Investigation of the binding determinants of phosphopeptides targeted to the Src homology 2 domain of the signal transducer and activator of transcription 3. Development of a high-affinity peptide inhibitor

Article abstract of DOI:10.1021/jm050513m

Signal transducer and activator of transcription 3 (Stat3) is a cytosolic transcription factor that relates signals from the cell membrane directly to the nucleus where it, in complex with other proteins, initiates the transcription of antiapoptotic and c

Full text of DOI:10.1021/jm050513m

J. Med. Chem. 2005, 48, 6661-6670
6661
Investigation of the Binding Determinants of Phosphopeptides Targeted to the
Src Homology 2 Domain of the Signal Transducer and Activator of
Transcription 3. Development of a High-Affinity Peptide Inhibitor
David R. Coleman, IV,†,‡ Zhiyong Ren,‡,§ Pijus K. Mandal,† Arlin G. Cameron,† Garrett A. Dyer,†
Seema Muranjan,† Martin Campbell,# Xiaomin Chen,§ and John S. McMurray*,†
Department of Neuro-Oncology, Department of Biochemistry and Molecular Biology, Department of Molecular Pathology, and
The Graduate School of Biomedical Sciences, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe
Boulevard, Houston, Texas 77030
Received June 1, 2005
Signal transducer and activator of transcription 3 (Stat3) is a cytosolic transcription factor
that relates signals from the cell membrane directly to the nucleus where it, in complex with
other proteins, initiates the transcription of antiapoptotic and cell cycling genes, e.g., Bcl-xL
and cyclin D1. In normal cells Stat3 transduces signals from cytokines such as IL-6 and growth
factors such as the epidermal growth factor. Stat3 is constitutively activated in a number of
human tumors. Antisense and dominant negative gene delivery result in apoptosis and reduced
cell growth, thus this protein is an attractive target for anticancer drug design. As part of our
research on the design of Src homology 2 (SH2) directed peptidomimetic inhibitors of Stat3, in
this paper we describe structure-activity relationship studies that provide information on the
nature of peptide-protein interactions of a high-affinity phosphopeptide inhibitor of Stat3
dimerization and DNA binding, Ac-Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2, peptide 1. There is
a hydrophobic surface on the SH2 domain that can accommodate lipophilic groups on the
N-terminus. Of the amino acids tested, leucine provided the highest affinity at pY+1 and its
main chain NH is involved with a hydrogen bond with Stat3, presumably Ser636. cis-3,4-
Methanoproline is optimal as a backbone constraint at pY+2. The side chain amide protons of
Gln are required for high-affinity interactions. The C-terminal dipeptide, Thr-Val, can be
replaced with groups ranging in size from methyl to benzyl. We synthesized a phosphopeptide
incorporating groups that provided increases in affinity at each position. Thus, hydrocinnamoyl-
Tyr(PO3H2)-Leu-cis-3,4-methanoPro-Gln-NHBn, 50, was the highest affinity peptide, exhibiting
an IC50 of 125 nM versus 290 nM for peptide 1 in a fluorescence polarization assay.
Signal transducer and activator of transcription 3
(Stat3) is a member of the signal transducer and
activator of transcription (STAT) family of transcription
factors that relate signals from extracellular signaling
protein receptors on the plasma membrane directly to
the nucleus (reviewed in refs 1a-c). Stat3 was discov-
ered as a major component in the acute phase response
to inflammation2 and a key mediator of interleukin 6
(IL-6) and epidermal growth factor signaling.3 Like all
STATs, Stat3 is composed of an amino-terminal oligo-
merization domain, a coiled coil domain, a DNA binding
domain, a linker domain, an Src homology 2 (SH2)
domain, and a C-terminal transactivation domain. On
binding of IL-6 to its receptor, JAK kinases-2 (JAK-2)
phosphorylates the coreceptor gp130 on several ty-
rosines.4,5 Stat3, via its SH2 domain, binds to these
phosphotyrosine residues. It is then phosphorylated on
Tyr705, a conserved tyrosine just C-terminal to the SH2
domain, by JAK-2. Upon phosphorylation, termed ac-
tivation, Stat3 forms homodimers (and/or heterodimers
with Stat1) via reciprocol interactions between the SH2
domains and pTyr705. The dimers then translocate to
the nucleus and bind specific DNA sequences where
they, in cooperation with other transcription factors,
regulate gene expression.1 Downstream targets of Stat3
include Bcl-xL, a member of the Bcl-2 family of anti-
apoptotic proteins, cell cycle regulators such as cyclin
D1 and p21WAF1/CIP1, and other transcription factors
including c-myc and c-fos. In epidermal growth factor
(EGF) signaling, Stat3 has been reported to bind directly
to phosphotyrosine residues on the epidermal growth
factor receptor (EGFR) and to be activated by the kinase
activity of the receptor.6 Further studies imply that Src
kinases first bind to EGFR via their SH2 domains and
recruit Stat3 via SH3 domain interactions with poly-
proline helices.7
Stat3 has been shown to be constitutively activated
in cancers of the head and neck,8 breast,9 brain,10
prostate,11 lung,12 leukemia,13 multiple myeloma,14 lym-
phoma,15 pancreas,16 and others.17 Inhibition of Stat3
activity by the introduction of antisense oligonucleotides
or dominant negative constructs has been shown to
induce apoptosis and reduce cell growth and soft agar
colony formation in several tumor cell lines exhibiting
constitutively active Stat3.14,18 In those cells driven by
EGF signaling, introduction of small-molecule EGFR
inhibitors reduced Stat3 acivation and resulted in cell
* To whom correspondence should be addressed. Phone: 713-745-
3763. Fax: 713-834-6230. E-mail: jsmcmur@mdanderson.org.
† Department of Neuro-Oncology.
‡ The Graduate School of Biomedical Sciences.
§ Department of Biochemistry and Molecular Biology.
# Department of Molecular Pathology.
10.1021/jm050513m CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/16/2005

6662 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21
Coleman et al.
Chart 1. Structure of Peptide 1,
Ac-Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2
death.19 A small-molecule inhibitor of the Src kinase
also inhibits Stat3 activation, induces apoptosis, and
inhibits cell growth in a breast cancer cell line.20 Thus
Stat3 is an attractive target for cancer drug design.17
To develop inhibitors capable of inhibiting Stat3
activity, we are targeting the SH2 domain with a
phosphopeptide-based peptidomimetic approach.21 Such
compounds are expected to block the binding of Stat3
to cytokine or growth factor receptors and thereby
prevent phosphorylation, dimerization, and nuclear
translocation. For dimerized Stat3, SH2-directed inhibi-
tors would also inhibit dimerization and DNA binding,
thereby reducing the expression of antiapoptosis and cell
cycling genes. SH2 domains are specific for phospho-
tyrosine-containing amino acid sequences with specific-
ity for individual proteins residing C-terminal to this
residue.22,23 Phosphopeptides based on the sequence
surrounding phosphotyrosine-705 of Stat3,24-26
EGFR,26,27 and gp13026 were shown to inhibit dimer-
ization and DNA binding in electrophoretic mobility
shift assays (EMSA). To find a lead peptide, we screened
a series of phosphopeptides derived from the amino acid
sequences of receptor docking sites of Stat3 for their
ability to inhibit dimer formation and DNA binding
using EMSA.26 Sequences from Stat3, gp130, EGFR, IL-
10R, and GM-CSF were tested. Of this group, Ac-
Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2 (1, Chart 1) was
found to be the most potent inhibitor (IC50 ) 150 nM)
and was selected as the lead compound. Peptide 1 is
composed of residues 904-909 of gp130, the cocompo-
nent of the IL-6 receptor, and possesses the consensus
sequence for Stat3 binding, Tyr(PO3H2)-Xxx-Xxx-Gln
reported by Stahl at al.4 and Gerhartz et al.5 Truncation
and alanine scanning experiments showed that the
C-terminal valine was not necessary for binding, leucine
at pY+1 contributed to affinity, proline was necessary
Figure 1. Structure of the phosphopeptide, residues 702-
711 (green), of Stat3 bound to the SH2 domain of the opposing
protein in the Stat3â dimer/DNA complex.29 Phospho-
tyrosine705 interacts with the side chains of Lys591, Arg609,
Ser611, and Ser613 and the main chain NH of Glu612 (yellow).
The hydrogen bond between the RNH of Leu706 and the
backbone CdO of Ser636 on the opposing protein is depicted
as a dotted line. Coordinates were kindly provided by C.
