Utilization of 3′-carboxy-containing tyrosine derivatives as a new class of phosphotyrosyl mimetics in the preparation of novel non-phosphorylated cyclic peptide inhibitors of the Grb2-SH2 domain

Article abstract of DOI:10.1039/b515432d

A new class of phosphotyrosyl (pTyr) mimetics, distinct from the conventional pTyr mimetic design of adding non-hydrolyzable acidic functionalities to the 4′-position of phenylalanine, was created by introducing carboxy-containing groups to the 3′-positio

Full text of DOI:10.1039/b515432d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Utilization of 3ꢀ-carboxy-containing tyrosine derivatives as a new class of
phosphotyrosyl mimetics in the preparation of novel non-phosphorylated
cyclic peptide inhibitors of the Grb2–SH2 domain
Yan-Li Song,a Jinzhi Tan,b Xiao-Min Luob and Ya-Qiu Long*a
Received 31st October 2005, Accepted 6th December 2005
First published as an Advance Article on the web 4th January 2006
DOI: 10.1039/b515432d
A new class of phosphotyrosyl (pTyr) mimetics, distinct from the conventional pTyr mimetic design of
adding non-hydrolyzable acidic functionalities to the 4ꢀ-position of phenylalanine, was created by
introducing carboxy-containing groups to the 3ꢀ-position of tyrosine. The effect of the chain length of
the carboxy substituent was examined. Reported herein is the chiral pool synthesis of the new pTyr
mimetics, and their first use in a novel non-phosphorylated Grb2–SH2 domain binding motif with the
5-amino-acid sequence Xx1-Leu-(3ꢀ-substituted-Tyr)-Ac6c-Asn. The highest affinity was exhibited by
the 3-L-(2-carboxyethyl)tyrosine-containing sulfoxide-cyclized peptide 3c, with an IC50 = 1.1 lM,
providing a promising new template for further development of potent Grb2–SH2 domain inhibitors
with reduced charge and peptidic nature, but improved selectivity and bioavailability.
In combination with the recent finding that substitution at the
Introduction
3ꢀ-position of the tyrosine greatly affects the activity of this non-
Growth factor receptor-bound protein 2 (Grb2) mediates intra-
phosphorylated peptide ligand of the Grb2–SH2 domain (Fig. 1,
2),14 we were intrigued to design a new class of pTyr mimetics
cellular signaling by its Src homology 2 (SH2) domain binding to
by introducing one carboxyl group at the 3ꢀ-position of tyrosine
and to incorporate it into a fully elaborated cyclic Grb2–SH2
domain-binding platform with reduced peptidic character (Fig. 1,
3). This is a brand new strategy, distinct from previously reported
pTyr mimetics in which the efforts were concentrated on the
replacement of the 4ꢀ-phosphate of pTyr with phosphonate or
dicarboxylate.15 Although an earlier attempt to synthesize pTyr
mimetics against the Grb2–SH2 domain by putting a carboxyl
group at the 3ꢀ-position of the tyrosine proved unsuccessful when
incorporated into an open-chain tripeptide platform,16 we wanted
to examine the new concept by trying various 3ꢀ-carboxyalkyl
groups in the context of our novel non-phosphorylated Grb2–
SH2 ligands. Reported herein is the chiral pool synthesis of
an orthogonally protected L-3-(carboxyalkyl)tyrosine suitable for
solid-phase peptide synthesis, and its first use in a new structural
platform binding to the Grb2–SH2 domain, with only a 5-
amino-acid sequence devoid of any pTyr or pTyr mimetics. The
highest affinity was exhibited by the L-3-(2-carboxyethyl)tyrosine-
containing sulfoxide-cyclized pentapeptide (3c) with an IC50 of
1.1 lM, providing a new solution to the trade-off between the
need for negatively charged pTyr mimetics for optimal affinity
and the need to minimize charge to facilitate cell entry for all SH2
domain inhibitors.6
phosphotyrosyl (pTyr) motifs on growth factor receptors, thereby
leading todownstreamactivation ofthe mitogenic Ras pathways.1,2
An aberrant Grb2–SH2 dependent Ras activation pathway has
been shown to contribute to cellular processes important to
cancer development and progression, including cell proliferation,
apoptosis and metastasis.3,4 Thus, compounds which selectively
antagonize the SH2 domain of Grb2 should interrupt these
signaling pathways and thus are attractive targets for drug therapy
in oncology.5,6
Based on the unique structural feature that Grb2–SH2 do-
main preferentially associates with the peptide or protein lig-
ands bearing the “pTyr-X-Asn” consensus sequence in a b-
turn conformation,7 many efforts in the development of Grb2-
dependent signaling inhibitors have been focused on the creation
ofpTyrmimetics andpre-inducedb-turngeometries.5 Thesestrate-
gies are applicable to the phage library-based non-phosphorylated
cyclic peptide antagonists of the Grb2–SH2 domain, which do not
contain phosphortyrosine, yet can specifically bind to the target
protein with high affinity (Fig. 1, lead compound 1).8 This new
type of pTyr-independent Grb2–SH2 binding motif functions by
a multi-point binding pattern involving well-defined side chains
and specific conformations.9–13 However, the phosphorylation of
the Tyr residue within the consensus sequence of Tyr-Xx-Asn
improved the binding affinity 100-fold with respect to G1TE.10
aState Key Laboratory of Drug Research, Shanghai Institute of Materia
Medica, Shanghai Institutes for Biological Sciences, Graduate School of
the Chinese Academy of Sciences, CAS, 555 Zuchongzhi Road, Shanghai,
201203, China. E-mail: yqlong@mail.shcnc.ac.cn; Fax: +86-21-50806876;
Tel: +86-21-50806876
Synthesis
Three new phosphotyrosyl mimetics bearing one carboxyl group
with different chain lengths at the 3-position of L-tyrosine
were synthesized in protected forms with O-Boc and COO-
tBu protection on the side chain and N-a-Fmoc protection on
bDrug Discovery and Design Center, Shanghai Institute of Materia Medica,
Shanghai Institutes for Biological Sciences, Graduate School of the Chinese
Academy of Sciences, CAS, 555 Zuchongzhi Road, Shanghai, 201203, China
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 659–666 | 659
©

View Article Online
Fig. 1 The structures of synthetic non-phosphorylated peptide ligands binding to the Grb2–SH2 domain.
