Synthesis and biological evaluation of rhodanine derivatives as PRL-3 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2006.02.060

A series of rhodanine derivatives was synthesized and evaluated for their ability to inhibit PRL-3. Benzylidene rhodanine derivative showed good biological activity, while compound 5e was the most active in this series exhibiting an IC₅₀ value of 0.9 μM in vitro and showed a reduced invasion in cell-based assay.

Full text of DOI:10.1016/j.bmcl.2006.02.060

Bioorganic & Medicinal Chemistry Letters 16 (2006) 2996–2999
Synthesis and biological evaluation of rhodanine derivatives
as PRL-3 inhibitors
Jin Hee Ahn,a, Seung Jun Kim,b, Woul Seong Park,a Sung Yun Cho,a Jae Du Ha,a
Sung Soo Kim,a Seung Kyu Kang,a Dae Gwin Jeong,b Suk-Kyeong Jung,b
Sang-Hyeup Lee,b Hwan Mook Kim,c Song Kyu Park,c Ki Ho Lee,c
Chang Woo Lee,c Seong Eon Ryub,* and Joong-Kwon Choia,*
aBio-Organic Science Division, Korea Research Institute of Chemical Technology, 100 Jang-Dong,
Yuseong-Gu, Daejeon 305-343, Republic of Korea
bSystemic Proteomics Research Center, Korea Research Institute of Bioscience and Biotechnology,
52 Eoeun-Dong, Yuseong-Gu, Daejeon 305-333, Republic of Korea
cBioevaluation Center, Korea Research Institute of Bioscience and Biotechnology,
52 Eoeun-Dong, Yuseong-Gu, Daejeon 305-333, Republic of Korea
Received 7 January 2006; revised 16 February 2006; accepted 22 February 2006
Available online 10 March 2006
Abstract—A series of rhodanine derivatives was synthesized and evaluated for their ability to inhibit PRL-3. Benzylidene rhodanine
derivative showed good biological activity, while compound 5e was the most active in this series exhibiting IC50 value of 0.9 lM
in vitro and showed a reduced invasion in cell-based assay.
ꢀ 2006 Elsevier Ltd. All rights reserved.
The protein tyrosine phosphatases (PTPs) constitute a
family of closely related key regulatory enzymes that
dephosphorylate phosphotyrosine residues in their pro-
tein substrates. Malfunctions in PTP activity are linked
to various diseases, ranging from cancer to neurological
disorders and diabetes. Consequently, PTPs have
emerged as promising targets for therapeutic interven-
tion in recent years.1
It is recently identified as a metastasis-related enzyme.
It is located at the cytoplasmic membrane when preny-
lated and in the nucleus when nonprenylated. Overex-
pression of PRL-3 has been found to transform
human embryonic kidney cell HEK293 and increase
HEK293 cell growth.4 Saha et al. reported that PRL-
3 mRNA expression was consistently increased in the
liver metastasis of colorectal cancers, suggesting that
PRL-3 was associated with colorectal cancer metasta-
sis.5 Furthermore, Zeng et al. showed that PRL pro-
teins promote cell motility, invasion activity and
metastasis, which were directly dependent on their cat-
alytic activity.6 These suggest that PRL-3 is not only a
putative prognostic marker but also a therapeutic tar-
get for metastastic tumors.
Among the PTPs, the phosphatase of regenerating liver
(PRL) family tyrosine phosphatases (PRL-1, PRL-2,
and PRL-3) are closely related intracellular enzymes
that possess the PTP active-site signature sequence
CX5R.2,3 All are proteins of about 20 kDa with at least
75% amino acid sequence homology.
PRL-3 gene codes a 22 kDa nonclassical protein tyro-
sine phosphatase with a C-terminal prenylation motif.
Thus far, only pentamidine7 was reported as a PRL fam-
ily inhibitor with anti-cancer potential via inactivation
of PRL-1, -2, and -3. This prompted us to attempt
systematic discovery of potent PRL-3 inhibitors. In the
course of the search for PRL-3 inhibitors through high
throughput screening (HTS) using chemical library of
Korea Chemical Bank, rhodanine skeleton was discov-
ered as a hit. We now report the synthesis and their
Keywords: Rhodanine; Cancer; Metastasis; PRL-3; Inhibitor.
*
Corresponding authors. Tel.: +82 42 860 4149; fax: +82 42 860 4598
(S.E.R.); tel.: +82 42 860 7070; fax: +82 42 860 7160 (J.-K.C.); e-mail
addresses: ryuse@kribb.re.kr; jkchoi@krict.re.kr
These authors contributed equally to this work.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.02.060