Mueller.29 They can be obtained from the Protein Data Bank,
entry 1BG1. Figure was generated using PyMol.48
of Src,28,30a,b Grb2,30c Lck,30d p85 PI3K,30e,f Stat1,31 and
others. A hydrogen bond exists between the RNH of
Leu706 and the CdO of Ser636 (âD2) on the opposing
protein. Similar interactions are noted between the
pY+1 residue of phosphopeptide ligands and the âD4
residue of other SH2 domains. (The architecture of Stat3
is somewhat different from the other signaling proteins,
hence the variation in the designation of the âD2 and
âD4 residues.) Thus, the pY and pY+1 residues in the
Stat3 dimer bind similarly to analogous residues in
other SH2 domains. However, the X-ray structure does
not provide a direct model of the interaction of the Pro-
Gln-Thr-Val residues of peptide 1 and Stat3. Therefore,
SAR studies were undertaken to further understand the
contributions to affinity of backbone and side chain
groups of our lead peptide.32
Chemistry
Phosphopeptides were synthesized using either manual
or automated solid-phase methods using the Fmoc
protection scheme and carbodiimide/HOBt or PyBOP/
HOBt/DIPEA coupling techniques. Unless otherwise
noted, phosphotyrosine was introduced as the unpro-
tected phosphate33 using PyBOP/HOBt/DIPEA coupling
conditions. The amino termini were capped with excess
acetic anhydride and Et3N. Peptides were cleaved with
TFA/TIS/H2O, 95:2.5:2.5,34 and purified by reverse-
phase HPLC before assay.
Methylation of the Side Chain Amide of Gln. The
synthesis of Ac-Tyr(PO3H2)-Leu-Pro-Glu(NR1R2)-Thr-
NH2 (27, 28) is depicted in Scheme 1. Ac-Tyr(PO-
[NMe2]2)-Leu-Pro-Glu(OAll)-Thr(tBu) (51) was synthe-
sized on Rink amide resin using automated protocols.
The allyl group on the side chain of glutamic acid was
removed with tetrakistriphenylphosphinylpalladium in
for high-affinity activity, and glutamine was essential
28
for binding.26 Similar to other SH2 domains,
Stat3
ligands bind in a (pY)-(pY+3) “two-pronged” motif.
SH2 domain inhibitor development efforts against
Src, Grb2, and others have relied on X-ray crystal-
lographic or NMR-derived structures to provide under-
standing of peptide-protein interactions and to guide
the conversion of peptide into non-peptide. The crystal
structure of the dimer of Stat3â bound to DNA29 reveals
the interactions between the phosphopeptide segment
Tyr(PO3H2)705-Leu-Lys-Thr-Lys-Phe of one protein mol-
ecule and the SH2 domain of the other (Figure 1). The
phosphate is in contact with the side chains of Lys591
(RA2), Arg609 (âB5), Ser611 (âB7), and Ser613 (BC2)
and the main chain NH of Glu612 (BC1) in a manner
nearly identical with that of phosphopeptide complexes

Peptide Inhibitors of Stat3
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21 6663
Scheme 1. Synthesis of Methylated Glutamine-Containing Peptidesa
a Reagents and conditions: (i) [Ph3P]4Pd, NMM, AcOH; (ii) amine, PyBOP, DIPEA, HOBt; (iii) (a) TFA/TIS/H2O, 1 h, (b) 10% H2O/
TFA, overnight.
Scheme 2. Synthesis of Ac-Tyr(PO3H2)-Leu-Pro-Gln-Xxxa
a Reagents and conditions: (i) [Ph3P]4Pd, NMM, AcOH; (ii) amine, PyBOP, DIPEA, HOBt; (iii) (a) TFA/TIS/H2O, 1 h, (b) 10% H2O/
TFA, overnight.
the presence of acetic acid and N-methylmorpholine.35
Derivatization of the carboxyl group was carried out by
the addition of 5 equiv of NH2Me or NHMe2 with
PyBOP/HOBt/DIPEA (53, 54). The use of phospho-
tyrosine with unprotected phosphate33 or protected with
diphenylmethylsilylethyl groups36 resulted in poor yields
and very impure products. Replacement of the phos-
phate group on the tyrosine with bis(dimethylamino)-
phosphoramidate37 resulted in better quality crude
products. Conversion of the phosphoramidate to the
phosphate requires aqueous acidic conditions.37 So after
cleavage from the resin, 10% H2O was added and
acidolysis was allowed to proceed overnight. It was also
found that the Dmab protection on the side chain of Glu,
with removal with 2% hydrazine in DMF,38 resulted in
unacceptable levels of side products that were not there
with the allyl/Pd[0] protecting group scheme.
C-Terminal Derivization. To synthesize tetra-
peptides in which the C-terminal Thr was replaced
(39-47), Ac-Tyr(PO[NMe2]2)-Leu-Pro-Glu-OAll (55) was
assembled using automated methods starting with
attachment of the side chain of Fmoc-Glu-OAll to Rink
resin (Scheme 2). The allyl group was cleaved using
(PPh3)4Pd/AcOH/NMM35 to give the free carboxyl group
of the C-terminal Gln (56). The amines were added in
excess, and coupling was mediated with DIPCDI/HOBt
to minimize racemization (57). TFA cleavage gave the
desired peptides, 39-47, which were purified by reverse-
phase HPLC. As above, use of nonprotected phospho-
tyrosine resulted in crude peptides of low purity.

6664 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21
Coleman et al.
Table 1. IC50 Values of Peptide Inhibitors of Stat3
IC50 (nM)
Coupling of pyrrolidine was incomplete, and the result-
ing tetrapeptide, Ac-Tyr(PO3H2)-Leu-Pro-Gln-OH (43),
was obtained and assayed.
peptide
EMSAa
FP
Peptides 31, 32, and 34-38, in which Gln was
replaced by Gaba or cyclic amino acids, were synthesized
by coupling commercially available Fmoc-protected
amino acids to Rink amide resin using manual solid-
phase peptide synthesis. For the synthesis of 38, Fmoc-
pyrollidine-3-carboxylic acid, 58, was prepared by stand-
ard coupling of Fmoc-OSu to commercially available
pyrollidine-3-carboxylic acid under aqueous alkaline
conditions. The final peptide, hydrocinnamoyl-Tyr-
(PO3H2)-Leu-3,4-cis-methanoPro-Gln-NHBn (isomers 49
and 50), was synthesized by manual solid-phase peptide
synthesis. Fmoc-Glu-NHBn, 59, was attached to Rink
amide resin via the side chain carboxyl group using
PyBOP/HOBt/DIPEA coupling methodology. The re-
mainder of the residues were coupled in the same
manner. The final product consisted of two diastereo-
mers that were readily resolvable by reverse-phase
HPLC using a gradient of MeOH in 0.01 M NH4OAc,
49 and 50. Fmoc-Glu-NHBn (59) was prepared by
coupling benzylamine to commercially available Fmoc-
Glu(OtBu)-OH using water-soluble carbodiimide fol-
lowed by removal of the side chain tert-butyl group with
TFA.
1 Ac-Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2
2 Ac-Tyr(PO3H2)-Leu-Pro-Gln-Thr-NH2
3 Ac-Tyr(PO3H2)-Leu-Pro-Gln-NH2
a IC50 values from ref 26.
150
600
290 ( 63
739 ( 31
856 ( 41
1000
Table 2. Amino Terminal Substitutions in Peptide 2
peptide
R
IC50 (µM)
2
4
5
6
7
Ac
0.739 ( 0.031
0.814 ( 0.078
0.497 ( 0.010
0.750 ( 0.072
0.276 ( 0.012
PhCO
CH3CH2CO
PhCH2OCO
PhCH2CH2CO
phoretic mobility shift assays.26 Peptide 1 and C-
terminal truncated peptides 2 and 3 were reassayed
using fluorescence polarization. The IC50 values ob-
tained by both methods followed the same trend of
decreasing affinity with decreasing length, although the
values varied somewhat (Table 1). Thus, the C-terminal
residues play a role in binding.
The peptides used in this study were analogues of
peptide 2. The C-terminal of Val in 1 was removed to
increase synthesis efficiency and to simplify interpreta-
tion of the results.
Amino-Terminal Substitutions. To test for poten-
tial binding affinity not afforded by the amino-terminal
acetyl group of our lead peptide, a small survey of acyl
groups was conducted at pY-1 (Table 2). Of the groups
tested, the propionyl and hydrocinnamoyl groups (5 and
7) exhibited increases in affinity. Interestingly, the
benzyloxycarbonyl group (Cbz) of 6 did not change the
apparent affinity. The Cbz and hydrocinnamoyl groups
are isosteric. The gain in affinity due to the hydro-
cinnamoyl group of 7 was offset by the R-oxygen atom
in the Cbz group. During their studies of phospho-
peptide inhibitors based on the Stat3 Tyr705 sequence,
Turkson et al.25 noted no significant difference in
affinity in peptides with a proline or an alanine that is
just amino-terminal to the phosphotyrosine (position
pY-1). However, significant changes were observed
when substituted benzoyl groups or 3-pyridinecarbonyl
groups were employed in a series of dipeptides of the
structure Xxx-Tyr(PO3H2)-Leu.41 Taken together with
our results, it appears that there is a hydrophobic patch
on the surface of Stat3 SH2 domain that is just amino-
terminal to the Tyr(PO3H2) binding pocket.
Results
Haan et al.39 assayed several phosphopeptides based
on gp130 and LIFR as well as the sequence surrounding
Stat3 Tyr705 for binding to the isolated SH2 domain
from Stat3. At pH 7.5 no complex formation was
observed, whereas binding occurred at pH 5.5. It was
noted that at the higher pH the SH2 domain existed in
a dimerized state, which may have precluded phospho-
peptide binding. Binding constants determined at pH
5.5 must be considered tenuous because the conforma-
tion of the protein may be altered by the acidic environ-
ment. Therefore, full length Stat3 was used to assay our
compounds. This protein was not phosphorylated on
Tyr705 prior to use.
