Cyclic Peptides Incorporating Phosphotyrosine Mimetics as Potent and Specific Inhibitors of the Grb7 Breast Cancer Target

Article abstract of DOI:10.1021/acs.jmedchem.5b00609

The Grb7 adaptor protein is a therapeutic target for both TNBC and HER2+ breast cancer. A nonphosphorylated cyclic peptide 1 (known as G7-18NATE) inhibits Grb7 via targeting the Grb7-SH2 domain, but requires the presence of phosphate ions for both affinit

Full text of DOI:10.1021/acs.jmedchem.5b00609

Article
pubs.acs.org/jmc
Cyclic Peptides Incorporating Phosphotyrosine Mimetics as Potent
and Specific Inhibitors of the Grb7 Breast Cancer Target
Gabrielle M. Watson,† Menachem J. Gunzburg,† Nigus D. Ambaye,† William A. H. Lucas,†
Daouda A. Traore,† Ketav Kulkarni,†,‡ Katie M. Cergol,§ Richard J. Payne,§ Santosh Panjikar,†,∥
Stephanie C. Pero,⊥ Patrick Perlmutter,‡ Matthew C. J. Wilce,† and Jacqueline A. Wilce*,†
‡
†Department of Biochemistry and Molecular Biology and School of Chemistry, Monash University, Melbourne, Victoria 3800,
Australia
§School of Chemistry, The University of Sydney, Sydney, New South Wales 2006, Australia
∥Australian Synchrotron, 800 Blackburn Road, Clayton, Victoria 3168, Australia
⊥Department of Surgery and Vermont Cancer Center, University of Vermont, Burlington, Vermont 05401, United States
S
* Supporting Information
ABSTRACT: The Grb7 adaptor protein is a therapeutic
target for both TNBC and HER2+ breast cancer. A
nonphosphorylated cyclic peptide 1 (known as G7−
18NATE) inhibits Grb7 via targeting the Grb7-SH2 domain,
but requires the presence of phosphate ions for both affinity
and specificity. Here we report the discovery of malonate
bound in the phosphotyrosine binding pocket of the apo-
Grb7-SH2 structure. Based on this, we carried out the rational
design and synthesis of two analogues of peptide 1 that
incorporate carboxymethylphenylalanine (cmF) and carbox-
yphenylalanine (cF) as mimics of phosphotyrosine (pY).
Binding studies using SPR confirmed that affinity for Grb7-
SH2 domain is improved up to 9-fold over peptide 1 under physiological phosphate conditions (KD = 2.1−5.7 μM) and that
binding is specific for Grb7-SH2 over closely related domains (low or no detectable binding to Grb2-SH2 and Grb10-SH2). X-
ray crystallographic structural analysis of the analogue bearing a cmF moiety in complex with Grb7-SH2 has identified the precise
contacts conferred by the pY mimic that underpin this improved molecular interaction. Together this study identifies and
characterizes the tightest specific inhibitor of Grb7 to date, representing a significant development toward a new Grb7-targeted
therapeutic.
Furthermore, in a mouse model of pancreatic cancer, Grb7
inhibition reduced cancer cell metastasis.10 Clinical inves-
tigations have shown that Grb7 overexpression is significantly
associated with a higher clinical stage and larger tumor size for
patients with breast cancer, and is a significant predictor for
reduced cancer-free periods.11 Thus, Grb7, which acts at the
nexus of both growth and migratory signaling pathways, has
been identified as a therapeutic target for the treatment of
several cancer types including both HER2+ and TNBC.3,12
Grb7 is a 532 amino acid multidomain protein possessing an
N-terminal proline-rich domain, a Ras-associating-like (RA)
domain, a pleckstrin homology (PH) domain, and a C-terminal
src-homology 2 (SH2) domain. It is through the C-terminal
SH2 domain that Grb7 interacts with its upstream receptor
tyrosine kinases and other signaling molecules including HER2,
HER3, EGFR, and FAK, allowing Grb7 to propagate
downstream signaling.8,13−15 It is the SH2 domain, therefore,
INTRODUCTION
■
Tumor development and progression into a metastatic and
malignant state involves deregulation of signaling pathways
governing cell proliferation, migration, and survival.1 Key
signaling proteins that are aberrantly overexpressed or act at
critical junctions in cancer are therefore of interest for their
potential as therapeutic targets. The multiadaptor protein Grb7
(growth factor receptor bound protein 7) operates via distinct
signaling pathways to induce cell proliferation and migration in
the progression of a variety of cancer types including the breast
cancer subclasses HER2+ and triple negative breast cancer
(TNBC).2,3 In HER2+ cell lines, Grb7 is co-overexpressed with
HER2, the well-known predictor for a poor prognosis in breast
cancer patients, with Grb7 facilitating aberrant proliferative
signaling.4,5 The removal or inhibition of Grb7 has been shown
to reduce breast cancer cell viability and increase the activity of
anti-HER2 cancer therapeutics.6,7 Grb7 has also been shown to
play an important role in the migratory potential of cancer
cells.8 In both HER2+ and TNBC cell lines, inhibition of Grb7
impairs cell migration and invasion as well as colony growth.3,9
Received: April 20, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 1. Summary of Crystallographic Information
data collection
Apo Grb7-SH2
0.954
Grb7-SH2: 2
wavelength
0.954
space group
P63
P212121
unit cell dimensions
a, b, c (Å)
83.99 83.99, 75.95
90, 90, 120
55.48, 68.89, 77.04
90, 90, 90
α, β, γ (deg)
Resolution (Å)
28.17−1.80 (1.9−1.8)
10.6 (54.9)
43.21−2.55 (2.641−2.55)
11.07 (65.92)
11.71 (3.23)
a
Rmerge (%)
I/σI
16.7 (4.9)
unique reflections measured
completeness (%)
multiplicity
28306 (4125)
100 (100)
10038 (996)
99.53 (99.80)
8.9 (9.3)
11.3 (11.3)
Refinement
Rwork (%)
17.60 (20.49)
20.63 (22.52)
21.17 (28.31)
25.81 (34.72)
Rfree (%)
No. of atoms
protein
1809
7
1810
50
ligand (MLA)
solvent
Mean B-factors (Å2)
130
29
protein
20.70
15.90
26.60
45.80
52.80
38.70
ligand (MLA)
solvent
RMSDs
bond lengths (Å)
bond angles (deg)
Ramachandran plot (%)
favored regions
0.007
1.27
0.010
1.27
98.7
0.9
98.0
0
allowed regions
a
Rmerge = Σhkl Σi|Ii(hkl) - ⟨I(hkl)⟩|/Σhkl ΣiIi(hkl). where Ii(hkl) is the ith intensity measurement of reflection hkl, ⟨I(hkl)⟩ its average. Values given in
brackets are for the high resolution shell.
that has been the target for inhibitor development to date.16,17
SH2 domains are ∼100 amino acid domains that recognize
phosphotyrosine (pY) residues of upstream binding partners
with surrounding amino acid residues contributing to substrate
specificity.18 In the case of the Grb7-SH2 domain (residues
416−532), sequences with a pYXN motif that occur in a turn
conformation are preferentially targeted.19 This feature guided
the development of a nonphosphorylated cyclic peptide 1
(termed G7−18NATE), that targets the SH2 domain of Grb7
over closely related Grb2- and Grb14-SH2 domains.20 The 11-
residue peptide (WFEGYDNTFPC), cyclized via a thioether
bond from the N-terminus to the C-terminal thiol side chain of
cysteine, binds the Grb7-SH2 domain with micromolar affinity,
with a reported equilibrium dissociation constant (KD) of 347−
4.1 μM depending upon the buffer conditions utilized.21−23
Promisingly, in cellular assays, peptide 1 conjugated to the cell
permeability sequence Penetratin, reduced cell migration in
both HER2+ and TNBC cell lines, and showed co-operative
effects with the anticancer therapeutics doxorubicin and
Herceptin in HER2+ cell lines.2,3,7 The cell permeable peptide
1 has also been tested in a mouse model of pancreatic cancer,
resulting in reduction in the number and size of pancreatic
tumors.10 Hence, peptide 1 is a successful peptide for inhibiting
Grb7 in breast cancer assays, and a promising starting point for
therapeutics targeting TNBC and HER2+ breast cancer.