J. H. Ahn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2996–2999
2997
structure–activity relationship (SAR) study of rhoda-
nine derivatives as PRL-3 inhibitors.
PRL-3 protein was overexpressed as His-tag fusion pro-
tein in Escherichia coli and purified. Assays were
performed
using
6,8-difluoro-4-methylumbelliferyl
A series of rhodanine derivatives was synthesized
according to the Schemes 1 and 2. 5-Bromosalicylalde-
hyde (1a, X = Br) was reacted with rhodanine in the
presence of ammonium acetate to produce the corre-
sponding rhodanine derivative (2). Also salicylaldehydes
(1, X = H or Br) was benzylated and then coupled with
rhodanine, N-methylrhodanine, or thiazolidinedione to
afford 3a–d. Reduction of 3a by lithium borohydride
in pyridine gave compound 4. Next, 5-bromosalicylalde-
hyde (1a, X = Br) was treated with methanesulfonyl
chloride in pyridine or the appropriate benzyl halide
equivalents to provide the corresponding aldehydes,
which was converted to 5a–e by treatment with rhoda-
nine. Also, 1 was coupled with benzyl bromide to
produce 6, which adopted diverse aryl groups at 5-posi-
tion by Suzuki type coupling, followed by attaching rho-
danine moiety to afford the desired compounds 7a–d.
Rhodanine derivatives with naphthalene skeleton were
prepared as outlined in Scheme 2. 3-Hydroxynaphtha-
lene-2-carbaldehyde (8) was benzylated by several ben-
zyl bromides, and condensed with rhodanine to give
the corresponding naphthalydene rhodanine derivatives
(9a–d).
phosphate (DIFMUP) as a substrate at 25 ꢁC for
5 min in 20 mM Tris–HCl (pH 8.0), 5 mM DTT, in
the presence or absence of the inhibitor. After the addi-
tion of purified PRL-3 (0.3 lM) and DIFMUP (5 lM),
the reaction mixture was incubated for 5 min. The reac-
tion was stopped by the addition of sodium orthovana-
date (20 mM). The phosphatase activities were
measured by the absorbance changes caused by hydroly-
sis of the substrate at 460 nm. IC50 values were an aver-
age of triplicate experiments as determined from direct
regression curve analysis. Pentamidine was used as a
reference.
The result for the rhodanine derivatives is shown in
Table 1. While 5-bromosalicylaldehyde (1b, X = Br)
was not active, introduction of a rhodanine group pro-
vided an active PRL-3 inhibitor with an IC50 of
9.5 lM (2). Benzyl substitution of OH at 2-position
exhibited enhanced potency (3.0 lM, 3a), and was al-
most 20-fold more potent than reference pentamidine.
Either elimination of Br at 5-position (3b) or introduc-
tion of thiazolidinedione instead of rhodanine abolished
the activity (3c). N-Methylation was also detrimental to
the in vitro activity (3d). Compound 4 produced by
reduction of double bond of compound 3a showed sim-
ilar activity (4.0 lM).
All compounds prepared were evaluated for their in vitro
inhibitory activity against recombinant human PRL-3.
These data suggest that substituents at 2- and 5-position
of benzene ring influence the in vitro inhibitory activity.
Introduction of sulfonyl group or pyridinylmethyl group
at R2 position showed no inhibitory activities (5a and
5b). Introduction of 2-chlorobenzyl group as R1 demon-
strated comparable potency with 3a (5c, 2.4 lM). Intro-
duction of 2-bromobenzyl substituent (5e) exhibited
enhanced potency, and is the first of our compounds
to break the micromolar barrier with an IC50 value of
0.9 lM. Various substitution at R2 position resulted in
IC50 values in the range of 1.2–3.7 lM. Compound 7d
was another submicromolar inhibitor toward PRL-3
with an IC50 value of 0.9 lM.
OH
O
NH
S
S
a
S
Br
O
Bn
Bn
OH
X
O
O
O
O
H
b, a
c
Z
N
NH
S
Y
S
X
Br
b
d, a
O
R1
R1
R1
O
O
O
O
O
Naphthalene based rhodanine derivatives were another
possibility to show PRL-3 inhibition comparable to that
of the benzene as shown in Table 2. The effect of R lipo-
philicity showed similar trend as benzylidene series.
Among them, compound 9d showed an IC50 of 1.7 lM.
e, a
H
NH
NH
S
S
S
S
Br
R2
X
Scheme 1. Reagents and conditions: (a) rhodanine, N-methylrhoda-
nine or thiazolidinedione, NH4OAc, AcOH, benzene, reflux; (b) benzyl
bromide, K2CO3, KI, acetone, reflux; (c) LiBH4, pyridine, THF, and
reflux; (d) methanesulfonyl chloride, pyridine, rt or R1Br, K2CO3, KI,
acetone, reflux; (e) R2B(OH)2, Pd(dppf)2Cl2, Na2CO3, 100 ꢁC.
Recently, we reported the crystal structure of PRL-1
protein for the first time.8 Our crystal structure revealed
a well-ordered active-site structure with catalytically
important residues in active conformations, providing
the insight into the mode of interaction between rhoda-
nine compounds and PRL proteins. NH-Group in
rhodanine ring may be deprotonated, mimicing nega-
tively charged substrates. This moiety would be directed
toward the active-site P-loop that are surrounded by
positively charged guanidine group from Arg110 and
several amide groups from main-chain atoms. As stated
previously, N-methylation in rhodanine resulted in no
inhibitory activity, supporting this notion. Further, the
R
OH
O
O
O
H
NH
S
S
Scheme 2. Reagents and conditions: (a) RX, K2CO3, KI, acetone,
reflux; (b) rhodanine, NH4OAc, AcOH, benzene, reflux.

