Design and synthesis of small chemical inhibitors containing different scaffolds for lck SH2 domain

Article abstract of DOI:10.1016/S0960-894X(03)00735-2

On the basis of the structure of (R)-rosmarinic acid, a series of small chemical compounds with a different scaffold was synthesized as inhibitors for lck SH2 domain. From ELISA results, most of all chemical compounds showed a similar or a little lower bi

Full text of DOI:10.1016/S0960-894X(03)00735-2

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3455–3459
Design and Synthesis of Small Chemical Inhibitors Containing
Different Scaffolds for lck SH2 Domain
See-Hyoung Park,a Sun-Hee Kang,b Sang-Hyeong Lim,b Hyun-Sik Ohb
and Keun-Hyeung Leeb,*
aSignal Transduction Laboratory, Mogam Biotechnology Research Institute, 341 Pojung-Ri, Koosung-Myun, Yongin-City,
Kyunggi-Do 449-910, South Korea
bDepartment of Chemistry, Inha University, 253 Younghyong-Dong, Nam-Gu, Inchon-City 402-751, South Korea
Received 3 June 2003; revised 14 July 2003; accepted 14 July 2003
Abstract—On the basis of the structure of (R)-rosmarinic acid, a series of small chemical compounds with a different scaffold was
synthesized as inhibitors for lck SH2 domain. From ELISA results, most of all chemical compounds showed a similar or a little
lower binding activity for lck SH2 domain compared to the lead compound, (R)-rosmarinic acid. It was characterized that the
backbone rigidity between two catechol substructures was required for the full activity and acid substructure of the lead compound
was important for the activity. We successfully identified novel lead compounds that did not contain phosphotyrosine moiety and
might have an improved bioavailability asinhibitor for lck SH2 domain.
# 2003 Elsevier Ltd. All rights reserved.
Many of intracellular signaling pathways are mediated
by phosphorylation and dephosphorylation of tyrosine
residue of intracellular protein, controlled by kinase and
phosphatase.1 SH2 domains in kinase and phosphatase
play pivotal rolesin organizing coherent isgnal tran-s
ducing complexes that are essential for the appropriate
cellular response to extracellular stimuli.1 Thus, ligands
that are able to disrupt these inappropriately hyper-
stimulated pathways, by blocking SH2 domain-depen-
dent interactions, may be developed to therapeutic
agentsfor cancer, autoimmune diesaes, and chronic
inflammatory disease.2ꢀ4
Asan alternative way, screening of natural productsto
identify small chemical inhibitors must afford useful and
conceptually straightforward starting points in the
development of SH2 domain inhibitorsfor therapeutic
agents. Previously, we reported that a small chemical
compound isolated from Prunella vulgaris had con-
siderable inhibitory activities on lck SH2-pYEEIE-
interaction, T-cell antigen receptor (TCR)-induced
interleukin (IL)-2 expression, and subsequent T-cell
proliferation in vitro.6,7 Interestingly, the small chemical
SH2 domain inhibitor wasproved to be ( R)-rosmarinic
acid, frequently found in herbal plants.8,9 Even though
(R)-rosmarinic acid that did not contain phosphotyr-
osine moiety as a small chemical compound had several
advantagesfor the lead compound over the phopsho-
peptide-based inhibitors, there are still some obstacles
for in vivo efficacy test and for structure–activity rela-
tionship study as follows. Rosmarinic acid was not
facilely isolated from natural sources as g scale and a
total chemical synthesis of the compound was not eco-
nomic because of several tedious syntheses and the low
yield.10,11 Furthermore, rosmarinic acid, dihydroxyl–
phenyl ester compound, was reported to be hydrolyzed
readily in vivo because of labile ester bond.12
Recent several studies about peptide inhibitors for SH2
domainsrevealed that SH2 domain including lck and
src exhibited a marked preference for the sequence
-pYEEIE-, and that short peptides bearing this
sequence exhibited a reasonably high affinity for the
SH2 domains.3ꢀ5 Even though moderately high-affinity
phosphopeptide-based inhibitors for SH2 domains have
been reported, their utility astherapeutic agentswas
limited mainly by their low stability against tyrosine
phosphatase and protease and by low cell permeability.
In an attempt to identify new lead compound which have
more improved in vivo stability and cell penetration
*Corresponding author. Tel.: +82-32-860-7674; fax:+82-32-867-
5604; e-mail: leekh@inha.ac.kr
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00735-2