concentrations of phosphate (50 mM).24 In the absence of
phosphate, the affinity of peptide 1 for the Grb7-SH2 domain
was reduced to KD > 200 μM and selectivity for Grb7-SH2 over
Grb2-, Grb10-, and Grb14-SH2 was lost. It was speculated that
the phosphate from the buffer was acting at the pY binding
pocket of the Grb7-SH2 domain, to stabilize the BC loop at the
peptide binding site (note that the nomenclature used is as
derived by Eck et al. for SH2 domain secondary structure).25
Indeed, a previously reported structure of the apo-Grb7-SH2
domain revealed a sulfate ion bound at this site that appeared to
constrain the BC loop in an optimal conformation for peptide
binding.21 Thus, in order to improve the potential of peptide 1
for therapeutic use, we predicted that the addition of a
negatively charged or electronegative group appended to the
inhibitor peptide tyrosine would improve peptide 1 binding
selectivity and affinity under physiological phosphate con-
ditions. Such phosphotyrosine (pY) mimetics have been used
in the development of other SH2 domain inhibitors and are
favored over pY itself, since the phosphate group would likely
render the inhibitor peptide impermeable to cellular mem-
branes and unstable due to constitutive phosphatase activity
inside cells and therefore not efficacious in cellular or animal
studies.26
Here we report the design, synthesis, and characterization of
two peptide 1 analogues that incorporate carboxylic acid-based
phosphotyrosine (pY) mimetics (carboxymethylphenylalanine
(cmF) and carboxyphenylalanine (cF)) and show improved
binding of these analogues to Grb7-SH2 under physiological
conditions. Our choice of pY mimetic was guided by the high-
Upon further investigation of the dependence of the peptide
1 interaction with Grb7-SH2 on the buffer conditions used, the
low micromolar binding (KD = 4 μM) and specificity of peptide
1 for Grb7 was discovered to only occur in the presence of high
B
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Structure based rationale for a functionalized peptide 2 derivative. (A) Structural overlay of the 2.4 Å Grb7-SH2:peptide 1 domain complex
(PDB ID 2QMS: green sticks and purple cartoon, respectively) and the 1.8 Å apo-Grb7-SH2 structure (PDB ID 4WWQ: blue-white cartoon) with
malonic acid (orange sticks) in the pY binding pocket. The inset shows the 2Fo-Fc map surrounding malonic acid contoured at 2σ; (B−E) Detail of
the pY binding pocket: (B) Malonic acid bound Grb7-SH2, Chain A PDB ID: 4WWQ; (C) Sulfate bound Grb7-SH2, PDB: 2QMS; (D) Ligand free
Grb7-SH2, Chain B PDB: 4WWQ; (E) peptide 1 bound Grb7-SH2, PDB: 3PQZ. Key amino acids are shown as sticks and hydrogen bonds as
dashed lines.
resolution X-ray crystal structure of the Grb7-SH2 domain that
was crystallized in the presence of the dicarboxylic acid,
malonate. The malonate ion was discovered to reside precisely
in the Grb7-SH2 pY binding pocket making similar contacts to
surrounding residues as a sulfate or phosphate ion would. This
inspired the design and synthesis of cmF peptide analogue 2
and cF peptide analogue 3. The pY mimicking residues within
the cyclic peptides would be expected to provide a negative
charge for forming interactions in the pY binding pocket at
physiological pH, but would neither interfere with the ability of
the peptide to cross cell membranes nor compromise in vivo
stability of the analogue.
We demonstrate that peptides 2 and 3 indeed bind with
enhanced affinity to the Grb7-SH2 domain compared with
peptide 1 under physiological phosphate concentrations, and
that increased phosphate levels compete with the peptides for
Grb7-SH2 domain binding. We also demonstrate that they
retain their preferential binding to the Grb7-SH2 domain
compared to the SH2 domains of the closely related Grb2 and
Grb10. We furthermore report the X-ray crystal structure of
cmF bearing analogue 2 in complex with the Grb7-SH2
domain. This demonstrates the binding mode of peptide 2 with
the target and, in particular, the way in which the cmF group
extends into the phosphotyrosine binding cleft forming
additional contacts with core binding pocket residues. The
cmF and cF containing peptides 2 and 3 thus represent the
highest affinity inhibitors of Grb7 to date that possess
specificity for Grb7 under physiological conditions. Cell
permeable forms of these peptides will provide both a powerful
tool for studies of Grb7 inhibition in cells and underpin future
steps toward the development of a new therapeutic targeted to
Grb7 for the treatment of HER2+ and TNBC.
domain, we grew high quality crystals of the apoprotein for
structure determination using crystallization conditions that
included the dicarboxylic acid malonate. The protein
crystallization and data collection statistics have been reported
previously.27 The apo-Grb7-SH2 domain crystal structure was
solved to 1.8 Å using molecular replacement with PDB ID
2QMS as the search model. The refinement statistics are
provided in Table 1 and the coordinates are deposited in the
Protein database (PDB ID: 4WWQ). Interestingly, the
asymmetric unit contains two chains of Grb7-SH2 (residues
421−532 in chain A and 415−532 in chain B visible in the
electron density): chain A which contains malonate in the
phosphotyrosine binding site, and chain B that contains no
ligand and is thus a truly “apo” structure.
The chains both adopt the typical SH2 domain fold, with
αββββα topology (Figure 1A). Each chain is part of a
noncovalent dimer with a chain from an adjacent asymmetric
unit (chain A from one asymmetric unit to chain B of another).
The noncovalent dimer forms through an interface centered on
F511 (αB8) as has also been observed in all other crystal
structures of Grb7 family member SH2 domains and has also
been shown to form in solution.28−30 Interestingly, chains A
and B within the asymmetric unit are also covalently linked
through a disulfide bond that occurs between C526 in chain A
and C527 in chain B. This occurrence is assumed to be a
crystallization artifact and not physiologically relevant, since the
cellular environment is reducing, but it may have played a role
in stabilizing the crystal and contributed to the quality of the
diffraction. It has not been seen previously in crystal forms of
the Grb7-SH2 domain.
The presence of malonate is an outcome of the crystallization
condition that contained tacsimate, and it is unsurprising to
observe this anion accommodated in the positively charged pY
binding pocket. More surprising is the absence of this ligand in
the second chain; however, the occurrence of ligand-bound
alongside ligand-free protein chains in crystal structures is the
result of the protein packing arrangement in which the ligand at
RESULTS
■
Apo-Grb7-SH2 Domain Crystal Structures Reveals
Both a Bound Malonate as well as a Truly “apo” SH2
Domain. In the course of our investigation of the Grb7-SH2
C
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. Chemical structures and synthesis of peptide 2. Shown is the synthetic pathway used for the preparation of the cmF containing peptide 2.
The suitably protected carboxymethylphenylalanine building block (4), for incorporation into the peptide target, is highlighted in blue. Also shown,
in the inset, are the chemical structures of peptide 1 and the cF containing peptide 3.
one binding site partially occludes the binding site of the
adjacent molecule, an occurrence that has been observed
previously.31 Thus, as well as providing a higher resolution
structure for apo-Grb7-SH2 than previously achieved (PDB ID:
2QMS; 2.1 Å resolution) this structure reveals the way in which
a carboxylic acid group can be bound in the pY binding pocket
of the Grb7-SH2 domain.
Malonate Mimics Phosphate in the pY Binding Pocket
of the Grb7-SH2 Domain. In chain A of the apo-Grb7-SH2
model, clear electron density allows malonate to be placed
D
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 3. SPR binding analysis of peptides 1, 2, and 3 binding to Grb7-, Grb2-, and Grb10-SH2 domains. (A) Sensorgrams of peptide 1 (bottom
left), peptide 2 (top left), and peptide 3 (top right) at varying concentrations binding the Grb7-SH2 domain at 1 mM phosphate concentration, and
the corresponding equilibrium binding curves (bottom right). (B) Equilibrium binding curves for peptide 2 (left) and peptide 3 (right) binding the
Grb7-SH2 domain under buffers containing different concentrations of phosphate. All buffers contained 150 mM NaCl and 1 mM DTT. (C)
Equilibrium binding curves for peptide 1 (right), peptide 2 (left), and peptide 3 (center) binding the SH2 domains of Grb7, Grb2, and Grb10 under
1 mM phosphate concentration. Sensorgrams for all experiments are supplied as Supporting Information.
unambiguously in the pY binding pocket (Figure 1A - inset).
The malonate forms ionic interactions and hydrogen bonds
with Arg458 (βB5) and Arg438 (αA2) guanidino side chains
and the backbone NH of Gln461 (BC1) (Figure 1B). This
appears to hold the BC loop (residues 460−466) in a rigid
position, similarly to the Grb7-SH2 domain solved with a
sulfate ion bound in the pY binding pocket (PDB ID: 2QMS;
the BC loop of the apo-Grb7 SH2 chain A and the BC loop of
the 2QMS chain A have a Cα atom RMSD of 0.358 Å.) (Figure
1C).21 In the 2QMS structure the sulfate was engaged in salt
bridge interactions with the guanadino side chains of Arg458
(βB5) and Arg462 (BC2), and H-bonding interactions with the
Gln461 backbone NH (BC1) and also the Ser460 side chain
hydroxyl (βB7). The fact that Arg462 (BC2) is observed to
engage the sulfate, but not the malonate ion, is likely an artifact
of the malonate-containing crystal form with Arg438 (αA2) of a
symmetry-related chain B forming salt bridge interactions with
the second carboxylic acid of the malonate instead of Arg462
(BC2). Overall, although there are some differences in the
amino acids involved in the interactions, both of these anionic
ligands are able to hold the BC loop in a position that is
understood to represent the ligand binding conformation.32
E
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
pY Binding Site of the apo-Grb7-SH2 Domain Adopts
an “Open” Conformation. In chain B of the apo-Grb7-SH2
domain structure the BC loop adopts an alternative
conformation (Figure 1D). With an empty pY binding pocket,
the side chains of Arg458 (βB5), Arg462 (BC2), and Ser460
(βB7) are all directed into the solvent, and Arg438 (αA2)
forms salt bridge interactions with the symmetry related chain
A (explaining why the ligand binding site of chain B is empty).
