Phosphotyrosine-containing dipeptides as high-affinity ligands for the p56(lck) SH2 domain

Article abstract of DOI:10.1021/jm980612i

Src homology-2 (SH2) domains are noncatalytic motifs containing approximately 100 amino acid residues that are involved in intracellular signal transduction. The phosphotyrosine-containing tetrapeptide Ac-pYEEI binds to the SH2 domain of p56(lck) (Lck) with an affinity of 0.1 μM. Starting from Ac-pYEEI, we have designed potent antagonists of the Lck SH2 domain which are reduced in peptidic character and in which the three carboxyl groups have been eliminated. The two C-terminal amino acids (EI) have been replaced by benzylamine derivatives and the pY + 1 glutamic acid has been substituted with leucine. The best C-terminal fragment identified, (S)-1-(4-isopropylphenyl)ethylamine, binds to the Lck SH2 domain better than the C-terminal dipeptide EI. Molecular modeling suggests that the substituents at the 4-position of the phenyl ring occupy the pY + 3 lipophilic pocket in the SH2 domain originally occupied by the isoleucine side chain. This new series of phosphotyrosine-containing dipeptides binds to the Lck SH2 domain with potencies comparable to that of tetrapeptide 1.

Full text of DOI:10.1021/jm980612i

722
J . Med. Chem. 1999, 42, 722-729
P h osp h otyr osin e-Con ta in in g Dip ep tid es a s High -Affin ity Liga n d s for th e p 56lck
SH2 Dom a in
Montse Llina`s-Brunet,* Pierre L. Beaulieu, Dale R. Cameron, J ean-Marie Ferland, J ean Gauthier, Elise Ghiro,
J ames Gillard, Vida Gorys, Martin Poirier, J ean Rancourt, and Dominik Wernic
Boehringer Ingelheim (Canada) Ltd., Bio-Me´ga Research Division, 2100 Cunard Street, Laval, Que´bec H7S 2G5, Canada
Raj Betageri, Mario Cardozo, Scott J akes, Suzanne Lukas, Usha Patel, J ohn Proudfoot, and Neil Moss
Boehringer Ingelheim Pharmaceuticals Inc., 175 Briar Ridge Road, Ridgefield, Connecticut 06877
Received October 30, 1998
Src homology-2 (SH2) domains are noncatalytic motifs containing approximately 100 amino
acid residues that are involved in intracellular signal transduction. The phosphotyrosine-
containing tetrapeptide Ac-pYEEI binds to the SH2 domain of p56lck (Lck) with an affinity of
0.1 µM. Starting from Ac-pYEEI, we have designed potent antagonists of the Lck SH2 domain
which are reduced in peptidic character and in which the three carboxyl groups have been
eliminated. The two C-terminal amino acids (EI) have been replaced by benzylamine derivatives
and the pY + 1 glutamic acid has been substituted with leucine. The best C-terminal fragment
identified, (S)-1-(4-isopropylphenyl)ethylamine, binds to the Lck SH2 domain better than the
C-terminal dipeptide EI. Molecular modeling suggests that the substituents at the 4-position
of the phenyl ring occupy the pY + 3 lipophilic pocket in the SH2 domain originally occupied
by the isoleucine side chain. This new series of phosphotyrosine-containing dipeptides binds
to the Lck SH2 domain with potencies comparable to that of tetrapeptide 1.
Ta ble 1
In tr od u ction
A universal feature of intracellular signal transduc-
tion and transmission in mammalian cells is the tran-
sient association between specific proteins mediated by
distinct functional modules such as Src homology-2
(SH2) domains.1 SH2 domains contain about 100 amino
acid residues and are present in many intracellular
signal transduction proteins.2 Their physiological func-
tion is to bind phosphotyrosine (pY)-containing peptides
which they recognized in the context of immediately
adjacent amino acid sequences.3
The protein p56lck (Lck) is a Src-like lymphocyte-
specific tyrosine kinase that plays a key role in the
initiation of T cell receptor (TCR) signaling events.4 Lck
is also thought to play a crucial role in T cell develop-
ment since mice deficient in Lck or expressing a
dominant-negative mutant form of Lck exhibit a severe
defect in T cell function.5,6 We believe that selective
inhibitors of Lck must be of benefit in the treatment of
T cell leukemias and lymphomas and autoimmune
diseases such as rheumatoid arthritis.
Like other Src kinases, Lck is made up of five
domains: N-terminal, SH3, SH2, kinase, and C-termi-
nal noncatalytic domain.2 The SH2 domain of Lck is
essential for the initiation of signaling events following
TCR stimulation, probably by mediating an interaction
between Lck and the Zap-70 tyrosine kinase and/or the
ú subunit of the T cell receptor.7 In addition to a role in
interacting with other proteins, the SH2 domain of Src-
like kinases also regulates the kinase activity of these
enzymes by intramolecularly binding the negative regu-
latory pY residue located in the C-terminus domain.8
A phosphopeptide library screen used to generate a
preferred binding sequence for the Lck SH2 domain
showed a preference for the sequence Glu-Glu-Ile in the
three positions C-terminal to the phosphotyrosine.9
Crystal structures show that peptides containing the
pYEEI motif bind to the Src and Lck SH2 domains in
an extended conformation10,11 and make the most
extensive contacts through the phosphotyrosine residue
and the isoleucine side chain. The tetrapeptide Ac-
pYEEI (compound 1, Table 1) binds to the Lck SH2
domain with high affinity (KD ) 0.1 µM). However,
tetrapeptide 1, in addition to the phosphate group,
contains three carboxylic acid residues that will be
negatively charged under physiological conditions. In
10.1021/jm980612i CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/04/1999

High-Affinity Ligands for the p56lck SH2 Domain
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 723
Ta ble 2
order for Lck SH2 domain antagonists to be efficacious
in vivo, they have to be able to get inside T cells, and
molecules containing multiple anionic groups such as
carboxylates are not generally considered good at pen-
etrating the lipophilic cell membrane. A series of di- and
tripeptides containing a lipophilic substituent at the
C-terminus have been described as antagonists of other
SH2 domains of the Src family such as the pp60src SH2
domain.12-17 However, the majority of these dipeptide
antagonists still contain the Ac-pYE sequence.13-16 Two
of the studies also describe the replacement of the pY
+ 1 glutamic acid with a non-carboxylic-acid-containing
residue.12,17 This report concentrates on our efforts to
replace the three C-terminal amino acids (EEI) of
tetrapeptide 1 with non-carboxyl-containing substitu-
tions while maintaining inhibitory potency.
