibitor SI-109 as control compounds for further
investigation of the activity, specificity and cellular molecular mechanism of SD-36.
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Design of SD-36Me as a control compound.
Evaluation of representative degraders in additional in the SU-DHL-1 lymphoma cell line
The SU-DHL-1 anaplastic large cell lymphoma (ALCL) cell line has a high level of
pSTAT3Y705 protein and, we evaluated the ability of several representative STAT3 degraders in
the SU-DHL-1 cell line and obtained the data shown in Table 6 and Figure S9.
SD-36 effectively and potently reduces the levels of STAT3 and pSTAT3Y705 with DC50
of 28 nM in the SU-DHL-1 cell line with a 16 h treatment time (Table 6 and Figure S9). SI-109
and SD-36Me had no effect on the levels of STAT3 and activated pSTAT3Y705 at concentrations
up to 10 µM in SU-DHL-1 cells.
26
Compound 19, which potently induces degradation of total STAT3 and pSTAT3Y705
proteins in Molm-16 cells with DC50 = 50 nM, is also potent and effective in reducing STAT3
and pSTAT3Y705 protein levels in SU-DHL-1 cells with a DC50 value of 11 nM.
Compounds 18 and 30 are 4 times less potent than SD-36 in degradation of STAT3
protein in Molm-16 cells, and both compounds are 2 times less potent than SD-36.
Compounds 17 and 31, which are >10-times less potent than SD-36 in Molm-16 cells, are
3
0 times less potent than SD-36 in SU-DHL-1 cells.
18
ACS Paragon Plus Environment

Page 19 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compounds 27 and 34 have DC50 values of 3-5 µM in the Molm-16 cell line, but in the
SU-DHL-1 cell line, both compounds are ineffective in reducing the levels of STAT3 and
pSTAT3Y705 proteins in SU-DHL-1 cells up to 5 µM.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Table 6. Degradation potencies of STAT3 protein and cell growth-inhibitory activities of STAT3
degraders in the Molm-16 and SU-DHL-1cell lines.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Molm-16
SU-DHL-1
DC50 (STAT3, µM)
Cmpds
DC50 (STAT3, µM),
IC ± SD (µM)
IC ± SD (µM)
50
50
4h
4 days
16 h
4 days
SI-109 (8)
>10
2.63
0.26
0.06
0.09
0.15
0.65
0.24
0.05
0.18
0.25
0.18
2.47
0.38
0.16
0.09
3.54
0.41
0.36
0.22
0.64
>10
6.45 ± 4.16
> 10
> 10
n.t.a
1
2
3
0.95± 0.081
0.078 ± 0.029
0.013 ± 0.010
0.025 ± 0.007
0.042 ± 0.004
0.147 ± 0.027
0.128 ± 0.050
0.091 ± 0.081
0.241 ± 0.246
0.232 ± 0.175
0.099 ± 0.054
0.36 ± 0.056
0.067 ± 0.032
0.063 ± 0.016
0.059 ± 0.016
0.582 ± 0.654
0.104 ± 0.071
0.711 ± 0.262
0.210 ± 0.109
0.226 ± 0.114
>2
n.t.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
n.t.
0.028
n.t.
n.t.
SD-36 (14)
0.61 ± 0.014
n.t.
1
5
6
1
n.t.
n.t.
17
0.919
0.049
0.011
n.t.
> 10
18
19
20
21
2.37 ± 0.70
0.43 ± 0.16
n.t.
n.t.
n.t.
22
n.t.
n.t.
23
n.t.
n.t.
24
n.t.
n.t.
25
26
27
28
n.t.
n.t.
n.t.
n.t.
> 10
n.t.
> 10
n.t.
29
n.t.
n.t.
30
0.055
0.89
n.t.
3.24 ± 0.57
> 10
31
32
33
34
n.t.
>10
>2
n.t.
n.t.
4.45
>10
1.25 ± 0.71
> 10
> 10
> 10
SD-36Me (35)
5.21 ± 6.04
> 10
a
not tested.