For the SAR studies reported here, peptides were
assayed for competition of a fluorescently tagged version
of our lead peptide using fluorescence polarization
(FP).40 Thus, Ala-Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2
was labeled at the amino terminus with the isomeric
mixture fluorescein-5(6) carboxylic acid. Synthesis of
this probe produced two isomers, and the second com-
pound to elute by reverse-phase HPLC was found to
bind to Stat3 with higher affinity (data not shown). The
individual 5- and 6-carboxyfluorescein isomers were
purchased and coupled to the N-terminus of Ala-
Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2. The peptide
capped with 5-carboxyfluorescein coeluted with the
high-affinity isomer on reverse-phase HPLC. Thus,
the probe used for FP was 5-carboxyfluoresceinyl-
Ala-Tyr(PO3H2)-Leu-Pro-Gln-Thr-Val-NH2. Subsequent
batches of probe were synthesized with the 5-isomer.
In FP titration assays, the bacterially expressed protein
was identical to that expressed in the baculovirus-Sf9
system employed for our earlier study24,26 when assayed
by FP (data not shown).
Substitutions at pY+1. In the crystal structure of
dimerized Stat3â dimer bound to DNA, the side chain
of Leu706 at the pY+1 position of Tyr(PO3H2)705-Leu-
Lys-Thr-Lys-Phe resides on a nonpolar surface of the
opposing SH2 domain. A survey of hydrophobic amino
acids was conducted to find which would provide
optimum binding (Table 3, peptides 8-13). The â-sub-
stituted residues, Val and Ile, decreased activity. In-
The initial screen phosphopeptides based on known
receptor docking sites for Stat3 was done using electro-

Peptide Inhibitors of Stat3
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21 6665
Table 3. Inhibition of Stat3 with Analogues of 2 Substituted
at Position pY+1: Ac-Tyr(PO3H2)-Xxx-Pro-Gln-Thr-NH2
the correct molecular mass (18, 19). The first eluting
isomer (18) enhanced affinity of the peptide 2-fold (IC50
) 0.265 µM), and the second (19) reduced affinity (IC50
) 3.11 µM). Fmoc-cis-3,4-methanoproline is sold as a
racemic mixture of (2S,3R,4S)- and (2R,3S,4R)-enanti-
omers. Thus, 18 and 19 are diastereomers.
peptide
pY+1
IC50 (µM)
2
8
Leu
0.739 ( 0.031
1.54 ( 0.079
1.07 ( 0.059
1.89 ( 0.20
0.91 ( 0.12
0.717 ( 0.080
>100
Val
9
Ile
10
11
12
13
14
Phe
The homologous cyclic amino acids, four-membered
azetidine-2-carboxylate and six-membered pipecolic acid,
resulted in lower affinity peptides (20 and 21, respec-
tively). The structural isomers of pipecolic acid nipecotic
acid (Nip) and isonipecotic acid (Inp) were also incor-
porated. Moving the position of the carbonyl group from
the R- to the â- (Nip) or the γ-carbon (Inp) alters the
spatial the relationship of the glutamine to the rest of
the peptide, most importantly tyrosine. Nip was ob-
tained commercially as a racemic mixture of the R and
S isomers. During purification by preparative HPLC,
we were able to isolate the two diastereomers of Ac-
Tyr(PO3H2)-Leu-(R,S)-Nip-Gln-Thr-NH2 (peptides 22
and 23), though we do not know which contains the R
or S isomer. Both exhibited a 50-fold reduction in
affinity, as did the peptide containing Inp (24). 4-Car-
boxymethylpiperazine (Cmpi) was also substituted for
proline, and the resulting peptide, 25, was very weak
(IC50 ) 57.1 µM). The unnatural amino acid analogue,
R-aminocyclohexane carboxylic acid (1-Ac6c), was re-
ported to be a substitute for proline and to induce a
â-turn or a 310 helix.42 It was incorporated into Grb2
SH2 domain antagonists to induce a â-turn at pY+1.43
In peptide 26, 1-Ac6c resulted in a 20-fold reduction in
affinity, suggesting that the peptide may not bind to
Stat3 in a turn structure. Thus, cis-methanoproline is
the scaffold that provides the optimum disposition of
contact groups for interaction with Stat3.
Cha
Nle
Ph2Ala
NMe-Leu
>100
terestingly, Phe was 2-fold less active than the fully
reduced analogue cyclohexylalanine. Nle resulted in
activity closest to that of Leu. Diphenylalanine resulted
in an inactive compound.
As mentioned, the Leu706 NH at pY+1 participates
in a hydrogen-bond interaction with the backbone
carbonyl of Ser636 of the opposing protein (Figure 1).
Replacement of the Leu amide proton with a methyl
group would preclude formation of this hydrogen bond
and would also sterically prevent tight association of
the peptide with the protein surface. Incorporation of
N-methylleucine at pY+1, peptide 14, destroyed activity
of the peptide (IC50 > 100 µM). Thus, it seems likely
that the Leu residue of our peptide binds similarly to
the pY+1 residues found in the Stat3 crystal structure29
and other SH2 domain ligands.28,30,31
Substitutions at pY+2. Our intitial screen of recep-
tor docking sites and the alanine scan established that
proline at position pY+2 is very important for affinity.26
This amino acid experiences no rotation about the NR-
CR bond, which may be important in constraining the
peptide to a population of conformations resembling the
bound conformation. To further probe structural effects
at pY+2, cyclic amino acids were substituted for proline
(Table 4).
Gln Side Chain Amide Protons Are Required for
High-Affinity Interaction with Stat3. Glutamine at
position pY+3 was found early to be a binding deter-
minant for the Stat3 SH2 domain,4,5 and most known
receptor docking sites for Stat3 possess the consensus
sequence Tyr-Xxx-Xxx-Gln.4,5,26,44 This was confirmed
in our screening of receptor docking sites26 and by
screening of a combinatorial phosphopeptide library45
Not surprisingly, the D-amino acid (15) severely
curtailed binding. R-Methyl substitution reduced activ-
ity 3- to 4-fold (16). 4-Hydroxyproline (17) resulted in a
slight increase in affinity. Incorporation of cis-3,4-
methanoproline resulted in two peptides that were
readily separable by reverse-phase HPLC and that gave
Table 4. Inhibition of Stat3 with Analogues of 2 Substituted at Position pY+2: Ac-Tyr(PO3H2)-Leu-Xxx-Gln-Thr-NH2

6666 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21
Coleman et al.
Table 5. Effect of Modifications of the Side Chain of Gln in
Ac-Tyr(PO3H2)-Leu-Pro-Xxx-Thr-NH2
Table 6. Effect of Gln Mimics in the Peptide
Ac-Tyr(PO3H2)-Leu-Pro-Xxx
peptide
pY+3
Gln
IC50 (µM)
2
27
28
29
30
0.739 ( 0.031
18.1 ( 2.60
80.0 ( 15.0
9.87 ( 1.36
13.9 ( 0.72
Gln(Me)
Gln(Me2)
Met(O)
Met(O2)
for Stat3 ligands. Previously we found that glutamic
acid at pY+3 abrogated activity and that asparagine
attenuated activity.26 The side chain amide protons
could serve as H-bond donors, either in intermolecular
interactions or in an intramolecular manner in confor-
mations resembling the bound state prior to binding.
To determine the role of these protons, one or both were
substituted with methyl groups. With the goal of
membrane permeability in mind, these substitutions,
if tolerated, would serve to make the peptide more
hydrophobic, thus enhancing biological activity. Sub-
stitution of one of the side chain amide protons reduced
binding by 25-fold (27, Table 5, IC50 ) 18 µM). Addition
of a second methyl group brought the IC50 to 80 µM
(peptide 28). The reduction in affinity may be due to
loss of important inter- or intramolecular hydrogen
bonding, steric crowding in a binding pocket accom-
modating the Gln side chain, or both. Thus, these amide
protons are required for affinity.
The role of the Gln carbonyl oxygen was probed by
substitution of methionine sulfoxide and methionine
sulfone, peptides 29 and 30, respectively. In both cases
inhibition was curtailed 20- to 30-fold (Table 5). Me-
thionine sulfoxide is isosteric with Gln; the sulfonyl
group mimics the side chain carbonyl, and the methyl
group replaces the NH2. The fact that Met(O) resulted
in a dramatic loss of inhibition underscores the contri-
bution of the carboxamide NH2 group on the side chain
of the pY+3 residue. However, the retention of modest
activity suggests that the carbonyl group may also be
involved in intermolecular interactions.