the backbone, suitable for the solid-phase peptide synthesis. As
outlined in Scheme 1, our synthesis begins with commercially
available N-Boc-L-tyrosine. Formylation of the phenol using a
modified Reimer–Tiemann process, namely treatment of the N-
Boc-L-tyrosine with chloroform and solid sodium hydroxide in
the presence of a small amount of water, gave the desired 3-formyl
compound 4.17 Transformation of the N-Boc protection into N-
Fmoc followed by the benzylation of the phenol group resulted in
the key intermediate 5. Compound 5 is a common intermediate
for the synthesis of 7 through an oxidation pathway, and 10
and 11 through a Wittig sequence. The oxidation of the formyl
group of 5 with Ag2O under basic conditions and subsequent
esterification of the resulting carboxylic acid with tert-butyl 2,2,2-
trichloroacetimidate afforded the fully protected compound 6.
Hydrogenation removed the benzyl protecting groups, and was
followed by treatment with (Boc)2O, yielding the desired product
7, orthogonally protected for solid-phase peptide synthesis. In the
other route, treatment of 5 with different Wittig reagents provided
compounds 8 and 9. Hydrogenation followed by treatment with
(Boc)2O furnished the desired compounds 10 and 11, which were
suitably protected for solid-phase peptide synthesis.
The 3ꢀ-(carboxyalkyl)tyrosine-containing cyclic peptides 3a–f
were synthesized manually using Fmoc-based SPS protocols on
PAL amide resin.18 Key intermediates included the N-terminally
chloroacetylated derivative of the linear 5-mer precursor hav-
ing cysteine at the C-terminal position with a x-aminovaleric
Scheme 1 Synthetic route towards globally protected L-3-carboxyl-substituted tyrosines for solid-phase peptide synthesis. Reagents and conditions:
(a) 6 eq. NaOH, 2 eq. H2O, CHCl3, reflux; (b) 20% TFA in 10 mL DCM, rt; (c) 1.2 eq. FmocOSu, 10% Na2CO3, 0 ◦C to rt; (d) 2.5 eq. BnBr, 2.5 eq.
n
K2CO3, Bu4NI, DMF, rt; (e) 1.0 eq. Ph3PCHR, DCM, rt; (f) 2.0 eq. Ag2O, 5% NaOH; (g) 3 eq. tert-butyl 2,2,2-trichloroacetimidate, THF–hexane;
(h) H2, 10% Pd/C, EtOH, rt; (i) 2.0 eq. (Boc)2O, cat. DMAP, 2.0 eq. Et3N, DCM, rt.
660 | Org. Biomol. Chem., 2006, 4, 659–666
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Table 1 The inhibitory activity of 3ꢀ-(carboxyalkyl)tyrosine-containing
acid linker in between. Peptide cyclization was achieved by
intramolecular nucleophilic displacement of the chloro group
by the cysteine side-chain thiol functionality. The oxidation
of the thioether linkage into sulfoxide was readily achieved
using 5% aqueous H2O2. The resulting sulfoxide diastereoiso-
mers were easily separated by RP-HPLC. The stereochemistry
of the sulfoxide diastereoisomers was identified by CD spec-
troscopy. The faster-eluting sulfoxide was the R-diastereomer,
and the slower-eluting sulfoxide the S-diastereomer.19 All final
products were purified by RP-HPLC, their identities assessed
by mass spectroscopy, and their purity checked by analytical
HPLC.
sulfoxide-bridged pentapeptidesa
Rꢀ
IC50/lMb
Results and discussion
Compd.
R
Because of the important role of SH2 domains in signal trans-
duction pathways, SH2-mediated protein–protein interactions are
attractive therapeutic targets. However, a major difficulty in
designing SH2 domain-binding antagonists is the concentrated
negative charge of SH2 ligands (from phosphate and acidic side
chains), which would confer poor cellular activity and selectivity.20
Therefore, phosphotyrosyl mimetics that retain high binding
affinity yet exhibit reduced net anionic charge is recognized and
pursued as an effective strategy to tackle such problems.
3a
3b
3c
3d
3e
3f
-H
-COOH
-CH2CH2COOH
-CH2CH2CO2Et
-CH2CH2COOH
-CH2CH2COOH
-CH2(CH2)2COOH
-CH2(CH2)2COOH
-CH2(CH2)2COOH
-CH2(CH2)2COOH
-CH2CH(COOH)2
-CH3
77.0 13.0
16.5 3.5
1.1 0.1
40.0 7.0
44.5 16.5
14.0 2.0
a The competitive binding affinity of ligands for the Grb2–SH2 domain
protein was assessed using Biacore surface plasmon resonance (SPR)
methods. IC50 values were determined by mixing the inhibitor with
recombinant Grb2–SH2 protein and measuring the amount of binding
at equilibrium to an immobilized SHC (pTyr-317) phosphopeptide in
a manner similar to that reported previously.8 b Values are calculated
from at least two independent experiments, and are expressed as the
concentration at which half-maximal competition was observed (IC50).
Errors are expressed as the standard deviation.