As a consequence, the BC loop adopts a more open structure
with a Cα RMSD of 2.503 Å between the chain B BC loop
(residues 460−466) compared with the chain A BC loop. This
is similar to the peptide 1 bound form of Grb7-SH2 that
contains no anion in the pY site (PDB ID: 3PQZ; 2.4 Å
resolution), in which these pY binding pocket amino acid
residues are not engaged in hydrogen bonding except for
Arg438 (αA2) that forms a bond with the backbone carbonyl of
G4 of peptide 1. (Note that in discussing protein−ligand
interactions, the protein will be referred to in three-letter code
and the peptide ligand in one-letter code to distinguish them).
In this structure the BC loop is also in an “open” configuration
and displays some disorder with Asn463 (BC3) not visible in
the electron density (Figure 1E).
cmF and cF residues Are Predicted as pY Mimetics for
Grb7-SH2 Inhibition. Thus, the current Grb7-SH2 domain
structure reveals that an anionic ligand can form several
interactions in the pY binding site. We therefore predicted that
either a cmF or cF residue could serve to enhance the binding
of peptide 1 to the Grb7-SH2 domain and this designed
peptides 2 and 3 (Figure 2). CmF has been previously observed
to enhance SH2 domain-binding affinity in the case of a Grb2-
targeted peptide and cF has also been used to enhance binding
to an SH2 domain at the Y+3 position.33,34 From our structural
analysis of the separately determined position of peptide 1 Y5
and malonate in the pY binding site of the Grb7-SH2 domain
we can see that malonate is positioned 2.5 Å beyond the Y5
hydroxyl oxygen at an approximate angle of 130° (as depicted
in the superposed structures shown in Figure 1A). Thus, a cmF
could mimic (or a cF closely mimic) this spatial relationship
and potentially enhance the binding of peptide 1.
Synthesis of Target Cyclic Peptides 2 and 3. Based
upon this structural information a suitably protected cmF
building block, Fmoc-cmF(OMe)-OH 4, was synthesized that
could be directly incorporated into Fmoc-solid-phase peptide
synthesis to serve as a pY mimetic. Specifically, the building
block was incorporated into peptide 1 at the Y5 position
resulting in target cyclic peptide 2. In this target the thioether
linkage and remaining 10 amino acids were unchanged to allow
for direct comparison with peptide 1. Synthesis of the target
amino acid Fmoc-cmF(OMe)-OH 4, whereby the side chain
carboxylic acid was protected as the corresponding methyl
ester, was carried out from Boc-Tyr-OtBu using slight
modifications to a method previously reported by Tilley and
co-workers.35 Target cyclic peptide 2 was then prepared using
Fmoc-SPPS on a Rink amide resin and subject to N-terminal
chloroacetylation after the removal of the N-terminal Fmoc
group (Figure 2). Resin bound hexapeptide 5 was first
assembled using an excess of the commercially available
protected amino acids with HBTU and HOBt as the coupling
reagents and 20% piperidine in DMF for Fmoc-deprotection.
Fmoc-cmF(OMe)-OH 4 was next coupled using only a slight
excess of the building block and coupling reagents to afford
resin bound 6. From here, elongation of the desired peptide
sequence and N-terminal chloroacetylation provided the resin
bound peptide 7. After cleavage from the resin, thioether
cyclization between the N-terminal chloroacetyl moiety and the
C-terminal thiol was achieved in aqueous basic conditions to
produce peptide 8 and the methyl ester side chain of the cmF
residue was gently saponified using dilute aqueous sodium
hydroxide. The target cyclic peptide 2 was purified by reverse-
phase HPLC and produced in a 7% yield based on the original
resin loading. The peptide was shown to possess suitable purity
(>95%) by LC-MS analysis.
Cyclic peptide 3, bearing a cF residue in place of the cmF
present in 2, was produced using solid-phase peptide synthesis
by Wuxi Nordisk Biotech to >95% purity based on LC-MS
analysis.
Target Cyclic Peptides 2 and 3 Show Enhanced
Binding to the Grb7-SH2 Domain under Physiological
Phosphate Conditions. Surface plasmon resonance (SPR)
was utilized to measure differences in affinity for the Grb7-SH2
domain between peptide 1 and the target cyclic peptides 2 and
3. GST-fused Grb7-SH2 domain was captured on the chip
surface using anti-GST antibodies, as previously established for
measuring Grb7-SH2:peptide interactions.22 Figure 3A shows
that peptides 1, 2, and 3 each rapidly reach GST-Grb7-SH2
binding equilibrium upon peptide injection. This allows the
response at equilibrium to be determined and binding curves
constructed for KD determination by fitting a single-site binding
model (Figure 3A - bottom right).
In a buffer containing 1 mM phosphate at pH 7.4, peptide 2
bound to Grb7-SH2 with an affinity of KD = 5.7 μM and
peptide 3 bound to Grb7-SH2 with an affinity of KD = 2.1 μM.
Under the same conditions peptide 1 bound at a KD = 18.1 μM
(Table 2). The substitution of cmF and cF for Y thus confer a
3-fold and 9-fold enhancement of binding over peptide 1,
respectively, under physiologically relevant phosphate con-
ditions.
Table 2. SPR Binding Studies to Investigate the Cyclic
Peptides 2 and 3: Grb7-SH2 Interaction
b
affinity dissociation constant (μM)
a
buffer
20 mM Tris
1
2
3
c
347
3.9 0.05 Not measured
31.1 0.07 6.57 0.02
50 mM NaPO4
350 mM NaPO4
2.9 0.004
2.6 0.01
77.3 0.8
Not measured
20 mM Tris, 1 mM NaPO4 18.1 0.1
5.7 0.03 2.1 0.006
a
b
All buffers contained 150 mM NaCl and 1 mM DTT. KD was
derived from fits to single-site saturation model. Errors are standard
c
errors arising from the fits. Reported by Gunzburg et al. (2012) under
the same experimental conditions.
Phosphate is Not Required for the Binding of Cyclic
Peptides 2 and 3 to the Grb7-SH2 Domain. We have
previously determined that peptide 1 binding to the Grb7-SH2
domain is enhanced by the presence of phosphate in the
buffer.24 This can be rationalized by the presence of a
phosphate ion in the pY binding site that enhances peptide
binding by stabilizing the BC loop. In the case of peptide 1,
binding to Grb7-SH2 was previously reported to range from KD
= 4.1 μM in 100 mM phosphate to 347 μM in the absence of
any phosphate.24 In the case of peptides 2 and 3 it was
anticipated that there would no longer be a requirement for
phosphate for high-affinity binding to the Grb7-SH2 domain.
Figure 3B shows a series of binding curves measured for
peptides 2 and 3 in a range of phosphate concentrations. As
F
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
expected, the highest affinity measurement for peptide 2
binding to Grb7-SH2 was observed in the absence of phosphate
in the buffer (KD = 3.9 μM) and, with increasing concentrations
of phosphate in the buffer the binding was reduced. The affinity
of peptide 2 was determined at KD = 31.1 μM in 50 mM
phosphate and KD = 77.3 μM in 350 mM phosphate-containing
buffer. Likewise, the affinity of peptide 3 was lowered to KD =
6.6 μM in 50 mM phosphate-containing buffer (Table 2; SPR
sensorgrams provided as Supporting Information). This was in
stark contrast with peptide 1 binding of KD = 2.9 and 2.6 μM at
high phosphate concentrations (Table 2; SPR sensorgrams
provided as Supporting Information). Thus, the cmF and cF
substitutions successfully enhance peptide binding to the Grb7-
SH2 domain, effectively acting as a substitute for phosphate in
the pY binding site.
deposited in the Protein database (PDB ID: 4X6S). The
asymmetric unit contains two chains of the Grb7-SH2 domain,
both in complex with cyclic peptide 2. The electron density for
the Grb7 chains encompasses residues 424−529 for chain A
and 423−529 for chain B. In the peptide binding site, clear
electron density is visible for the peptide ligand including
density showing the position of the cmF group (Figure 4A).
The Grb7-SH2 domain adopts the canonical αββββα SH2
domain topology (Figure 4A) with Cα RMSDs of 0.35 and 0.39
for chains A and B, respectively, compared to chain A of the
apo-Grb7 SH2 structure. The interchain covalent linkage
observed in the apo-Grb7 structure was not present in this
complex structure; but the typical Grb7 family dimerization
interface was present, centering on F511 (αB8) from each
protomer.
Cyclic Peptides 2 and 3 Are Specific for the Grb7-SH2
Domain over Other Closely Related SH2 Domains. SPR
was further utilized to determine whether peptides 2 and 3
retain specificity for the Grb7-SH2 domain over closely related
SH2 domains (Grb2-SH2 domain (34% identity) and Grb10-
SH2 domain (70% identity)). Using the same SPR method,
under physiological phosphate concentrations (1 mM phos-
phate), peptides 2 and 3 displayed low binding affinities for the
Grb2-SH2 domain (KD = 440 and 560 μM, respectively).