Substitutions at pY+3. Glutamine can be consid-
ered to be γ-carboxy-γ-aminobutyric acid (γ-carboxy-
Gaba). To test the effect of the main chain carboxamide,
glutamine was substituted with 4-aminobutyramide
(peptide 31) and was compared to peptide 3. As shown
in Table 6 this substitution reduced affinity 2-fold (IC50
) 1.8 µM). Thus, the main chain carboxamide of Gln
enhances activity by inter- or intramolecular interac-
tions. The Gaba homologue, ꢀ-aminohexanoic acid amide
(Ahx), further decreased inhibition (peptide 32).
Gln was substituted with nonpeptidic, cyclic amino
acid amides to constrain the rotation of the side chain
(Table 6). 3-Carboxamidoaniline is a highly constrained
mimic in which the carboxamido group is spaced from
the proline carbonyl by the same number of atoms as
Gln, but all atoms are coplanar. Peptide 33 was 45-fold
less active than peptide 3, suggesting that in the bound
state the Gln side chain is not coplanar. The saturated
analogue aminocyclohexane-3-carboxamide (3Ac6c-NH2)
resulted in approximately 4-fold higher affinity than the
aniline mimic, IC50 ) 9.81 µM for 34 versus IC50 ) 36
µM for 33. However, 3Ac6c-NH2 resulted in a 10-fold
reduction in affinity relative to peptide 3. Interestingly,
nipecotamide, piperazine-1-acetamide, and pyrrolidine-
1-acetamide (peptides 35, 37, and 38, respectively),
Table 7. Affinities of Peptides with the Structure
Ac-Tyr(PO3H2)-Leu-Pro-Gln-R
peptide
R
IC50 (µM)
3
39
40
41
42
43
44
45
46
47
48
NH2
NHMe
NHBn
NHCH(Me)Ph (S)
NHCH(Me)Ph (R)
OH
NMe2
NEt2
pyrrolidine
piperidine
N-methyl-Thr-NH2
0.856 ( 0.041
0.882 ( 0.077
0.409 ( 0.015
0.596 ( 0.082
0.801 ( 0.142
0.748 ( 0.085
1.85 ( 0.675
4.22 ( 0.074
2.11 ( 0.401
7.00 ( 0.988
1.22 ( 0.051
possessing endocyclic “backbone” nitrogens, decreased
activity by only 2- to 4-fold compared to peptide 3. This
reduced activity may be due to the lack of the “main
chain” NH proton from the Gln surrogate or from a
slightly imperfect fit. Isonipecotamide reduced activity
14-fold (36).
Substitutions at pY+4. Tetrapeptides were pre-
pared in which the C-terminal Thr of 2 was replaced
with non-amino-acid groups (Table 7). This was done
to determine (1) the necessity of the pY+4 residue and
(2) if the C-terminus could be capped as a 3° amide, thus
reducing polarity and increasing membrane perme-
ability. The methyl group (39) had the same affinity as
the nonsubstituted amide (3). The benzyl group in
peptide 40 results in somewhat higher affinity than that
provided by threonine in peptide 2 (409 vs 739 nM). The
differences between the benzyl (40) and the (S)-(41) and
(R)-(42) methylbenzyl groups suggest that the aromatic
groups are interacting with the protein and that the
latter, more structured substituents somewhat impede
the interaction. Thus, it seems likely that there is a
hydrophobic surface on the Stat3 SH2 domain that
accommodates the pY+4 region of the peptide inhibitor.

Peptide Inhibitors of Stat3
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21 6667
benzyl and (S)-R-methylbenzyl groups suggest a hydro-
phobic binding pocket for the pY+4 residue. With the
exception of 3-carboxamidoaniline, constrained cyclic
Gln mimics showed modest losses in activity. Thus, it
may be possible to constrain the CONH2 group to a
conformation close to that of the bound peptide.
The purpose of these studies was to gain information
for the development of peptidomimetic inhibitors of
Stat3. Our results suggest that the design of such
compounds should include the aryl phosphate, the R-NH
of pY+1, an intact side chain CONH2 at pY+3, and N-
and C-terminal hydrophobic groups. That peptide 50,
containing these features, was higher in affinity than
our lead peptide 1 confirms our conclusions.
Figure 2. Structure and IC50 values of PhCH2CH2CO-
Tyr(OPO3H2)-Leu-3,4-cis-methanoPro-Gln-NHBn.
The low affinities of the disubstituted amide ana-
logues suggest that the C-terminal amide bond is a
hydrogen bond donor. That replacing Gln with Gaba,
which does not possess that carboxamide group, results
in lower affinity is in keeping with important inter- or
intramolecular interactions involving the Gln backbone
carboxamide. N-Methylthreonine resulted in a slight
loss in activit; IC50 is 1.2 µM for peptide 48 versus 0.74
µM for peptide 2.
Development of a High-Affinity Tetrapeptide.
A peptide was synthesized that incorporated the
amino acids that produced the highest affinity at each
position: hydrocinnamoyl on the N-terminus, cis-3,4-
methanoproline at pY+2, and benzyl at pY+4 (Figure
2). As in peptides 18 and 19, two diastereomers were
isolated by preparative HPLC: 49 and 50. In this case
the order of affinity was reversed, the first eluting
isomer, 49, had an IC50 of 18.8 ( 2.4 µM. However, the
second isomer, 50, was a highly potent inhibitor with
an IC50 of 0.125 ( 0.027 µM, 6-fold more active than
pentapeptide 2, the lead peptide for this paper, and 2.3-
fold more active than the original hexapeptide peptide
lead 1. The hydrophobic dihydrocinnamoyl group on the
N-terminus, the methanoPro, and the C-terminal ben-
zylamide resulted in additive effects on affinity.
Experimental Section
NR-Protected amino acids were purchased from Advanced
Chemtech, NovaBiochem, ChemImpex, or AnaSpec. HOBt was
from ChemImpex. Rink amide resin was from Advanced
Chemtech, loaded between 0.6 and 0.7 mmol/g. PL-DMA, a
polydimethylacrylamide-based resin, was purchased from
Polymer Laboratories, Ltd. Anhydrous DMF for amino acid
solutions was from Baker. Other solvents were reagent grade
and were used without further purification.
Peptides were purified by reverse-phase HPLC on a Rainin
rabbit HPLC using a Vydac 2.5 cm × 25 cm C18 peptide and
protein column. Gradients of ACN in H2O (both containing
0.1% TFA) or MeOH or ACN in 0.01 M NH4OAc (pH 6.5) at
10 mL/min were employed. Peptides were tested for purity by
reverse-phase HPLC on a Hewlett-Packard 1090 HPLC or an
Agilent 1100 HPLC using a Vydac 4.6 mm × 250 mm C18
peptide/protein column in two systems, one with 0.1% TFA in
both H2O and ACN (systems A-C) and the other with
increasing gradients of ACN or MeOH in 0.01 M NH4OAc, pH
6.5 (systems D-G).
A
B
C
D
E
F
0-40% ACN (0.1%TFA)
10-50% ACN (0.1%TFA)
10-80% ACN (0.1%TFA)
5-45% MeOH in 0.01 M NH4OAc, pH 6.5
15-55% MeOH in 0.01 M NH4OAc, pH 6.5
5-45% ACN in 0.01 M NH4OAc, pH 6.5
15-55% ACN in 0.01 M NH4OAc, pH 6.5
Conclusions
G
Our results suggest that a hydrogen bond exists
between the RNH of Leu at pY+1 of our peptides and
the CdO of Ser 636 of Stat3 as seen in the crystal
structure.29 Thus, we are confidant the Tyr(PO3H2)-Leu
part of our lead peptide binds to Stat3 in a “standard”
manner for phosphopeptide/SH2 domain complexes
(Figure 1). There appears to be a hydrophobic sur-
face on Stat3 that binds pY-1 substituents. cis-3,4-
Methanoproline was the optimum scaffold at position
pY+2. The stereochemistry of the enantiomer that
produces the high affinity is not known. Whether the
methano group makes important binding contact with
Stat3 or causes a conformational effect is unclear and
is under investigation. Glutamine at position pY+3 is
very important for affinity. Attempts to replace the side
chain amide protons with methyl groups or to replace
the amide nitrogen with a methyl group (Met(O)) were
deleterious, suggesting a hydrogen-bonding role. The
lower affinity of the larger methyl-substituted amide
suggests a tight fit of the CONH2 into a binding pocket
on the surface of Stat3. The intact structure of glutamine
is optimal. Removal of the R-carboxyl group resulted in
a loss of activity, suggesting a role for this group in
restriction of rotation of the bulk of the residue or a role
in hydrogen binding. Slight increases in activity of
All gradients were run at 1.5 mL/min and detection at 230
and 275 nm was performed simultaneously.
Fluorescence Polarization Assays. A 50 µL aliquot of a
solution of 0.8 µg of full length Stat3R (160 nM) and 20 nM of
probe in 50 mM NaCl, 10 mM Hepes, 1 mM Na4EDTA, 2 mM
DTT, and 1% NP-40 was placed in wells of a 96-well microtiter
plate. To each well was added 50 µL of a peptide solution in
the same buffer. Fluorescence polarization was then read in a
Tecan Polarian plate reader. With use of Prizm, version 4, from
GraphPad Software, Inc., the mP was plotted against the log
of the peptide concentration and IC50 values were obtained
from non-linear regression analysis in the one-site competition
mode. Peptides were assayed three times using three separate
Stat3-probe preparations. IC50 values are reported as the mean
of three IC50 values ( the standard deviation. Full length Stat3
was a gift from Dr. Xiaomin Chen.