Our starting point is
a phage library derived non-
phosphorylated thioether-bridged cyclic decapeptide ligand
(Fig. 1, 1).8 In order to reduce the peptidic character, a new cyclic
Grb2–SH2 domain-binding platform was developed in which the
peptide chain was shortened to a 5-amino-acid sequence, and a
turn-inducing amino acid, namely a-aminocyclohexanecarboxylic
acid (Ac6c), incorporated at position 4. The cyclization linkage
adopted was the favored (R)-configured sulfoxide12 in the context
of non-phosphorylated cyclic Grb2–SH2 binding peptides. The
resulting (R)-sulfoxide-cyclized pentapeptide 3a displayed moder-
ate inhibitory activity (IC50 = 77 lM), serving as the structural
platform for the incorporation of new pTyr mimetics bearing one
carboxy group at the 3-position of tyrosine (Table 1).
pTyr mimetics containing phosphonate- and carboxylate-
based structures have been reported in the literature.14 To date,
monoacidic phosphoryl mimetics have tended to exhibit less
affinity than diacidic mimetics, due in part to the presence of
two highly conserved positively charged Arg residues within the
SH2 domain pTyr binding pocket. In particular, the monocarboxy
series, e.g., carboxymethylphenylalanine, while useful in decreasing
the net negative charge relative to pTyr, decreased affinity almost
100-fold, underscoring the importance of two negative charges for
molecular recognition by SH2 domains in the context of regular
linear phosphopeptides.21
Unlike the phosphopeptide ligands of Grb2–SH2 in which
phosphotyrosine is a major contributor to the binding, our
phage library based non-phosphorylated cyclic peptide antagonist
requires essentially all amino acids (except Gly7), as well as the
constrained conformation, to retain high affinity binding with
the Grb2–SH2 domain.11–13 It is hoped that this multipoint bind-
ing pattern and conformational constraint may allow improved
interaction with the Grb2–SH2 domain by the incorporation
of 3-carboxytyrosine-based pTyr mimetics with monocarboxy
functionality.
To our delight, when Tyr3 was replaced with 3-carboxytyrosine,
the resulting peptide 3b exhibited a 4.6-fold increase in the binding
potency (IC50 = 16.5 3.5 lM) relative to the parent compound 3a
(Table 1). The chain length of the monocarboxy-based substitutent
at the 3ꢀ-position of Tyr was found to play an important role in the
Grb2–SH2 binding. Just extending the side-chain of the 3ꢀ-carboxy
substituent with two methylene units, i.e. substitution of Tyr with
3ꢀ-(2-carboxyethyl)tyrosine, provided peptide 3c, displaying a 14-
fold and 70-fold enhancement in Grb2–SH2 domain antagonist
potency (IC50 = 1.1 0.1 lM) relative to its zero-linker congener
3b and the parent compound 3a, respectively (Table 1). The gain
in activity caused by the introduction of the linker might be
attributed to an optimal positioning of the carboxyl group into
the binding pocket by the flexible long side-chain. As expected, the
loss of the free carboxylic acid by formation of the ethyl ester 3d,
i.e. replacement of Tyr3 with L-3-(2-(ethylcarboxy)ethyl)tyrosine
resulted in a marked decrease in the activity (IC50 = 40.0 7.0
lM) (Table 1), suggesting that the acidic groups at the side chain
of the tyrosine provide the key ionic interactions with the charged
guanidine groups of the Arg67 (aA-helix) and Arg86 (bC-strand)
residues within the pTyr binding pocket. However, the ethyl ester
3d could be used as a potential prodrug of 3c in the future.
A previous SAR study has disclosed that the Glu1 side chain
in G1TE (1) compensates for the absence of Tyr3 phosphory-
lation in securing effective binding to the Grb2–SH2 domain.9
Thus, replacement of Glu1 with c-carboxyglutamic acid (Gla)
prominently improves binding affinity.11–14 However, in our case,
in the presence of the L-3ꢀ-(2-carboxyethyl)tyrosine as a pTyr
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 659–666 | 661
©

View Article Online
mimetic, the introduction of Gla at position 1 (to give 3e)
adversely affects the binding, with a 40-fold decrease in the potency
(IC50 = 44.5 16.5 lM) compared to its precursor 3c (Table 1).
This negative result may indicate that the dicarboxyl side-chain
of Gla1 competes for the conformational space occupied by the
carboxyl functionality of the pTyr mimetics, thus the crowded
acidic groups interfere with the proper interactions with the
binding pocket. This observation is similar to that found in pTyr-
containing G1TE analogs, in which the acidic Glu1 is not favorable
for binding.9 Consistently, the replacement of Gla1 with Ala1
(to give 3e), resulted in a 3-fold increase of the binding affinity
(IC50 = 14.0 2.0 lM) relative to the precursor 3d (Table 1), while
keeping a low net negative charge. These results are in agreement
with the unique and important molecular mechanism disclosed
by our previous SAR studies9–14 for the non-phosphorylated cyclic
Grb2–SH2 domain-binding motif with a 9-amino-acid sequence.
increase of the binding potency, which might be attributable to
an enhanced electrostatic interaction originating from the 3ꢀ-
carboxy group of the tyrosine. The highest affinity was exhibited
by the L-3-(2-carboxyethyl)tyrosine-containing pentapeptide (3c)
with an IC50 = 1.1 lM, providing a promising new strategy for the
development of potent Grb2–SH2 domain inhibitors with reduced
charge and peptidic nature. This strategy may potentially find
value in SH2 domain systems other than Grb2, where pTyr serves
as a critical recognition determinant for effective binding.