Evaluation of degradation selectivity of SD-36 on other STAT proteins in cells
20
ACS Paragon Plus Environment

Page 21 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
There are seven STAT family members, STAT1, 2, 3, 4, 5A, 5B and 6, which share a
highly homologous SH2 domain. To investigate the degradation selectivity of SD-36, we evaluated
SD-36 for its selectivity in both Molm-16 and SU-DHL-1 cells with SI-109 and SD-36Me included
as controls.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Molm-16, 4 h
SD-36
SU-DHL-1, 18 h
SD-36
10 µM
10 µM
A
B
STAT3
STAT3
pSTAT3
pSTAT3
(Y705)
(
Y705)
STAT1
STAT1
STAT2
STAT4
STAT2
STAT5A/B
STAT5A/B
STAT6
STAT6
β-actin
β-actin
Figure 5. Western blotting analysis of STAT proteins in Molm-16 and SU-DHL-1 cells treated
with SD-36 or control compounds SD-36Me and SI-109. (A) Molm-16 cells were treated with SD-
3
6 at 0.01-10 µM or SD-36Me and SI-109 at 10 µM for 4 h. (B) SU-DHL-1 cells were treated with
SD-36 at 0.01-10 µM or SD-36Me and SI-109 at 10 µM for 18 h.
As shown in Figure 5, SD-36 is very effective in reducing the level of STAT3 protein at
concentrations as low as 0.1 µM but it has little or no effect on STAT1, STAT2, STAT4,
STAT5A/B and STAT6 proteins in both Molm-16 cell and SU-DHL-1 cell lines up to 10 µM.
Hence, SD-36 demonstrates >100-fold selectivity for degradation of STAT3 over other STAT
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
members in both Molm-16 cell and SU-DHL-1 cell lines. The control compound SD-36Me has no
effect on the levels of any STAT protein at 10 µM. Similarly, the STAT3 inhibitor SI-109 has no
effect on the levels of any STAT protein.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Our data therefore show that SD-36 achieves an exceptional degradation selectivity for
STAT3 over other STATs (Figure 5).
To gain a further insight into the degradation selectivity of SD-36 for STAT3 over other
2
6
STAT proteins, we determined its binding affinities to different STAT proteins. Our binding data
showed that while SD-36 has a high affinity for STAT3 with a K value of 44 nM, it also displays
a good affinity to STAT1 and STAT4 with K values of approximately 1 µM. SD-36 has a much
d
2
6
d
2
6
weaker affinity to other STAT proteins tested. Hence, although SD-36 binds to STAT1 and
STAT4 with a good affinity, it fails to degrade these two STAT proteins, as well as other STAT
proteins in cells. We posit that the high degradation selectivity for SD-36 is achieved through the
formation of a stable and productive degradation ternary complex consisting of STAT3:SD-
3
6:CRBN/Cullin4A, as compared to much less stable and productive ternary complexes for other
2
9
STAT proteins. Further investigations are needed to fully understand the exceptional STAT3
degradation selectivity for SD-36.
Evaluation of STAT3 degraders for their growth-inhibitory activities
As SD-36 is able to effectively and selectively degrade STAT3 protein in Molm-16 and
SU-DHL-1 cells, we therefore tested all of our STAT3 degraders reported in this study for their
ability to inhibit cell growth in the Molm-16 cell line and a number of representative STAT3
degraders in the SU-DHL-1 cell line. SI-109 and SD-36me were included as control compounds
in the cell growth experiments. The data are summarized in Table 6.
22
ACS Paragon Plus Environment

Page 23 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In the Molm-16 cell line, SD-36 exerts a potent growth-inhibitory activity with IC50 = 13
nM after a 4-day treatment. The control compound SD-36Me shows modest growth-inhibitory
activity with IC50 = 5.21 µM, and is 400-times less potent than SD-36.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
There is an excellent correlation between the STAT3 degradation potencies and cell
growth inhibitory activities for these STAT3 degraders in the Molm-16 cell line. All compounds
which effectively and potently degrade STAT3 protein also potently inhibit cell growth in the
Molm-16 cell line. Compounds such as 32 and 33, which fail to degrade STAT3 protein up to 10
µM, also display very weak activity in inhibition of cell growth.