Resu lts a n d Discu ssion
The phosphotyrosine-containing tetrapeptide 1 (Table
1) binds with high affinity to the recombinant SH2
domain of Lck (KD ) 0.1 µM).18 A crystal structure of
compound 1 bound to the Lck SH2 domain11 shows the
peptide bound in a relatively extended conformation and
having the following ligand-protein interactions: there
are H-bonds between the protein and the oxygen atom
of the Ac-group (pY - 1) and the hydrogen atom of the
amide bond between phosphotyrosine (pY) and glutamic
acid (pY + 1); the phosphate group is situated in a
highly hydrophilic binding pocket and is involved in an
intricate network of hydrogen-bonding and charge-
charge interactions. The importance of the phosphate
group for binding is evidenced by the fact that Ac-YEEI
has no measurable binding affinity toward the Lck SH2
domain at concentrations up to 1 mM.19 One face of the
aromatic ring of the pY side chain lies against the
aliphatic portion of a Lys side chain. The other face
interacts with the guanidinium group of an Arg residue.
The remaining significant interaction between com-
pound 1 and the SH2 domain is between the isoleucine
side chain and a lipophilic binding pocket in the protein.
Consistent with the X-ray structure of tetrapeptide
1 complexed with the Lck SH2 domain, early structure-
activity relationship (SAR) studies indicated that a wide
range of amino acid substitutions are independently
well-tolerated at both pY + 1 and pY + 2 positions.18
In particular, leucine, phenylalanine, and glutamine,
amino acids with uncharged side chains, are all good
replacements for glutamic acid at pY + 1 or pY + 2.
Also, the isoleucine at pY + 3 can be replaced by a
simple (S)-2-methylbutylamine without loss in binding
affinity. Leucine-containing tetrapeptides Ac-pYLEI (KD
) 0.1 µM) and Ac-pYELI (KD ) 0.1 µM) display similar
binding affinities and a potency comparable to that of
compound 1. However, when two changes are intro-
duced in the same molecule, then a substantial drop in
potency results. Simultaneous introduction of Leu at pY
+ 1 and pY + 2 (Ac-pYLLI) produced a 10-fold drop in
binding affinity toward the Lck SH2 domain. These
early studies on compound 1 suggested that we were
not going to be able to replace all the carboxyl groups
and maintain potency by incorporating lipophilic amino
acid derivatives.
1) with a KD of 27 µM. Compound 2 was used as a
starting point to identify new C-terminal appendages
that could reach and occupy the hydrophobic binding
pocket originally occupied by the isoleucine side chain.
Toward that end, a number of aliphatic and aromatic
residues were introduced at the C-terminus of dipeptide
2.
Table 2 shows the effect of positioning a phenyl ring
at the C-terminus of compound 2. Benzyl derivative 4
attracted our attention because of its simplicity and also
because two carboxylic-acid-containing amino acids (EI)
could be eliminated despite a 15-fold reduction in
potency from tetrapeptide 1. This study provided us
with a new starting point for further SAR. To test the
effect of eliminating the remaining carboxylic acid
residue, a number of amino acids with lipophilic side
chains were introduced at the pY + 1 position. Unfor-
tunately, all the substitutions tried resulted in a sub-
stantial loss in potency. The best lipophilic replacement
for the pY + 1 glutamic acid residue was leucine
(compound 6) which gave a 10-fold decrease in binding
C-Terminal truncation on compound 1 (Table 1)
yielded phosphotyrosine-containing dipeptide 2 (Table

724 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Llina`s-Brunet et al.
Ta ble 3
phenyl ring had no effect on potency. However, when
the methoxy substituent was introduced at the 4-posi-
tion (compound 12), a greater than 10-fold increase in
binding affinity was observed. Interestingly, the ben-
eficial effect of the 4-methoxy substituent was lost when
an additional methoxy group was introduced at the
3-position (compound 13). Because of the observed
increase in potency by the addition of a 4-methoxy
substituent on the C-terminal phenyl ring of our ligands,
we explored additional substitutions at this position.
Compound 14 (Table 3) containing a 4-methyl group had
potency similar to that of compound 12. However, when
the size of the alkyl group was increased to an isopropyl
group (compound 15), the potency of the ligand in-
creased by 3-fold. A further increase in the size of the
alkyl substituent to a tert-butyl group (compound 16)
was detrimental to potency (KD ) 8 µM). The interaction
of the 4-substituent on the aromatic ring with the SH2
domain appears to be very sensitive to small structural
changes as evidenced by the 26-fold loss in potency upon
replacement of the 4-isopropyl group in compound 15
by a 4-dimethylamino group (compound 17, KD ) 20
µM).
We have been unsuccessful in obtaining a crystal
structure of any of this new series of ligands bound to
the Lck SH2 domain. Computational chemistry studies
have been performed with this series of analogues, and
suitable models for protein-ligand complexes have been
derived. In particular, docking and molecular dynamic
(MD) refinement have been carried out for compound
12 and the corresponding pY + 1 glutamic acid analogue
18 which, as expected, binds to the Lck SH2 domain
with higher affinity (KD ) 0.2 µM).