3456
S.-H. Park et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3455–3459
Table 1. The binding affinity of compoundsfor lck SH2 domain
activity and can be easily synthesized, we designed and
synthesized rosmarinic acid analogues, which had a dif-
ferent scaffold, and successfully identified novel lead
compoundsasinhibitorsfor lck SH2 domain.
Compd No.
Structure
IC50 (mM)
24
Rosmarinic acid
Chemistry
In order to explore novel lead compound with more
improved bioavailability, we designed the analogues of
rosmarinic acid in which ester bond was replaced with
amide bond, thioamide bond, N-methyl amide bond,
reduced amide bond, and urethane bond (Table 1). In
addition, to improve cell penetration, we prepared the
analogues that had a methylester structure instead of an
acid structure.
1
121
20
1a
The syntheses of compounds 1 and 2 are shown in
Scheme 1. Compound 1 wasysntheiszed by coupling
reaction of commercially available caffeic acid with 3,4-
dihydroxylphenyl-d-alanine (DOPA) methylester 7.
Alternatively, compound 1 wasprepared by two tseps
including the coupling reaction of 3,4-dimethoxy-
cinnamic acid with DOPA methylester 7 and deprotec-
tion with BBr3 (data not shown). However, both
synthetic pathways gave the similar total yield (70%)
regardless of the number of synthesis step. The protec-
tion of four hydroxyl groupsof compound 1 and the
reaction of compound 8 with Lawesson reagent,13 fol-
lowed by deprotection, provided compound 2 in 25%
yield.
2
162
149
3
4
5
98
The synthesis of compound 3 wasdescribed in Scheme
2. The coupling reaction between DOPA methylester
and 3,4-dimethoxycinnamic acid gave compound 10 in
87% yield. Protection of compound 10 and N-methyla-
tion, followed by deprotection with BBr3 yielded
compound 12 in a reasonable yield. However, treat-
ment with NaH, followed by quenching in 1 M HCl
>500
Average IC50 valueswere calculated from three independent experi-
mentsperformed in duplicate, which provided a tsandard deviation
below 20%. IC50 value of control peptide (Ac-pYEEIE) was1.5 mM.
Scheme 1. Syntheses of compounds 1 and 2. Reagent: (a) MeOH, SOCl2, 0 ꢁC; (b) caffeic acid, PyBOP, TEA, DMF, DCM; (c) TBDMSOTf, TEA,
DCM; (d) Lawesson reagent, THF, reflux; (e) TBAF, THF.

S.-H. Park et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3455–3459
3457
hydrolyzed methylester bond of the compound. Thus,
objective compound 3 wasobtained by reacting com-
pound 12 with methanol using SOCl2 asactivating
agent. To consider the application of the synthesis
scheme in solid-phase synthesis, we alternatively syn-
thesized compound 3, ashsown in Scheme 3. In this
case, we first synthesized N-methyl-3,4-dihydroxyl-
phenyl-d-alanine (N-methyl DOPA) by reductive alkyl-
ation.14 And then compound 3 wasprepared by
coupling N-methyl DOPA with caffeic acid. We
obtained the best yield by using PyBroP among the
several coupling agents such as DCC, DIPC, PyBOP,
HBTU, and BOP, however the coupling yield (<20%)
was still too low presumably because of steric hindrance
of N-methyl group.
Compound 4 was synthesized as shown in Scheme 4.
Compound 16 wasconverted into the correpsonding
isocyanate by Curtius rearrangement.15 The reaction of
the corresponding isocyanate with DOPA methylester,
followed by hydrolysis of dimethylether of 18 provided
compound 4 in a reasonable yield. As shown in Scheme
5, compound 5 was synthesized in two steps; reductive
alkylation of DOPA methylester with the aldehyde and
Scheme 2. Synthesis of compound 3. Reagent: (a) 3,4-dimethoxycinnamic acid, PyBOP, TEA, DCM; (b) TBDMSOTf, TEA, DCM; (c) NaH, CH3I,
THF, quenching 1 M HCl (0 ꢁC); (d) TBAF, THF; (e) 1.0 M BBr3 in DCM, ꢀ30 ꢁC ! rt, DCM; (f) SOCl2, MeOH.
Scheme 3. Synthesis of compound 3. Reagent: (a) CH2Cl2, 24 h; (b) NaCNBH3, AcOH, THF, 50 min; (c) NaCNBH3, AcOH, HCOH, 6 h; (d) H2,
Pd/C, MeOH, 6 h; (e) PyBroP, 3,4-dihydroxycinammic acid, TEA, DMAP, DMF, 6 h.
Scheme 4. Synthesis of compound 4. Reagent: (a) SOCl2, 0 ꢁC, DMF 2–3 drops, 5 h; (b) NaN3, 0 ꢁC, acetone/H2O, 4 h; (c) benzene reflux, 5 h;
(d) DCM, DMF, TEA, rt for 13 h; (e) BBr3 in DCM 1.0 M solution, DCM, rt for 18 h.