Experimental
Binding affinity measurement of peptides to Grb2–SH2 domain
using Surface Plasmon Resonance (SPR)
The competitive binding affinity of ligands for the Grb2–SH2
protein was assessed using Biacore SPR methods on a BIAcore
3000 instrument (Pharmacia Biosensor, Uppsala, Sweden). IC50
values were determined by mixing various concentrations of
inhibitors with the recombinant GST–Grb2–SH2 domain protein
and measuring the amount of binding at equilibrium to the
immobilized SHC phosphopeptide (pTyr317), i.e. biotinyl-DDPS-
pY-VNVQ, in a manner described previously.8 Briefly, the bi-
otinylated phosphopeptide was attached to a streptavidin-coated
SA5 Biosensor chip, and the binding assays were conducted in
pH 7.4 PBS buffer containing 0.01% P-20 surfactant (Pharmacia
Biosensor).
Molecular modeling
The above observations can be nicely rationalized from the
results of molecular modeling experiments. Based on the X-
ray crystal structure of Grb2–SH2 domain having a bound
cyclic phosphopeptide ligand,22 we undertook molecular modeling
studies to examine the interactions of peptides 3b, 3c and 3d
with the Grb2–SH2 domain protein, respectively. The automated
molecular docking can produce several optional conformations
of the inhibitors. The conformation corresponding to the lowest
binding energy was selected as the possible binding conformation,
as shown in Fig. 2.
Molecular modeling
In the final models (Fig. 2), it is evident that the carboxyl
group introduced at the 3ꢀ-position of tyrosine makes key hydrogen
bonding interactions with residues Arg67, Arg86 and Ser96 within
the pTyr binding pocket of the Grb2–SH2 protein. For peptide
3b, the acidic hydroxyl group of the 3ꢀ-carboxytryosine residue
makes two H-bonding interactions with each guanidino group
of Agr67 and Arg86, while the 4ꢀ-hydroxy group of the phenyl
ring forms an H-bond with Ser90. As a comparison, when
the side-chain of the 3ꢀ-carboxy substituent was extended by
two CH2 units, the L-3-(2-carboxyethyl)tyrosine-containing cyclic
peptide 3c gained two additional H-bonding interactions with
the hydroxyl side chain of Ser96 and the guanidino side chain
of Arg86, provided by the hydroxyl and the carbonyl group of
the 3ꢀ-(2-carboxyethyl) side chain, respectively. However, for the
L-3-(2-ethylcarboxyethyl)tyrosine-containing peptide 3d, the loss
of the free carboxyl acid group at the 3-position of Tyr resulted
in a complete loss of the H-bonding interactions. The proposed
binding mode is consistent with the activity data.
The advanced docking program Autodock 3.0.324,25 was used to
automatically dock each of the three inhibitors to the binding
pocket of the Grb2–SH2 protein. The crystal structure of Grb2–
SH2 complexed with a cyclic phosphopeptide of cyclo(Ac-thiaLys-
pTyr-Val-Asn-Val-Pro) (PDB entry code 1BM2)21 was recovered
from Brookhaven Protein Database (PDB). The missed side-chain
atoms of Met55 were modeled by Sybyl 6.8 software.26 The initial
conformations of the three inhibitors were modeled based on the
peptide in the crystal structure. The kinds of atomic charges were
taken as Kollman-united-atom27 for the protein and Gasteiger–
Marsili28 for the inhibitors.
The whole docking operation used in this study is as follows.
First, the protein was checked for polar hydrogen and assigned
for partial atomic charges, and then a PDBQ file was created,
and the atomic solvation parameters assigned for the protein.
Meanwhile, some of the torsion angles of the inhibitor that
would be explored during the molecular docking stage were
defined, allowing the conformation search for the ligand during
the docking process. Second, a grid map with 80 × 80 × 80 points
Conclusions
˚
and a spacing of 0.375 A was calculated using the AutoGrid
In this study, we designed a new class of phosphotyrosyl mimetics
by introducing carboxy-containing groups at the 3ꢀ-position of
tyrosine for enhanced binding to Grb2–SH2 domain of non-
phosphorylated cyclic peptide ligands with reduced charge. A chi-
ral pool synthetic approach was developed to prepare orthogonally
protected L-3-(carboxyalkyl)tyrosines suitable for solid-phase pep-
tide synthesis. The incorporation of the new pTyr mimetics into
a phage library-derived non-phosphorylated structural platform
bearing only a 5-amino-acid sequence resulted in a significant
program,24 to evaluate the binding energies between the inhibitor
and the protein. Third, the Lamarckian genetic algorithm (LGA)24
was applied to deal with the inhibitor–protein interaction. Some
important parameters for LGA calculations were set as follows:
an initial population of random individuals with a size of 500; a
maximum number of 1.5 × 106 energy evaluations; a maximum
number of generations of 5.0 × 104. Default values were used for
the other parameters. The energy minimization was performed by
a Solis and Wets local search on a user-specified proportion of the
662 | Org. Biomol. Chem., 2006, 4, 659–666
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
Fig.
2
Two-dimensional representations of the interaction models of the 3-carboxytyrosine-containing cyclic peptide 3b, the 3-(2-car-
boxyethyl)tyrosine-containing cyclic peptide 3c, and the 3-(2-(ethylcarboxy)ethyl)tryosine-containing cyclic peptide 3d with the Grb2–SH2 protein.
The images were generated with LIGPLOT program.23
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 659–666 | 663
©

View Article Online
population. Finally, the docked conformations of the inhibitor
were selected according to the criterion of interaction energy
combined with geometrical matching quality after a reasonable
number of evaluations.
(m, 2H). EI-MS: m/z 611 (M), 178 (100), 91 (24). HRMS: calcd
for C33H39O6: 611.2308; found: 611.2315. [a]2D2 = +3.2 (c = 1.7,
CHCl3).