Peptide 3 also had low affinity binding (KD = 420 μM) for the
Grb10-SH2 domain, whereas binding measurements could not
be determined for peptide 2 (Figure 3C; Table 3). Peptide 1
Peptide 2 is positioned across the βD strand in both chains,
making contacts with residues from the BC, EF, and BG loops,
analogously to peptide 1 in the Grb7-SH2:peptide 1 complex
structure (PDB ID: 3PQZ; Figure 4B). Peptide 2 has a Cα
RMSD of 0.16 between the two chains in the asymmetric unit,
and 0.27 and 0.21 for chains A and B, respectively, compared to
peptide 1 chain L across the 11 amino acid residues.
The peptide 2 residues G4, D6, and N7 display the highest
level of similarity between the two peptides, with these residues
forming the main binding interface with the Grb7-SH2 domain
via electrostatic and van der Waals (vdW) interactions (Figure
4B). The F2 and F9 aromatic rings form hydrophobic
interactions with the Grb7-SH2 domain surface and the G4
backbone carbonyl forms H-bonding interactions with the
Arg438 (αA2) side chain that are equivalent to those seen in
the Grb7-SH2:peptide 1 complex structure. In contrast, the
flexible solvent exposed residues (W1, E3, T8, and C11) adopt
varied rotamer conformations compared with those in peptide
1.
Table 3. SPR Binding Studies to Determine Specificity of
Cyclic Peptides 1, 2, and 3
a
affinity dissociation constant (μM)
protein
1
2
3
Grb7-SH2
Grb2-SH2
18.1 0.1
200 70
5.7 0.03
440 70
2.1 0.01
560 40
As predicted, the cmF residue in peptide 2 probes the pY
binding pocket of Grb7 mediating additional interactions to
those made by peptide 1. The cmF group is oriented slightly
differently between the two chains with an approximate 11°
twist at the Cβ position. The average temperature factor of the
cmF residue in the two chains, 47 Å2 for chain A and 52 Å2 for
chain B, is higher than the average temperature factor for the
peptide chains, 43 Å2 for chain A and 49 Å2 for chain B,
suggesting the phosphotyrosine mimetic is relatively mobile
compared to the peptide as a whole. This shift between the two
cmF groups slightly alters the H-bond network with BC loop
residues in the pY binding pocket, but essentially the same
contacts are made in chains A and B (Figure 4C and D). The
cmF carboxylate O1 forms H-bond or vdW (where the distance
is slightly long for H-bond formation) contacts with the Ser460
(βB7) side chain and the backbone amides of Gln461 (BC1)
and Arg462 (BC2), similarly to the mode of interaction seen in
the sulfate containing Grb7-SH2 domain structure (PDB ID:
2QMS). The cmF O2 forms H-bond or vdW interactions with
residues at the base of the pY binding pocket including the
Arg438 (αA2) guanidinium side chain and bipartite interactions
with the Arg458 (βB5) guanidinium group. These are the same
amino acids that were observed to bind to malonate; however,
the cmF O2 interaction with Arg438 (αA2) is via the δNH
group (since the ωNH2 groups extends to the G4 backbone
carbonyl in peptide 2). In both chains the BC loop is
consistently held in the closed conformation for peptide
binding, similarly to the sulfate and malonate-bound Grb7-SH2
structures.
Grb10-SH2 No detectable binding No detectable binding 420 60
a
KD was derived from fits to single-site saturation model. Errors are
standard errors arising from the fits. SPR specificity experiments were
conducted in 20 mM Tris, 1 mM NaPO4, 150 mM NaCl, and 1 mM
DTT
also displayed low affinity for the Grb2-SH2 domain (KD of 200
μM) and no detectable binding for the Grb10-SH2 domain
(Figure 3C; Table 3). Thus, peptides 2 and 3 successfully
maintain specificity for Grb7-SH2 over closely related SH2
domains compared with peptide 1 under physiological
phosphate conditions.
Crystal Structure of the Grb7-SH2:Peptide 2 Complex
Reveals the Basis for Binding Affinity. To determine the
structural basis of the enhanced binding affinity of 2 and 3 for
the Grb7-SH2 domain, cocrystallization of the peptides with
the Grb7-SH2 domain was pursued. Only the Grb7-
SH2:peptide 2 complex gave rise to crystals suitable for
diffraction experiments. The best crystal obtained from the
PEG/Ion HT screen diffracted to 2.55 Å with the reservoir
solution containing 12% w/v PEG3350, 0.05 M sodium HEPES
and 1% w/v tryptone (parameters provided in Table 1). The
complex crystallized in the P212121 space group with unit cell
dimensions of a = 55.55, b = 68.96, c = 77.12, α = β = γ = 90
°C. Molecular replacement was used to determine initial phases
using a monomer of the apo-Grb7-SH2 domain as a search
model (PDB ID: 4WWQ). The data collection and refinement
statistics are provided in Table 1 and the coordinates are
G
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Structure of the Grb7-SH2:peptide 2 complex. (A) Structure of Grb7 SH2 (blue-white cartoon) in complex with peptide 2 (yellow sticks)
(PDB ID: 4X6S). 2Fo-Fc map surrounding peptide 2 shown contoured at 1.2σ. (B) Structural overlay of peptide 2 (yellow) and peptide 1 (green)
from PDB ID: 3PQZ. Grb7 SH2 shown as blue-white surface representation. (C−D) Close-up of the pY binding pocket of chain A and chain B. Key
amino acids are shown as sticks and hydrogen bonds as black dashed lines (or gray if close to being defined as a hydrogen bond).
The crystal structure also reveals that the basis for peptide 2
specificity for the Grb7-SH2 domain is the same as previously
reported for peptide 1.31 Critical interactions are made with the
nonconserved residue Leu481 (βD6) that is known to play a
central role in Grb7 binding specificity.36 The peptide 2 N7
side chain carbonyl and amine groups form bipartate hydrogen
bonds with Grb7 Leu481 backbone amino and carbonyl groups,
respectively. This positions the peptide such that the aromatic
rings of both cmF5 and F2 form van der Waals interactions
against the Leu481 side chain that could not occur in other
Grb-SH2 domains. Another interaction between peptide 2 and
Grb7 that may contribute to binding specificity is a hydrogen
bond between the P10 carbonyl group and the nonconserved
Grb7 Gln499 (EF4) side chain amine. In addition, due to the
subtle effects of shape complementarity, van der Waals
contributions to binding specificity are also likely to be at
play in the same way as for peptide 1. Overall, it can be seen
that the incorporation of the cmF group in peptide 2 does not
negatively impact on the molecular interactions that underlie
specificity of binding to the Grb7-SH2 domain.
Peptides 2 and 3 Represent Improved Inhibitors of
the Grb7-SH2 Domain. The cmF group has thus successfully
been utilized as a pY group replacement, effecting the
interactions with the Grb7-SH2 domain similar to those
previously observed for sulfate and malonate binding. Although
structural data has not yet been obtained for the Grb7-
SH2:peptide 3 complex, the cF group in peptide 3 likely makes
similar contacts in the pY binding site. As well as providing
enhanced binding affinity through direct electrostatic and
hydrogen bond contributions, it is possible that the BC loop
stabilization also contributes to peptide binding through subtle
adjustments to the peptide binding site. The structure helps to
explain the increased binding affinity of peptides 2 and 3 over
peptide 1 under physiological phosphate concentrations as well
as competition for peptide binding by phosphate at higher
concentrations.
DISCUSSION
■
The role of Grb7 in cancer progression through migratory and
proliferative signaling, and its overexpression in a myriad of
cancers, has highlighted Grb7 as a viable target for developing
novel anticancer agents.2,3,17 Initial success has been made with
the discovery of the cyclic nonphosphorylated peptide 1, with
cell-permeable forms inhibiting cell migration, invasion,
proliferation, and colony formation in HER2+ and TNBC
cell lines.2,3,9 To further the potential of peptide 1 in
therapeutic development, derivatives of this peptide need to
be designed, synthesized, and characterized.
The current study has identified two new cyclic derivatives,
peptides 2 and 3, with enhanced affinity for the Grb7-SH2
domain compared with peptide 1. The introduction of the cmF
and cF moieties at the Y5 position of peptide 1 was inspired by
the observation of a malonate positioned in the pY binding site
of the apo-Grb7-SH2 domain crystal structure. The target cyclic
peptides were successfully prepared via solid-phase peptide
synthesis and binding studies using SPR determined that
peptide 2 improves binding 3-fold and peptide 3 9-fold for the
Grb7-SH2 domain compared with peptide 1. Importantly, these
measurements are in the presence of 1 mM phosphate, which
reflects a physiological phosphate concentration.37 Further-
more, peptides 2 and 3 show greater than 75-fold higher
binding for the Grb7-SH2 domain over the closely related
Grb2- and Grb10-SH2 domains under the same conditions.
While these studies were not conducted on the full-length Grb
proteins, it is anticipated that this specificity will be retained in
the cellular environment. In studies conducted to test the
binding of peptides 1, 2, and 3 to full-length Grb7 (using the
same SPR strategy as for the SH2 domain alone) we were able
to confirm binding affinities in the micromolar range (peptides
2 and 3 binding with approximately 4-fold and 8-fold higher
affinity than peptide 1, respectively), though the instability of
full-length Grb7 over time prevented the determination of
accurate values for peptide binding (the data and a description
of these experiments are supplied as Supporting Information).
