Small ligands interacting with the phosphotyrosine binding pocket of the Src SH2 protein

Article abstract of DOI:10.1016/S0960-894X(02)00140-3

Various small fragments bearing phosphate, phosphonate or phosphonic acid moieties have been prepared through parallel synthesis and their binding potencies evaluated on the Src SH₂ protein using a BIAcore assay. This provided us insight into the requirement of the Src SH₂ pTyr binding pocket and some promising small ligands have been characterised.

Full text of DOI:10.1016/S0960-894X(02)00140-3

Bioorganic & Medicinal Chemistry Letters 12 (2002) 1295–1298
Small Ligands Interacting with the Phosphotyrosine Binding
Pocket of the Src SH2 Protein
Pierre Deprez,* Eliane Mandine, Dominique Gofflo, Stephane Meunier and
Dominique Lesuisse
Aventis Pharma, Paris Research Center, Medicinal Chemistry, 102 route de Noisy, 93235 Romainville Cedex, France
Received 23October 2001; accepted 19 February 2002
Abstract—Various small fragments bearing phosphate, phosphonate or phosphonic acid moieties have been prepared through
parallel synthesis and their binding potencies evaluated on the Src SH2 protein using a BIAcore assay. This provided us insight into
the requirement of the Src SH2 pTyr binding pocket and some promising small ligands have been characterised. # 2002 Elsevier
Science Ltd. All rights reserved.
Many of the key signaling pathways are mediated by
phosphorylation/dephosphorylation of tyrosine residues
found in intracellular proteins.1 Thus tyrosine kinases
can bind to phosphotyrosine sequences via their Src
homology 2 (SH2) domain2 thus affording molecular
association and translocation. Tyrosine kinases are
involved in many pathologies such as cancer, inflam-
mation or osteroporosis and are consequently important
targets for the pharmaceutical industry.3
The pTyr binding pocket is a positively charged pocket
interacting with the phenyl phosphate pTyr residue and
is formed in part by Arg 12, Arg 32 and Lys 60. Thus
strong ionic interactions between the doubly negatively
charged phosphate and the two Arg are observed.
Additional polar interactions also exists with the phos-
phate and Ser 34, Glu 35 and Thr 36. Moreover, the
hydrophobic alkyl side chain of Lys 60 interacts with
the phenyl ring. Thus, it appears that the phenyl phos-
phate group is a tight binder of this pocket. However, to
overcome the problems associated with the phosphate
moiety of pTyr, many groups have searched phosphate
surrogates, resulting in the identification of phosphon-
ate, difluorophosphonate or carboxyl based mimetics
amongst others.10
In our research program on Src SH2 inhibitors, we were
looking for peptidomimetics of the pYEEI tetrapeptide.
Many X-ray co-structures of pYEEI or other ligands
within SH2 domains have been reported,4À6 indicating
that there are two major binding pockets, one interact-
ing with the phosphotyrosine itself, and the other with
the hydrophobic iLeu residue.
Our purpose in this study was to focus on the Src SH2
protein and to find out whether it could be possible to
improve the binding affinity of phenyl phosphate by
modifying the phenyl ring (with substitution on the
phenyl ring or by its replacement with another aromatic
ring) and/or by replacing the phosphate with a non
hydrolysable group that retained potent Src SH2 binding
affinity.
Our strategy was to use a modular approach and to try
to optimise each binding site independently, and we
would like to disclose in this letter some of our results
on the search of pTyr surrogates. A similar modular
strategy with the evaluation of small organic molecules
has also been used for the search of new inhibitors of
protein tyrosine phosphatase CD 457 and PTP1B.8 Very
recently, Fesik et al. also used a small fragment
approach with a NMR based screen for the discovery of
novel phosphotyrosine mimetics that bind to the Lck
Chemistry
9
SH2 domain.