3458
S.-H. Park et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3455–3459
Scheme 5. Synthesis of compound 5. Reagent: (a) 4-hydroxy-3-methoxycinnamaldehyde, NaCNBH3, AcOH (1%), THF, 4 h; (b) BBr3 in DCM
1.0 M solution, ꢀ30 ꢁC ! rt, DCM, 5 h.
then deprotection of methyl ether bond of the phenyl
ring. All synthesized analogues employed in this study
were further purified by preparative reverse-phase
HPLC and the productswere obtained in high purity
have improved bioavailability because thioamide bond
wasreported to be more tsable than amide bond in
vivo.18 However, compound 2 showed the lowest inhi-
bition activity among the active analogue series, which
can be explained by the fact that the larger and less
electronegative sulfur atom than oxygen atom must
induce some conformational distortions and/or may
form hydrogen bonding to the protein less tightly.
(generally >95% by RP analytical HPLC, UV214 nm
before binding assay for lck SH2 domain.
)
Results and Discussion
Considering the inhibition activity as well as stability
and cell penetration activity, compound 3 must be bet-
ter candidate for in vivo efficacy test than the com-
pounds 1, 2, and 4 because N-methyl amide bond had a
potent resistance against various intracellular enzymes
and are more hydrophobic than amide bond and ure-
thane bond.
The inhibition activity of the derivativesof roms arinic
acid on lck SH2-AcpYEEIE- interaction wasinvetsi-
gated by using the previously reported ELISA
method.16 All synthetic analogues except compound 5
inhibited the binding of EPQpYEEIPIYL with lck SH2
domain in a concentration-dependent manner (data not
shown). Table 1 summarized IC50 valuesof compounds
1–5 obtained in the ELISA assay.
In the present study, we successfully identified novel
lead inhibitorswith a different csaffold from that of
rosmarnic acid for lck SH2 domain and characterized the
role of backbone and acid substructure of the inhibitors
on the binding activity. On the basis of the structure of
compound 1a as a lead chemical, several non-phospho-
peptide inhibitors for SH2 domains were successfully
synthesized from solid phase parallel synthesis17 and
structure–activity relationships studies of (R)-rosmari-
nic acid and compound 1 are in progress.
Ashsown in Table 1, we successfully identified novel
small chemical inhibitors with a different scaffold from
that of rosmarinic acid. However, most of the analogues
exhibited lower binding affinity than rosmarinic acid. As
all analogues had the same structure as the lead com-
pound except the backbone scaffold and methylester
substructure, backbone scaffold or acid substructure
must play an important role in the binding activity. To
investigate which modification is more dominant for the
activity, compound 1a that had acid substructure was
prepared by hydrolysis of compound 1 and itsbinding
affinity with lck SH2 domain wasmeasured. Compound
1a had a similar binding affinity to the lead compound,
rosmarinic acid, which indicated that the decrease of
binding affinity of the analogueswasmainly due to the
modification of acid substructure. As shown in Table 1,
the comparison between IC50 value of compounds 1–5
indicated that the binding activity must rely on the
rigidity of the scaffold between two catechol sub-
structures because inactive analogue, compound 5, had
the most flexible scaffold among the analogue series.
Compound 4 showed the best binding affinity among
the series, which indicated the double bond of the com-
poundsisnot required for the full activity and if some
chemicalshad conformational contsrain between two
catechol substructures, the compound must exhibited
considerable binding affinity with lck SH2 domain.
Acknowledgements
Thiswork wassupported by grant (No. R01-2001-000-
00057-0) from the Basic Research Program of the Korea
Science & Engineering Foundation.
References and Notes
1. Cantley, L. C.; Auger, K. R.; Carpenter, C.; Duckworth,
B.; Graziani, A.; Kappeller, R.; Soltoff, S. Cell 1991, 64, 241.
2. Botfield, M. C.; Green, J. Ann. Rep. Med. Chem. 1995, 30, 227.
3. (a) Garcia-Echeverria, C. Curr. Med. Chem. 2001, 13, 1589.
(b) Broadbridge, R. J.; Sharma, R. P. Curr. Drug Targets
2000, 1, 365. (c) Beaulieu, P. L.; Cameron, D. R.; Ferland,
J. M.; Gauthier, J.; Ghiro, E.; Gillard, J.; Gorys, V.; Poirier,
M.; Rancourt, J.; Wernic, D.; Llinas-Brunet, M.; Betageri, R.;
Cardozo, M.; Hickey, E. R.; Ingraham, R.; Jakes, S.; Kabce-
nell, A.; Kirrane, T.; Lukas, S.; Patel, U.; Proudfoot, J.;
Sharma, R.; Tong, L.; Moss, N. J. Med. Chem. 1999, 42, 1757.
(d) Morelock, M. M.; Ingraham, R. H.; Betageri, R.; Jakes, S.
J. Med. Chem. 1995, 38, 1309.
We expect that compound 1 containing amide bond
must have different bioavailability from that of ros-
marinic acid. However, it isdifficult to anticipate that
compound 1 may exhibit improved bioavailability
because amide bond did not have a considerable resis-
tance against peptidase but esterase, whereas rosmarinic
acid had a resistance against peptidase rather than
esterase. Compound 2 containing thioamide bond must
4. Vu, C. B. Curr. Med. Chem. 2000, 10, 1081.
5. Songyang, Z.; Shoelson, S. E.; Chaudhuri, M.; Gish, G.;
Pawson, T.; Haser, W. G.; King, F.; Roberts, T.; Patnofsky,
S.; Lechleider, R. J.; Neel, B. G.; Birge, B.; Cantley, L. C. Cell
1993, 72, 767.