Two low-energy conformations have been derived for
compound 12 bound to the SH2 domain of Lck. The only
difference between these models is in the bound con-
formation adopted by the 4-methoxy substituent (Figure
1). In both models, the C-terminal 4-methoxybenzyl
group is oriented toward the pY + 3 hydrophobic pocket,
with the 4-methoxy substituent located deep in the
pocket. Figure 2 highlights the closest contacts of the
4-methoxyphenyl portion of compound 12 with the
amino acid residues of the protein. In these models, the
aromatic ring is located perpendicular to the plane of
Tyr181, in a favorable position for an edge-to-face
aromatic-aromatic interaction which may explain the
fact that the C-terminal benzyl group was optimal
compared to phenyl and phenethyl (Table 2). Moreover,
the methyl group of the 4-methoxy substituent is in van
der Waals contact with Ile193, Ser194, Leu216, and
Gly215. The binding models shown in Figure 2 are
consistent with most of the SAR reported in Table 3 and
explain the importance of the methoxy substituent at
the 4-position of the aromatic ring. Since the two models
differ primarily in the positioning of the 4-methoxy
substituent, there is evidently enough space to accom-
modate larger groups consistent with the fact that
compound 15, the 4-isopropyl analogue, has slightly
better affinity. The increase in the hydrophobic contacts
for the 4-isopropyl group may explain the enhancement
of the binding affinity of compound 15. In addition, a
larger substituent such as 4-tert-butyl (compound 16)
and disubstitued analogues such as the 3,4-dimethoxy
affinity compared to 4. Compound 6 with a KD of 16 µM,
but containing no carboxyl groups, became a new frame
of reference for further optimization.
Va r ia tion in th e Na tu r e of RHS. Initial investiga-
tions showed that replacement of the C-terminal phenyl
ring in compound 6 by a pyridyl derivative (compounds
7-9, Table 3) had little effect on potency. Likewise,
introduction of a methoxy substituent at the 2- or
3-position (compounds 10 and 11, respectively) of the

High-Affinity Ligands for the p56lck SH2 Domain
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 725
F igu r e 1. Two models of compound 12 bound to the p56lck SH2 domain. The binding pocket of the SH2 domain is highlighted.
The pY binding pocket is located between loop BC and sheet âD. The pY + 3 binding pocket is located between loops BG and EF.
The protein backbone is highlighted as a blue ribbon with no side chains shown. Two superimposed positions for ligand 12 are
shown, and the ligands are colored by atom type as follows: carbon, gray; oxygen, red; nitrogen, dark blue; phosphorus, violet;
hydrogen, white.
F igu r e 3. Model of compound 18 bound to the p56lck SH2
domain. The protein and compound 18 are colored as in Figure
2. Hydrogen bonds between the pY + 1 carboxylate and protein
side chains are shown as dashed lines.
F igu r e 2. 4-Methoxylbenzylamine binding pocket for two
bound models of compound 12. The protein backbone is
well as in MD simulations, the pY + 1 Glu is not
involved in hydrogen bonding with Lys179 or Tyr181.
depicted as a ribbon and is colored violet where direct contact
with ligand 12 is seen. A potential edge-face aromatic
interaction is shown as a dotted line. Protein side chains and
the ligand 12 are colored as in Figure 1.
analogue 13 appear too bulky to properly bind in the
hydrophobic pocket.
Our computational study also provides a rational for
the significant difference observed in binding affinity
between compounds 12 and the corresponding pY + 1
glutamic acid derivative 18. Compound 18 is a much
better ligand than the corresponding leucine analogue
12. Molecular dynamic results indicate that the car-
boxylic group of pY + 1 Glu in compound 18 is able to
form hydrogen bonds to Lys179 and Tyr181 (Figure 3).
The lack of these hydrogen bonds for pY + 1 Leu may
explain the 10-fold difference in binding affinity between
these analogues. Interestingly, as mentioned previously
for tetrapeptide 1, this difference in binding affinity
between Glu and Leu at pY + 1 position is not observed,
and in the crystal structure of the 1-SH2 complex, as
A possible explanation for this difference between the
tetrapeptide and dipeptide series is that in the former,
the pY + 1 glutamic acid side chain and its carbonyl
backbone have some water-mediated interactions with
amino acid residues on the side of the protein opposite
to Lys179 and Tyr181. This network of polar interac-

726 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Llina`s-Brunet et al.
Ta ble 4
fold less potent than the corresponding unsubstituted
analogue. When an (S)-R-methyl group was introduced
along with a 4-isopropyl group in the C-terminal ben-
zylamine (compound 22), a 4-fold increase in potency
was observed compared to unsubstituted analogue 15
(Table 3). The increase in potency that is gained from
R-substitution is dependent on the configuration but is
independent of the nature or size of the group at the
R-position. Indeed, similar results have been obtained
with groups such as hydroxymethyl and phenyl with the
appropriate configuration (data not shown). With com-
pound 22 (KD ) 0.2 µM) we succeeded in eliminating
all the carboxyl groups in the molecule while maintain-
ing a potency comparable to that of our original lead
tetrapeptide compound 1 (KD ) 0.1 µM).
Con clu sion
This report reveals a new series of Lck SH2 domain
ligands that do not contain any carboxylic acid groups.
The starting point in our studies was Ac-pYEEI (com-
pound 1) which in addition to the phosphotyrosine
contains three carboxylic-acid-containing residues. We
have demonstrated that the two C-terminal residues in
1 (EI) could be replaced by 4-methoxybenzylamine.
Compound 18 containing a single carboxylic acid group
and a 4-methoxybenzylamine has a binding affinity
toward the Lck SH2 domain almost equivalent to that
of tetrapeptide 1. The last carboxylic-acid-containing
residue, pY + 1 Glu, was replaced by leucine with a 10-
fold loss in potency. However, potency could be regained
by introducing an isopropyl substituent at the 4-position
of the C-terminal phenyl ring and a configurationally
defined methyl group at the R-position of the benzyl-
amine. Compound 22 containing an (S)-1-(4-isopropyl-
phenyl)ethylamine has a binding potency of 0.2 µM and
is thus almost equipotent to the initial tetrapeptide 1.
More significantly, the three carboxylic acid residues
have been eliminated and the peptidic character of the
ligand has been reduced. Studies directed toward find-
ing a metabolically stable and less charged replacement
for the phosphate group in this series of ligands will be
reported shortly.
Exp er im en ta l Section
Gen er a l. NMR spectra were recorded on a Bruker AMX400
(400 MHz for 1H NMR) spectrometer and were referenced to
TMS as an internal standard (0 ppm δ scale). Data are
reported as follows: chemical shift (ppm), multiplicity (s )
singlet, d ) doublet, t ) triplet, q ) quartet, br ) broad, m )
multiplet), coupling constant (J , reported to the nearest 0.5
Hz), and integration. FAB mass spectra were obtained from a
MF 50 TATC instrument operating at 6 kV and 1 mA by using
thioglycerol or nitrobenzyl alcohol (NBA) as a matrix support.