In the SU-DHL-1 line, SD-36 effectively inhibits cell growth and achieves IC50 values of
0.61 µM, while STAT3 inhibitors SI-109 and SD-36Me have no effect on cell viability up to 10
µM. Compound 19, which effectively degrades STAT3 with DC50 = 28 nM, displays an IC50
value of 0.43 µM. Compounds 18 and 30, which achieve DC50 values of 49 nM and 55 nM,
respectively, have IC50 values of 2.37 µM and 3.24 µM, respectively in the SU-DHL-1 line.
Compounds 17 and 31, which have DC50 values of 0.92 µM and 0.89 µM, respectively, have
minimal activity in inhibition of cell growth with IC50 values > 10 µM. Compound 34, which
fails to degrade STAT3 up to 10 µM, is also ineffective in inhibition of cell growth (IC50 >10
µM). In general, STAT3 degraders show weaker cell growth inhibitory activity in the SU-DHL-1
line than in the Molm-16 cell line.
Hence, effective degradation of STAT3 results in robust growth inhibition in both Molm-
16 and SU-DHL-1 cell lines, and there is a good correlation between STAT3 degradation and
cell growth inhibition in both cell lines.
Pharmacodynamics evaluation of SD-36 in native and xenograft tumor tissues in mice
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Our cellular data demonstrate that SD-36 and 19 are two potent STAT3 degraders. We
performed pharmacodynamics (PD) experiments to evaluate the ability of SD-36 and 19 to degrade
STAT3 protein in mice.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We first investigated SD-36 and 19 for their ability to degrade STAT3 protein in native
mouse tissue. Western blotting analysis showed that a single intravenous dose of both SD-36 and
1
9 at 25 mg/kg are very effective in reducing STAT3 protein in native mouse spleen tissue with
the effect persisting for more than 48 h, and SD-36 appears to be more effective than 19 (Figure
A). We thus chose to further evaluate SD-36 in vivo.
6
A
SD-36 (25 mg/kg)
19 (25 mg/kg)
Veh 3h 24h 48h 72h
Veh 3h 24h 48h 72h
Mouse #
1
2
3
4
5
6
7
8
9 10
1 2 11 12 13 14 15 16 17 18
STAT3
Spleen
GAPDH
SD-36 (50 mg/kg)
B
Veh 1 h 3 h 24 h 48 h 72 h
6 7 9 10 11 12
#
1
2
3
4
5
8
STAT3
Molm-16
xenograft tumor
pSTAT3 (Y705)
β-actin
Figure 6. Pharmacodynamics study of SD-36 and 19 in native mouse spleen tissue in CD-1 mice (A) and SD-36 in
the Molm-16 xenograft tumors in SCID mice (B). CD-1 and SCID mice were treated intravenously with a single
dose of SD-36 or 19 as indicated. Native spleen tissue or tumor lysates were analyzed by immunoblotting.
24
ACS Paragon Plus Environment

Page 25 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
We evaluated SD-36 for its ability to degrade STAT3 in the Molm-16 xenograft tumor
tissue in mice (Figure 6B). Our western blotting data showed that a single dose of SD-36 at 50
mg/kg is very effective in reducing the levels of both total STAT3 and pSTAT3Y705 proteins in
mouse tumor tissue. Both total STAT3 and pSTAT3Y705 proteins were reduced by >95% after 1 h,
indicating a rapid degradation of STAT3 in vivo. Even 72 h (3 days) after dosing, SD-36 still
reduced the levels of total STAT3 and pSTAT3Y705 proteins by >90% as compared to the control.
Our PD data thus demonstrate that a single dose of SD-36 is highly effective in inducing near
complete elimination of STAT3 protein in tumor tissue in mice with the effect persisting for >3
days.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Antitumor activity of SD-36 in the Molm-16 xenograft model in mice
Based upon its impressive PD data in the Molm-16 xenograft tumor tissue, we evaluated
SD-36 for its efficacy in the Molm-16 xenograft model in mice at different doses and schedules.
