Identification of she Src homology 2 domain-binding peptoid-peptide hybrids

Article abstract of DOI:10.1021/jm800789h

A fluorescence anisotropy (FA) competition-based Shc Src homology 2 (SH2) domain-binding was established using the high affinity fluorescein isothiocyanate (FITC) containing peptide, FITC-NH-(CH₂) ₄-CO-pY-Q- G-L-S-amide (8; K_d

Full text of DOI:10.1021/jm800789h

1612
J. Med. Chem. 2009, 52, 1612–1618
Identification of Shc Src Homology 2 Domain-Binding Peptoid-Peptide Hybrids
Won Jun Choi,† Sung-Eun Kim,† Andrew G. Stephen,‡ Iwona Weidlich,†,§ Alessio Giubellino,| Fa Liu,† Karen M. Worthy,‡
Lakshman Bindu,‡ Matthew J. Fivash, Marc C. Nicklaus,† Donald P. Bottaro,| Robert J. Fisher,‡ and Terrence R. Burke, Jr*,†
Laboratory of Medicinal Chemistry, Center for Cancer Research, National Cancer Institute, National Insitutes of Health, Frederick, Maryland
21702, Protein Chemistry Laboratory, AdVanced Technology Program, SAICsFrederick, Frederick, Maryland 21702, Urologic Oncology
Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, and Data
Management SerVices, Inc., NCIsFrederick, Frederick, Maryland 21702
ReceiVed June 30, 2008
A fluorescence anisotropy (FA) competition-based Shc Src homology 2 (SH2) domain-binding was established
using the high affinity fluorescein isothiocyanate (FITC) containing peptide, FITC-NH-(CH2)4-CO-pY-Q-
G-L-S-amide (8; Kd ) 0.35 µM). Examination of a series of open-chain bis-alkenylamide containing peptides,
prepared as ring-closing metathesis precursors, showed that the highest affinities were obtained by replacement
of the original Gly residue with NR-substituted Gly (NSG) “peptoid” residues. This provided peptoid-peptide
hybrids of the form “Ac-pY-Q-[NSG]-L-amide.” Depending on the NSG substituent, certain of these hybrids
exhibited up to 40-fold higher Shc SH2 domain-binding affinity than the parent Gly-containing peptide
(IC50 ) 248 µM) (for example, for N-homoallyl analogue 50, IC50 ) 6 µM). To our knowledge, this work
represents the first successful example of the application of peptoid-peptide hybrids in the design of SH2
domain-binding antagonists. These results could provide a foundation for further structural optimization of
Shc SH2 domain-binding peptide mimetics.
Introduction
inhibitory motifs that grew out of unsuccessful attempts to
develop macrocyclic Shc SH2 domain binding inhibitors.
Disrupting protein-protein interactions is evolving as a
challenging but potentially rewarding approach to therapeutic
development.1 Src homology 2 (SH2a) domains facilitate the
formation of signaling complexes by mediating the recognition
and binding to phosphotyrosyl (pTyr) containing sequences,2-4
and SH2 domain-binding antagonists may down-regulate
protein-tyrosine kinase signaling in a fashion that is comple-
mentary to kinase-directed inhibitors.5,6 Although significant
effort has focused on development of p56lck, Src, and growth
factor receptor bound protein 2 (Grb2) SH2 domain-binding
antagonists,7-11 relatively little has been reported on agents
directed against the Shc SH2 domain.12,13 Members of the Shc
family of noncatalytic SH2 domain-containing adapter proteins
serve as transforming elements involved with mitogenic signal
transduction,14,15 and they have been shown to be essential for
induction of the proangiogenic vascular endothelial growth
factor (VEGF).16 Pathways downstream of Shc or Grb2 have
alsobeenshowntobecriticalfortransformationandmetastasis.17,18
The importance of Shc in Ras activation pathways has made
antagonists of Shc signaling potentially attractive therapeutic
targets. The current report details unexpected finding of potent
Results and Discussion
Development of Shc SH2 Domain-Binding Fluorescence
Anisotropy Assays: Identification of a High Affinity
Fluorescein Isothiocyanate Containing Peptide. Previous
analyses of Shc SH2 domain-binding constants have relied on
relatively labor-intensive NMR or surface plasmon resonance
techniques.13 Assays based on fluorescence anisotropy (FA)
afford attractive alternatives for determining binding constants,
but they can be limited by the need to append fluorescein
isothiocyanate (FITC) to every peptide examined. However, this
requirement can be circumvented by the availability of a high
affinity reference FITC-containing peptide that can be used in
FA assays to determine affinities of non-FITC labeled peptides
by competition techniques. Therefore, our first objective of the
current study was to identify a FITC-containing reference
peptide having high Shc SH2 domain binding affinity for use
in FA competition assays.
Modeling based on NMR solution studies of the ꢀ chain
T-cell-receptor-derived 14-mer peptide “GHDGLpYQGL-
STATK” (1) bound to the Shc SH2 domain19,20 indicated that
significant interactions only occur for residues “pYQGL.”
Accordingly, the slightly longer “LpYQGLS” sequence (2) was
used as a starting point for structural exploration of FITC-
containing peptides. Initial studies were conducted to determine
the optimal introduction of an N-terminal FITC group using a
variety of spacer lengths (3-14, Table 1). Direct N-terminal
attachment of the FITC group (3) resulted in a complete loss
of binding affinity. However, deletion of the N-terminal Leu
residue from 2 prior to addition of the FITC tag gave high
affinity (4, Kd ≈ 12 µM). Adding the FITC label to the latter
“pY-Q-G-L-S-amide” sequence by means of a series of progres-
sively longer amino-n-alkylcarboxylic acid amide linkers (5-9)
gave the highest affinity using a pentanoyl chain (8, Kd ) 0.35
µM). The best affinities from a similar series of FITC-containing
* To whom correspondence should be addressed. Phone: (301) 846-5906.
