Design, synthesis, and evaluation of peptidomimetics containing Freidinger lactams as STAT3 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.01.091

The STAT3 oncogene is a promising molecular target for the design of a new class of anticancer drugs. In this letter, we describe the design, synthesis, and evaluation of peptidomimetics containing Freidinger lactams as novel STAT3 inhibitors. Compound 3

Full text of DOI:10.1016/j.bmcl.2009.01.091

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1733–1736
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis, and evaluation of peptidomimetics containing
Freidinger lactams as STAT3 inhibitors
Cindy Gomez a,b, Longchuan Bai a,c, Jian Zhang a,c, Zaneta Nikolovska-Coleska a,d
,
Jianyong Chen a,c, Han Yi a,c, Shaomeng Wang a,b,c,e,
*
a Cancer Center, University of Michigan, Ann Arbor, MI 48109, USA
b Department of Medicinal Chemistry, University of Michigan, Ann Arbor, MI 48109, USA
c Department of Internal Medicine, University of Michigan, Ann Arbor, MI 48109, USA
d Department of Pathology, University of Michigan, Ann Arbor, MI 48109, USA
e Department of Pharmacology, University of Michigan, Ann Arbor, MI 48109, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The STAT3 oncogene is a promising molecular target for the design of a new class of anticancer drugs. In
this letter, we describe the design, synthesis, and evaluation of peptidomimetics containing Freidinger
lactams as novel STAT3 inhibitors. Compound 3 binds to STAT3 with a Ki value of 190 nM and is a prom-
ising lead compound for further design and optimization.
Received 16 December 2008
Accepted 26 January 2009
Available online 31 January 2009
Ó 2009 Published by Elsevier Ltd.
Keywords:
STAT3
Small-molecule inhibitors
Peptidomimetics
Freidinger lactams
Constitutive activation of the Signal Transducers and Activators
of Transcription 3 (STAT3) is frequently detected in specimens from
cancer patients with advanced disease and human cancer cell lines,
but not in normal epithelial cells.1–4 Persistent activation of STAT3
signaling has been demonstrated to contribute directly to oncogen-
esis by stimulating cell proliferation and preventing apoptosis in
human cancer cells.1–4 STAT3 activation not only provides a growth
advantage to tumor cells, allowing their accumulation, but may also
confer resistance to conventional therapies that rely on apoptotic
machinery to eliminate tumor cells.1–4 In cells, STAT3 is recruited
from the cytosol by transmembrane cytokine and growth factor
receptors and participates in specific interactions through its Src
homology 2 (SH2) domain with the phosphotyrosine docking sites
displayed by the receptors. It then is phosphorylated on tyrosine,
Tyr705.1–4 Tyrosine phosphorylation of STAT3 causes it to dimerize
and translocate to the nucleus, where it binds to specific promoter
sequences on its target genes.1–4 Dimerization of STAT3 is a decisive
event for its activation and its transcriptional activity.1–5 Blocking
STAT3 dimerization using a small-molecule inhibitor has been pro-
posed as a potentially promising therapeutic approach to the devel-
opment of molecularly targeted therapies for the treatment of
human cancers with constitutively activated STAT3.4–10
Two approaches are currently being pursued in the design of
small-molecule antagonists which inhibit STAT3 dimerization.4–13
The first is the design of peptidomimetics6–10 and the second is a
search for non-peptidic small-molecules.11–13 While a number of
peptide-based ligands can achieve quite high binding affinities to
STAT3,6–10 they generally have no or very poor cellular activity
due to their peptidic nature and/or the negatively charged phos-
photyrosine amino acid residue. The major advantage of non-pep-
tide small-molecule inhibitors is that they are cell permeable, but
the inhibitors reported to date typically have weak binding affini-
ties to STAT3 and may not be very specific against STAT3.11–13
Therefore, there is a great need for potent and cell-permeable
STAT3 inhibitors. In this letter, we present our structure-based de-
sign, synthesis, and biochemical evaluation of peptidomimetic
inhibitors of STAT3. Incorporation of Freidinger lactams into our
designed compounds has reduced their peptidic character and
afforded potent STAT3 inhibitors.
We previously used the phosphorylated peptide (pYLKTKF)
derived from STAT3 as the basis of our design of a macrocyclic
peptidomimetic inhibitor of STAT3.8 One significant drawback
of this phosphopeptide is that it has a relatively weak affinity
to the STAT3 protein (Ki = 25.9 l
M)8 and consequently, in this
study, we employed a gp130 protein derived phosphopeptide
pYLPQTV, which binds to STAT3 with a much higher affinity than
the STAT3 phosphorylated peptide.7,8,14
* Corresponding author. Tel.: +1 734 615 0362; fax: +1 734 647 9647.
E-mail address: shaomeng@umich.edu (S. Wang).
0960-894X/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.bmcl.2009.01.091