2-Benzyloxy-5-[2-benzyloxycarbonyl-2-(9H-fluoren-9-ylmethoxy-
carbonylamino)ethyl]benzoic acid tert-butyl ester (6). Compound
5 (220 mg, 0.36 mmol) was added to a stirred suspension of
Ag2O (158 mg, 0.68 mmol) in 5% aq. NaOH (4 mL). After
being stirred for 24 h, the suspension was filtered and the solid
Synthesis
The1H-NMR spectra were recorded on a Varian Mercury-
400 MHz spectrometer. The data are reported in parts per million
relative to TMS and referenced to the solvent in which they
were run. Elemental analyses were obtained using a Vario EL
spectrometer. Melting points (uncorrected) were determined on
a Buchi-510 capillary apparatus. EI-MS spectra were obtained
on a Finnigan MAT 95 mass spectrometer. ESI-MS spectra were
obtained on a Finnigan LCQ Deca mass spectrometer. Specific
rotations (uncorrected) were determined in a Perkin–Elmer 341
polarimeter. The solvent was removed by rotary evaporation
under reduced pressure, and flash column chromatography was
performed on silica gel H (10–40 lm). Anhydrous solvents were
obtained by redistillation over sodium wire.
◦
was washed with H2O. The filtrate was cooled to 0 C, acidified
with 1 N HCl, extracted with Et2O, and the extracts dried and
concentrated. To the solution of the crude product and tert-butyl
2,2,2-trichloroacetimidate (398 mg, 1.35 mmol) in dry THF
(10 mL), was added BF3·Et2O (50 lL). The solution was stirred
at rt for 16 h, and then concentrated. The residue was diluted
with saturated NaHCO3 and ethyl acetate, and the aqueous
layer back-extracted with ethyl acetate. The organic extracts were
washed with H2O, brine, dried over Na2SO4, and concentrated.
Flash column chromatography (silica gel, 6 : 1 PE–EtOAc)
afforded the desired product 6 as a white solid (150 mg, 61%). 1H
NMR (CDCl3, 400 MHz): d 7.76–7.26 (m, 19H), 7.16 (d, 1H, J =
8.4 Hz), 6.87 (d, 1H, J = 8.4 Hz), 5.36 (m, 1H), 5.18 (s, 2H), 5,14
(s, 2H), 4.72–4.10 (m, 4H), 3.13–3.04 (m, 2H), 1.50 (s, 9H).
N-[(1,1-Dimethylethoxy)carbonyl]-3-(3-formyl-4-hydroxyphenyl)-
L-alanine (4). Powdered sodium hydroxide (1.71 g, 42.72 mmol)
was added to a suspension of N-Boc-L-tyrosine (2 g, 7.12 mmol),
water (0.256 mL, 14.13 mmol), and chloroform (30 mL). The
mixture was refluxed for 4 h. Additional sodium hydroxide was
added (0.84 g, 21.36 mmol) in two portions after 1 h and 1.5 h,
respectively. The reaction was then diluted with water and ethyl
acetate, and the aqueous layer was acidified to pH 1 with 1 N HCl
and back-extracted with ethyl acetate. The organic extracts were
washed with brine, dried over Na2SO4, and concentrated. Flash
column chromatography (silica gel, 12 : 1 CHCl3–MeOH with 1%
acetic acid as eluent) afforded the desired product 4 as a brown
oil (720 mg, 33%). 1H NMR (CDCl3, 400 MHz): d 10.92 (1H, s),
9.84 (1H, s), 7.37 (d, 1H, J = 1.6 Hz), 7.33 (1H, d, J = 8.4 Hz),
6.74 (1H, d, J = 7.8 Hz), 5.28 (1H, bs), 4.96 (1H, m), 3.06–3.03
(2H, m), 1.40 (9H, s).
2-tert-Butoxycarbonyloxy-5-[2-carboxy-2-(9H-fluoren-9-ylmeth-
oxycarbonylamino)ethyl]benzoic acid tert-butyl ester (7). 10%
Pd/C (10 mg) and 6 (315 mg, 0.46 mmol) was dissolved in dry
EtOH (10 mL). The solution was stirred at rt overnight under
H2. The Pd/C was filtered off and the filtrate concentrated in
vacuo to afford a colorless oil (181 mg, 0.36 mmol). This residue,
Boc2O (156 mg, 0.72 mmol), Et3N (20 mL), and cat. DMAP were
dissolved in DCM (18 mL), and the solution stirred at rt for 2 h.
The solvent was removed in vacuo. Flash column chromatography
(silica gel, 1 : 4 EtOAc–PE as eluent) afforded the desired product
7 as a white foam (110 mg, 51%). 1H NMR (CDCl3, 400 MHz): d
7.76–7.26 (m, 8H), 7.18–7.02 (m, 3H), 5.59 (d, 1H, J = 8.0 Hz),
4.74–4.69 (m, 1H), 4.64–4.18 (m, 3H), 3.21–3.13 (m, 2H), 1.56 (s,
t
9H), 1.49 (s, 9H). ESI-MS: m/z 602 (M − H)−, 546 (M − Bu),
3-(4-Benzyloxy-3-formylphenyl)-2-(9H-fluoren-9-ylmethoxycar-
530 (M − OtBu). Anal. calcd for C34H37NO9: C, 67.65; H, 6.18;
N, 2.32. Found: C, 67.76; H, 6.37; N, 2.06. [a]2D2 = +19.8 (c = 0.5,
CHCl3).
bonylamino)propionic acid benzyl ester (5). Compound
4
(600 mg, 1.94 mmol) was dissolved in 20% TFA in DCM (10 mL),
and the solution stirred at rt for 30 min. The solvent was then
concentrated, and the residue dissolved in 10% Na2CO3 (10 mL)
and a solution of FmocOSu (786 mg, 2.3 mmol) in dioxane
(10 mL) added dropwise, while stirring the mixture at rt overnight.