Fax: (301) 846-6033. E-mail: tburke@helix.nih.gov.
†
Laboratory of Medicinal Chemistry, National Cancer Institute.
SAICsFrederick.
On leave from the Poznan University of Medical Sciences, Faculty of
‡
§
Pharmacy, Poland.
|
Urologic Oncology Branch, National Cancer Institute.
Data Management Services, Inc.
a Abbreviations: FA, fluorescence anisotropy; FITC, fluorescein isothio-
cyanate; SH2, Src homology 2; NSG, NR-substituted glycine; pTyr,
phosphotyrosyl; VEGF, vascular endothelial growth factor; Grb2, growth
factor receptor bound protein 2; PEG, polyethylene glycol; HGF, hepatocyte
growth factor; RCM, ring-closing metathesis; FEB, free energy of binding;
NMP, N-methyl-2-pyrrolidone; DIPCDI, N,N′-diisopropylcarbodiimide;
HOBT, 1-hydroxybenzotriazole; PyBOP, (benzotriazol-1-yloxy)tripyrroli-
dinophosphonium hexafluorophosphate; DIPEA, N,N-diisopropylethylamine;
EDT, 1,2-ethanedithiol.
10.1021/jm800789h CCC: $40.75
2009 American Chemical Society
Published on Web 02/18/2009

Peptoid-Peptide Hybrids
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6 1613
Table 1. Shc SH2 Domain-Binding Affinities Determined Using a
Fluorescence Anisotropy Assay
To examine the possible effects in binding affinity of the
linker structure other than chain length, analogues were prepared
based on the FITC-X-pY-Q-G-L-S-amide sequence (16-28,
Table 2). These more structurally complex linkers could
potentially affect binding affinity by interacting directly with
the SH2 domain protein or by positioning the FITC ring system.
Most of the peptides exhibited binding affinities in the range
5-25 µM. However, peptides 24 (Kd ≈ 0.8 µM) and 27 (Kd ≈
0.5 µM) exhibited approximately 10-fold higher affinities than
other members of the series. Of particular note was the 50-fold
higher affinity of 27 relative to its trans isomer (26). Since none
of the peptides in Table 2 exhibited higher affinity than the
original FITC-containing 8, this latter peptide was chosen for
use as a reference in subsequent FA competition assays.
compd
sequence
Kd ( SD (µM)a
3
4
5
6
7
8
9
10
11
12
13
14
15
FITC-L-pY-Q-G-L-S-amide
FITC-pY-Q-G-L-S-amide
nbb
12.4 ( 2.6
FITC-NH-(CH2)-CO-pY-Q-G-L-S-amide
FITC-NH-(CH2)2-CO-pY-Q-G-L-S-amide
FITC-NH-(CH2)3-CO-pY-Q-G-L-S-amide
FITC-NH-(CH2)4-CO-pY-Q-G-L-S-amide
FITC-NH-(CH2)5-CO-pY-Q-G-L-S-amide
FITC-NH-(CH2)-CO-L-pY-Q-G-L-S-amide
FITC-NH-(CH2)2-CO-L-pY-Q-G-L-S-amide
FITC-NH-(CH2)3-CO-L-pY-Q-G-L-S-amide
FITC-NH-(CH2)4-CO-L-pY-Q-G-L-S-amide
FITC-NH-(CH2)5-CO-L-pY-Q-G-L-S-amide
Ac-NH-(CH2)4-CO-pY-Q-G-L-S-amide
nbb
1.09 ( 0.04
7.2 ( 0.2
0.35 ( 0.13
4.7 ( 0.1
nbb
7.6 ( 0.5
8.6 ( 0.3
8.9 ( 0.3
nbb
158 ( 10c
a Except as noted, determined by fluorescene anisotropy techniques as
described in the Experiment Section. b No binding. c IC50 values determined
by FA competition experiments using peptide 8 as the reference peptide.
Blockade of Shc Binding to c-Met Tyrosine Kinase in
Whole Cells. In order to confirm that the highest affinity FITC-
labeled peptide (8) is capable of binding to the Shc SH2 domain
in a physiologically relevant manner, studies were performed
that examined the ability of 8 to inhibit the binding of
intracellular Shc protein to the cytoplasmic domain of the c-Met
protein-tyrosine kinase in intact human mammary B5/589 cells,
which express high levels of c-Met. For this purpose, Universal
PEG Novatag resin21 was used to prepare peptides 29 and 30
(Figure 1), which contained the membrane carrier peptide
peptides prepared using the longer “L-pY-Q-G-L-S-amide”
sequence (10-14) were approximately 20-fold less potent.
Replacement of the FITC group in 8 with N-acetyl (15, IC50
)
158 µM) resulted in a significant loss of binding affinity as
determined using a FA competition assay against 8 as the
reference peptide. This indicated that the FITC moiety in 8
contributes substantially to the binding affinity.
Table 2. Shc SH2 Domain-Binding Affinities for FITC-X-pY-Q-G-L-S-amide
a Determined by a direct binding fluorescence anisotropy assay as described in the Experimental Section.

1614 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6
Choi et al.