Solid-Phase Peptide Synthesis: Automated Method.
Individual reactor vessels on an Advanced Chemtech 348
multiple peptide synthesizer were charged with 0.200 g of Rink
amide resin. Amino acids were dissolved to a concentration of
0.4 M in DMF containing 0.4 M HOBt. For coupling, 2 mL of
amino acid solution and 2 mL of 0.4 M DIPCDI (in CH2Cl2)
were transferred to the resin, and the mixture was vortexed
for 1 h. The resin was drained and washed 5× with 4.5 mL of
DMF/CH2Cl2. Fmoc removal was accomplished with two treat-
ments of 4.5 mL of 20% piperidine in DMF for 5 and 10 min
each. Resins were washed 5× with DMF/CH2Cl2. Cleavage was
accomplished with three treatments of the resins with 5 mL

6668 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21
Coleman et al.
of TFA/TIS/H2O (95:2.5:2.5)34 for 10 min each. The combined
filtrates sat at room temperature for 2 h. The volumes of the
solvents were reduced, and the solutions were dropped into
ice-cold Et2O. After 30 min, the precipitates were collected by
filtration and were washed 2× more with Et2O. Crude peptides
were dried, and peptides were purified by reverse-phase HPLC.
All peptides were dried over P2O5 in vacuo at 37 °C, and
peptide contents were calculated from the absorbance at 268
NH4OAc and purified to give 47.1 mg of peptide. HPLC tR (A-
D) 12.63; ESI-MS (M + H) calcd 756.75, found 756.5.
Ac-pTyr-Leu-Pro-Gln(Me2)-Thr-NH2, 28. Synthesized as
per 27 except that dimethylamine was coupled to the glutamic
acid side chain. Yield 31.1 mg. HPLC tR (A-D) 14.17; ESI-MS
(M + H) calcd 770.78, found 770.5.
Ac-Tyr(PO3H2)-Leu-Pro-Gln-NHMe, 39. Resin-bound Ac-
Tyr(PO[NMe2]2)-Leu-Pro-Gln-OH was treated with 0.6 mmol
each of MeNH3Cl, DIPCDI, and HOBt in DMF/CH2Cl2 plus
0.6 mmol of DIEA in 4 mL of DMF/CH2Cl2. After 5 h the resin
was drained and washed with DMF/CH2Cl2 followed by
CH2Cl2. The peptide was cleaved and purified as described
above for 27. Yield 30.3 mg. HPLC tR (A) 13.05, (D) 12.16; ESI-
MS (M + H) calcd 655.28, found 655.4. Peptides 40-49 were
synthesized as per 39 using the appropriate base.
nM using the extinction coefficient ꢀ ) 695 M-1 cm-1 46
. Peptide
yields, HPLC retention times, and mass spectra are tabulated
in the Supporting Information.
Solid-Phase Peptide Synthesis: Manual Method.
Manual solid-phase synthesis was carried out on aliquots of
0.20 g of Rink resin (0.6 mmol/g) or Rink-handle derivatized
PL-DMA resin. Fmoc groups were removed with 2 × 5 mL of
20% piperidine for 3 and 7 min each. For coupling, 3-fold
excesses of Fmoc-amino acids, PyBop, and HOBt were used
along with 6-fold excesses of DIPEA in 4 mL of DMF/CH2Cl2.
After coupling and deprotection steps, resins were washed with
3 × 5 mL of DMF/CH2Cl2. For the introduction of phospho-
tyrosine, Fmoc-Tyr(PO[NMe2]2)-OH was used. Cleavage was
accomplished with three treatments of the resins with 5 mL
of TFA/TIS/H2O (95:2.5:2.5)34 for 10 min each. The filtrates
were combined. After 2 h, 3 mL of H2O was added and the
reaction sat at room temperature overnight. The volumes of
the solvents were reduced, and the solutions were dropped into
ice-cold Et2O. After 30 min the precipitates were collected by
filtration and were washed 2× more with Et2O. Crude peptides
were dried, and peptides were purified by reverse-phase HPLC.
All peptides were dried over P2O5 in vacuo at 37 °C, and
peptide contents were calculated from the absorbance at 268
nM using the extinction coefficient ꢀ ) 695. Peptide yields,
HPLC retention times, and mass spectra are tabulated in
Table 1.
Preparation of PL-DMA Resin. Polydimethylacrylamide-
based PL-DMA resin was treated overnight with neat ethyl-
enediamine as described in Arshady et al.49 After thoroughly
washing the resin with DMF/CH2Cl2, Fmoc-Rink linker was
added in 3-fold excess, as calculated from the nominal loading
of 1 mmol/g. Coupling was mediated with 3-fold excesses of
PyBOP, HOBt, and a 6-fold excess of DIEA. On completion of
the coupling as judged by negative ninhydrin tests, the resin
was drained, washed with DMF/CH2Cl2 and CH2Cl2, and then
dried under vacuum and stored.
Solid-Phase Synthesis of Peptides 27, 28, 39-47. These
peptides were synthesized in parallel on aliquots of 0.20 g of
Rink amide resin (0.6 mmol/g, 0.12 mmol) on an Advanced
Chemtech 348 multiple synthesizer employing a 40-well reac-
tor block. For peptides 27 and 28, the sequence Ac-Tyr(PO-
[NMe2]2)-Leu-Pro-Glu(OAll)-Thr(tBu) (51) was synthesized in
two wells. For peptides 39-47, the sequence Ac-Tyr(PO-
[NMe2]2)-Leu-Pro-Gln-OAll (55) was synthesized in 10 wells.
Note that the latter peptide was attached to the resin via the
side chain of the C-terminal Gln by using Fmoc-Glu-OAll. On
completion of the syntheses, the resins were washed with 5 ×
4.5 mL of CH2Cl2. A solution of 30 mM Pd[PPh3]4 in CHCl3/
AcOH/NMM (37:2:1), 4 mL, was added to each well by syringe
using the automated synthesizer. The reactor was rotated for
2 h. The resin was drained and then washed with 5 × 4 mL of
DMF/CH2Cl2, 3 × 4 mL of 2% NaS2CNMe2 in DMF, 3 × 4 mL
of 0.5% HOBt in DMF, then 5 × 4 mL of DMF/CH2Cl2. Resins
were then derivatized in parallel as follows.
Ac-Tyr(PO3H2)-Leu-Pro-Gln(Me)-Thr-NH2, 27. A solu-
tion of 0.6 mmol each of MeNH3Cl, PyBOP, and HOBt plus
1.2 mmol of DIEA in 4 mL of DMF was added to resin-bound
Ac-Tyr(PO[NMe2]2)-Leu-Pro-Glu-Thr(tBu) in one reactor well.
After 5 h the resin was drained and washed with DMF/CH2Cl2
followed by CH2Cl2. The peptide was cleaved with 3 × 5 mL
of TFA/TIS/H2O (95:2.5:2.5) for 10 min each. The combined
cleavage solutions were mixed with 1.5 mL of H2O, and the
solution sat overnight. The TFA volume was reduced, and the
residue was dropped into ice-cold Et2O to precipitate the
product. After the peptide was washed in Et2O, it was purified
by reverse-phase HPLC using a gradient of MeOH in 0.01M
Synthesis of Ac-Tyr(PO3H2)-Leu-Pro-3-carboxamido-
anilide, 33. Boc-proline (0.869 g, 4.04 mmol), 3-aminobenz-
amide (0.500 g, 3.67 mmol), HBTU (1.39 g, 3.67 mmol), and
DBU (1.10 mL, 7.34 mmol) were stirred in 7 mL of DMF
overnight. The solvent was removed in vacuo, and the residue
was taken up in EtOAc and washed with 2% KHSO4, 5%
NaHCO3, and brine. After drying (MgSO4) and evaporation,
the yield was 0.621 g (1.94 mmol, 53%). The residue was
stirred in TFA for 1 h, and the acid was removed in vacuo.
Toluene was added and evaporated off twice to remove excess
TFA. The residue was stirred in 7 mL of DMF/CH2Cl2 (1:1).
Boc-Leu-OH (0.492 g, 1.95 mmol), EDC (0.378 g, 1.95 mmol),
and NMM (0.214 mL, 1.95 mmol) were added, and the mixture
stirred overnight. The solvents were removed in vacuo, and
the residue was taken up in EtOAc and washed as before. After
the mixture was dried (MgSO4), the solvent was removed to
yield 0.866 g of a white solid (1.94 mmol, 100%). The product
was treated with 20 mL of TFA for 1 h, precipitated in Et2O,
collected by centrifugation, and washed with Et2O twice more.