This shows, however, that the inhibitor peptides do bind to the
H
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Grb7 in the context of the whole protein and not just to the
isolated SH2 domain.
the Grb7-SH2 domain is thus likely to occur via other more
subtle interactions across the surface of the protein where it
makes contact across the βD strand, BG loop, and EF loop as
previously observed for peptide 1.31
Our structural studies confirm that the cmF group of peptide
2 reaches into the Grb7 pY-binding pocket mediating a
combination of interactions previously observed in the
malonate- and sulfate-bound Grb7-SH2 domain structures,
effectively holding the BC loop in a closed and favorable
conformation for peptide binding. The cmF is thus a successful
pY mimetic for Grb7-SH2 inhibitors. Structural data have not
been obtained for the Grb7-SH2:peptide 3 complex, so it is not
yet known how the cF group interacts in the pY-binding pocket
to effect the higher affinity interaction. In the case of the cF
group, it is shorter than cmF by one methylene group and
seemingly less ideal for placement in the binding site from
inspection of Y in peptide 1 and malonate when bound to the
Grb7-SH2 domain (Figure 1A). Nevertheless, this study has
shown it makes additional productive interactions with Grb7-
SH2 domain target and is also a successful pY mimetic for
Grb7-SH2 inhibitors.
The cmF group has been utilized as a pY mimetic in previous
studies of SH2 inhibition.33,38,39 Incorporation of the cmF
group to Src SH2 and Grb2-targeted peptides improved
binding >2-fold and from undetectable to low μM binding,
respectively, compared to Y alone.33,38,40 It is the case, however,
that cmF derivatives have reduced binding compared to their
pY peptide counterparts: 940-fold for Src, 50-fold for Grb2, and
467-fold for a p56lck directed cmF peptide.33,38,39 It has been
determined that the phosphate group can contribute up to half
the total binding energy to SH2 domain binding.41 Thus, the
cmF group only partially mimics the pY group as may be
expected for a single dicarboxylic acid with less capacity for H-
bond formation than a phosphate. A further increase in binding
affinity may be achieved with the use of a higher valency pY
mimetic or with the use of a protected phosphate group as
developed by McMurray et al.16 This study represents the first
time, to our knowledge, that the cF group has been
incorporated and tested as a pY mimetic for targeting an
SH2 domain binding pocket. The cF group has previously been
substituted for the Y+3 phenylalanine in Src targeting peptides.
This single substitution showed a 180-fold improvement in
inhibiting the substrate polyGlu4Tyr compared to the parent
phenylalanine containing peptide.34
The current study has demonstrated that peptides 2 and 3
maintain specificity for Grb7-SH2 under physiological con-
ditions. While the cF and cmF groups confer greater affinity for
the Grb7-SH2 domain, binding to Grb2 and Grb10 is only
marginally affected. Peptides 2 and 3, in fact, showed decreased
binding to Grb2-SH2 compared to peptide 1, though peptide 3
showed a detectable interaction with the Grb10-SH2 domain
where peptides 1 and 2 had shown no interaction. This is
similar to the effect observed upon addition of phosphate to the
buffer on peptide 1 binding. While peptide 1 binding to Grb7-
SH2 domain was greatly enhanced in 100 mM phosphate,
binding to Grb10-SH2 and Grb2-SH2 domain was either not
enhanced or only marginally enhanced.24 Grb10-SH2 possesses
a significant sequence identity of ∼70% with Grb7-SH2, and
although Grb2 possesses a low sequence identity of ∼34%, the
SH2 domain shares the same substrate as Grb7, the ErbB2
receptor, and thus recognizes the same pYXN interaction motif.
Furthermore, the amino acid side chains shown to be important
for binding to the cmF group (of Ser460 (βB7), Arg438 (αA2),
Arg458 (βB5)) are conserved between the Grb7 family
members and Grb2. Specificity of cyclic peptides 2 and 3 for
In the development of SH2 domain inhibitors it is an
achievement to develop a specific inhibitor suitable for cellular
studies due to the large number of SH2 domains that exist in
the cell with similar substrate interactions. Peptide 1 was the
first molecule to be discovered with specificity for Grb7, but its
affinity depends upon the presence of phosphate ions. Other
peptides and small molecules have been discovered that bind to
Grb7 but have not been shown to possess binding
specificity.42−44 This study thus reveals an important new
class of inhibitor that is specific for the Grb7-SH2 domain
under physiological conditions. Future work will focus on
improving the inhibitor affinity. With a KD = 2.1 μM binding
affinity for the Grb7-SH2 domain, cyclic peptide 3 represents
an improvement on peptide 1, but still higher affinities are
required for its therapeutic use. The structural information
obtained for the Grb7-SH2:peptide 2 complex will facilitate the
design of further modifications for improvement of this first
generation inhibitor of Grb7. These may include additional
hydrogen bonding functionalities and rigidification of the
peptide backbone as have previously shown to be successful in
peptide inhibitor design.26,45 The effective synthesis and
characterization of cyclic peptides 2 and 3 thus represents an
important step toward a potent Grb7-SH2 domain inhibitor
that will be used as a tool both for better understanding the
downstream effects of Grb7 inside cells and for future potential
therapeutic use in the treatment of HER2+ and TNBC.
METHODS
■
Protein Expression and Purification. The pGex2T plasmids
containing the SH2 domain inserts of Grb2 (encoding residues 58−
160), Grb7 (encoding residues 415−532) and Grb10 (encoding
residues 471−594) were kindly provided by Roger Daly. The SH2
domains were expressed as GST fusion proteins in Escherichia coli
strain BL21(DE3)pLysS and purified using glutathione affinity
chromatography and size exclusion chromatography as described
previously.24,30 To obtain Grb7-SH2 domain alone for crystallization
experiments, the GST tag was cleaved off with thrombin following
glutathione affinity chromatography, and further purified with cation
exchange and size exclusion chromatography as also described
previously.30 The GST alone, used as a control for SPR, was expressed
from the pGex2T plasmid (GE Healthcare) and purified similarly to
the GST-Grb proteins excluding size exclusion chromatography. The
final concentrations of GST, GST-Grb7-SH2, GST-Grb2-SH2, GST-
Grb10-SH2, and Grb7-SH2 were determined spectrophotometrically
at A280 using extinction coefficients of 42860, 51340, 58330, 49850,
and 8480 M−1 cm−1, respectively.46
Preparation of Fmoc-cmF(methyl ester)-OH. Fmoc-cmF-
(OMe)-OH 4 was synthesized from Boc-cmF(OMe)-OtBu that was
in turn prepared using a procedure previously reported by Tilley et
al.35 Briefly, Boc-cmF(OMe)-OtBu (108 mg, 0.27 mmol, 1 equiv) was
treated with an acidic cocktail comprising a 90:5:5 v/v/v mixture of
TFA:triisopropylsilane:H2O (2 mL) and allowed to stir at rt for 3 h
before removal of the solution in vacuo. The crude product was used in
the next step without further purification. A solution of the resulting
amino acid in a 2:1 v/v mixture of 1,4-dioxane/sat. aq. NaHCO3
solution (3 mL) was treated with Fmoc-OSu (102 mg, 0.3 mmol, 1.1
equiv) and allowed to stir at rt for 16 h. The resulting mixture was
poured onto water, and acidified to pH 1−2 with 1 M HCl, extracted
with EtOAc (3 times), dried (Na2SO4), and concentrated in vacuo.
The crude product was purified by column chromatography (1:1 v/v
EtOAc/hexanes +1% to 3% MeOH), to afford Fmoc-cmF(OMe)-OH
4 as a colorless oil (107 mg, 86%). 1H NMR (d6-acetone, 400 MHz) δ
I
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(ppm) 7.86 (d, J = 7.20 Hz, 2H), 7.70−7.63 (m, 2H), 7.45−7.20 (m,
8H), 6.72 (d, J = 8.34 Hz), 4.60−4.50 (m, 1H), 4.38−4.16 (m, 3H),
3.65−3.59 (m, 5H), 3.27 (dd, J = 4.78, 14.03 Hz, 1H), 3.06 (dd, J =
9.56, 14.03 Hz, 1H); 13C NMR (d6-acetone, 100 MHz) δ (ppm)
173.3, 172.3, 156.8, 144.9, 142.0, 137.0, 133.8, 130.2, 130.1, 128.5,
127.9, 126.2, 120.8, 67.2, 56.1, 51.9, 47.9, 40.9, 37.7. HRMS: (Mexpected
= 459.1682 g/mol, Mmeasured = 459.1682 g/mol).
carried out using COOT, PHENIX, and REFMAC5.51−53 Solvent
molecules were automatically added during refinement, and then
manually edited in COOT. The final apo-Grb7-SH2 model had an
overall Rwork of 17.60% and an Rfree of 20.63%.