29 and 30 as determined by extracellular FA competition assays
were Kd ) 5 and 1 µM, respectively. Following incubation of
cells with the peptides in the presence and absence of hepatocyte
growth factor (HGF, a ligand for c-Met receptor), the cells were
washed and lysed and binding of Shc to activated c-Met was
measured as described in the Experimental Section. At 10 µM,
both 29 and 30 were able to achieve a consistent blockade of
intracellular binding of Shc to c-Met (Figure 1).
Examination of Shc SH2 Domain-Binding Affinties of
N-Alkenyl-Containing Peptides (31-44) Designed as
Ring-Closing Metathesis Precursors. We had previously
shown that induction of conformational constraint through
macrocyclization using ring-closing metathesis (RCM)24 could
improve the binding affinity of Grb2 SH2 domain-binding
peptides.25 In the current study we performed molecular
modeling experiments to explore the potential utility of RCM
macrocyclization in the design of Shc SH2 domain-binding
peptides. However, these modeling studies failed to identify
obvious macrocyclic candidates. Nonetheless, we undertook
work directed at the preparation of a library of macrocycles.
This was intended as an empirical alternative to a “structure-
based design” approach. For this purpose, two series of peptides
were prepared, predicated on the sequences Ac-pY-Q-G-L-S-
NHR and Ac-pY-Q-G-L-NHR (A and B, respectively, Table
3), where R ) allyl and homoallyl. For each series of peptides,
additional N-allyl groups were placed at various nitrogen
positions within the peptide backbone to form bis-alkenyl open-
chain RCM precursors. Shc SH2 domain binding affinities of
these putative precursors were determined prior to macrocy-
clization studies.
In the A series, the highest affinities were obtained by
N-allylation of the Gln-Gly amide nitrogen, with the N-terminal
homoallyl peptide (37; IC50 ) 8 µM) exhibiting higher affinity
Figure 1. Inhibition by peptides 29 (A) and 30 (B) of binding of Shc to
hepatocyte growth factor receptor (c-Met) tyrosine kinase in intact B5/
589 cells as described in Experimental Section. Cells were incubated for
1 h in the presence (10 µM) or absence (0 µM) of the indicated peptides,
then untreated (control) or briefly stimulated with HGF (+HGF). Cells
were lysed and the amount of Shc bound to c-Met were determined (upper
panels) and the amount c-Met recovered (lower panels).
than the corresponding N-terminal allyl congener (33; IC50
)
48 µM). Placement of the backbone N-alkenyl group at other
positions in the peptide chain (position 1 or 3, Table 3) or
deletion entirely (peptides 31 and 35) gave generally lower
affinities. Similar results were found in the B series. The data
presented in Table 3 were of note in that the highest affinities
were derived by replacing the original Gly residues with NR-
substituted glycines (NSG). The NSG groups are the funda-
mental units of the “peptoid” class of peptide mimetics, and
sequences “RRRRRRRR”22 and “RQIKIWFQNRRMKWKK”23
attached to the C-terminus of 8 by means of a polyethylene
glycol (PEG) spacer. The Shc SH2 doman-binding affinities of
Table 3. Shc SH2 Domain-Binding Affinities of Bis-N-Olefin Containing Peptides
peptide A analogues
peptide B analogues
compd
positiona
n
IC50 ( SD (µM)b
compd
positiona
n
IC50 ( SD (µM)b
31
32
33
34
35
36
37
0
1
2
3
0
1
2
1
1
1
1
2
2
2
189 ( 15
353 ( 34
48 ( 2
38
39
40
41
42
43
44
0
1
2
3
0
1
2
1
1
1
1
2
2
2
91 ( 12
81 ( 6
47 ( 2
41 ( 7
450 ( 38
84 ( 4
84 ( 6
14 ( 3
165 ( 4
89 ( 5
8 ( 2
a Site of N-allyl substitution. A designation of “0” indicates the lack of N-allyl substitution. b Determined by a fluorescence anisotropy based competition
assay using peptide 8 as a reference as described in the Experimental Section.

Peptoid-Peptide Hybrids
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6 1615
Table 4. Shc SH2 Domain-Binding Affinities of N-Alkylglycine Containing Peptides
a Determined by a fluorescence anisotropy based competition assay using peptide 8 as the reference peptide as described in the Experimental Section.
peptides such as 37 and 44 represent peptoid-peptide hybrids.
Toourknowledge,thisisthefirstidentificationofpeptoid-peptide
hybrids as high affinity SH2 domain-binding inhibitors.
As indicated above, the bis-alkenylamine-containing peptides
were originally intended as precursors for the preparation of
macrocyclic peptide mimetics. However, RCM macrocyclization
studies carried out using second generation Grubbs catalyst26 were
largely unsuccessful and only a small number of ring-closed
products could be obtained. Consistent with preliminary molecular
modeling studies that failed to predict binding enhancement through

1616 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6
Choi et al.
Figure 2. Hydrophobic and hydrophilic surface maps for two conformations (A and B) of peptide ligand 50 binding to Shc SH2 domain formed
after 10 ps of molecular dynamics simulations. The coloring of atoms is as follows: carbon, green; nitrogen, blue; oxygen, red. The N-homoallyl
group interacts with hydrophobic pockets. Hydrophobic and hydrophilic surface maps are colored in orange and blue, respectively. Peptide 50 is
displayed in ball and stick representation.
macrocyclization, none of the obtained final cyclic peptides showed
significantly enhanced affinity (data not shown), and further work
with peptide macrocyclization was not pursued.
that both complexes were energetically similar (conformation
A, -21 957.656 kJ/mol; conformation B, -21 957.133 kJ/mol).