S.-H. Park et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3455–3459
3459
6. Hur, E. M.; Choi; B.; Park; C.; Lee; J.; Park; D.; Yun; Y.;
Lee; K. H.; Oh; J. E.; Ahn; C. S.; Lee; H. S.; Ahn; J. S; Jung;
S. I. US 6,140,363, Oct. 31, 2000.
7. Won, J.; Hur, E. M.; Hur, Y.; Park, S.; Kang, M.; Choi,
Y.; Park, C.; Lee, K. H.; Yun, Y. Eur. J. Immunology 2003,
33, 870.
8. Petersen, M.; Simmonds, M. S. Phytochemistry 2003, 62,
121.
9. Al-Sereiti, M. R.; Abu-Amer, K. M.; Sen, P. Indian J. Exp.
Biol. 1999, 37, 124.
10. Bogucki, D. E.; Charlton, J. L. Can. J. Chem. 1997, 75,
1783.
11. Eicher, T.; Ott, Markus; Speicher, A. Synthesis 1996, 755.
12. Nakazawa, T.; Ohsawa, K. J. Nat. Prod. 1998, 61, 993.
13. Caca, M. P.; Levinson, M. I. Tetrahedron 1985, 41, 5061.
14. Jason, J.; Chruma, D. S.; Robin, P. Tetrahedron Lett.
1997, 38, 5085.
15. Gilles, G.; Vincent, S.; Marc, R.; Jean-Paul, B. Tetra-
hedron Lett. 2000, 41, 1553.
16. Gilmer, T.; Rodriguez, M.; Jordan, S.; Crosby, R.; Alli-
good, K.; Green, K.; Kimery, M.; Wagner, C.; Kinder, D.;
Charifson, P.; Hassell, A. M.; Willard, D.; Luther, M.; Rus-
nak, D.; Sternbach, D. D.; Mehrotra, M.; Peel, M.; Shampine,
L.; Davis, R.; Robbins, J.; Patel, I. R.; Kassel, D.; Burkhart,
W.; Moyer, M.; Bradshaw, T.; Berman, J. J. Biol. Chem. 1994,
269, 31711.
17. Kang, S. H.; Park, S. H.; Shim, H. S.; Lee, K. H. Bull
Korean Chem. Soc. 2003, 24, 664.
18. Zacharie, B.; Lagraoui, M.; Dimarco, M.; Penney, C. L.;
Gagnon, L. J. Med. Chem. 1999, 42, 2046.