Subsequently, the reaction mixture was poured into water and
extracted with Et2O. The aqueous phase was acidified with 1 N
HCl to pH 3, and the aqueous phase was back-extracted with
EtOAc. The organic phase was dried over Na2SO4, and the solvent
was removed in vacuo. The residue was dissolved in 10 mL DMF,
and BnBr (0.56 mL, 4.64 mmol), K2CO3 (640 mg, 4.64 mmol),
and nBu4NI (81 mg, 0.125 mmol) added. The mixture was stirred
at rt overnight. Subsequently, the reaction mixture was poured
into water and extracted with Et2O. The organic phase was dried
over Na2SO4, and the solvent removed in vacuo. Flash column
chromatography (silica gel, 1 : 4 EtOAc–PE as eluent) afforded
3-{2-Benzyloxy-5-[2-benzyloxycarbonyl-2-(9H-fluoren-9-ylmeth-
oxycarbonylamino)ethyl]phenyl}acrylic acid tert-butyl ester (8).
t
=
Ph3P CHCO2 Bu (273 mg, 0.80 mmol) was added to a solution
of compound 5 (402 mg, 0.65 mmol) in DCM (15 mL), and the
mixture stirred at rt for 10 h. The solvent was removed in vacuo.
Flash column chromatography (silica gel, 1 : 6 EtOAc–PE as
eluent) afforded the desired product 8 as a white foam (282 mg,
64%). 1H NMR (CDCl3, 400 MHz): d 7.69 (d, 1H, J = 16.0 Hz),
7.90–7.25 (m, 19H), 6.92 (dd, 1H, J1 = 8.4 Hz, J2 = 1.6 Hz),
6.76 (d, 1H, J = 8.4 Hz), 6.39 (d, 1H, J = 16.4 Hz), 5.31 (d, 1H,
J = 8.4 Hz), 5.17 (d, 2H, J = 12.0 Hz), 5,09 (m, 1H), 4.71–4.67
(m, 1H), 4.46–4.41 (m, 3H), 3.08–3.05 (m, 2H), 1.49 (s, 9H).
EI-MS: m/z (%) 709 (M+), 91 (100). HRMS: calcd for C43H39NO7:
709.3040; found: 709.3028. [a]2D2 = −3.6 (c = 1.0, CHCl3).
1
the desired product 5 as a white foam (758 mg, 64%). H NMR
3-{2-Benzyloxy-5-[2-benzyloxycarbonyl-2-(9H-fluoren-9-ylmeth-
oxycarbonylamino)ethyl]phenyl}acrylic acid ethyl ester (9). Pre-
pared from compound 5 (409 mg, 0.67 mmol) by a procedure
similar to that used for 8, to give 9 as a white foam (342 mg,
(CDCl3, 400 MHz): d 10.49 (s, 1H), 7.77–7.26 (m, 19H), 7.16 (d,
1H, J = 8.4 Hz), 6.87 (d, 1H, J = 8.4 Hz), 5.36 (m, 1H), 5.18 (s,
2H), 5,14 (s, 2H), 4.70–4.64 (m, 1H), 4.45–4.11(m, 3H), 3.13–3.04
664 | Org. Biomol. Chem., 2006, 4, 659–666
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
75%). 1H NMR (CDCl3, 400 MHz): d 7.98 (d, 1H, J = 16.4 Hz),
7.76–7.23 (m, 19H), 6.92 (d, 1H, J = 8.4 Hz), 6.76 (d, 1H, J =
8.4 Hz), 6.47 (d, 1H, J = 16.4 Hz), 5.31 (d, 1H, J = 8.4 Hz),
5.19–5.10 (d, 2H, J = 12.4 Hz), 5.10 (s, 2H), 4.71–4.17 (m,
4H), 4.23 (q, 2H, J = 7.2 Hz), 3.08–3.05 (m, 2H), 1.31 (t, 3H,
J = 7.2 Hz). EI-MS: 681 (M+). HRMS: calcd for C43H39NO7:
681.2727; found: 681.2735. [a]2D2 = −2.5 (c = 1.0, CHCl3).
Cyclo-[Ac-Adi-Leu-(3-CO2H-Tyr)-Ac6c-Asn-Ava-Cys (O)-(R)]-
amide (3b). ESI-MS, m/z: calc. 978.4 (M + H)+, found 978.4.
tR = 15.1 min (10–70% of solvent B in 30 min, purity 99%); tR =
21.6 min (10–70% of solvent C in 30 min, purity 96%).
Cyclo-[Ac-Adi-Leu-(3-CH2CH2CO2H-Tyr)-Ac6c-Asn-Ava-Cys
(O)-(R)]-amide (3c). ESI-MS, m/z: calc. 1004.4 (M − H)−, found
1004.7. tR = 15.3 min (10–70% of solvent B in 30 min, purity 98%);
tR = 21.4 min (10–70% of solvent C in 30 min, purity 95%).