NMR and crystallographic studies have previously indicated that
interactions of peptide ligands with Shc SH2 domain protein
can vary significantly.19,27 Therefore, it was not surprising that
binding of peptide 50 differed from that reported for the 14-
mer T-cell-derived parent peptide (1).19 For peptide 50, the
slightly more favorable conformation A (Figure 2) has the
N-homoallyl group interacting with a hydrophobic pocket
created by residues Leu54, Leu76, Ile87, and Ile88. In confor-
mation B, the Ala90 residue has shifted position, allowing the
N-homoallyl group to bind within a newly created hydrophobic
pocket. In both conformations, the homoallyl group makes
significant binding interactions that may explain the relatively
high affinity of the peptoid-peptide hybrids.
Examination of N-Substituted Glycine Residues in
Peptides Lacking C-Terminal Alkenylamide Substituents.
Because N-alkenylation of the Gln-Gly amide nitrogen appeared
to be advantageous in the open-chain series bearing C-terminal
alkenylamides (for example, 37 and 44), studies were undertaken
to examine an array of NSG residues in a simplified peptide
platform lacking any C-terminal carboxamide substituents (Table
4). For the parent nonsubstituted Gly-containing peptides 38
and 42, deletion of the C-terminal allyl and homoallyl amide
functionalities, respectively, gave “Ac-pY-Q-G-L-amide”, which
resulted in an approximate 2- to 3-fold loss of binding affinity
(45, Table 4; IC50 ) 248 µM). Relative to 45, addition of a
C-terminal Ser residue (“Ac-pY-Q-G-L-S-amide”) resulted in
an approximate doubling of binding affinity (IC50 ) 112.5 (
0.7 µM), while shortening 45 by removing the N-terminal Leu
residue abrogated nearly all binding affinity (“Ac-pY-Q-G-
amide” IC50 > 500 µM). Therefore, peptide 45 was taken as a
starting point for peptoid-peptide synthesis because it repre-
sented a balance between size and binding affinity.
Conclusions
By use of a FA-based competition assay based on the high
affinity FITC-containing peptide 8, evaluation of a number of
bis-alkenylamide peptides for Shc SH2 domain-binding affinity
resulted in the discovery of pTyr + 2 N-alkyl-Gly residues as
favorable structural motifs. Deletion of the C-terminal alkenyl-
amide could be achieved without overall loss of binding affinity.
Molecular modeling studies identified specific pockets within
the protein surface that appear to afford regions of hydrophobic
interactions with the NSG side chain. These findings represent
the first examples of peptoid-peptide hybrids as preferred SH2
domain-binding motifs. As such, this work could provide the
foundation for further structure-based optimization of Shc SH2
domain-binding peptide mimetics.
Starting from 45, formation of NSG residues by alkylation
of the Gln-Gly amide bond enhanced binding affinity by an order
of magnitude or more (Table 4). The highest affinities were
obtained with 4-carbon linear chains (n-butyl (49), IC50 ) 8.9
µM and homoallyl (50), IC50 ) 6.0 µM). Increasing chain length
beyond 4 units resulted in a loss of affinity (n-pentyl (58) IC50
) 10 µM and n-hexyl (63), IC50 ) 20 µM), as did insertion of
heteroatoms (oxygen (56), IC50 ) 56 µM; sulfur (62), IC50
)
34 µM or nitrogen (65), IC50 ) 44 µM). Although an
approximate 2- to 3-fold loss of potency was incurred by the
original deletion of C-terminal alkenylamide functionality,
subsequent introduction of a homoallyl-NSG residue (50) more
than compensated for the potency loss to yield a peptide that
exibited twice the affinity of the parent bis-homoallyl-containing
44.
Molecular Modeling. Molecular dynamics (MD) simulations
of N-homoallyl-containing 50 bound to the Shc SH2 domain
protein were conducted. The two conformations with the two
lowest energy complexes are depicted in Figure 2. Subjecting
these complexes to FEB (free energy of binding) calculations
using the program eMBrAcE developed by Schro¨dinger showed
Experimental Section
Preparation of Shc SH2 Domain Protein. Shc protein was
prepared using previously published protocols.28 A total of 1.2 mg
of lyophilized Shc protein was resuspended in a 17 µL of 100%
(v/v) dimethyl sulfoxide (DMSO), and then 159 µL of a buffer
solution (BS) consisting of 10 mM HEPES, 150 mM NaCl, pH
7.5, was added for a stock of 500 µM protein.
Steady State Fluorescence Anisotropy Assays. Aliquots of a
solution of 5 nM peptide labeled with fluorescein at the N-terminus
in 1× BS, 10% (v/v) dimethyl sulfoxide were placed in 96-well
Costar polypropylene plates (Corning, NY). Shc protein was serially
diluted in the peptides with a starting concentration of 100 µM.
Then 40 µL was removed and transferred into 384-well Costar

Peptoid-Peptide Hybrids
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6 1617
polypropylene plates (Corning, NY). Samples were excited at 485
nm, and the emission intensities at 535nm from the parallel and
perpendicular planes were measured using a Tecan Ultra plate reader
(Durham, NC).
Curves were fit assuming a single binding site per peptide, using
the relation
each simulation, and the two conformations with the two lowest
energy complexes are depicted in Figure 2. These two complexes
were subjected to FEB (free energy of binding) calculations using
the program eMBrAcE developed by Schro¨dinger. Both complexes
were energetically similar (conformation A, -21 957.656 kJ/mol;
conformation B, -21 957.133 kJ/mol).