HPLC homogeneities were measured by using an analytical
C18 reversed-phase column and two different eluting sys-
tems: method a, 0.06% TFA in water-0.06% TFA in aceto-
nitrile gradient (20-100% acetonitrile over 30 min); method
b, 20 mM aqueous Na2HPO4 (pH 7.4)-acetonitrile gradient
(20-75% acetonitrile over 25 min).
Ma ter ia ls. N-Acetyl-L-tyrosine, N-Fmoc-L-tyrosine, common
amines, and N-Boc-L-amino acids were obtained from com-
mercial sources. The synthesis of N-Fmoc-L-Tyr(PO3Bn2)-OH
was done as described in ref 20.
(S)-1-(4-Isop r op ylp h en yl)eth yla m in e Hyd r och lor id e
Sa lt (26).21 The enantioselective synthesis is outlined in
Scheme 1. (S)-(-)-4-Isopropyl-2-oxazolidinone (2.7 g, 21 mmol)
was dissolved in dry THF (40 mL), and the resultant solution
was cooled to -40 °C under an argon atmosphere. n-BuLi (1.6
M in hexane, 13 mL, 21 mmol) was added dropwise, and the
tions may prevent the tetrapeptide from moving in the
direction of Lys179 and Tyr181, and the pY + 1 Glu is
too far from any hydrogen bond donor groups on the
protein surface.
These molecular modeling studies also suggested that
alkylation of the amido or the adjacent methylene group
at the C-terminus of compound 12 should not interfere
with binding to the Lck SH2 domain. In addition, these
modifications have been reported to increase potency
in a similar series of peptidomimetic ligands for the
pp60src SH2 domain.17 In our series of ligands, N-
methylation of the C-terminal amide bond (compound
19, Table 4) was well-tolerated but did not improve
potency over the unsubstituted analogue. A configura-
tionally defined methyl group was also introduced at
the R-position of the benzyl amide. Compound 20 having
an R-methyl group with the S-configuration (KD ) 1.5
µM) was slightly more potent than the corresponding
unsubstituted analogue (compound 14, KD ) 2.5 µM).
On the other hand, compound 21 having the R-methyl
group with the R-configuration (KD ) 27 µM) was 10-

High-Affinity Ligands for the p56lck SH2 Domain
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 727
Removal of the chiral auxiliary was accomplished by treat-
ing a solution of compound 23 (2.6 g, 8.5 mmol) in water (15
mL)-THF (40 mL) with LiOH monohydrate (0.40 g, 10 mmol)
and H2O2 (30% aqueous solution, 4.75 mL, 42.2 mmol). After
the mixture was stirred at ambient temperature for 3 h,
aqueous Na2S2O3 (5.3 g, 37 mmol in 10 mL of water) and
saturated aqueous NaHCO3 (5 mL) were added. The THF was
removed under vacuum, and the basic aqueous solution was
washed with CH2Cl2. The aqueous layer was acidified with
10% aqueous HCl and extracted with ethyl acetate (2 × 50
mL). The organic solution was washed with brine (75 mL),
dried (MgSO4), and concentrated to afford the desired acid 24
Sch em e 1
as a white solid (1.6 g, quantitative yield): mp 55-56 °C; [R]22
D
1
+65.3° (c 1.35, MeOH); H NMR (400 MHz, CDCl3) δ 7.24 (d,
J ) 8 Hz, 2 H), 7.18 (d, J ) 8 Hz, 2 H), 3.71 (q, J ) 7 Hz, 1 H),
2.94-2.82 (m, 1 H), 1.50 (d, J ) 7 Hz, 3 H), 1.23 (d, J ) 7 Hz,
6 H); HRMS calcd for
C
12H17O2 (MH+) 193.1222, found
193.1228. Anal. Calcd for C12H16O2: C, 74.97; H, 8.39. Found:
C, 74.88; H, 8.78.
The conversion of carboxylic acid 24 to amine 26 was
achieved via a Curtius rearrangement:22 a solution of acid 24
(0.24 g, 1.2 mmol), diphenylphosphorylazide (DPPA) (0.29 mL,
1.3 mmol), and Et3N (0.19 mL, 1.4 mmol) in THF (5 mL) was
heated at reflux for 2 h. Benzyl alcohol (0.38 mL, 3.7 mmol)
was added, and the mixture was heated to reflux for an
additional 16 h. The reaction mixture was allowed to cool to
ambient temperature, and ethyl acetate (100 mL) was added.
The organic solution was washed with 10% aqueous citric acid
(2 × 50 mL), saturated aqueous NaHCO3 (2 × 50 mL), and
brine (50 mL), dried (MgSO4), filtered, and concentrated. Flash
chromatography of the residue on silica gel eluting with
hexanes-ethyl acetate (8:2) gave carbamate 25 as a white solid
(0.22 g, 60% yield): mp 95-96 °C; [R]22D -52.6° (c 1.12, MeOH);
1H NMR (400 MHz, CDCl3) δ 7.35-7.25 (m, 4 H), 7.23-7.17
(m, 5 H), 5.11 (d, J ) 12 Hz, 1 H), 5.06 (d, J ) 12 Hz, 1 H),
4.82-4.88 (m, 1 H), 2.93-2.86 (m, 1 H), 1.48 (d, J ) 6.5 Hz, 3
H), 1.24 (d, J ) 7 Hz, 6 H); HRMS calcd for C19H24NO2 (MH+)
298.1807, found 298.1790. Anal. Calcd for C19H23NO2: C,
76.73; H, 7.79; N, 4.71. Found: C, 76.52; H, 8.02; N, 4.90.