1734
C. Gomez et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1733–1736
Figure 1. Design of STAT3 peptidomimetics by incorporating Freidinger lactams.
To quantitatively determine the binding affinity of our designed
gp130 pYLPQTV peptide, it still has an appreciable binding affinity
to STAT3 and is a good lead structure for further optimization.
The crystal structure of mouse STAT3 homodimer bound to DNA
has been determined (PDB entry: 1BG1),5 but no crystal structure
has been reported for any gp130 based phosphopeptide in a com-
plex with STAT3. To assist our design efforts, computational dock-
ing was performed to predict the binding mode of compound 1 in a
complex with STAT3. The mouse and human STAT3 protein se-
quences differ by just one residue: Glu760 in human STAT3 is re-
placed by Asp760 in mouse STAT3. This residue is at the N-
terminus of STAT3, distant from the phosphopeptide binding site
within STAT3’s SH2 domain. We have therefore used the 1BG1
crystal coordinates in our modeling studies as an acceptable
approximation.
compounds to STAT3, we developed a fluorescence-polarization
(FP) based binding assay which is similar to our previously re-
ported FP assay.14 The gp130 pYLPQTV peptide binds to STAT3
with a high affinity,7,8,14 and accordingly we incorporated this se-
quence into our fluorescently labeled probe. We synthesized pep-
tide, b(Ala)-b(Ala)-G-pYLPQTV, and labeled it at the N-terminus
with a fluorescent tag, 5-carboxyfluorescein (5-FAM). A saturation
experiment determined that this fluorescently tagged gp130 pep-
tide binds to recombinant human STAT3 protein (residues 127–
722) with Kd = 35 2.3 nM.
It has been shown that the threonine and valine residues in the
gp130 phosphopeptide can be replaced by a benzyl amide without
a significant loss in its STAT3 binding affinity.6 Accordingly, we first
designed and synthesized compound 1, which represents the
shortened gp130 sequence (Fig. 1). Our competitive FP binding as-
say shows this compound binds to STAT3 with Ki = 350 nM. Thus,
although compound 1 binds to STAT3 less potently than the
Computational docking was performed using the GOLD pro-
gram (version 3.1).16,17 The binding site for STAT3 was centered
at Glu638 and given a radius of 13 Å, sufficient to cover the entire
binding pocket. The docking simulations were terminated after 10
Figure 2. Predicted binding models of gp130 phospho-peptide, compounds 1, 2, and 3 in complex with STAT3. STAT3 protein is shown in surface model, with key binding
residues displayed in stick model and colored in green for carbon atoms, blue for nitrogen atoms, and red for oxygen atoms, respectively. STAT3 ligands were shown in stick
model and colored in yellow for carbon atoms, blue for nitrogen atoms, red for oxygen atoms, and orange for phosphorous atoms, respectively. Figures were generated using
pymol (http://www.pymol.org/).