General Synthetic Procedures. Except where noted, solid-phase
peptide synthesis was performed using NovaSyn TGR resin (Nova-
biochem, Inc., catalog number 01-64-0060) in 5 or 10 mL disposable
polypropylene columns (Pierce) fitted with porous polyethylene filter
disks. Resin mixing was achieved by means of 360° rotation on a
Barnstead-Thermolyne Labquake shaker. Protected-peptide resins were
manually constructed using Fmoc-based solid-phase protocols. Trityl
protection was employed for Gln and Ser side chain protection. The
pTyr residue was incorporated in its PO(NMe2)2 form. Fmoc amino
acids were coupled in NMP by treatment with Fmoc amino acid (5
equiv) and coupling reagents [DIPCDI (5 equiv)/HOBt (5 equiv) for
2 h; PyBroP (5 equiv)/DIPEA (10 equiv) for 2 h]. Double couplings
were used for secondary amines. After completion of the coupling
reactions, resins were washed (DMF, 6×), and then Fmoc deprotection
was achieved using 20% piperidine in DMF (20 min) followed by
DMF wash (6×). Final N-terminal acetylation was performed in DMF
by treating with 1-acetylimidazole (for primary amines) or acetic
anhydride/DIPEA (for secondary amines). Introduction of N-terminal
fluorescein was achieved by reaction of deprotected resins with
fluorescein isothiocyanate (“isomer I”, Sigma Aldrich Corp) in DMF
containing 3 equiv of diisopropylamine (shaken overnight and shielded
from light). Finished resins were washed with DMF, methanol,
dichloromethane, and ether and then dried under vacuum. Peptides
and peptoid-peptide hybrids were cleaved from the resin by treatment
(3 h) with 5 mL of trifluoroacetic acid/EDT/H2O (76:20:4). An
additional 0.5 mL of H2O was added, and mixing was continued
(overnight). Resin was removed by filtration, and the filtrate was
concentrated under vacuum. Crude peptide precipitated by the addition
of diethyl ether, and the precipitate was washed with diethyl ether (2×).
The resulting solids were dissolved in aqueous acetonitrile and purified
by reverse phase preparative HPLC using a Phenomenex C18 column
(21.20 mm diameter × 250 mm) with a linear gradient from 0%
aqueous acetonitrile (0.1% trifluoroacetic acid) to 100% acetonitrile
(0.1% trifluoroacetic acid) over 30 min at a flow rate of 10.0 mL/min.
Lyophilization provided products as amorphous solids.
An(C) ) ((RAnbound - Anfree)({(Kd + C + Lt) - [(Kd + C +
Lt)2 - 4LtC]1⁄2} ⁄ (2Lt)) + Anfree) ⁄ ((R - 1)({(Kd + C + Lt) -
[(Kd + C + Lt)2 - 4LtC]1⁄2} ⁄ (2Lt)) + 1) (1)
where An(C) is the measured anisotropy as a function of
concentration, Lt is the peptide concentration, Anfree is the anisotropy
of unbound peptide, Anbound is the anisotropy of the completely
bound peptide solution, R is the ratio of the fluorescence intensity
of saturated bound peptide relative to free peptide, C is the protein
concentration, and Kd is the measured dissociation constant resulting
from the fit.
Fluorescence Anisotropy Competition Assays. A complex of
2 µM Shc and 10 nM peptide 8 in 1× BS, 10% (v/v) DMSO was
placed in 96-well Costar polypropylene plates (Corning, NY).
Unlabeled competitor peptides were serially diluted into the complex
with starting concentrations of 500 µM. A total of 40 µL was
removed and transferred into 384-well Costar polypropylene plates
(Corning, NY). Samples were excited at 485 nm, and the emission
intensities at 535 nm from the parallel and perpendicular planes
were measured using a Tecan Ultra plate reader (Durham, NC).
To estimate the IC50 values for these competition experiments, the
fractional binding of the competitor ligand was determined from
the binding part of the anisotropy signal using estimates of the
maximum and minimum anisotropy signal.29 The fractional binding
signal is described using a least-squares spline approximate.30 Three
knots in a spline of order 3 were used, with the value at which the
spline achieved one-half being taken as an estimate of the IC50 value.
Inhibition of Shc SH2 Domain Binding to Cytoplasmic
c-Met Tyrosine Kinase Receptor in Intact Cells. Intact B5/589
human mammary epithelial cells were serum-deprived (48 h), then
incubated (1 h) in presence or absence of peptides 29 and 30. Cells
were treated with hepatocyte growth factor (R&D Systems, Min-
neapolis, MN; 50 ng/mL, 20 min) and then lysed in cold buffer (pH
7.4) containing nonionic detergent and protease and phosphatase
inhibitors. After incubation with a biotin-tagged, affinity-purified c-Met
antibody (BAF358, R&D Systems, Minneapolis, MN; 1 h on ice),
immunocomplexes were captured using a streptavidin resin (Pierce,
Rockford, IL, 1 h at 4 °C with rotation). The beads were washed with
cold lysis buffer (3×), and samples were eluted with SDS sample
buffer and resolved by SDS-PAGE prior to electrophoretic transfer
to PVDF membranes (Immobilon P; Millipore, Billerica, MA).