resultant solution was stirred for 30 min at -40 °C. The
reaction mixture was cooled to -78 °C, and a solution of (4-
isopropylphenyl)acetyl chloride (prepared from the correspond-
ing carboxylic acid and oxalyl chloride under standard condi-
tions) in THF (10 mL) was added dropwise. The mixture was
stirred at -78 °C for 30 min and at 0 °C for 1 h. A saturated
aqueous solution of NH4Cl (10 mL) was added, and the
volatiles were removed under vacuum. The resulting solution
was diluted with ethyl acetate (200 mL) and washed with 10%
aqueous citric acid (2 × 100 mL), saturated aqueous NaHCO3
(2 × 100 mL), and brine (100 mL). The organic layer was dried
(MgSO4), filtered, and concentrated. Flash chromatography of
the residue on silica gel eluting with hexanes-ethyl acetate
(7:3) gave the desired acyloxazolidinone derivative as a yellow
To a solution of benzyl carbamate 25 (1.09 g, 3.7 mmol) in
EtOH (20 mL) was added 10% Pd/C (100 mg), and the mixture
was stirred under 1 atm of H2 for 45 min. The catalyst was
removed by filtration through a Millex HV 0.45-µm filter, and
to the clear solution was added 1 N HCl solution in ether (7.3
mL). The mixture was concentrated to yield the desired amine
hydrochloride salt 26 as an off-white solid (0.68 g, 82% yield):
mp 180 °C dec; [R]22D -7.5° (c 2.20, MeOH); 1H NMR (400 MHz,
CDCl3) δ 8.67 (br s, 3 H), 7.39 (d, J ) 8 Hz, 2 H), 7.21 (d, J )
8 Hz, 2 H), 4.40-4.25 (m, 1 H), 2.91-2.84 (m, 1 H), 1.63 (d, J
solid (4.9 g, 80% yield): mp 54.5-55.5 °C; [R]25 +61° (c 1.14,
D
MeOH); 1H NMR (400 MHz, CDCl3) δ 7.23 (d, J ) 8 Hz, 2 H),
7.20 (d, J ) 8 Hz, 2 H), 4.45 (td, J ) 8 and 4 Hz, 1 H), 4.30 (d,
J ) 15.5 Hz, 1 H), 4.25 (t, J ) 8 Hz, 1 H), 4.22 (d, J ) 15.5 Hz,
1 H), 4.19 (dd, J ) 8 and 3 Hz, 1 H), 2.92-2.85 (m, 1 H), 2.39-
2.31 (m, 1 H) 1.23 (d, J ) 7 Hz, 6 H), 0.87 (d, J ) 7 Hz, 3 H),
0.79 (d, J ) 7 Hz, 3 H); HRMS calcd for C17H24NO3 (MH+)
290.1756, found 290.1741. Anal. Calcd for C17H24NO3: C,
70.56; H, 8.01; N, 4.84. Found: C, 70.37; H, 8.16; N, 4.89.
) 7 Hz, 3 H), 1.22 (d, J ) 7 Hz, 6 H); HRMS calcd for C11H18
N
(MH+) 164.1439, found 164.1448.
N-Acetyl-L-Tyr (P O3Bn 2)-OH (27). To a solution of N-
acetyl-L-tyrosine (5.0 g, 22 mmol) in THF (150 mL) were added
tert-butyldimethylsilyl chloride (3.4 g, 22 mmol) and N-
methylmorpholine (2.5 mL, 22 mmol). After 5 min, 1 H-
tetrazole (6.3 g, 90 mmol) and dibenzyldiisopropylamine
phosphoramidite (11.3 mL, 33 mmol) were added, and the
solution was stirred for an additional 3 h. The reaction mixture
was cooled to 0 °C, and 14% aqueous tert-butyl hydroperoxide
(16.0 mL, 51 mmol) was added. After 30 min at 0 °C and 1 h
at ambient temperature, the solution was cooled again to 0
°C, and 10% aqueous sodium thiosulfate (200 mL) was added.
The solution was rapidly stirred for 1 h at ambient tempera-
ture followed by removal of THF under reduced pressure and
addition of ether-ethyl acetate (1:1, 500 mL). The organic
solution was extracted with saturated aqueous NaHCO3 (2 ×
250 mL). The aqueous layer was acidified with 6 N HCl and
extracted with ethyl acetate (2 × 200 mL). The organic solution
was washed with brine, dried (MgSO4), and concentrated to
give the desired compound 27 as a colorless foam (9.2 g, 85%
yield): [R]25D +74° (c 1.37, CHCl3); 1H NMR (400 MHz, DMSO-
d6) δ 8.17 (d, J ) 8 Hz, 1 H), 7.40-7.32 (m, 10 H), 7.22 (d, J
) 8.5 Hz, 2 H), 7.09 (d, J ) 8.5 Hz, 2 H), 5.15 (s, 2 H) 5.13 (s,
2 H), 4.41-4.36 (m, 1 H), 3.02 (dd, J ) 14, 5 Hz, 1 H), 2.82
A solution of lithium hexamethyldisilazane in THF (1 M,
15.1 mL) was slowly added to a -78 °C solution of the above
compound (4.0 g, 13 mmol) in THF (100 mL). After the solution
was stirred at -78 °C for 30 min, MeI (0.85 mL, 13 mmol)
was added and the mixture was allowed to slowly warm to
ambient temperature. After the reaction mixture was stirred
for 2 h, 10% aqueous citric acid was added (150 mL). The THF
was removed under vacuum, and the aqueous solution was
extracted with ethyl acetate (2 × 100 mL). The organic phase
was washed with saturated aqueous NaHCO3 (2 × 100 mL)
and brine (100 mL), dried (MgSO4), and concentrated. Flash
chromatography of the residue on silica gel eluting with
hexanes-ethyl acetate (8:2) gave the desired compound 23 as
a yellow solid (3.0 g, 71% yield): mp 86-87 °C; [R]22D +137.5°
(c 0.96, MeOH); 1H NMR (400 MHz, CDCl3) δ 7.26 (d, J ) 8.5
Hz, 2 H), 7.15 (d, J ) 8.5 Hz, 2 H), 5.11 (q, J ) 7 Hz, 1 H),
4.37-4.33 (m, 1 H), 4.16-4.13 (m, 2 H), 2.92-2.82 (m, 1 H),
2.47-2.31 (m, 1 H) 1.50 (d, J ) 7 Hz, 3 H), 1.22 (d, J ) 7 Hz,
6 H), 0.92 (d, J ) 7 Hz, 3 H), 0.91 (d, J ) 7 Hz, 3 H); HRMS
calcd for C18H26NO3 (MH+) 304.1912, found 304.1899. Anal.