The Grb7-SH2: peptide 2 complex was formed using 10 mg/mL
protein in a 1:1.5 molar ratio of protein:peptide. Crystals were
obtained using 2 μL hanging drops composed of 1:1 protein: reservoir
with the reservoir solution containing 0.05 M HEPES, pH 7.0, 1%
tryptone and 12% PEG3350 from the Hampton screen PEG/Ion HT
Screen. Harvestable crystals were soaked in 15% v/v glycerol/mother
liquor as a cryoprotectant and flash frozen in liquid nitrogen prior to
data collection held in a stream of nitrogen gas at 100 K. Diffraction
data were collected using the microfocus beamline (MX2) at the
Australian Synchrotron at a wavelength of 0.954 Å using an ADSC
Quantum 315r detector and BLU-ICE software for data acquisition.54
Images were recorded with an oscillation angle of 1° and exposure
time of 3 s. The diffraction images were integrated and scaled with the
software pipeline XIA2, with the data restricted to a high-resolution
limit of 2.55 Å.55 5% of the total reflections were randomly assigned as
an Rfree set and excluded from refinement. Phaser was used for
molecular replacement with one chain of the apo Grb7-SH2 domain
structure (PDB ID: 4WWQ) used as a search model.56 Subsequent
model building and refinement rounds occurred as for the apo Grb7
SH2 structure. The peptide 2 restraint files were generated using
phenix.elbow from SMILES strings.57 The final Grb7 SH2:peptide 1
model had an overall Rwork of 21.17% and an Rfree of 25.81%.
MOLPROBITY was used to assess the quality of the final
structures, and figures were generated using PyMOL.58 The data
collection, processing, and refinement statistics for both structures are
provided in Table 1.
Synthesis of Peptides 1, 2, and 3. Peptide 2 (cyclo-(CH2CO-
WFEGcmFDNTFPC)-CONH2) was synthesized on a 0.1 mmol scale
using standard Fmoc-chemistry on Rink amide resin (0.7 mmol/g).
Commercially available amino acids (3.1 equiv with respect to the
resin loading), HBTU (3.0 equiv with respect to the resin loading),
HOBt (3.0 equiv to resin loading), and DIPEA (4.5 equiv to resin
loading) were dissolved in DMF (3 mL) and added to the resin with
shaking for 45 min. These reactions were repeated twice to ensure
complete coupling. Fmoc-cmF(OMe)-OH 4 was only coupled once in
slight excess of 1.3 equiv to resin loading, along with HBTU (1.3 equiv
with respect to the resin loading), HOBt (1.3 equiv to resin loading),
and DIPEA (∼2 equiv to resin loading) in DMF (2 mL). After
removal of the terminal Fmoc protecting group on the resin-bound
peptide through the treatment with 20 vol % piperidine in DMF, the
resin was treated with chloroacetic anhydride (1 mmol) and DIPEA
(0.1 mmol) in DMF (2 mL) to afford a chloroacetyl-capped N-
terminus. The resin was subsequently washed with DMF, CH2Cl2, and
Et2O and dried. Cleavage was performed on 0.15 mmol of the resin by
treating the resin with a cleavage solution comprising distilled water
(2.5% v/v), triisopropylsilane (2.5% v/v), and ethanedithiol (0.5% v/
v) in TFA (9.45 mL) for 1.5 h. The TFA was then evaporated under a
stream of N2 and the peptide precipitated by addition of cold Et2O.
The precipitate was filtered and redissolved in H2O/CH3CN (1:1) for
lyophilization.
Binding Studies Using Surface Plasmon Resonance. Experi-
ments were conducted on a BIAcore T100 using BIAcore CM5 series
S sensor chips as previously reported.22 The immobilization buffer
consisted of 50 mM NaPO4, 150 mM NaCl, and 1 mM DTT (pH
7.4). Polyclonal anti-GST antibody was immobilized on the reference
and active flow cells using amine coupling of the antibody to the
surface of the chip (GE Healthcare GST capture kit). For this, the
sensor chip was activated with 1-ethyl-3-(3-(dimethylamino)propyl)-
carbodiimide hydrochloride and N-hydroxysuccinimide, then poly-
clonal anti-GST antibody (at 30 μg/mL) in 10 mM sodium acetate pH
5.0 injected over the chip surface at 5 μL/min for 7 min. The flow cells
were then blocked with 1 M ethanolamine, resulting in anti-GST
antibody immobilization levels between 2755 RU and 6692 RU.
GST alone was immobilized on the reference flow cell and the
GST-fusion proteins of Grb2-SH2, Grb7-SH2, and Grb10-SH2
immobilized on the active flow cells so that binding in the four flow
cells could be simultaneously assessed. GST, Grb7-SH2, Grb2-SH2,
and Grb10-SH2 (all at 0.7−0.9 μM in immobilization buffer) were
injected over the corresponding flow cells for 7 min at 5 μL/min with
486−1272 RU, 767−1521 RU, 800−972 RU, and 811−1065 RU
immobilized, respectively. Triplicate or duplicate samples of peptide
were injected for 60 s at 30 μL/min, with 300−600 s dissociation in a
range of buffers as described in the main text. The data were analyzed
using Scrubber2.0 (BioLogic Software, Campbell, ACT, Australia) and
SigmaPlot v 12.0 (Systat Software, Inc., Chicago, IL, USA) to
determine KD values.
Thioether formation was effected by dissolving the peptide (2 mg/
mL concentration) in 50 mM NH4HCO3 solution (made up in 1:1 v/v
CH3CN/H2O), at pH 8.0, for 1.5 h at room temperature. The cyclized
peptide was then purified using reverse-phase HPLC. Hydrolysis of the
methyl ester protecting group on the side chain of the cmF residue was
carried out by dissolving the peptide (1 mg/mL concentration) in 10−
11 mM NaOH solution (0.5 mL of CH3CN in 6.5 mL of H2O), at pH
8.0−9.0, for 30 h at room temperature. After final purification (to
≥95% purity based on LC) the peptide mass was verified by ion-trap
mass spectrometry (Mexpected (C69H81N14O20S)− = 1457.6; Mmeasured
(C69H81N14O20S)− = 1457.6; Mexpected (C69H81N14O20S)2− = 728.3;
Mmeasured (C69H81N14O20S)2− = 728.5).
The synthesis of cyclic peptide
3 (cyclo-(CH2CO-
WFEGcFDNTFPC)-CONH2) was carried out using a similar protocol
to that for peptide 2 by Wuxi Nordisk Biotech (China) to >95% purity
based on LC-MS. In this case, the Fmoc-protected cF residue (Fmoc-
p-carboxyphenylalanine(OtBu)-OH) was commercially available
(Chem-Impex). After purification using reverse-phase HPLC the
peptide mass was verified by ion-trap mass spectrometry (Mexpected
(C68H79N14O20S)− = 1443.6; Mmeasured (C68H79N14O20S)− = 1443.9;
Mexpected (C68H79N14O20S)2− = 721.8; Mmeasured (C68H79N14O20S)2−
=
721.6).
Peptide 1 (cyclo-(CH2CO-WFEGYDNTFPC)-amide) was pre-
pared as described previously.21 The final concentrations of peptides 1,
2, and 3 were determined spectrophotometrically at A280 using
extinction coefficients of 6990 M−1 cm−1 and 5809 M−1 cm−1,
respectively, with the extinction coefficient for the latter two calculated
using the DirectDetect (Millipore) system.
Protein Crystallization, X-ray Diffraction Data Collection,
and Structure Determination. Apo-Grb7-SH2 crystals were
prepared and data collected as previously described.27 The diffraction
images were indexed and integrated in the space group P63 using
IMOSFLM scaled with SCALA from the CCP4 suite and cut to a high
resolution limit of 1.80 Å.47−49 5% of the total reflections were
randomly assigned as an Rfree set using UNIQUEIFY (CCP4 suite)
and subsequently excluded from refinement. Molecular replacement
was performed using MOLREP with a monomer from the previous
Grb7-SH2 domain apo structure used as a search model (PDB ID:
2QMS).21,50 Rounds of structure refinement and model building were
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.5b00609.
Validation of compounds prepared for this study are
provided, as are SPR sensorgrams used for KD
determination. I. Supporting Information and Methods.
II. 1H and 13C NMR spectra of Fmoc-cmF(OMe)-OH 4.
III. MS for Fmoc-cmF(OMe)-OH 4. IV. LC chromato-
gram for peptide 2. V. MS for peptide 2. VI. LC
J
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(8) Han, D. C.; Guan, J. L. Association of focal adhesion kinase with
Grb7 and its role in cell migration. J. Biol. Chem. 1999, 274, 24425−30.
(9) Lim, R. C.; Price, J. T.; Wilce, J. A. Context-dependent role of
Grb7 in HER2+ve and triple-negative breast cancer cell lines. Breast
Cancer Res. Treat. 2014, 143, 593−603.
(10) Tanaka, S.; Pero, S. C.; Taguchi, K.; Shimada, M.; Mori, M.;
Krag, D. N.; Arii, S. Specific peptide ligand for Grb7 signal
transduction protein and pancreatic cancer metastasis. J. Natl. Cancer
Inst. 2006, 98, 491−8.
(11) Ramsey, B.; Bai, T.; Hanlon Newell, A.; Troxell, M.; Park, B.;
Olson, S.; Keenan, E.; Luoh, S. W. GRB7 protein over-expression and
clinical outcome in breast cancer. Breast Cancer Res. Treat. 2011, 127,
659−69.