Chemiluminescent detection of anti-Shc (Santa Cruz Biotechnology)
or anti-c-Met (AF276, R&D Systems, Minneapolis, MN) antibodies
was performed using ECL (GE Healthcare).
FITC-NH-(CH2)4-CO-pTyr-Gln-Gly-Leu-Ser-amide (8). 1H
NMR (400 MHz, DMSO-d6): δ 10.09 (s, 1H), 9.93 (s, 1H), 8.25
(br, 1H), 8.15 (d, 1H), 8.06 (t, 1H), 7.93 (d, 1H), 7.81 (d, J ) 4
Hz,1H), 7.73 (d, J ) 4 Hz, 1H), 7.17 (d, 1H), 7.11 (d, 1H), 7.01
(d, J ) 4 Hz, 2H), 6.73 (br, 1H), 6.62 (d, 2H), 6.54-6.50 (m, 3H),
4.50-4.45 (m, 1H), 4.33-4.27 (m, 1H), 4.18 (dd, J1 ) 8 Hz, J2 )
4 Hz, 1H), 4.13-4.10 (m, 1H), 3.76-3.67 (m, 2H), 3.53-3.46 (m,
2H), 3.39 (s, 2H), 2.98 (d, J ) 8 Hz, 1H), 2.68-2.62 (m, 1H),
2.06 (dd, J1 ) 4 Hz, J2 ) 4 Hz, 4H), 1.88-1.83 (m, 1H), 1.77-1.70
(m, 1H), 1.51-1.59 (m, 1H), 1.46-1.37 (m, 2H), 0.82 (d, J ) 4
Hz, 3H), 0.79 (d, J ) 8 Hz, 3H). MALDI-MS calcd for
C51H60N9O17PS: 1134.11; found 1134.82.
Molecular Modeling. Calculations were performed on Silicon
Graphics computer running the IRIX operating system. All simula-
tions were performed using the molecular modeling tools available
within Macromodel 9.1 (Schro¨dinger Inc.). The peptide GHDG-
LyQGLSTATK (1) extracted from the Shc SH2 domain NMR
structure (PDB code 1TCE) was used as a starting geometry for
the modeling study. The structure of the bound ligand was modified
using the 3D-sketcher tools available within Macromodel 9.1 to
yield initial structures of peptide 50 complexed with the protein.
The resulting protein-ligand complex was then subjected to
minimization using the OPLS2005 method until a gradient con-
vergence threshold of 0.05 was reached. This was followed by 10
ps of equilibration and 50 ps of molecular dynamics simulation
using the OPLS2005 force field. During minimization, the phos-
photyrosine and all atoms of the protein were held fixed. Two bound
conformations of 50 were obtained and subjected to molecular
dynamics simulations, during which the phosphotyrosine and all
protein residues were held fixed except for those within a 10 Å
sphere around the ligand. A total of 100 structures were saved for
FITC-[3-aminomethylbenzoyl]-pTyr-Gln-Gly-Leu-Ser-amide
(23). H NMR (400 MHz, DMSO-d6): δ 10.11 (s, 1H), 10.07 (s,
1
1H), 8.53 (d, J ) 8 Hz, 1H), 8.29 (d, J ) 8 Hz, 1H), 8.25 (s, 1H),
8.15 (t, J ) 8 Hz, 1H), 8.00 (d, 1H), 7.86 (d, 1H), 7.76-7.79 (m,
3H), 7.41 (d, J ) 8 Hz, 2H), 7.31 (d, J ) 8 Hz, 2H), 7.25 (br, 1H),
7.19 (d, J ) 4 Hz, 2H), 7.07 (s, 1H), 6.76 (br, 1H), 6.68 (d, J ) 4
Hz, 2H), 6.58 (s, 1H), 6.55 (d, J ) 4 Hz, 2H), 4.84 (d, J ) 4 Hz,
2H), 4.72-4.66 (m, 1H), 4.38-4.25 (m, 2H), 4.17 (dd, J1 ) 4 Hz,
J2 ) 8 Hz, 1H), 3.77 (d, 1H), 3.73 (d, 1H), 3.60 (t, J ) 4 Hz, 2H),
3.51 (s, 2H), 3.13 (d, J ) 4 Hz, 1H), 2.96 (t, J ) 16 Hz, 1H),
2.66-2.67 (m, 1H), 2.33-2.32 (m, 1H), 2.14 (t, J ) 8 Hz, 2H),
1.93-1.90 (m, 1H), 1.82-1.77 (m, 1H), 1.62-1.57 (m, 1H), 1.48
(br, 1H), 0.87 (d, J ) 4 Hz, 3H), 0.84 (d, J ) 2 Hz, 3H). MALDI-
MS calcd for C54H58N9O17PS: 1168.13; found 1168.71.
Peptides Bearing C-Terminal N-Alkenylamides (31-65).
Peptides bearing C-terminal N-alkenylamides were prepared by
adding the appropriate N-alkenylamine (10 equiv) and NaBH(OAc)3
(10 equiv) to a suspension of 4-(4-formyl)-3-methoxyphenoxy)bu-

1618 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 6
Choi et al.
(10) Garcia-Echeverria, C. Inhibitors of signaling interfaces: targeting Src
homology 2 domains in drug discovery. Protein Tyrosine Kinases
2006, 31–52.
(11) Burke, T. R., Jr. Development of Grb2 SH2 domain signaling
antagonists: a potential new class of antiproliferative agents. Int. J.
Pept. Res. Ther. 2006, 12, 33–48.