Calcd for C18H25NO3: C, 71.26; H, 8.31; N, 4.26. Found: C,
70.81; H, 8.55; N, 4.56.

728 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4
Llina`s-Brunet et al.
(dd, J ) 14, 10 Hz, 1 H), 1.78 (s, 3 H); HRMS calcd for C25H27
-
Sch em e 2
NO7P (MH+) 484.1525, found 484.1538.
In h ibitor Syn th esis. 1. Syn th esis of In h ibitor s in
Ta bles 1 a n d 2 (excep t Com p ou n d 6) a n d Com p ou n d 18.
All these inhibitors were prepared by sequentially coupling
simple amines, N-Boc-amino acid derivatives, or N-Fmoc-Tyr-
(PO3Bn)-OH from C- to N-terminus by using benzotriazol-1-
yl-1,1,3,3-tetramethyluronium tetrafluoroborate (TBTU) as the
coupling agent. Removal of the N-Boc protective group was
effected with 4 N HCl in dioxane. The N-Fmoc protective group
was removed by treatment with DBU in DMF. The following
procedure is representative.
To a solution of N-Boc-amino acid or N-Fmoc-Tyr(PO3Bn)-
OH (1 mmol) in dry acetonitrile (2.5 mL) were added TBTU
(1 mmol) and N-methylmorpholine (1 mmol). After ca. 5 min,
this solution was added to a solution of amino ester hydro-
chloride salt, peptide hydrochloride salt, or C-terminal amine
(1 mmol) in dry acetonitrile (2.5 mL) containing N-methyl-
morpholine (2 mmol for the hydrochloride salts or 1 mmol for
free amines). The reaction mixture was stirred at ambient
temperature for 2-6 h (reaction monitored by TLC) and then
poured into a mixture of EtOAc (50 mL) and saturated aqueous
NaHCO3 (50 mL). The organic phase was further washed with
saturated aqueous NaHCO3 (50 mL), 1 N aqueous HCl (2 ×
50 mL), and brine (50 mL). Drying (MgSO4), filtration, and
concentration provided the peptide, usually of sufficient purity
to continue to the next step without further purification. N-Boc
or N-Fmoc derivatives could be purified if necessary by
conventional flash chromatography. The N-Boc-peptide prod-
uct (1 mmol) was then treated with 4 N HCl in dioxane (5
mL) for 45 min. The solvent was removed under vacuum, and
the resulting hydrochloride salt was subjected to high vacuum
before use in the next coupling reaction. The N-Fmoc-peptide
product was treated with DBU (1.2 mmol) in DMF (5 mL) for
15 min. Saturated aqueous NaHCO3 (4 mL) and ethyl acetate
(10 mL) were added. The organic phase was dried (MgSO4),
filtered, and concentrated under vacuum, and the resulting
amine was used in the next coupling reaction without further
purification.
Subsequently, the protected peptide derivatives were puri-
fied by flash chromatography, and the benzyl-protecting groups
of the phosphotyrosine residue and other acid residues were
removed by catalytic hydrogenolysis using 10% P/C (10 wt %)
in ethanol and 1 atm of H2 for 2-3 h (reaction monitored by
HPLC). The resultant product was usually obtained in greater
than 95% purity (HPLC and NMR), but if necessary it was
purified by preparative HPLC on a C-18 reversed-phase
column (Vydac, 15-µm particle size) eluting with 0.06% aque-
ous TFA-0.06% TFA in acetonitrile gradients.
2. Syn th esis of In h ibitor s in Ta bles 3 a n d 4. These
inhibitors were prepared by solution coupling of the corre-
sponding C-terminal amine to a preformed dipeptide fragment
as outlined in Scheme 2. The acetylated dipeptide was
synthesized as follows: to a solution of Ac-Tyr(PO3Bn2)-OH
(27) (4.65 g, 9.6 mmol) in THF (30 mL) were added 1-hydroxy-
benzotriazole‚H2O (1.49 g, 11.1 mmol) and 1,3-dicyclohexyl-
carbodiimide (2.18 g, 10.6 mmol), and the mixture was stirred
at ambient temperature for 30 min. A suspension of leucine
allyl ester hydrochloride salt (28) and N-methylmorpholine (1.1
mL, 9.3 mmol) in THF (30 mL) were added to the above
mixture. After stirring at ambient temperature for 4 h, the
mixture was filtered, the filtrate concentrated, and the residue
dissolved in ethyl acetate (75 mL). The organic solution was
washed with 10% aqueous citric acid (2 × 50 mL), saturated
aqueous NaHCO3 (2 × 50 mL), and brine (50 mL), dried
(MgSO4), and concentrated. Flash chromatography of the
residue on silica gel eluting with a gradient from hexanes-
ethyl acetate (7:3) containing 1% MeOH to ethyl acetate-1%
MeOH gave the desired dipeptide 29 as a colorless gum (4.94
g, 80% yield): 1H NMR (400 MHz, CDCl3) δ 7.38-7.30 (m, 10
H), 7.14 (d, J ) 8.5 Hz, 2 H), 7.07 (d, J ) 8.5 Hz, 2 H), 6.23 (d,
J ) 8 Hz, 1 H), 6.10 (d, J ) 7.5 Hz, 1 H), 5.95-5.85 (m, 1 H),
5.31 (br d, J ) 17.5 Hz, 1 H), 5.25 (br d, J ) 10.5 Hz, 1 H),
5.12 (s, 2 H), 5.10 (s, 2 H), 4.66-4.51 (m, 4 H), 3.08-2.98 (m,
2 H), 1.96 (s, 3 H), 1.68-1.47 (m, 3 H), 0.90 (d, J ) 6 Hz, 6 H);
HRMS calcd for C34H42N2O8P (MH+) 637.2679, found 637.2693.