C. Gomez et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1733–1736
1735
Scheme 1. Synthesis of compounds 2 and 3. Reagents and conditions: (i) EDCI, HOBt, DIPEA; (ii) DEA/CH3CN; (iii) Cbz-phosphotyrosine, di-tert-butyl ester, EDCI, HOBt,
DIPEA; (iv) 1:1 TFA/DCM, TES.
runs for each compound. GoldScore, implemented in Gold 3.1 was
used as the fitness function with which to evaluate the docked con-
formations. The conformation ranked highest by the fitness func-
tion in each run was saved as potential docking modes. The
highest ranked conformation of each compound among the 10 runs
was reported as the predicted binding model (Fig. 2).
Compound (Ki SD, µM)
1 (0.35 0.15)
2 (3.13 0.42)
3 (0.19 0.09)
100
80
60
40
20
0
The predicted binding model for compound 1 (Fig. 2) showed
that phosphotyrosine has significant charge–charge interactions
with Lys591, Arg609, Ser611, and Ser613, highlighting the critical
importance of the phosphate for binding to STAT3. While the
hydrophobic side chain of the Leu residue has no specific contacts
with STAT3, its backbone amide forms a hydrogen bond with the
backbone carbonyl of Ser636 in STAT3. Although the Pro residue
in compound 1 does not have specific interactions with the protein,
it plays a key role in controlling the conformation of the peptide.
The amide side chain of the peptide’s Gln residue forms a hydrogen
bond with the backbone carbonyl of Glu 638 within STAT3. This
may be significant because the Gln residue has been shown to play
a critical role in the specific recognition of gp130 based phospho-
peptides by STAT3 over other STAT proteins and may also be
important for high-affinity binding to STAT3.10
Analysis of the predicted binding model for compound 1 in
complex with STAT3 also reveals that directed by the proline resi-
due, it assumes a b-turn conformation. We sought to redesign com-
pound 1 by incorporating a Freidinger lactam into the structure.
The 6- and 7-membered Freidinger lactam structures assume a II0
b-turn conformation and have been extensively used in the design
of conformationally constrained peptidomimetics.15 This led to the
design of compounds 2 and 3, which in addition to being conform-
ationally constrained also have reduced peptide character. Overlay
of our predicted binding models of compounds 1, 2, and 3 showed
that while compound 3 nicely mimics compound 1 in its binding to
STAT3, compound 2 shows significant deviations.
The synthesis of compounds 2 and 3 is described in Scheme 1.
Compounds 2 and 3 were synthesized in solution from the respec-
tive b-turn mimetic with standard (EDCI/HOBt/DIPEA) peptide
chemistry. A di-orthogonal protection strategy was employed,
where the N-termini were protected with base-labile Fmoc and
side-chains were protected with acid-labile trityl and tert-butyl es-
ters. The final products were obtained using known peptide chem-
istry (EDCI/HOBt/DIPEA; 1:1 DCM/TFA) for final deprotection.
The binding affinities of compounds 2 and 3 to STAT3 were
determined using our FP-based, quantitative and competitive
binding assay and the results are shown in Figure 3. Compound 2
containing a 6-membered Freidinger lactam binds to STAT3 with
-3
-2
-1
0
1
2
[Compounds] (log µM)
Figure 3. Binding affinities of designed peptidomimetics to STAT3, as determined
using a fluorescence-polarization-based, competitive binding assay.
Ki = 3.13 lM, nine times less than compound 1. By comparison,
compound 3, with a 7-membered Freidinger lactam binds to STAT3
with Ki = 190 nM, and is twice as potent as compound 1 and 18
times as potent as compound 2. Thus, the binding data for com-
pound 3 showed that incorporation of a 7-membered Freidinger
lactam structure into compound 1 results in a novel peptidomi-
metic compound with a high binding affinity to STAT3. The data
also suggest that the hydrogen bond between the Leu main chain
amino group in compound 1 and the carbonyl group of Ser 636
in STAT3 is not required for high-affinity binding since compound
3 no longer has the corresponding carbonyl group in its structure
but still binds to STAT3 with a high affinity.
In summary, using the STAT3 recruitment sequence from the
gp130 protein as our basis, we have carried out structure-based de-
sign of conformationally constrained peptidomimetic inhibitors of
STAT3, which incorporate Freidinger lactams. While compound 2,
containing a 6-membered lactam structure, is much less potent
than compound 1, compound 3 containing a 7-membered lactam
structure binds to STAT3 with Ki = 190 nM. Compound 3 has re-
duced peptide character compared to compound 1 and the gp130
phosphopeptide, and is a promising lead compound for further de-
sign and optimization toward developing potent and cell-perme-
able STAT3 inhibitors as a new class of anticancer drugs.
References and notes
1. Darnell, J. E., Jr. Nat. Rev. Cancer 2002, 2, 740.
2. Yu, H.; Jove, R. Nat. Rev. Cancer 2004, 4, 97.

1736
C. Gomez et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1733–1736
11. Song, H.; Wang, R.; Wang, S.; Lin, J. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 4700.
3. Turkson, J.; Jove, R. Oncogene 2000, 19, 6613.
4. Siddiquee, K.; Turkson, J. J. Cell Res. 2008, 18, 254.
12. Schust, J.; Sperl, B.; Hollis, A.; Mayer, T. U.; Berg, T. Chem. Biol. 2006, 13, 1235.
13. Siddiquee, K.; Zhang, S.; Guida, W. C.; Blaskovich, M. A.; Greedy, B.; Lawrence,
H. R.; Yip, M. L.; Jove, R.; McLaughlin, M. M.; Lawrence, N. J.; Sebti, S. M.;
Turkson, J. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 7391.
14. Schust, J.; Berg, T. Anal. Biochem. 2004, 330, 114.
15. For a review on Freidinger lactams, please see: Freidinger, R. M. J. Med. Chem.
2003, 46, 5553.
16. Jones, G.; Willett, P.; Glen, R. C.; Leach, A. R.; Taylor, R. J. Mol. Biol. 1997, 267,
727.
17. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
2003, 52, 609.
5. Becker, S.; Groner, B.; Muller, C. W. Nature 1998, 394, 145.
6. Coleman, D. R., IV; Ren, Z.; Mandal, P. K.; Cameron, A. G.; Dyer, G. A.; Muranjan,
S.; Campbell, M.; Chen, X.; McMurray, J. S. J. Med. Chem. 2005, 48, 6661.
7. Ren, Z.; Cabell, L. A.; Schaefer, T. S.; McMurray, J. S. Bioorg. Med. Chem. Lett.
2003, 13, 633.
8. Chen, J.; Nikolovska-Coleska, Z.; Yang, C.-Y.; Gomez, C.; Gao, W.; Krajewski, K.;
Jiang, S.; Roller, P.; Wang, S. Bioorg. Med. Chem. Lett. 2007, 17, 3939.
9. Siddiquee, K. A.; Gunning, P. T.; Glenn, M.; Katt, W. P.; Zhang, S.; Schroeck, C.;
Sebti, S. M.; Jove, R.; Hamilton, A. D.; Turkson, J. ACS Chem. Biol. 2007, 2, 787.
10. McMurray, J. C. Biopolymers (Peptide Science) 2007, 90, 69.