Phosphate compounds have been prepared using paral-
lel synthesis, either by solid-phase or solution-phase
chemistry. In both cases, the desired phosphate is
obtained in two steps from the corresponding alcohol
*Corresponding author. Tel.: +33-1-4991-3069; fax: +33-1-4991-
3089; e-mail: pierre.deprez@aventis.com
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00140-3

1296
P. Deprez et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1295–1298
using a phosphite (bound on the resin or not) as the
phosphorylating agent.
with DBU generated the anion which was then trapped
by an aromatic aldehyde (Scheme 2). Similarly, we
describe here the preparation of amido phosphonic
acids using an aromatic isocyanate as the electrophilic
agent. Here again, depending on the nature of R group
(tBu does not withstand TFA treatment), we also can
have access to amido phosphonatate mono esters or
amido phosphonic acids 4.
For solution-phase chemistry, we adapted the phos-
phorylation procedure described by Silvenberg et al.11
to a parallel synthesis approach (Scheme 1). Thus, in
our procedure, 0.95 equiv of alcohol was reacted with 1
equiv of dibenzyl phosphite in the presence of CCl4, and
the resulting dibenzyl phosphate was purified by liquid
liquid extraction, followed by SPE on a silica cartridge.
Debenzylation was performed in parallel either by
hydrogenation over Pd/C or TFA/CH2Cl2 treatment
depending of the compound, which led to the desired
pure free phosphate 1 after an aqueous extraction and
lyophilisation with an overall yield of 50–60%.
Results and Discussion
In the search for phenyl phosphate surrogates, around
120 compounds have been prepared and can be classi-
fied as phosphates 1, a-hydroxy phosphonic acids 3
and a-amido phosphonic acids C. The corresponding
monoesters of these three classes have also been
synthesised and a selection of representative com-
pounds are described in Tables 1 and 2 (phosphate and
phosphonate).
For the preparation of phosphates 1 and 2, we have also
developed12 a solid-phase version of this synthesis using
the phosphite bound on the Wang resin A13 (Scheme 1).
This methodology has the advantage over the solution-
phase version of allowing the quick synthesis of a larger
number of compounds, the final phosphates being puri-
fied with SPE on reverse-phase cartridges. Using this
pathway with various R groups (different to tBu),
phosphate monoesters 2 can also be prepared easily.
Our methodology to evaluate binding affinity of these
small ligands was based on BIAcore technology.14 In
this assay,15 the compound to be evaluated is in the
presence of the Src SH2 protein and has to compete for
binding with the pYEEI tetrapeptide bound on the sen-
sor chip. This assay is sensitive enough to detect milli-
molar binding affinity and, as a reference, the IC50 of
phenyl phosphate 1a has been measured at 2.5 mM.
a-Hydroxy phosphonates 3 and amido phosphonates 4
have also been prepared by solid phase from the same
starting bound phosphite following a methodology
published by Mjalli et al.12 Activation of the phosphite
Substitution in the ortho, meta or para position of the
phenyl ring (1b–1q, Table 1) in general gives compounds
with lower binding affinity to the Src SH2 protein.
Exceptions have been observed with the introduction
Table 1. Biacore Src SH2 binding affinity of aryl phosphate
fragments
Compd
Ar–OH
Ar–X, X=
IC50
(mM)
Scheme 1. Solid-phase and solution-phase phosphate syntheses: (a)
ArOH (0.95 equiv), dibenzylphosphite HPO(OBn)2 (1 equiv), DIEA (2
equiv), CCl4 (5 equiv); DMAP (0.1 equiv) in CH3CN, À10 ꢀC, 2 h,
then KH2PO4 treatment and AcOEt extraction followed by SPE on
silica (70 cyclohexane/30 AcOEt); (b) H2, Pd/C cat in MeOH, 2 h or
TFA/CH2Cl2 1/1 overnight, then aqueous extraction (H2O/toluene);
(c) 1 M solution of ArOH (5 equiv), DIEA (6.5 equiv), CCl4 (5 equiv);
DMAP (0.5 equiv) in CH3CN, 23 ꢀC, 1 h, then filtration and washing;
(d) 15% TFA/CH2Cl2 30 min.