At 0 °C, to a solution of palladium tetrakis(triphenylphos-
phine) (Pd(PPh3)4) (0.18 g, 0.15 mmol) in acetonitrile-meth-
ylene chloride (1:1, 5 mL) was added pyrrolidine (0.78 mL),
and the solution was stirred for 5 min. A suspension of allyl
ester 29 in acetonitrile-methylene chloride (1:1, 35 mL) was
slowly added to the above solution. After stirring the mixture
at ambient temperature for 3 h, the volatiles were removed
under vacuum. The residue was dissolved in saturated aqueous
NaHCO3 (60 mL), and the basic solution was washed with
Et2O (2 × 50 mL) and then acidified with 10% aqueous HCl.
The acidic solution was extracted with ethyl acetate (2 × 75
mL). The organic layer was washed with brine (100 mL), dried
(MgSO4), and concentrated to give the desired carboxylic acid
as an off-white solid: mp 130-138 °C dec; 1H NMR (400 MHz,
CDCl3) δ 7.38 (br s, 10 H), 7.14 (d, J ) 8 Hz, 2 H), 6.97 (d, J
) 8 Hz, 2 H), 6.56 (d, J ) 8 Hz, 1 H), 5.68 (d, J ) 9 Hz, 1 H),
5.25-5.14 (m, 4 H), 4.54-4.49 (m, 1 H), 4.42-4.38 (m, 1 H),
3.18 (dd, J ) 13 and 5.5 Hz, 1 H), 2.72 (t, J ) 11.5 Hz, 1 H),
4.66-4.51 (m, 4 H), 2.03 (s, 3 H), 1.59-1.52 (m, 1 H), 1.48-
1.33 (m, 2 H), 0.89 (d, J ) 6.5 Hz, 3 H), 0.88 (d, J ) 6.5 Hz, 3
H); HRMS calcd for C31H38N2O8P (MH+) 597.2382, found
597.2382.
The following procedure is representative for the synthesis
of the C-terminal amides: At -30 °C, to a solution of the above
carboxylic acid (0.15 mmol) in THF (3 mL) were added
N-methylmorpholine (0.30 mmol) and isobutyl chloroformate
(0.18 mmol). The solution was stirred at -30 °C for 30 min,
the corresponding C-terminal amine was added (0.15 mmol),
and the stirring was continued for an additional hour. A 10%
aqueous solution of citric acid (5 mL) was added, and the
mixture was warmed to ambient temperature. The mixture
was diluted with EtOAc (30 mL) and washed with 10%
aqueous citric acid (2 × 20 mL), saturated aqueous NaHCO3
(2 × 20 mL), and brine (20 mL). The organic solution was dried
(MgSO4) and concentrated. The protected peptide was purified
by flash. Removal of benzyl-protecting groups and purification

High-Affinity Ligands for the p56lck SH2 Domain
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 4 729
(5) Molina, T. J .; Kishihara, K.; Siderovski, D. P.; van Ewijk, W.;
Narendran, A.; Timms, E.; Wakeham, A.; Paige, C. J .; Hartman,
K.-U.; Veillette, A.; Davison, D.; Mark, T. W. A Dominant-
Negative Transgene Defines a Role for p56lck in Thymopoiesis.
Nature 1992, 357, 161-164.
(6) Levin, D. S.; Anderson, S. J .; Forbush, K. A.; Perlmutter, R. M.
Profound Block in Thymocyte Development in Mice Lacking
p56lck. EMBO J . 1993, 12, 1671-1680.
of final compounds were carried out as described in the
previous example.
3. In h ibitor Ch a r a cter iza tion a n d P u r ity. All peptide
derivatives showed satisfactory 400 MHz 1H NMR spectra,
FAB mass spectra (M+ + H) and/or (M+ + Na), amino acid
analyses including peptide recovery, and HPLC purity in two
solvent systems (>95%).
Molecu la r Mod elin g. Compounds 12 and 18 were modeled
using the crystal structure of the SH2 Lck Ac-Tyr(P)-Glu-Glu-
Ile complex11 as starting geometry. The corresponding pY + 1
residue and the 4-methoxybenzyl substituents were initially
positioned in the protein surface using manual docking
procedures with the hydrophobic C-terminus located in a
similar spatial position as the pY + 3 Ile side chain. Subse-
quently, a quenching molecular dynamics (QMD) protocol was
used to explore alternative docking orientations. The QMD was
performed using the program CharmM23 with a production run
of 100 ps and a temperature of 600 K. Following the QMD, a
sampling of the molecular dynamics trajectory was performed
at every picosecond, and each configuration was energy-
minimized using conjugate gradients protocol. The electrostatic
Coulombic potential was computed using a distance-dependent
dielectric constant of 4 with CharmM charges for the protein
and MNDO ESP charges for the ligand, using MOPAC.24 After
energy minimization, a cluster analysis of the ligand atoms
was carried out in coordinate space, and the lowest intermo-
lecular energy minimum, for each cluster, was selected as
representative of the cluster. Analysis of the structure for the
final complex of each cluster obtained identified two suitable
protein-inhibitor models for compound 12. These models were
further optimized by performing 400 ps of MD with explicit
solvent representation, using a TIP3P25 solvent cap of 25 Å
around the inhibitor. In this MD protocol, all atoms within 13
Å radius around the inhibitor were allowed to move freely.
The rest of the protein was tethered as follows: (1) the protein
was constrained using a potential harmonic constraint of 5
kcal (Å2)-1; (2) the remaining solvent molecules were main-
tained within the sphere limit by applying the CharmM DROP
restraint option.
(7) Strauss, D. B.; Chan, A. C.; Patai, B.; Weiss, A. SH2 Domain
Function Is Essential for the Role of the Lck Tyrosine Kinase
in T Cell Receptor Signal Transduction. J . Biol. Chem. 1996,
271, 9976-9981.
(8) Couture, C.; Songyang, Z.; J ascur, T.; Williams, S.; Tailor, P.;
Cantley, L. C.; Mustelin, T. Regulation of the Lck SH2 Domain
by Tyrosine Phosphorylation. J . Biol. Chem. 1996, 271, 24880-
24884.
(9) 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.
(10) Eck, M. J .; Atwell, S. K.; Shoelson, S. E.; Harrison, S. C. Crystal
Structure of the Regulatory Domains of the Src Family Tyrosine
Kinase P56lck. Nature (London) 1994, 368, 764-769.