(12) Nadler, Y.; Gonzalez, A. M.; Camp, R. L.; Rimm, D. L.; Kluger,
H. M.; Kluger, Y. Growth factor receptor-bound protein-7 (Grb7) as a
prognostic marker and therapeutic target in breast cancer. Ann. Oncol.
2010, 21, 466−73.
chromatogram for peptide 3. VII. MS for peptide 3. VIII.
SPR sensorgrams for peptides 1, 2, and 3 binding to the
SH2 domains of Grb2, Grb7, and Grb10. IX. SPR
sensorgrams and equilibrium binding curves for peptides
1, 2, and 3 binding to full-length Grb7 and Grb7-SH2
(PDF)
Accession Codes
Coordinates and structure factors have been deposited in the
PDB with accession numbers 4WWQ for apo-Grb7-SH2 and
4X6S for Grb7-SH2:2 complex.
AUTHOR INFORMATION
Corresponding Author
*E-mail: jackie.wilce@monash.edu. Phone: + 613 9902 9226.
Fax: + 613 9902 9500.
■
(13) Stein, D.; Wu, J.; Fuqua, S. A.; Roonprapunt, C.; Yajnik, V.;
D’Eustachio, P.; Moskow, J. J.; Buchberg, A. M.; Osborne, C. K.;
Margolis, B. The SH2 domain protein GRB-7 is co-amplified,
overexpressed and in a tight complex with HER2 in breast cancer.
EMBO J. 1994, 13, 1331−40.
(14) Fiddes, R. J.; Campbell, D. H.; Janes, P. W.; Sivertsen, S. P.;
Sasaki, H.; Wallasch, C.; Daly, R. J. Analysis of Grb7 recruitment by
heregulin-activated erbB receptors reveals a novel target selectivity for
erbB3. J. Biol. Chem. 1998, 273, 7717−24.
(15) Margolis, B.; Silvennoinen, O.; Comoglio, F.; Roonprapunt, C.;
Skolnik, E.; Ullrich, A.; Schlessinger, J. High-efficiency expression/
cloning of epidermal growth factor-receptor-binding proteins with Src
homology 2 domains. Proc. Natl. Acad. Sci. U. S. A. 1992, 89, 8894−8.
(16) McMurray, J. S.; Mandal, P. K.; Liao, W. S.; Klostergaard, J.;
Robertson, F. M. The consequences of selective inhibition of signal
transducer and activator of transcription 3 (STAT3) tyrosine705
phosphorylation by phosphopeptide mimetic prodrugs targeting the
Src homology 2 (SH2) domain. Jak-Stat 2012, 1, 263−347.
(17) Kraskouskaya, D.; Duodu, E.; Arpin, C. C.; Gunning, P. T.
Progress towards the development of SH2 domain inhibitors. Chem.
Soc. Rev. 2013, 42, 3337−70.
(18) Zhou, S.; Shoelson, S. E.; Chaudhuri, M.; Gish, G.; Pawson, T.;
Haser, W. G.; King, F.; Roberts, T.; Ratnofsky, S.; Lechleider, R. J.;
Neel, B. G.; Birge, R. B.; Fajardo, J. E.; Chou, M. M.; Hanafusa, H.;
Schaffhausen, B.; Cantley, L. C. SH2 domains recognize specific
phosphopeptide sequences. Cell 1993, 72, 767−778.
(19) Pero, S. C.; Daly, R. J.; Krag, D. N. Grb7-based molecular
therapeutics in cancer. Expert Rev. Mol. Med. 2003, 5, 1−11.
(20) Pero, S. C.; Oligino, L.; Daly, R. J.; Soden, A. L.; Liu, C.; Roller,
P. P.; Li, P.; Krag, D. N. Identification of novel non-phosphorylated
ligands, which bind selectively to the SH2 domain of Grb7. J. Biol.
Chem. 2002, 277, 11918−26.
(21) Porter, C. J.; Matthews, J. M.; Mackay, J. P.; Pursglove, S. E.;
Schmidberger, J. W.; Leedman, P. J.; Pero, S. C.; Krag, D. N.; Wilce,
M. C. J.; Wilce, J. A. Grb7 SH2 domain structure and interactions with
a cyclic peptide inhibitor of cancer cell migration and proliferation.
BMC Struct. Biol. 2007, 7, 58.
(22) Gunzburg, M. J.; Ambaye, N. D.; Hertzog, J. T.; del Borgo, M.
P.; Pero, S. C.; Krag, D. N.; Wilce, M. C. J.; Aguilar, M.-I.; Perlmutter,
P.; Wilce, J. A. Use of SPR to Study the Interaction of G7−18NATE
Peptide with the Grb7-SH2 Domain. Int. J. Pept. Res. Ther. 2010, 16,
177−184.
(23) Spuches, A. M.; Argiros, H. J.; Lee, K. H.; Haas, L. L.; Pero, S.
C.; Krag, D. N.; Roller, P. P.; Wilcox, D. E.; Lyons, B. A. Calorimetric
investigation of phosphorylated and non-phosphorylated peptide
ligand binding to the human Grb7-SH2 domain. J. Mol. Recognit.
2007, 20, 245−52.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We wish to thank all the beamline staff at the MX2 at the
Australian Synchrotron, Victoria, Australia where the diffraction
data were collected. This work was supported by a grant from
the National Health and Medical Research Council awarded to
J.A.W., P.P., and S.C.P., a National Health and Medical
Research Senior Research Fellowship awarded to M.C.J.W. as
well as the Victorian Government’s Operational Infrastructure
Support Program.
ABBREVIATIONS
■
cmF, carboxymethylphenylalanine; cF, carboxyphenylalanine;
Grb, growth factor receptor bound protein; HER2, human
epidermal growth factor receptor 2; KD, equilibrium dissocia-
tion constant; LC, liquid chromatography; MLA, malonic acid;
MS, mass spectrometry; PH, pleckstrin homology; pY,
phosphotyrosine; RA, ras-associating; RU, response units;
SH2, src homology 2; SPR, surface plasmon resonance;
TNBC, triple negative breast cancer
REFERENCES
■
(1) Hanahan, D.; Weinberg, R. A. Hallmarks of cancer: the next
generation. Cell 2011, 144, 646−74.
(2) Pradip, D.; Bouzyk, M.; Dey, N.; Leyland-Jones, B. Dissecting
GRB7-mediated signals for proliferation and migration in HER2
overexpressing breast tumor cells: GTP-ase rules. Am. J. Cancer Res.
2013, 3, 173−95.
(3) Giricz, O.; Calvo, V.; Pero, S. C.; Krag, D. N.; Sparano, J. A.;
Kenny, P. A. GRB7 is required for triple-negative breast cancer cell
invasion and survival. Breast Cancer Res. Treat. 2012, 133, 607−15.
(4) Bai, T.; Luoh, S. W. GRB-7 facilitates HER-2/Neu-mediated
signal transduction and tumor formation. Carcinogenesis 2007, 29,
473−479.
(5) Chu, P. Y.; Li, T. K.; Ding, S. T.; Lai, I. R.; Shen, T. L. EGF-
induced Grb7 recruits and promotes Ras activity essential for the
tumorigenicity of Sk-Br3 breast cancer cells. J. Biol. Chem. 2010, 285,
29279−85.
(6) Nencioni, A.; Cea, M.; Garuti, A.; Passalacqua, M.; Raffaghello,
L.; Soncini, D.; Moran, E.; Zoppoli, G.; Pistoia, V.; Patrone, F.;
Ballestrero, A. Grb7 upregulation is a molecular adaptation to HER2
signaling inhibition due to removal of Akt-mediated gene repression.
PLoS One 2010, 5, e9024.
(24) Gunzburg, M. J.; Ambaye, N. D.; Del Borgo, M. P.; Pero, S. C.;
Krag, D. N.; Wilce, M. C.; Wilce, J. A. Interaction of the non-
phosphorylated peptide G7−18NATE with Grb7-SH2 domain
requires phosphate for enhanced affinity and specificity. J. Mol.
Recognit. 2012, 25, 57−67.
(7) Pero, S. C.; Shukla, G. S.; Cookson, M. M.; Flemer, S., Jr.; Krag,
D. N. Combination treatment with Grb7 peptide and Doxorubicin or
Trastuzumab (Herceptin) results in cooperative cell growth inhibition
in breast cancer cells. Br. J. Cancer 2007, 96, 1520−5.
K
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(25) Eck, M. J.; Shoelson, S. E.; Harrison, S. C. Recognition of a
high-affinity phosphotyrosyl peptide by the Src homology-2 domain of
p56lck. Nature 1993, 362, 87−91.
(26) Burke, T. R., Jr.; Lee, K. Phosphotyrosyl mimetics in the
development of signal transduction inhibitors. Acc. Chem. Res. 2003,
36, 426−33.
SH2 domain of Grb7 and Grb2 using fluorescence polarization. J.
Biomol. Screening 2008, 13, 112−9.