(12) Batzer, A. G.; Rotin, D.; Urena, J. M.; Skolnik, E. Y.; Schlessinger,
J. Hierarchy of binding sites for Grb2 and Shc on the epidermal growth
factor receptor. Mol. Cell. Biol. 1994, 14, 5192–201.
(13) Zhou, M. M.; Harlan, J. E.; Wade, W. S.; Crosby, S.; Ravichandran,
K. S.; Burakoff, S. J.; Fesik, S. W. Binding affinities of tyrosine-
phosphorylated peptides to the COOH-terminal SH2 and NH2-terminal
phosphotyrosine binding domains of Shc. J. Biol. Chem. 1995, 270,
31119–31123.
(14) Pelicci, G.; Lanfrancone, L.; Grignani, F.; McGlade, J.; Cavallo, F.;
Forni, G.; Nicoletti, I.; Grignani, F.; Pawson, T.; Pelicci, P. G. A novel
transforming protein (SHC) with an SH2 domain is implicated in
mitogenic signal transduction. Cell 1992, 70, 93–104.
(15) Ravichandran, K. S. Signaling via Shc family adapter proteins.
Oncogene 2001, 20, 6322–6330.
(16) Saucier, C.; Khoury, H.; Lai, K. M. V.; Peschard, P.; Dankort, D.; Naujokas,
M. A.; Holash, J.; Yancopoulos, G. D.; Muller, W. J.; Pawson, T.; Park, M.
The Shc adaptor protein is critical for VEGF induction by Met/HGF and
ErbB2 receptors and for early onset of tumor angiogenesis. Proc. Natl. Acad.
Sci. U.S.A. 2004, 101, 2345–2350.
(17) Saucier, C.; Papavasiliou, V.; Palazzo, A.; Naujokas, M. A.; Kremer, R.; Park,
M. Use of signal specific receptor tyrosine kinase oncoproteins reveals that
pathways downstream from Grb2 or Shc are sufficient for cell transformation
and metastasis. Oncogene 2002, 21, 1800–1811.
tyryl-NovaGel HL resin (Novabiochem, Inc.) in dry 1,2-dichloro-
ethane/trimethyl orthoformate (2:1) solution, followed by mixing
at room temperature (12 h). The resin was washed successively
with DMF, 10% i-Pr2NEt/DMF, and DMF. Sequential coupling of
amino acids was then achieved according to the general synthetic
procedures outlined above. For peptides bearing NR-alkenyl or NR-
alkyl residues, introduction of NR-alkenyl and NR-alkyl groups was
conducted on solid-phase resin by reaction of the corresponding
N-terminally deprotected peptide with the appropriate alkenyl or
alkyl bromide (1.2 equiv) and powdered K2CO3 in anhydrous DMF
at room temperature (overnight). Resins were washed with DMF
(2×), 50% aqueous MeOH (5×), and CH2Cl2 (3×), followed by
coupling of remaining amino acid residues. Cleavage of final
peptides from the resin afforded side-chain-deprotected peptide
products, which were purified by reverse phase preparative HPLC.
Ac-pTyr-Gln-Gly(N-allyl)-Leu-NH(homoallyl) (44). 1H NMR
(400 MHz, DMSO-d6): δ 8.24 (dd, J ) 8, 12 Hz, 1H), 7.97-7.91
(m, 2H), 7.20 (d, 1H), 7.09-7.05 (m, 2H), 6.98 (d, J ) 8 Hz, 2H),
5.75-5.56 (m, 2H), 5.16-5.08 (m, 1H), 5.04-5.01 (m, 1H),
4.96-4.91 (m, 1H), 4.65-4.61 (m, 1H), 4.47-4.44 (m, 2H),
4.18-4.16 (m, 2H), 4.06-4.02 (m, 2H), 3.63-3.55 (m, 2H), 3.12
(s, 1H), 3.10-2.99 (m, 1H), 2.65-2.62 (m, 1H), 2.04-1.98 (m,
1H), 1.73 (s, 3H), 1.53-1.46 (m, 1H), 1.40-1.38 (m, 2H), 0.82
(d, J ) 4 Hz, 3H), 0.77 (d, J ) 8 Hz, 3H). MALDI-MS calcd for
C31H47N6O10P: 694.71; found 717.25.
Ac-pTyr-Gln-Gly(N-ethyl)-Leu-amide (46). 1H NMR (400
MHz, DMSO-d6): δ 7.05 (d, J ) 8 Hz, 2H), 6.98 (d, J ) 8 Hz, 2H),
6.91 (br, 2H), 4.65-4.64 (m, 1H), 4.47-4.11 (q, J ) 8 Hz, 2H),
4.09-4.07 (m, 1H), 3.68-3.62 (m, 1H), 3.55-3.52 (m, 1H), 3.26-3.19
(m, 2H), 2.85-2.83 (m, 1H), 2.65-2.62 (m, 2H), 2.33-2.32 (m, 1H),
2.08-2.06 (m, 2H), 1.93 (d, J ) 8 Hz, 1H), 1.76 (d, 2H), 1.74 (s,
3H), 1.55-1.53 (m, 1H), 1.07 (t, J ) 8 Hz, 3H), 0.88-0.87 (m, 1H),
0.84 (d, J ) 4 Hz, 3H), 0.78 (d, J ) 4 Hz, 3H). MALDI-MS calcd
for C26H41N6O10P: 628.61; found 650.52.