1a
1b
1c
1d
1e
1f
1g
1h
1i
PhOH
3-COOMe PhOH
3-CHO PhOH
3-Me PhOH
4-Me PhOH
4-tBu PhOH
4-OMe PhOH
4-COOMe PhOH
4-NH2 PhOH
4-NO2 PhOH
4-SCF3 PhOH
2-Cl PhOH
2-COOH PhOH
2-CHO PhOH
2-CHO PhOH
2-CHO PhOH
2-CHO PhOH
2-Me 6-Me PhOH
2,6 CHO PhOH
2-CHO 6-OMe PhOH
1-Naphthol
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
2.5
4
2.4
4
2
4
4
4
0.6
1
1.2
3.5
2
1j
1k
1l
1m
1n
2a
2b
2c
1o
1p
1q
1r
1s
1t
0.2
3.3
2.3
2.1
>10
0.5
0.9
1
OPO3H(Me)
OPO3H(CH2 c-hexyl)
OPO3H(iPr)
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
OPO3H2
2-Naphthol
1
2-COCH3 1-Naphthol
3-COOMe2-Naphthol
8-OH Quinoline
2-CN, 8-OH Quinoline
1,2,3,4-Tetrahydro 8-OH quinoline
3.5
3.2
>10
>10
0.3
1u
1v
1w
1x
Scheme 2. Solid-phase synthesis of phosphonates and phosphonic
acids: (a) 1 M solution of DBU in CH3CN, 10 min; (b) 1 M solution of
the aldehyde (5 equiv) in CH3CN, 30 min; (c) 1 M solution of the
isocyanate (5 equiv) in CH3CN, 30 min; (d) 15% TFA/CH2Cl2 30 min.

P. Deprez et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1295–1298
1297
Table 2. Biacore Src SH2 binding affinity of aryl phosphonate
fragments
Replacement of the phenyl group by other aromatic
rings (especially bicyclic rings) has also been studied
(1r–x). As shown in Table 1, the naphthyl phosphates 1r
and 1s are twice as potent as the phosphate 1a. The
naphthyl ring has already been identified as a good
phenyl replacement for phosphatases and SH2
domains.17 More interestingly, tetrahydroquinoline 1x
appeared to be the most potent aromatic scaffold (10-
fold increase in binding) despite the lack of activity of
the corresponding quinoline 1v (or 1w).
Compd
Ar
Ar–X, X=
IC50 (mM)
3a
3b
3c
3d
3e
3f
3g
3h
4a
4b
4c
4d
4e
4f
Ph
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
CH(OH)PO3H2
NHCOPO3H2
NHCOPO3H2
NHCOPO3H(Bn)
NHCOPO3H2
NHCOPO3H2
NHCOPO3H2
NHCOPO3H2
NHCOPO3H2
NHCOPO3H2
>10
>10
>10
>10
>10
>5
2-OMe Ph
4iPr-Ph
4PhPh
Cyclohexyl
1-Naphthyl
2-Naphthyl
4-OPr Ph
4-Et Ph
4.4
2.4
4.4
EtPh
4-Et Ph
CH2Ph
>5
Knowing that some hydroxy phosphonic acids have
been found as inhibitors of tyrosine phosphatase
CD45,7 we were wondering whether the additional
hydroxy group could serve as hydrogen bonding med-
iator within the Src SH2 protein. Thus, hydroxy phos-
phonic acids 3 have been prepared and evaluated
against Src SH2 (Table 2). Generally speaking, these
compounds 3, if not inactive, are less active that the
corresponding phosphates 1 as indicated by at least a 5-
fold decrease in binding affinity between 1a and 3a (or
1r and 3f, 1s and 3g). This is in line with some recent
results18 on Src SH2 indicating that the hydroxyphos-
phonomethyl cannot be considered as a very potent
phosphate surrogate. In this series, all the corre-
sponding monoesters prepared are inactive (data not
shown).
> >5
4.4
3.6
4-COOEt Ph
4-OBu Ph
2-OCF3 Ph
3,4,5-OMe Ph
4-OBn Ph
3.9
4.0
2.1
0.8
4g
4h
4i
of the NH2, NO2 and SCF3 groups (1i, 1j and 1k) in
the para position which give a 2–3-fold increase in
binding affinity, but without clear explanation.
More significant was the 10-fold improvement with an
ortho aldehyde substitution (1n, 0.2 mM). This latter
result was not a surprise to us and confirmed the 15-fold
increase in potency already observed by Charifson et al.6
with the Src SH2 protein when they introduced an ortho
aldehyde on their peptidic inhibitors. This improvement
is explained by the presence of the cysteine 42 in the
pTyr binding site which is trapped by the aldehyde
group to form a covalent and reversible bond. This is
clearly visible on the X-ray co-structure of this fragment
within the Src SH2 protein.19 Confirmation of the
potency of the known ortho aldehyde fragment with this
small fragment approach was important because it vali-
dated this methodology.