T
T
r
r
e
e
a
a
t m
t m
e
e
n
n
t
q
W
D
A
B
t
q
( 1 , 3 , 5 )
M O LM 1 6 T u m o r G ro w th
M
O
L
M
1
6
%
W
e
ig
h
t
C
h
a
n g e
T r e a tm e n t q W
V
e
h
i c l e
C
o
n
t
r
o
l ,
I
V
, q D ( 1 , 3 , 5 )
T r e a tm e n t q D (1 ,3 ,5 )
2500
2
2
1
1
5
0
5
0
5
0
5
0
5
S
S
D
D
- 3
- 3
- 3
6
6
2
5
m
g
/ k
/ k
g
g
,
,
I
V
,
q
D
D
( 1 , 3 , 5
)
5
0
m
g
I
V
,
q
( 1 , 3 , 5 )
2
1
1
000
500
000
S
D
6
5
0
m
g
/ k
g
,
I
V
,
q
W
V e h ic le C o n tr o l, IV , q D (1 ,3 ,5 )
S D -3 6
S D -3 6
2 5
5 0
m
m
g /k g , I V
g /k g , I V
,
q D ( 1 ,3 ,5 )
q D ( 1 ,3 ,5 )
,
S D -3 6
5 0
m
g /k g , I V
,
q W
-
1
1
5
00
0
-
-
1
5
20
25
30
35
40
45
50
1
5
2
0
2
5
3
0
3
5
4
0
4
5
5 0
E F -2 0 1 8 -0 7
E F - 2 0 1 8 - 0 7
D a y s p o s t Im p la n ta tio n
D
a
y
s
p
o
s
t
Im
p
l a
n
t a t io n
Figure 7. In vivo antitumor activity of SD-36 in the MOLM-16 xenograft model in mice. (A) Tumor growth. (B)
Percentage of animal body weight change.
Administered to mice at either 25 or 50 mg/kg three times per week for three weeks
intravenously, SD-36 is capable of inducing complete tumor regression (Figure 7A). In fact,
complete tumor regression was achieved after dosing for only 2 weeks at either 25 or 50 mg/kg
and persisted for more than 2 weeks after termination of the treatment.
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Because a single dose of SD-36 at 50 mg/kg is capable of reducing STAT3 protein levels
for >3 days, we evaluated its efficacy with a weekly dosing schedule. SD-36 at 50 mg/kg,
administered once weekly for 3 weeks is very effective in inhibition of tumor growth (Figure 7A).
Furthermore, SD-36 at 100 mg/kg weekly dosing was capable of achieving complete tumor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
26
regression. In the efficacy experiment, SD-36 caused no weight loss or other signs of toxicity
Figure 7B).
Hence, SD-36 is highly efficacious in vivo and achieves complete and long-lasting tumor
(
regression in the MOLM16 cancer xenograft model at well-tolerated dose-schedules.
CHEMISTRY
Scheme 1. Synthesis of 5-((di-tert-butoxyphosphoryl)methyl)-1H-indole-2-carboxylic acid (41)
Reagents and conditions: a. NaH, Boc O, THF, 0 °C to rt; b. NBS, Bz O , CCl , reflux, 77% over 2 steps; c. (t-
2
2
2
4
o
2 2
BuO) PONa 1.0 equiv. NaH 2.5 equiv., THF, then reflux for 3 h; d. LiOH, MeOH-THF-H O, 60 C.
Scheme 1 outlines the synthesis of the protected phosphotyrosine mimetic, 5-((di-tert-
butoxyphosphoryl)methyl)-1H-indole-2-carboxylic acid (41). Boc protection of 36 followed by
benzylic bromination provided 38. Treatment of 38 with sodium di-tert-butyl phosphite yielded a
mixture of 39 and 40 which were hydrolyzed to afford compound 41.
Scheme 2. Synthesis of 3-((diethoxyphosphoryl)difluoromethyl)-1H-indole-2-carboxylic acid (46).
26
ACS Paragon Plus Environment

Page 27 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Reagents and conditions: a. P(OEt) , 100 °C, 84%; b. Ti(O-iPr) , BnOH, 100 °C, 83%; c. NaH, Cbz-Cl, THF, 0 °C to
3
4
rt, 88%; d. NFBS, NaHMDS, THF, -78 °C to rt, 95%; e. H /Pd-C, THF, 94%.