(18) Ursini-Siegel, J.; Hardy, W. R.; Zuo, D.; Lam, S. H. L.; Sanguin-
Gendreau, V.; Cardiff, R. D.; Pawson, T.; Muller, W. J. ShcA
signalling is essential for tumour progression in mouse models of
human breast cancer. EMBO J. 2008, 27, 910–920.
(19) Zhou, M.-M.; Meadows, R. P.; Logan, T. M.; Yoon, H. S.; Wade,
W. S.; Ravichandran, K. S.; Burakoff, S. J.; Fesik, S. W. Solution
structure of the Shc SH2 domain complexed with a tyrosine-
phosphorylated peptide from the T-cell receptor. Proc. Natl. Acad.
Sci. U.S.A. 1995, 92 (17), 7784–7788.
(20) Protein Data Bank code 1TCE.
(21) See NoVabiochem Letters, January 2004 (catalog number 04-12-3911).
(22) Kertesz, A.; Varadi, G.; Toth, G. K.; Fajka-Boja, R.; Monostori, E.;
Sarmay, G. Optimization of the cellular import of functionally active
SH2-domain-interacting phosphopeptides. Cell. Mol. Life Sci. 2006,
63, 2682–2693.
(23) Garcia-Echeverria, C.; Jiang, L.; Ramsey, T. M.; Sharma, S. K.; Chen,
Y. N. P. A new Antennapedia-derived vector for intracellular delivery
of exogenous compounds. Bioorg. Med. Chem. Lett. 2001, 11, 1363–
1366.
(24) Blackwell, H. E.; Sadowsky, J. D.; Howard, R. J.; Sampson, J. N.;
Chao, J. A.; Steinmetz, W. E.; O’Leary, D. J.; Grubbs, R. H. Ring-
closing metathesis of olefinic peptides: design, synthesis, and structural
characterization of macrocyclic helical peptides. J. Org. Chem. 2001,
66, 5291–5302.
(25) Oishi, S.; Shi, Z.-D.; Worthy, K. M.; Bindu, L. K.; Fisher, R. J.; Burke,
T. R., Jr. Ring-closing metathesis of allylglycines with ꢁ-vinyl-
substituted residues as an approach to novel macrocyclic ꢁ-bend
mimetics. ChemBioChem 2005, 6, 668–674.
(26) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Synthesis and activity
of a new generation of ruthenium-based olefin metathesis catalysts
coordinated with 1,3-dimesityl-4,5 dihydroimidazol-2-ylidene ligands.
Org. Lett. 1999, 1, 953–956.
(27) Mikol, V.; Baumann, G.; Zurini, M. G. M.; Hommel, U. Crystal
structure of the SH2 domain from the adaptor protein SHC: a model
for peptide binding based on X-ray and NMR data. J. Mol. Biol. 1995,
254, 86–95.
(28) Oishi, S.; Karki, R. G.; Shi, Z.-D.; Worthy, K. M.; Bindu, L.; Chertov,
O.; Esposito, D.; Frank, P.; Gillette, W. K.; Maderia, M.; Hartley, J.;
Nicklaus, M. C.; Barchi, J. J., Jr.; Fisher, R. J.; Burke, T. R., Jr.
Evaluation of macrocyclic Grb2 SH2 domain-binding peptide mimetics
prepared by ring-closing metathesis of C-terminal allylglycines with
an N-terminal beta-vinyl-substituted phosphotyrosyl mimetic. Bioorg.
Med. Chem. 2005, 13, 2431–2438.
(29) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.;
Kluwer Academic/Plenum Publishers: New York, 1999.
(30) Boor, C. D. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer
Academic/Plenum Publishers: New York, 1999.
Acknowledgment. Appreciation is expressed to Dr. Oleg
Chertov for purification of Shc SH2 domain protein. The work
was supported in part by the Intramural Research Program of
the NIH, Center for Cancer Research, NCIsFrederick, and the
National Cancer Institute, National Institutes of Health, under
Contract N01-CO-12400.
Supporting Information Available: Mass spectral data for
peptide products. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Wells, J. A.; McClendon, C. L. Reaching for high-hanging fruit in drug
discovery at protein-protein interfaces. Nature 2007, 450, 1001–1009.
(2) Bradshaw, J. M.; Waksman, G. Molecular recognition by SH2
domains. AdV. Protein Chem. 2003, 61, 161–210.
(3) Pawson, T. Specificity in signal transduction: from phosphotyrosine-
SH2 domain interactions to complex cellular systems. Cell 2004, 116,
191–203.
(4) Pawson, T.; Gish, G. D.; Nash, P. The SH2 domain: a prototype for
protein interaction modules. Modular Protein Domains 2005, 5–36.
(5) Ladbury, J. E. Protein-protein recognition in phosphotyrosine-
mediated intracellular signaling. Protein ReV. 2005, 3, 165–184.
(6) Machida, K.; Mayer, B. J. The SH2 domain: versatile signaling module
and pharmaceutical target. Biochim. Biophys. Acta, Proteins Proteom-
ics 2005, 1747, 1–25.
(7) Broadbridge, R. J.; Sharma, R. P. The Src homology-2 domains (SH2
domains) of the protein tyrosine kinase p56lck: structure, mechanism
and drug design. Curr. Drug Targets 2000, 1, 365–386.
(8) Shakespeare, W. C. SH2 domain inhibition: a problem solved. Curr.
Opin. Chem. Biol. 2001, 5, 409–415.
(9) Metcalf, C. A., III; Sawyer, T. Src homology-2 domains and structure-
based, small-molecule library approaches to drug discovery. Drug
DiscoVery Strategies Methods 2004, 23–59.
JM800789H