Finally, amido phosphonic acids 4 have been evaluated
(Table 2). Interestingly, many of them have similar
activity to phenyl phosphate 1a. Some substituted
phenyl hydroxyphosphonates displayed promising affi-
nity, the best compound in the series being the 4-OBn
substituted analogue 4i. To our knowledge, this is the
first time that amido phosphonic acids have been iden-
tified as good surrogates of phosphate and have the
advantage of not being so easily hydrolyzable.
In conclusion, different series of aryl phosphates and
aryl phosphonic acids (and their corresponding mono
esters) have been prepared and evaluated as potential
phosphotyrosine surrogates on the Src SH2 protein.
This study has given us an insight into the ability of
some fragments to be good replacements of the phenyl
phosphate group present in pTyr of the pYEEI tetra-
peptide. We confirmed the importance of the aldehyde
in the ortho position of the phenyl ring for the Src SH2,
thus validating the small fragment methodology, and we
identified the naphthyl and tetrahydroquinoline deriva-
tives as possible phenyl replacements. Amido phosphonate
is also a potential candidate for phosphate mimic.
Conversely, introduction of the aldehyde at the meta
position does not influence the binding (1c, 2.3mM)
indicating that the aldehyde is poorly placed in that case
to interact with the cysteine. It was of interest to reduce
the overall charge of the compound with the expectation
of improvement of cellular uptake of the inhibitor.
Potency of the ortho aldehyde phosphate monoesters
2a–c (one negative charge) is similar to that of the
phenyl phosphate 1a (two negative charges), indicating
that the aldehyde interaction is strong enough to com-
pensate the loss of one charge of the phosphate. How-
ever, it is also clear that the phosphate monoesters 2a–c
are 10 times less potent that the corresponding free
phosphate 1n. This result is consistent with the fact that
the two negative charges of the phosphate are useful for
potent binding. With the synthesis of phosphate esters,
we were wondering whether the loss in ionic binding
could be counterbalanced by some additional inter-
actions with the ester side chain. It appears not to be the
case whatever the ester group introduced.16 This sug-
gests that the ester group on the phosphate does not
interact with the pTyr pocket and probably lies outside
the protein.
To confirm how these selected fragments interacted with
the pTyr binding pocket and whether they adopt a
similar positioning to that for the phenyl phosphate of
the pYEEI tetrapeptide, some X-ray studies have thus
been undertaken. Some co-structures of fragments
within the Src SH2 protein have been obtained, con-
firming their interaction with the pTyr binding pocket
and giving essential information on the positioning of
these fragments. These data, as well as the incorpo-
ration of the most promising surrogates into full Src
SH2 ligand, will be reported shortly.19

1298
P. Deprez et al. / Bioorg. Med. Chem. Lett. 12 (2002) 1295–1298
6073. (b) Zhang, C.; Mjalli, A. M. M. Tetrahedron Lett. 1996,
37, 5457.
References and Notes
1. (a) Hunter, T.; Cooper, J. A. Ann Rev. Biochem. 1985, 54,
897. (b) Liu, D.; Wang, L.-H. J. Biomed. Sci. 1994, 1, 65.
2. For a recent review on SH2 domains, see: Sawyer, T. K.
Biopolymers 1998, 47, 243.
3. Brugge, J. S. Science 1993, 362, 87.
4. Waksman, G.; Shoelson, S. E.; Pant, N.; Cowburn, D.;
Kuriyan, J. Cell 1993, 72, 779.
5. (a) Eck, M. J.; Shoelson, S. E.; Harrison, S. C. Nature 1993,
362, 87. (b) Eck, M. J.; Atwell, S. K.; Shoelson, S. E.; Harri-
son, S. C. Nature 1994, 368, 764.
6. Charifson, P. S.; Shewchuk, L. M.; Rocque, W.; Hummel,
C. W.; Jordan, S. R.; Mohr, C.; Pacofsky, G. J.; Peel, M. R.;
Rodriguez, M.; Sternbach, D. D.; Consler, T. G. Biochemistry
1997, 36, 6283.