After the mixture was dried, the yield was 0.478 g (1.04 mmol,
53%). To 0.300 g (0.652 mmol) of residue in 5 mL of DMF/
CH2Cl2 (1:1) was added 0.385 g of Fmoc-Tyr(PO(NMe2)2)-OH
(0.716 mmol), 0.137 g of EDC (0.716 mmol), and 0.079 mL of
NMM (0.716 mmol). After overnight reaction, 0.160 mL of
NMM was added and the reaction allowed to proceed over-
night. The solvents were removed in vacuo, and the residue
was taken up in 30 mL of EtOAc. The organic phase was
washed with acid and base as before, dried, and evaporated
to yield 0.150 g of an off-white powder. This residue was stirred
in 1 mL of 2% DBU in DMF for 1 h. Ac2O, 40 µL, was added,
and after 2 h the product was precipitated by the addition of
Et2O. It was collected by centrifugation and washed with Et2O.
The product was treated with 3 mL of 90% (aqueous) TFA
overnight, precipitated in Et2O, centrifuged, and washed with
Et2O. After purification by reverse-phase HPLC, the yield was
19.1 mg. HPLC tR (A) 18.01, (D) 24.88; ESI-MS (M + H) calcd
695.31, found 695.3.
Synthesis of (R,S)-Fmoc-pyrrolidine-3-acetic Acid, 58.
To 20 mL of 1:1 acetone/H2O was added pyrrolidine-3-acetic
acid (1.0 g, 6.88 mmol), Fmoc-OSu (2.3 g, 6.88 mmol), and
NaHCO3 (2.19 g, 20.6 mmol), and the solution was stirred
overnight. The acetone was removed by evaporation in vacuo,
and the water layer was extracted with 3 × 15 mL of ethyl
acetate. The organic extracts were pooled, dried over MgSO4,
and filtered, and the solvent was evaporated in vacuo to yield
1.69 g (4.82 mmol, 70% yield) of a white solid. HPLC tR (A)
6.453; ESI-MS (M + 1) calcd 353.15, found 353.3; 1H NMR
(300 MHz, CDCl3) δ 7.76 (d, 2H, J ) 7.35 Hz), 7.60 (d, 2H, J
) 7.54 Hz), 7.39 (tr, 2H, J ) 7.34 Hz), 7.31 (tr, 2H, J ) 7.53
Hz), 4.40 (m, 2H), 4.24 (m, 1H), 3.73 (m, 1H, C(2)Ha), 3.54 (m,
1H, C(5)Ha), 3.40 (m, 1H, C(5)Hb), 3.09 (m, 1H, C(2)Hb), 2.63
(m, 1H, C(3)H), 2.48 (m, 2H, exoCH2), 2.16 (m, 1H, C(4)Ha),
1.64 (m, 1H, C(4)Hb); 13C NMR (75 MHz, CDCl3) δ (176.60,
176.34, doublet), 154.91, 144.10, 141.34, 127.67, 127.02, 125.10,
119.96, 67.64 (51.67, 51.32, doublet), 47.76 (45.93, 45.65,
doublet), 37.446 (35.75, 34.67, doublet), 31.89 and 31.10
(doublet).
Synthesis of Fmoc-Glu-NH-Bn, 59. Fmoc-glutamic acid
γ-tert-butyl ester (1.00 g, 2.26 mmol) was stirred in 20 mL of

Peptide Inhibitors of Stat3
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21 6669
(10) Schaefer, L. K.; Ren, Z.; Fuller, G. N.; Schaefer, T. S. Constitutive
activation of Stat3a in brain tumors: localization to tumor
endothelial cells and activation by the endothelial tyrosine
kinase receptor (VEGFR-2). Oncogene 2002, 21, 2058-2065.
(11) Dhir, R.; Ni, Z.; Lou, W.; DeMiguel, F.; Grandis, J. R.; Gao, A.
C. Stat3 activation in prostatic carcinomas. Prostate 2002, 51,
241-246.
(12) Seki, Y.; Suzuki, N.; Imaizumi, M.; Iwamoto, T.; Usami, N.;
Ueda, Y.; Hamaguchi, M. STAT3 and MAPK in human lung
cancer tissues and suppression of oncogenic growth by JAB and
dominant negative STAT3. Int. J. Oncol. 2004, 24, 931-934.
(13) Benekli, M.; Baer, M. R.; Baumann, H.; Wetzler, M. Signal
transducer and activator of transcription proteins in leukemias.
Blood 2003, 101, 2940-2954.
(14) Catlett-Falcone, R.; Landowski, T. H.; Oshiro, M. M.; Turkson,
J.; Levitzki, A.; Savino, R.; Ciliberto, G.; Moscinski, L.; Fernan-
dez-Luna, J. L.; Nunez, G.; Dalton, W. S.; Jove, R. Constitutive
activation of Stat3 signaling confers resistance to apoptosis
in human U266 myeloma cells. Immunity 1999, 10I, 105-
115.
(15) Weber-Nordt, R. M.; Egen, C.; Wehinger, J.; Ludwig, W.;
Gouilleux-Gruart, V.; Mertelsmann, R.; Finke, J. Constitutive
activation of STAT proteins in primary lymphoid and myeloid
leukemia cells and in Epstein-Barr virus (EBV)-related lym-
phoma cell lines. Blood 1996, 88, 809-816.
(16) Scholz, A.; Heinze, S.; Detjen, K. M.; Peters, M.; Welzel, M.;
Hauff, P.; Schirner, M.; Wiedenmann, B.; Rosewicz, S. Activated
signal transducer and activator of transcription 3 (STAT3)
supports the malignant phenotype of human pancreatic cancer.
Gastroenterology 2003, 125, 891-905.
(17) (a) Bowman, T.; Garcia, R.; Turkson, J.; Jove, R. STATs in
oncogenesis. Oncogene 2000, 19, 2474-2488. (b) Buettner, R.;
Mora, L. B.; Jove, R. Activated STAT signaling in human tumors
provides novel molecular targets for therapeutic intervention.
Clin. Cancer Res. 2002, 8, 945-954. (c) Bromberg, J. Stat
proteins and oncogenesis. J. Clin. Invest. 2002, 109, 1139-1142.
(d) Darnell, J. E., Jr. Transcription factors as targets for cancer
therapy. Nat. Rev. Cancer 2002, 2, 740-749. (e) Yu, H.; Jove,
R. The STATS of cancer. New molecular targets come of age.
Nat. Rev. Cancer 2004, 4, 97-105.
(18) (a) Burke, W. M.; Jin, X.; Lin, H.-J.; Huang, M.; Liu, R.;
Reynolds, R. K.; Lin, J. Inhibition of constitutively active Stat3
suppresses growth of human ovarian and breast cancer cells.
Oncogene 2001, 20, 7925-7934. (b) Niu, G.; Heller, R.; Catlett-
Falcone, R.; Coppola, D.; Jaroszeski, M.; Dalton, W.; Jove, R.;
Yu, H. Gene therapy with dominant-negative Stat3 suppresses
growth of the murine melanoma B16 tumor in vivo. Cancer Res.
1999, 59, 5059-5063. (c) Grandis, J. R.; Drenning, S. D.;
Chakraborty, A.; Zhou, M.-Y.; Zeng, Q.; Pitt, A. S.; Tweardy, D.
J. Requirement of Stat3 but not Stat1 activation for epidermal
growth factor receptor-mediated cell growth in vitro. J. Clin.
Invest. 1998, 102, 1385-1392. (d) Grandis, J. R.; Drenning, S.
D.; Zeng, Q.; Watkins, S. C.; Melhem, M. F.; Endo, S.; Johnson,
D. E.; Huang, L.; He, Y.; Kim, J. D. Constitutive activation of
Stat3 signaling abrogates apoptosis in squamous cell carcino-
genesis in vivo Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 4227-
4232.
(19) Real, P. J.; Sierra, A.; de Juan, A.; Segovia, J. C.; Lopez-Vega,
J. M.; Fernandez-Luna, J. L. Resistance to chemotherapy via
Stat3-dependent overexpression of Bcl-2 in metastatic breast
cancer cells. Oncogene 2002, 21, 7611-7618.
(20) Garcia, R.; Bowman, T. L.; Niu, G.; Yu, H.; Minton, S.; Muro-
Cacho, C. A.; Cox, C. E.; Falcone, R.; Fairclough, R.; Parsons,
S.; Laudano, A.; Gazit, A.; Levitzki, A.; Kraker, A.; Jove, R.
Constitutive activation of Stat3 by the Src and JAK tyrosine
kinases participates in growth regulation of human breast
carcinoma cells. Oncogene 2001, 20, 249-2513.
(21) (a) Sawyer, T. K. Src homology-2 domains: structure, mecha-
nisms, and drug discovery. Biopolymers 1998, 47, 243-261. (b)
Cody, W. L.; Lin, Z.; Panek, R. L.; Rose, D. W.; Rubin, J. R.
Progress in the development of inhibitors of SH2 domains. Curr.
Pharm. Des. 2000, 6, 59-98. (c) Muller, G. Peptidomimetic SH2
domain antagonists for targeting signal transduction. Top. Curr.