(11) Tong, L.; Warren, T. C.; King, J .; Betageri, R.; Rose, J .; J akes,
S. Crystal Structures of the Human p56lck SH2 Domain in
Complex with Two Short Phosphotyrosyl Peptides at 1.0 and
1.8 Å Resolution. J . Mol. Biol. 1996, 256, 601-610.
(12) Pacofsky, G. J .; Lackey, K.; Alligood, K. J .; Berman, J .; Charifson,
P. S.; Crosby, R. M.; Dorsey, G. F. J r.; Feldman P. L.; Gilmer,
T. M.; Hummel, C. W.; J ordan, S. R.; Mohr, C.; Shewchuk, L.
M.; Sternbach, D. D.; Rodriguez, M. Potent Dipeptide Inhibitors
of the pp60c-src SH2 Domain. J . Med. Chem. 1998, 41, 1894-
1908.
(13) Plummer, M. S.; Lunney, E. A.; Para, K. S.; Vara Prasad, J . V.
N.; Shahripour, A.; Singh, J .; Stankovic, C. J .; Humblet, C.;
Fergus, J . H.; Marks, J . S.; Sawyer, T. K. Hydrophobic D-Amino
Acids in the Design of Peptide Ligands for the pp60Src SH2
Domain. Drug Des. Discovery 1996, 13, 75-81.
(14) Plummer, M. S.; Holland, D. R.; Shahripour, A.; Lunney, E. A.;
Fergus, J . H.; Marks, J . S.; McConnell, P.; Mueller, W. T.;
Sawyer, T. K. Design, Synthesis, and Cocrystal Structure of a
Nonpeptide Src SH2 Domain Ligand. J . Med. Chem. 1997, 40,
3719-3725.
(15) Shahripour, A.; Para, K. S.; Plummer, M. S.; Lunney, E. A.;
Holland, D. R.; Rubin, J . R.; Humblet, C.; Fergus, J . H.; Marks,
J . S.; Saltiel, A. R.; Sawyer, T. K. Structure-Based Design of
Novel, Dipeptide Ligands Targeting the pp60src SH2 Domain.
Bioorg. Med. Chem. Lett. 1997, 7, 1107-1112.
(16) Rodriguez, M.; Crosby, R.; Alligood, K.; Gilmer, T.; Berman, J .
Tripeptides as Selective Inhibitors of Src-SH2 Phosphoprotein
Interactions. Lett. Pept. Sci. 1995, 2, 1-6 and references therein.
(17) Plummer, M. S.; Lunney, E. A.; Para, K. S.; Shahripour, A.;
Stankovic, C. J .; Humblet, C.; Fergus, J . H.; Marks, J . S.;
Herrera, R.; Hubbell, S.; Saltiel, A.; Sawyer, T. K. Design of
Peptidomimetic Ligands for the pp60src SH2 Domain. Bioorg.
Med. Chem. 1997, 7, 41-47.
Lck SH2 Dom a in Bin d in g Assa y. The binding affinities
of these ligands for the Lck SH2 domain were measured in a
competitive binding assay. The protein used is a glutathione-
S-transferase (GST) fusion protein, and the binding constant
is determined using surface plasmon resonance according to
a published protocol.15 The reported KD values are the mean
of at least four separate determinations with a standard
deviation of (20%.
Ackn owledgm en t. We are grateful to Colette Bouch-
er and Serge Valois for analytical support and to J anice
Kelland for critical review of the manuscript.
(18) Morelock, M. M.; Ingraham, R. H.; Betageri, R.; J akes, S.
Determination of Receptor-Ligand Kinetics and Equilibrium
Binding Constants Using Surface Plasmon Resonance: Applica-
tion to the Lck SH2 Domain and Phosphotyrosyl Peptides. J .
Med. Chem. 1995, 38, 1309-1318.
Su p p or tin g In for m a tion Ava ila ble: Full tabulation of
1H NMR, FAB mass spectra, amino acid analyses, and HPLC
purity data for new inhibitors. This material is available free
of charge via the Internet at http://pubs.acs.org.
(19) Unpublished results.
(20) Perich, J . W.; Reynolds, E. C. The Facile One-Pot Synthesis of
NR-(9-Fluorenylmethoxycarbonyl)-O-(O′,O”-Dialkylphosphoro)-
L-Tyrosines Using Dialkyl N,N-Diethylphophoramidites. Synlett
1991, 577-578.
(21) Saigo, K.; Kai, M.; Yonezawa, N.; Hasegawa, M. Synthesis and
Optical Resolution of 1-(4-Isopropylphenyl)-Ethylamine. Syn-
thesis 1985, 2, 214-216.
(22) Capson, T. L.; Poulter, C. D. A Facile Synthesis of Primary
Amines from Carboxylic Acids by the Curtius Rearrangement.
Tettrahedron Lett. 1984, 3515-3518.
(23) Molecular Simulations Inc., Scranton Rd, San Diego, CA 92121-
3752.
Refer en ces
(1) Beattie, J . SH2 Domain Protein Interaction and Possibilities for
Pharmacological Intervention. Cell. Signal 1996, 8, 75-86.
(2) Koch, C. A.; Anderson, D.; Moran, M. F.; Ellis, C.; Pawson, T.
SH2 and SH3 Domains: Elements that Control Interactions of
Cytoplasmic Signaling Proteins. Science 1991, 252, 668-674.
(3) Songyang, Z.; 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.;
Hidesaburo, H.; Schaffhausen, B.; Cantley, L. C. SH2 Domains
Recognize Specific Phosphopeptide Sequences. Cell 1993, 72,
767-778.
(24) Steward, J . J . P. MOPAC Manual, MOPAC 6.0; Frank J . Seiler
Research Laboratory, United State Air Force Academy, 1990.
(25) J orgensen, W. L.; Briggs, J . M.; Contreras, M. L. J . Phys. Chem.
1990, 94, 1683-1687.
(4) Weil, R.; Veillette, A. Signal Transduction by the Lymphocyte-
Specific Tyrosine Protein Kinase p56lck. Curr. Top. Microbiol.
Inmunol. 1996, 205, 63-87.
J M980612I