3-{2-tert-Butoxycarbonyloxy-5-[2-carboxy-2-(9H -fluoren-9-
ylmethoxycarbonylamino)ethyl]phenyl}acrylic acid tert-butyl ester
(10). Prepared from compound 8 (275 mg, 0.39 mmol) by a
procedure similar to that used for 7, to give 10 as a white foam
(135 mg, 55%). 1H NMR (CDCl3, 400 MHz): d 7.79–7.28 (m, 8H),
7.03–6.97 (m, 3H), 5.41 (d, 1H, J = 8.4 Hz), 4.65–4.60 (m, 1H),
4.43–4.18 (m, 3H), 3.10–3.07 (m, 2H), 2.81 (t, 2H, J = 8.0 Hz),
2.48 (t, 2H, J = 8.0 Hz), 1.53 (s, 9H), 1.44 (s, 9H). EI-MS: m/z
Cyclo-[Ac-Adi-Leu-(3-CH2CH2CO2Et-Tyr)-Ac6c-Asn-Ava-Cys
(O)-(R)]-amide (3d). ESI-MS, m/z: calc. 1032.4 (M − H)−,
found 1032.4. tR = 14.4 min (10–70% of solvent B in 30 min,
purity 98%); tR = 20.3 min (10–70% of solvent C in 30 min,
1
purity 95%). H-NMR (600 MHz, H2O + D2O): d 8.70 (s, 1H),
8.39 (d, 1H, J = 8.4 Hz), 8.28 (d, 1H, J = 6.0 Hz), 8.11 (d,
1H, J = 7.2 Hz), 7.69 (s, 1H), 7.62 (s, 1H), 7.54 (d, 1H, J =
6.0 Hz), 7.50–7.48 (m, 2H), 7.15 (s, 1H), 6.86 (s, 1H), 6.74 (d, 1H,
J = 6.6 Hz), 6.83 (s, 1H), 6.51 (d, 1H, J = 6.6 Hz), 4.39–4.36
(m, 2H), 4.09–4.01 (m, 3H), 3.60–3.46 (m, 3H), 3.28–3.05 (m,
2H), 3.03–2.71 (m, 10H), 2.28–2.10 (m, 6H), 1.67–1.63 (m, 1H),
1.53–1.44 (m, 4H), 1.47–1.37 (m, 10H), 1.36–1.16 (m, 6H), 0.76
(d, 3H, J = 6.6 Hz), 0.68 (d, 1H, J = 6.6 Hz).
t
(%) 530 (M − CO2 Bu), 178 (100), 161 (12). [a]2D2 = +13.6 (c = 0.5,
CHCl3). Anal. calcd for C36H41NO9: C, 68.45; H, 6.54; N, 2.22.
Found: C, 68.76; H, 6.34; N, 1.79.
3-[4-tert-Butoxycarbonyloxy-3-(2-(ethoxycarbonyl)ethyl)phenyl]-
2-(9H-fluoren-9-ylmethoxycarbonylamino)propionic acid (11).
Prepared from compound 9 (241 mg, 0.35 mmol) by a procedure
similar to that used for 7, to give 11 as a white foam (109 mg, 51%).
1H NMR (CDCl3, 400 MHz): d 7.77–7.29 (m, 9H), 7.03–6.99 (m,
2H), 5.45 (d, 1H, J = 8.0 Hz), 4.71–4.17 (m, 4H), 4.12 (q, 2H,
J = 7.6 Hz), 3.10–3.00 (m, 2H), 2.84 (t, 2H, J = 7.6 Hz), 2.56 (t,
2H, J = 7.6 Hz), 1.46 (s, 9H), 1.24 (t, 3H, J = 7.6 Hz). EI-MS:
m/z (%) 531 (M − CO2Et), 178 (100), 91 (45). [a]2D2 = +15.6 (c =
0.65, CHCl3). Anal. calcd for C34H37NO9: C, 67.65; H, 6.18; N,
2.32. Found: C, 67.95; H, 6.03; N, 1.88.
Cyclo-[Ac-Gla-Leu-(3-CH2CH2CO2H-Tyr)-Ac6c-Asn-Ava-Cys
(O)-(R)]-amide (3e). ESI-MS, m/z: calc. 1035.1 (M − H)−, found
1035.4. tR = 15.2 min (10–70% of solvent B in 30 min, purity 96%);
tR = 22.1 min (10–70% of solvent C in 30 min, purity 96%).
Cyclo-[Ac-Ala-Leu-(3-CH2CH2CO2H-Tyr)-Ac6c-Asn-Ava-Cys
(O)-(R)]-amide (3f). ESI-MS, m/z: calc. 932.4 (M − H)−, found
932.3. tR = 16.7 min (10–70% of solvent B in 30 min, purity 99%);
tR = 24.6 min (10–70% of solvent C in 30 min, purity 95%).
General procedure for the synthesis of peptides 3a–f
All peptides were synthesized manually using standard solid-
phase peptide chemistry with Fmoc-protected amino acids on
Pal resin on a 0.1 mmol scale. HOBt/DIPCDI activation of Na-
protected amino acids was employed for coupling, and 20% piperi-
dine in DMF was used for Fmoc deprotection. NH4Ac/HOAc
buffer was used for backbone cyclization. TFA–TES–H2O (9.5 :
0.25 : 0.25) was used for the resin cleavage and side-chain
deblocking. The oxidation of the thioether linkage into sulfox-
ide was accomplished using 5% aq. H2O2. The final product
was purified by semi-preparative reverse-phase HPLC. HPLC
conditions: Vydac C18 column (20 × 250 mm). Solvent gra-
dient system 1: A, 0.05% TFA in water, B, 0.05% TFA in
90% acetonitrile in water. Solvent gradient system 2: A, 0.05%
TFA in water; C, 0.05% TFA in 90% methanol in water with
gradient indicated below. Flow rate: 2.5 mL min−1. UV detector:
225 nm. ESI-MS was performed on a Finnigan LCQ Deca
mass spectrometer. The purity of products was characterized
by analytical RP-HPLC using two solvent systems: method 1,
gradient 10–70% B over 30 min; method 2, gradient 10–70% C over
30 min.
Acknowledgements
The Chinese Academy of Sciences (No. KSCX1-SW-11) and
the Ministry of Science and Technology of China (No.
2004CB518903) are greatly appreciated for their financial support.
The authors thank Professor Xu Shen and his colleagues of the
DDDC at SIMM for their kind assistance when using their Biacore
3000 instrument.
References
1 E. J. Lowenstein, R. J. Daly, W. L. Batzer, B. Margolis, R. Lanners, A.
Ulrich, E. Y. Skolnick, D. Bar-Sagi and J. Schlessinger, Cell, 1992, 70,
431–442.