Chem. 2001, 211, 17-59. (d) Shakespeare, W. C. SH2 domain
inhibition: a problem solved? Curr. Opin. Chem. Biol. 2001, 5,
409-415. (e) Metcalf, C. A., III; Sawyer, T. Src Homology-2
Domains and Structure-based, Small-molecule Library Ap-
proaches to Drug Discovery. In Drug Discovery Strategies and
Methods; Makriyannis, A., Biegel, D., Eds.; Marcel Dekker: New
York, 2004; pp 23-59.
ACN along with 0.520 g of 1-ethyl-3-(3-dimethylaminopropyl)-
carbodiimide hydrochloride (2.71 mmol), 191.7 g of HOBt (2.71
mmol), and 0.247 mL of benzylamine (2.26 mmol). After
overnight reaction, the solvent was removed in vacuo and the
residue was taken up in 100 mL of EtOAc. The organic solution
was washed with 2% KHSO4, 5% NaHCO3, and brine. After
the mixture was dried (MgSO4) and filtered, the solvent was
removed to give a white solid. The residue was treated with
30 mL of TFA/H2O (95:5) for 1 h, and the solvent was removed
in vacuo. Toluene was added and evaporated 2× to give 1.10
g of a white powder (100% yield), which was used without
further purification: mp 178-180; 1H NMR (300 MHz, DMSO-
d6) δ 8.38 (tr, 1H, NHBn), 7.89 (d, 2H, Fmoc arom, J ) 7.4
Hz), 7.74 (d, 2H, Fmoc arom, J ) 6.4 Hz), 7.57 (d, 1H, Gln
NH, J ) 8.1 Hz), 7.2-7.4 (m, 9H Fmoc, Bn arom), 4.2 (m, 5H,
Fmoc CH-CH2, Bn CH2), 4.06 (m, 1H, Glu RCH), 2.28 (br tr,
2H, Glu γCH2), 1.97 (m, 1H Glu âCH) 1.83(M, 1H, Glu âCH);
13C NMR (75.5 MHz) δ 174.72, 172.35, 156.87, 144.77, 144.63,
141.58, 140.24, 129.08, 128.50, 127.94, 127.58, 126.18, 120.96,
66.57, 55.00, 47.55, 42.95, 41.21, 31.17, 28.12, 21.60.
Acknowledgment. We are grateful to the National
Cancer Institute for support of this work (Grant
CA096652). We also acknowledge the NCI Cancer
Center Support Grant CA016672 for the support of our
NMR facility and both the M. D. Anderson Cancer
Center Proteomics Facility and the Structural Biology
Program for mass spectrometry.
Supporting Information Available: Tables of compound
1
characterization and H NMR assignments for peptides 1-3.
This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) (a) Bromberg, J.; Darnell, J. E., Jr. The role of STATs in
transcriptional control and their impact on cellular function.
Oncogene 2000, 19, 2468-2473. (b) Levy, D. E.; Darnell, J. E.,
Jr. Stats: transcriptional control and biological impact. Nat. Rev.
Mol. Cell Biol. 2002, 3, 651-662. (c) Stark, G. R.; Kerr, I. M.;
Williams, B. R.; Silverman, R. H.; Schreiber, R. D. How cells
respond to interferons. Annu. Rev. Biochem. 1998, 67, 227-264.
(2) Akira, S.; Nishio, Y.; Inoue, M.; Wang, X. J.; Wei, S.; Matsusaka,
T.; Yoshida, K.; Sudo, T.; Naruto, M.; Kishimoto, T. Molecular
cloning of APRF, a novel IFN-stimulated gene factor 3 p91-
related transcription factor involved in the gp130-mediated
signaling pathway. Cell 1994, 77, 63-71.
(3) (a) Zhong, Z.; Wen, Z.; Darnell, J. E., Jr. Stat3: A STAT family
member activated by tyrosine phosphorylation in response to
epidermal growth factor and interleukin-6. Science 1994, 264,
95-98. (b) Zhong, Z.; Wen, Z.; Darnell, J. E., Jr. Stat3 and
Stat4: Members of the Family of signal transducers and
activators of transcription. Proc. Nat. Acad. Sci. U.S.A. 1994,
91, 4806-4810.
(4) Stahl, N.; Farruggella, T. J.; Boulton, T. G.; Zhong, Z.; Darnell,
J. E.; Yancopoulos, G. D. Choice of STATs and other substrates
specified by modular tyrosine-based motifs in cytokine receptors.
Science 1995, 267, 1349-1353.
(5) Gerhartz, C.; Heesel, B.; Sasse, J.; Hemmann, U.; Landgraf, C.;
Schneider-Mergener, J.; Horn, F.; Heinrich, P. C.; Graeve, L.
Differential activation of acute phase response factor/STAT3 and
STAT1 via the cytoplasmic domain of the interleukin 6 signal
transducer gp130. I. Definition of a novel phosphotyrosine motif
mediating STAT1 activation. J. Biol. Chem. 1996, 271, 12991-
12998.
(6) (a) Coffer, P.; Kruijer, W. EGF receptor deletions define a region
specifically mediating STAT transcription factor activation.
Biochem. Biophys. Res. Commun. 1995, 210, 74-81. (b) Tong,
Z.; Jing, M.; Xinmin, C. Grb2 regulates Stat3 activation nega-
tively in epidermal growth factor signalling. Biochem. J. 2003,
376, 457-464.
(7) Schreiner, S. J.; Schiavone, A. P.; Smithgall, T. E. Activation of
STAT3 by the Src family kinase Hck requires a functional SH3
domain. J. Biol. Chem. 2002, 277, 45680-45687.
(8) Song, J. I.; Grandis, J. R. STAT signaling in head and neck
cancer. Oncogene 2000, 19, 2489-2495.
(9) Garcia, R.; Yu, C. L.; Hudnall, A.; Catlett, R.; Nelson, K. L.;
Smithgall, T.; Fujita, D. J.; Ethier, S. P.; Jove, R. Constitutive
activation of Stat3 in fibroblasts transformed by diverse onco-
proteins and in breast carcinoma cells. Cell Growth Differ. 1997,
812, 1267-1276.
(22) Pawson, T. Protein modules and signalling networks. Nature
1995, 373, 573-580.
(23) (a) Songyang, Z.; Shoelson, S. E.; Chaudhuri, M.; Gish, G.;
Pawson, T.; Haser, W. G.; King, F.; Roberts, T.; Ratnofsky, S.;
Lechleider, R. J. SH2 domains recognize specific phosphopeptide
sequences. Cell 1993 72, 767-778. (b) Songyang, Z.; Shoelson,

6670 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 21
Coleman et al.
S. E.; McGlade, J.; Olivier, P.; Pawson, T.; Bustelo, X. R.;
Barbacid, M.; Sabe, H.; Hanafusa, H.; Yi, T. Specific motifs
recognized by the SH2 domains of Csk, 3BP2, fps/fes, GRB-2,
HCP, SHC, Syk, and Vav. Mol. Cell. Biol. 1994, 14, 2777-2785.
(24) Park, O. K.; Schaefer, L. K.; Wang, W.; Schaefer, T. S. Dimer
stability as a determinant of differential DNA binding activity
of Stat3 isoforms. J. Biol. Chem. 2000, 275, 32244-32249.
(25) Turkson, J.; Ryan, D.; Kim, J. S.; Zhang, Y.; Chen, Z.; Haura,
E.; Laudano, A.; Sebti, S.; Hamilton, A. D.; Jove, R. Phospho-
tyrosyl peptides block Stat3-mediated DNA-binding activity,
gene regulation and cell transformation. J. Biol. Chem. 2001,
276, 45443-45455.
(36) Chao, H.-G.; Bernatowicz, M. S.; Reiss, P. D.; Matsueda, G. R.
Synthesis and application of bis-silylethyl-derived phosphate-
protected Fmoc-phosphotyrosine derivatives for peptide synthe-
sis. J. Org. Chem. 1994, 59, 6687-6691.
(37) (a) Chao, H.-G.; Leiting, B.; Reiss, P. D.; Burkhardt, A. L.;
Klimas, C. E.; Bolen, J. B.; Matseuda, G. R. Synthesis and
Application of Fmoc-O-[bis(dimethylamino)phosphono]tyrosine,
a versatile protected phosphotyrosine equivalent. J. Org. Chem.
1995, 60, 7710-7711. (b) Ueki, M.; Tachibana, J.; Ishii, Y.;
Okumura, J.; Goto, M. N,N′-Dialkyldiamide-type phosphate
protecting groups for Fmoc synthesis of phosphotyrosine-
containing peptides. Tetrahedron Lett. 1996, 37, 4953-4956. (c)
Ueki, M.; Goto, M.; Okumura, J.; Ishii, Y. N,N′-Dialkyldiamide-
type phosphate protecting groups for Fmoc synthesis of phos-
photyrosine-containing peptides: Optimization of the alkyl
group. Bull. Chem. Soc. Jpn. 1998, 71, 1887-1898.
(26) Ren, Z.; Cabell, L. A.; Schaefer, T. S.; McMurray, J. S. Identifica-
tion of a high affinity phosphopeptide inhibitor of Stat3. Bioorg.
Med. Chem. Lett. 2003, 13, 633-636.