7. Frechette, R. F.; Ackerman, C.; Beers, S.; Look, R.;
Moore, J. Bioorg. Med. Chem. Lett. 1997, 7, 2169.
8. Yao, Z.-J.; Ye, B.; Wu, X.-W.; Wang, S.; Wu, L.; Zhang,
Z.-Y.; Burke, T. R. Bioorg. Med. Chem. 1998, 6, 1799.
9. Hajduk, P. J.; Zhou, M. M.; Fesik, S. W. Bioorg. Med.
Chem. Lett. 1999, 9, 2403.
10. (a) Burke, T. R., Jr.; Yao, Z.-J.; Smyth, M. S.; Ye, B. Curr.
Pharm. Des. 1997, 3, 291. (b) Burke, T. R., Jr.; Luo, J.; Yao,
Z.-J.; Gao, Y.; Zhao, H.; Milne, G. W. A.; Guo, R.; Voigt,
J. H.; King, C. R.; Yang, D. Bioorg. Med. Chem. Lett. 1999, 9,
347. (c) Burke, T. R., Jr.; Ye, B.; Akamatsu, M.; Ford, H.;
Yan, X.; Kole, H. K.; Wolf, G.; Shoelson, S. E.; Roller, P. P.
J. Med. Chem. 1996, 39, 1021. (d) Burke, T. R., Jr.; Yao, Z.-J.;
Zhoa, H.; Milne, G. W. A.; Wu, L.; Zhang, Z.-Y.; Voigt, J. H.
Tetrahedron 1998, 54, 9981.
14. Ladbury, J. E.; Lemmon, M. A.; Zhou, M.; Green, J.;
Botfield, M. C.; Schlessinger, J. Proc. Nat. Acad Sci. U.S.A.
1995, 92, 3199.
15. The Src-specific phosphopeptide was immobilized to a
biosensor chip CM5 through injection of a 0.5 mM solution of
the phosphopeptide in 0.5 M NaHCO3, across the chip surface
previously activated according to the instructions of the
manufacturer. Injection of anti-phosphotyrosine antibody was
used to confirm that successful immobilization of the peptide
was achieved. For the competition binding assay 0.5 mM of the
Src SH2 protein was premixed in HBS (10 mM Hepes/150 mM
NaCl/0.05% surfactant P20, pH 7.4) at room temperature
with increasing concentrations (up to 5 mM) of competing
molecules and injected at a flow rate of 30 mL/mn for 4 mn
across a surface to which the phosphopeptide was immobil-
ized. Prior to each run the phosphopeptide surface of the chip
was regenerated using 2 M guanidium HCl. The amount of
bound Src SH2 domain was estimated from the surface plas-
mon resonance signal at a fixed time just before the end of the
injection and the percentage bound, relative to injection of Src
SH2 alone, calculated.
16. All phenyl phosphate monoesters prepared (>10 ester
groups introduced) are inactive (> >10 mM in the BIAcore
assay).
17. The naphthyl group has also been used successfully with
the protein tyrosine phosphatase PTP1B: see ref 8 and: Burke,
T. R., Jr.; Ye, B.; Yan, X.; Wang, S.; Jia, Z.; Chen, L.; Zhang,
Z.-Y.; Barford, D. Biochemistry 1996, 35, 15938.
18. Lai, J. H.; Marsilje, T. H.; Choi, S.; Nair, S. A.; Hang-
auer, D. G. J. Peptide Res. 1998, 51, 271.
19. Lesuisse, D.; Lange, G.; Dreprez, P.; Bernard, D.; Schoot,
B.; Delettre, G.; Marquette, J. P.; Broto, P.; Jean-Baptiste, V.;
Bichet, P.; Sarubbi, E.; Mandine, E. J. Med Chem. Submitted.
11. Silverberg, L. J.; Dillon, J. L.; Vemishetti, P. Tetrahedron
Lett. 1996, 37, 771.
12. Deprez, P. Tetrahedron Lett. To be submitted.
13. (a) Cao, X.; Mjalli, A. M. M. Tetrahedron Lett. 1996, 37,