2
Scheme 2 shows the construction of the 5-((diethoxyphosphoryl)-difluoromethyl)-1H-
2
6
indole-2-carboxylic acid (46), an α,α-difluoro derivative of 41. Compound 38 was heated in
P(OEt) at 100 °C for 12 h furnishing the phosphate (42), which was subject to transesterification,
affording the benzyl ester (43). The indole nitrogen of 40 was protected with a Cbz group, giving
4. Difluorination of the double-activated methylene of 44 provided 45, and finally, hydrogenation
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
of 45 removed two protecting groups, furnishing the carboxylic acid (46).
Scheme 2. Synthesis of compounds 2-4.
Reagents and conditions: a. EDC-HCl 2.0 equiv., HOBt 2.0 equiv., EtN(i-Pr) , DCM, rt, 3 h, 89% yield; b. CF CO H
2
3
2
2
mL, Et Si-H 0.1 mL, DCM 3 mL rt 1 h c. 1). Protected phosphotyrosine mimetics, EDC-HCl, HOBt, EtN(i-Pr) ,
3 , , ; 2
DCM, rt, 3 h; 2). TFA-DCM, rt, 3 h; or TMSI, 0 °C, 1 h.
2
6
The syntheses of compounds 2-4 are shown in Scheme 2. Amide condensation of the acid
(47) with the amine (48) gave compound 49 removal of whose Boc group yielded compound 50.
Condensation of 50 with corresponding protected phosphotyrosine mimetics followed by
hydrolysis of the dialkyl phosphate group afforded compounds 2-4.
2
6
30
Scheme 3 shows the syntheses of compounds 5 and 7. Using the acid (51) as starting
material, amide condensation reaction and subsequent selective removal of Boc group yielded an
amine (53) which, coupled with 46, gave 54. The common intermediate (56) was prepared by
deprotection of Cbz group in 51 and introduction of a methyl group followed by hydrolysis of the
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 71
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
tert-butyl ester. The corresponding amines were coupled to 56 and the resulting diethyl phosphates
were treated with TMSI to yield compounds 5 and 7.
Scheme 3. Synthesis of compounds 5 and 7.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Reagents and conditions: a. 1). HATU, EtN(i-Pr) , DMF, rt, 1 h; 2). TFA, DCM; b. 46, HATU, EtN(i-Pr) , DMF, rt,
2
2
1 h; c. H
2
, Pd/C, MeOH; d. 1). HCHO, Sodium triacetoxyborohydride, DCE; 2). TFA, DCM; e. R-NH
2
, HATU,
EtN(i-Pr) , DMF, rt, 1 h.
2
Scheme 4. Synthesis of compounds 8-11.
Reagents and conditions: a. HATU, EtN(i-Pr) , DMF, rt, 1 h; b. TFA, DCM; c. 51, HATU, EtN(i-Pr) , DMF, rt, 1 h;
2
2
2 2
d. 1) H , Pd/C, MeOH; 2) Ac O, EtN(i-Pr) , DCM; 3) TFA, DCM; e. 1) phosphotyrosine mimetics, HATU, EtN(i-
2
Pr) , DMF, rt, 1 h; 2). TFA-DCM, rt, 3 h; or TMSI, 0 °C, 1 h.
2
28
ACS Paragon Plus Environment

Page 29 of 71
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The syntheses of compounds 8-11 are shown in Scheme 4. The intermediate (60) was
obtained by an amide condensation reaction followed by removal of the Boc protecting group
under acidic conditions. Condensation of 60 with 51 led to 61, which following removal of the
Cbz and subsequent acetylation and Boc deprotection gave the common intermediate (62).
Amide condensation of 62 with corresponding protected phosphotyrosine mimetics and
hydrolysis of the diethyl phosphate group afforded compounds 8-11.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 5. Synthesis of SD-36.
Reagents and conditions: a. 1). TFA, DCM; 2) 46, HATU, DIEA, DMF; b. H
2
, Pd/C, MeOH; c. 68, HATU, DIEA,
DMF; d. TMSI, BSTFA, DCM . e. Pd(PPh )Cl , CuI, DMF/TEA, 80 °C , 1 h.
3
2
29
ACS Paragon Plus Environment