(27) Shao, H.; Cheng, H. Y.; Cook, R. G.; Tweardy, D. J. Identification
and characterization of signal transducer and activator of
transcription 3 recruitment sites within the epidermal growth
factor receptor. Cancer Res. 2003, 63, 3923-3930.
(28) Waksman, G.; Schoelson, S. E.; Pant, N.; Cowburn, D.; Kuriyan,
J. Binding of a high affinity phosphotyrosyl peptide to the Src
SH2 domain: crystal structures of the complexed and peptide
free forms. Cell 1993, 72, 779-790.
(29) Becker, S.; Gromer, B.; Muller, C. W. Three-dimensional struc-
ture of the Stat3â homodimer bound to DNA. Nature 1998, 394,
145-151.
(30) (a) Xu, R. X.; Word, J. M.; Davis, D. G.; Rink, M. J.; Willard, D.
H., Jr.; Gampe, R. T., Jr. Solution structure of the human pp60c-
src SH2 domain complexed with a phosphorylated tyrosine
pentapeptide. Biochemistry 1995, 34, 2107-2121. (b) Plumme,
R M. S.; Holland, D. R.; Shahripour, A.; Lunney, E. A.; Fergus,
J. H.; Marks, J. S.; McConnell, P.; Mueller, W. T.; Sawyer, T. K.
Design, synthesis, and cocrystal structure of a nonpeptide Src
SH2 domain ligand. J. Med. Chem. 1997, 40, 3719-3725. (c)
Ettmayer, P.; France, D.; Gounarides, J.; Jarosinski, M.; Martin,
M. S.; Rondeau, J. M.; Sabio, M.; Topiol, S.; Weidmann, B.;
Zurini, M.; Bair, K. W. Structural and conformational require-
ments for high-affinity binding to the SH2 domain of Grb2. J.
Med. Chem. 1999, 42, 971-980 (d) Tong, L.; Warren, T. C.; King,
J.; Betageri, R.; Rose, J.; Jakes, S. Crystal structures of the
human p56lck SH2 domain in complex with two short phospho-
tyrosyl peptides at 1.0 Å and 1.8 Å resolution. J. Mol. Biol. 1996,
256, 601-610. (e) Breeze, A. L.; Kara, B. V.; Barratt, D. G.;
Anderson, M.; Smith, J. C.; Luke, R. W.; Best, J. R.; Cartlidge,
S. A. Structure of a specific peptide complex of the carboxy-
terminal SH2 domain from the p85 alpha subunit of phospha-
tidylinositol 3-kinase. EMBO J. 1996, 15, 3579-89. (f) Pauptit,
R. A.; Dennis, C. A.; Derbyshire, D. J.; Breeze, A. L.; Weston, S.
A.; Rowsell, S.; Murshudov, G. N. NMR trial models: experiences
with the colicin immunity protein Im7 and the p85alpha
C-terminal SH2-peptide complex. Acta Crystallogr., Sect. D 2001,
57 (Part 10), 1397-1404.
(38) Chan, W. C.; Bycroft, B. W.; Evans, D. J.; White, P. D. A novel
4-aminobenzyl ester-based carboxy-protecting group for synthe-
sis of atypical peptides by Fmoc-But solid-phase chemistry. J.
Chem. Soc., Chem. Commun. 1995, 21, 2209-2210.
(39) Haan, S.; Hemmann, U.; Hassiepen, U.; Schaper, F.; Schneider-
Mergener, J.; Wollmer, A.; Heinrich, P. C.; Grotzinger, J.
Characterization and binding specificity of the monomeric
STAT3-SH2 domain. J. Biol. Chem. 1999, 274, 1342-1348.
(40) (a) Lynch, B. A.; Loiacono, K. A.; Tiong, C. L.; Adams, S. E.;
MacNeil, I. A. A fluorescence polarization based Src-SH2 binding
assay. Anal. Biochem. 1997, 247, 77-82. (b) Wu, P.; Brasseur,
M.; Schindler, U. A high-throughput STAT binding assay using
fluorescence polarization. Anal. Biochem. 1997, 249, 29-36. (c)
Schust, J.; Berg, T. A high-throughput fluorescence polarization
assay for signal transducer and activator of transcription 3. Anal.
Biochem. 2004, 330, 114-118.
(41) Turkson, J.; Kim, J. S.; Zhang, S.; Yuan, J.; Huang, M.; Glenn,
M.; Haura, E.; Sebti, S.; Hamilton, A. D.; Jove, R. Novel
peptidomimetic inhibitors of signal transducer and activator of
transcription 3 dimerization and biological activity. Mol. Cancer
Ther. 2004, 3, 261-269.
(42) Goodman, M.; Ro, S. Peptidomimetics for Drug Design. In
Burger’s Medicinal Chemistry and Drug Discovery, 5th ed.;
Wolff, M. E., Ed.; John Wiley & Sons Inc.: New York, 1995; pp
803-861.
(43) Garc´ıa-Echeverr´ıa, C.; Furet, P.; Gay, G.; Fretz, H.; Rahuel, J.;
Schoepfer, J.; Caravatti, G. Potent antagonists of the SH2
domain of Grb2: Optimization of the X+1 position of 3-amino-
Z-Tyr(PO3H2)-X+1-Asn-NH2. J. Med. Chem. 1998, 41, 1741-
1744.
(44) (a) Chakraborty, A.; Dyer, K. F.; Cascio, M.; Mietzner, T. A.;
Tweardy, D. J. Identification of a novel Stat3 recruitment and
activation motif within the granulocyte colony-stimulating factor
receptor. Blood 1999, 93, 15-24. (b) Weber-Nordt, R. M.; Riley,
J. K.; Greenlund, A. C.; Moore, K. W.; Darnell, J. E.; Schreiber,
B. D. Stat-3 recruitment by two distinct ligand-induced, tyrosine-
phosphorylated docking sites in the interleukin-10 receptor
intracellular domain. J. Biol. Chem. 1996a, 271, 27954-27961.
(c) Coffer, P.; Kruijer, W. EGF receptor deletions define a region
specifically mediating STAT transcription factor activation.
Biochem. Biophys. Res. Commun. 1995, 210, 74-81.
(45) Wiederkehr-Adam, M.; Ernst, P.; Muller, K.; Bieck, E.; Gombert,
F. O.; Ottl, J.; Graff, P.; Grossmuller, F.; Heim, M. H. Charac-
terization of phosphopeptide motifs specific for the Src homology
2 domains of signal transducer and activator of transcription 1
(STAT1) and STAT3. J. Biol. Chem. 2003, 278, 16117-16128.
(46) Bradshaw, J. M.; Waksman, G. Calorimetric investigation of
proton linkage by monitoring both the enthalpy and association
constant of binding: application to the interaction of the Src SH2
domain with a high-affinity tyrosyl phosphopeptide. Biochem-
istry 1998, 37, 15400-15407.
(31) (a) Chen, X.; Vinkemeier, U.; Zhao, Y.; Jeruzalmi, D.; Darnell,
J. E.; Kuriyan, J. Crystal structure of a tyrosine phosphorylated
STAT-1 dimer bound to DNA. Cell 1998, 93, 827-839. (b) Mao,
X.; Ren, Z.; Parker, G. N.; Sondermann, H.; Pastorello, M. A.;
Wang, W.; McMurray, J. S.; Demeler, B.; Darnell, J. E., Jr.;
Chen, X. Structural bases of unphosphorylated STAT1 associa-
tion and receptor binding. Mol. Cell 2005, 17, 761-771.
(32) Parts of this work were published in a preliminary communica-
tion: Ren, Z.; Coleman, D. R.; Cabell, L. A.; McMurray, J. S.
The Development of Peptide-Based Inhibitors of Stat3R. In
Peptide Revolution: Genomics, Proteomics
& Therapeutics,
Proceedings of the 18th American Peptide Symposium; Chorev,
M., Sawyer, T. K., Eds.; American Peptide Society: San Diego,
CA; 2004, pp 550-552.
(33) Ottinger, E. A.; Shekels, L. L.; Bernlohr, D. A.; Barany, G.
Synthesis of phosphotyrosine-containing peptides and their use
as substrates for protein tyrosine phosphatases. Biochemistry
1993, 32, 4354-4361.
(47) Arshady, R.; Atherton, E.; Clive, D. L. J.; Sheppard, R. C. Peptide
synthesis. Part 1. Preparation and use of polar supports based
on polydimethylacrylamide. J. Chem. Soc., Perkin Trans. 1 1981,
529-537.
(48) DeLano, W. L. The PyMOL Molecular Graphics System; DeLano
Scientific LLC: San Carlos, CA; http://www.pymol.org.
(34) Pearson, D. A.; Blanchette, M.; Baker, M. L.; Guindon, C. A
Trialkylsilanes as scavengers for the trifluoroacetic acid de-
blocking of protecting groups in peptide synthesis. Tetrahedron
Lett. 1989, 30, 2739-2742.
(35) Kates, S. A.; Daniels, S. B.; Albericio, F. Automated allyl cleavage
for continuous-flow synthesis of cyclic and branched peptides.
Anal. Biochem. 1993, 212, 303-310.
JM050513M