(43) Ambaye, N. D.; Gunzburg, M. J.; Lim, R. C.; Price, J. T.; Wilce,
M. C.; Wilce, J. A. Benzopyrazine derivatives: A novel class of growth
factor receptor bound protein 7 antagonists. Bioorg. Med. Chem. 2011,
19, 693−701.
(44) Ambaye, N. D.; Gunzburg, M. J.; Lim, R. C.; Price, J. T.; Wilce,
M. C.; Wilce, J. A. The discovery of phenylbenzamide derivatives as
Grb7-based antitumor agents. ChemMedChem 2013, 8, 280−8.
(45) Morlacchi, P.; Robertson, F. M.; Klostergaard, J.; McMurray, J.
S. Targeting SH2 domains in breast cancer. Future Med. Chem. 2014,
6, 1909−26.
(27) Ambaye, N. D.; Gunzburg, M. J.; Traore, D. A. K.; Del Borgo,
M. P.; Perlmutter, P.; Wilce, M. C. J.; Wilce, J. A. Preparation of
crystals for characterizing the Grb7 SH2 domain before and after
complex formation with a bicyclic peptide antagonist. Acta Crystallogr.,
Sect. F: Struct. Biol. Commun. 2014, 70, 182−186.
(28) Stein, E. G.; Ghirlando, R.; Hubbard, S. R. Structural basis for
dimerization of the Grb10 Src homology 2 domain. Implications for
ligand specificity. J. Biol. Chem. 2003, 278, 13257−64.
(46) Pace, C. N.; Vajdos, F.; Fee, L.; Grimsley, G.; Gray, T. How to
measure and predict the molar absorption coefficient of a protein.
Protein Sci. 1995, 4, 2411−23.
(29) Depetris, R. S.; Hu, J.; Gimpelevich, I.; Holt, L. J.; Daly, R. J.;
Hubbard, S. R. Structural basis for inhibition of the insulin receptor by
the adaptor protein Grb14. Mol. Cell 2005, 20, 325−33.
(30) Porter, C. J.; Wilce, M. C.; Mackay, J. P.; Leedman, P.; Wilce, J.
A. Grb7-SH2 domain dimerisation is affected by a single point
mutation. Eur. Biophys. J. 2005, 34, 454−60.
(47) Leslie, A. W.; Powell, H. Processing Diffraction Data with
Mosflm. In Evolving Methods for Macromolecular Crystallography, Read,
R.; Sussman, J., Eds.; Springer: Dordrecht, 2007; Vol. 245, pp 41−51.
(48) Evans, P. Scaling and assessment of data quality. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2006, 62, 72−82.
(49) Winn, M. D.; Ballard, C. C.; Cowtan, K. D.; Dodson, E. J.;
Emsley, P.; Evans, P. R.; Keegan, R. M.; Krissinel, E. B.; Leslie, A. G.;
McCoy, A.; McNicholas, S. J.; Murshudov, G. N.; Pannu, N. S.;
Potterton, E. A.; Powell, H. R.; Read, R. J.; Vagin, A.; Wilson, K. S.
Overview of the CCP4 suite and current developments. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 235−42.
(50) Vagin, A.; Teplyakov, A. Molecular replacement with MOLREP.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010, 66, 22−5.
(51) Emsley, P.; Cowtan, K. Coot: model-building tools for
molecular graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004,
60, 2126−32.
(52) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
macromolecular structures by the maximum-likelihood method. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240−55.
(53) Afonine, P. V.; Grosse-Kunstleve, R. W.; Echols, N.; Headd, J. J.;
Moriarty, N. W.; Mustyakimov, M.; Terwilliger, T. C.; Urzhumtsev, A.;
Zwart, P. H.; Adams, P. D. Towards automated crystallographic
structure refinement with phenix.refine. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2012, 68, 352−67.
(54) McPhillips, T. M.; McPhillips, S. E.; Chiu, H. J.; Cohen, A. E.;
Deacon, A. M.; Ellis, P. J.; Garman, E.; Gonzalez, A.; Sauter, N. K.;
Phizackerley, R. P.; Soltis, S. M.; Kuhn, P. Blu-Ice and the Distributed
Control System: software for data acquisition and instrument control
at macromolecular crystallography beamlines. J. Synchrotron Radiat.
2002, 9, 401−6.
(55) Winter, G. xia2: an expert system for macromolecular
crystallography data reduction. J. Appl. Crystallogr. 2010, 43, 186−190.
(56) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.
D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.
Crystallogr. 2007, 40, 658−674.
(31) Ambaye, N. D.; Pero, S. C.; Gunzburg, M. J.; Yap, M.; Clayton,
D. J.; Del Borgo, M. P.; Perlmutter, P.; Aguilar, M. I.; Shukla, G. S.;
Peletskaya, E.; Cookson, M. M.; Krag, D. N.; Wilce, M. C.; Wilce, J. A.
Structural basis of binding by cyclic nonphosphorylated peptide
antagonists of Grb7 implicated in breast cancer progression. J. Mol.
Biol. 2011, 412, 397−411.
(32) Waksman, G.; Shoelson, 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, 779−90.
(33) Burke, T. R., Jr.; Luo, J.; Yao, Z. J.; Gao, Y.; Zhao, H.; Milne, G.
W.; Guo, R.; Voigt, J. H.; King, C. R.; Yang, D. Monocarboxylic-based
phosphotyrosyl mimetics in the design of GRB2 SH2 domain
inhibitors. Bioorg. Med. Chem. Lett. 1999, 9, 347−52.
(34) Wang, W.; Ramdas, L.; Sun, G.; Ke, S.; Obeyesekere, N. U.;
Budde, R. J.; McMurray, J. S. Cyclic peptides incorporating 4-
carboxyphenylalanine and phosphotyrosine are potent inhibitors of
pp60(c-) (src). Biochemistry 2000, 39, 5221−8.
(35) Tilley, J. W.; Sarabu, R.; Wagner, R.; Mulkerins, K. Preparation
of carboalkoxyalkylphenylalanine derivatives from tyrosine. J. Org.
Chem. 1990, 55, 906−910.
(36) Janes, P. W.; Lackmann, M.; Church, W. B.; Sanderson, G. M.;
Sutherland, R. L.; Daly, R. J. Structural determinants of the interaction
between the erbB2 receptor and the Src homology 2 domain of Grb7.
J. Biol. Chem. 1997, 272, 8490−7.
(37) Bevington, A.; Mundy, K. I.; Yates, A. J.; Kanis, J. A.; Russell, R.
G.; Taylor, D. J.; Rajagopalan, B.; Radda, G. K. A study of intracellular
orthophosphate concentration in human muscle and erythrocytes by
31P nuclear magnetic resonance spectroscopy and selective chemical
assay. Clin. Sci. 1986, 71, 729−35.
(57) Moriarty, N. W.; Grosse-Kunstleve, R. W.; Adams, P. D.
electronic Ligand Builder and Optimization Workbench (eLBOW): a
tool for ligand coordinate and restraint generation. Acta Crystallogr.,
Sect. D: Biol. Crystallogr. 2009, 65, 1074−80.
(58) Chen, V. B.; Arendall, W. B., 3rd; Headd, J. J.; Keedy, D. A.;
Immormino, R. M.; Kapral, G. J.; Murray, L. W.; Richardson, J. S.;
Richardson, D. C. MolProbity: all-atom structure validation for
macromolecular crystallography. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 2010, 66, 12−21.
(38) Gilmer, T.; Rodriguez, M.; Jordan, S.; Crosby, R.; Alligood, K.;
Green, M.; Kimery, M.; Wagner, C.; Kinder, D.; Charifson, P.; Hassell,
A. M.; Willard, D.; Luther, M.; Rusnak, D.; Sternbach, D. D.;
Mehrotra, M.; Peel, M.; Shampine, L.; Davis, R.; Robbins, J.; Patel, I.
R.; Kassel, D.; Burkhart, W.; Moyer, M.; Bradshaw, T.; Berman, J.
Peptide inhibitors of src SH3-SH2-phosphoprotein interactions. J. Biol.
Chem. 1994, 269, 31711−31719.
(39) Tong, L.; Warren, T. C.; Lukas, S.; Schembri-King, J.; Betageri,
R.; Proudfoot, J. R.; Jakes, S. Carboxymethyl-phenylalanine as a
replacement for phosphotyrosine in SH2 domain binding. J. Biol.
Chem. 1998, 273, 20238−42.
(40) Yao, Z. J.; King, C. R.; Cao, T.; Kelley, J.; Milne, G. W.; Voigt, J.
H.; Burke, T. R., Jr. Potent inhibition of Grb2 SH2 domain binding by
non-phosphate-containing ligands. J. Med. Chem. 1999, 42, 25−35.
(41) Bradshaw, J. M.; Mitaxov, V.; Waksman, G. Investigation of
phosphotyrosine recognition by the SH2 domain of the Src kinase. J.
Mol. Biol. 1999, 293, 971−85.
(42) Luzy, J. P.; Chen, H.; Gril, B.; Liu, W. Q.; Vidal, M.; Perdereau,
D.; Burnol, A. F.; Garbay, C. Development of binding assays for the
L
DOI: 10.1021/acs.jmedchem.5b00609
J. Med. Chem. XXXX, XXX, XXX−XXX