2 T. Pawson and J. Schlessinger, Curr. Biol., 1993, 3, 434–442.
3 A. M. Tari, M. C. Hung, K. Li and G. Lopez-Berestein, Oncogene,
1999, 18, 1325–1332.
4 J. V. Soriano, N. F. Liu, Y. Gao, Z.-J. Yao, T. Ishibashi, C. Underhill,
T. R. Burke and D. P. Bottaro, Mol. Cancer Ther., 2004, 3, 1289–
1299.
5 H. Fretz, P. Furet, C. Garcia-Echeverria, J. Rahuel and J. Schoepfer,
Curr. Pharm. Des., 2000, 6, 1777–1796.
6 K. Machida and B. J. Mayer, Biochim. Biophys. Acta, 2005, 1747, 1–25.
7 J. Rahuel, B. Gay, D. Erdmann, A. Strauss, C. Garcia-Echeverria, P.
Furet, G. Caravatti, H. Fretz, J. Schoepfer and M. G. Grutter, Nat.
Struct. Biol., 1996, 3, 586–589.
8 L. Oligino, F.-D. T. Lung, L. Sastry, J. Bigelow, T. Cao, M. Curran,
T. R. Burke, S.-M. Wang, D. Krag, P. P. Roller and C. R. King, J. Biol.
Chem., 1997, 272, 29046–29052.
Cyclo-[Ac-Adi-Leu-Tyr-Ac6c-Asn-Ava-Cys (O)-(R)]-amide (3a).
ESI-MS, m/z: calc. 932.4 (M − H)−, found 932.3. tR = 14.9 min
(10–70% of solvent B in 30 min, purity 98%); tR = 20.9 min (10–
70% of solvent C in 30 min, purity 95%).
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 659–666 | 665
©

View Article Online
9 Y.-Q. Long, J. H. Voigt, F.-D. T. Lung, C. R. King and P. P. Roller,
Bioorg. Med. Chem. Lett., 1999, 9, 2267–2272.
10 Y.-Q. Long, Z.-J. Yao, J. H. Voigt, F.-D. T. Lung, J. H. Luo, T. R. Burke,
C. R. King, D. Yang and P. P. Roller, Biochem. Biophys. Res. Commun.,
1999, 264, 902–908.
11 F.-D. T. Lung, Y.-Q. Long, C. R. King, J. Varady, X.-W. Wu, S.-M.
Wang and P. P. Roller, J. Pept. Res., 2001, 57, 447–454.
12 Y.-Q. Long, F.-D. Lung and P. P. Roller, Bioorg. Med. Chem., 2003, 11,
3929–3936.
20 W. C. Shakespeare, Curr. Opin. Chem. Biol., 2001, 5, 409–415.
21 S.-K. Kang, Z.-D. Shi, K. M. Worthy, L. K. Bindu, P. G. Dharmawar-
dana, S. J. Choyke, D. P. Bottaro, R. J. Fisher and T. R. Burke, Jr.,
J. Med. Chem., 2005, 48, 3945–3948.
22 P. Ettmayer, D. France, J. Gounarides, M. Jarosinski, M.-S. Martin, J.-
M. Rondeau, M. Sabio, S. Topiol, B. Weidmann, M. Zurini and K. W.
Bair, J. Med. Chem., 1999, 42, 971–980.
23 A. C. Wallace, R. A. Laskowski and J. M. Thornton, Protein Eng.,
1995, 8, 127–134.
24 G. M. Morris, D. S. Goodsell, R. S. Halliday, R. Huey, W. E. Hart,
R. K. Belew and A. J. Olson, J. Comput. Chem., 1998, 19, 1639–
1662.
25 G. M. Morris, D. S. Goodsell, R. Huey, W. E. Hart, R. S. Halliday,
R. K. Belew and A. J. Olson, Autodock (Version 3.0.3), Molecular
Graphics Laboratory, Department of Molecular Biology, The Scripps
Research Institute, La Jolla, CA, 1999.
13 Y.-Q. Long, R. Guo, J. H. Luo, D. Yang and P. P. Roller, Biochem.
Biophys. Res. Commun., 2003, 310, 334–340.
14 Y.-L. Song, P. P. Roller and Y.-Q. Long, Bioorg. Med. Chem. Lett., 2004,
14, 3205–3208.
15 T. R. Burke, Jr. and K. Lee, Acc. Chem. Res., 2003, 36, 426–433.
16 Y. Gao, L. Wu, J. H. Luo, R. Guo, D. Yang, Z. Y. Zhang and T. R.
Burke, Bioorg. Med. Chem. Lett., 2000, 10, 923–927.
17 M. E. Jung and T. I. Lazarova, J. Org. Chem., 1997, 62, 1553–1555.
18 E. Atherton and R. C. Sheppard, Solid Phase Peptide Synthesis: A
Practical Approach, IRL, Oxford, 1989.
19 P. Li, M. Zhang, Y.-Q. Long, M. L. Peach, H. Liu, D. Yang, M. Nicklaus
and P. P. Roller, Bioorg. Med. Chem. Lett., 2003, 13, 2173–2177.
26 Sybyl (Version 6.8), Tripos Associates, St. Louis, MO, 2000.
27 S. J. Weiner, P. A. Kollman, D. A. Case, U. C. Singh, C. Ghio, G.
Alagona, S. Profeta and P. Weiner, J. Am. Chem. Soc., 1984, 106, 765–
784.
28 J. Gasteiger and M. Marsili, Tetrahedron, 1980, 36, 3219–3228.
666 | Org. Biomol. Chem., 2006, 4, 659–666
This journal is
The Royal Society of Chemistry 2006
©