2998
J. H. Ahn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2996–2999
Table 1. Inhibitory activity of compounds derived from salicylaldehydes against PRL-3
R1
X or R2
Y
Z
IC50 (lM)
a
Compound
1a
29
—
H
Benzyl
Br
Br
—
S
—
H
nab
9.5
3.0
na
3a10
Br
S
H
3b10
3c10
3d
Benzyl
Benzyl
H
Br
S
H
O
H
na
na
Benzyl
—
Br
—
S
CH3
—
—
—
—
—
—
—
—
—
—
4
5a
—
—
—
—
—
—
—
—
—
—
4.0
na
CH3SO2
Pyridin-3-yl-methyl
2-Chlorobenzyl
4-Bromobenzyl
2-Bromobenzyl
Benzyl
Br
Br
5b
5c
na
Br
2.4
1.6
0.9
1.7
3.7
1.2
0.9
53.6
5d
5e
7a10
H
Br
Phenyl
7b
7c
Benzyl
Benzyl
Benzyl
4-N,N-Dimethylaminophenyl
Benzofuran-3-yl
Benzo[b]thiophen-3-yl
7d
Pentamidine
a IC50 values were determined from direct regression curve analysis.
b Not active up to 50 lM.
in Figure 1. B16F10 cells (1 · 105) in routine medi-
um were seeded on the upper chamber of 24-well
BD Biocoat Matrigel invasion chambers with 8 lm
polycarbonate filters (BD Bioscience, San Jose,
CA). The bottom chamber contained conditioned
medium from HT1080 cells as a chemoattractant.
Cells were incubated in the presence and absence
of the compounds for 22 h at 37 ꢁC in a humidified
incubator with 5% CO2. Nonmigratory cells on the
upper surface of the filter were removed by wiping
with a cotton swab. Migrated cells to the underside
of the membrane were stained with 0.5% crystal vio-
let after fixation with methanol and observed under
the microscope.
Table 2. Inhibitory activity of naphthalydene rhodanine derivatives
against PRL-3
a
Compound
R
IC50 (lM)
9a
9b
Benzyl
2.0
3.1
3,5-Dimethoxybenzyl
4-Phenylbenzyl
2-Chloro-6-fluorobenzyl
9c
9d
1.9
1.7
Pentamidine
53.6
a IC50 values were determined from direct regression curve analysis.
surroundings from the active-site pocket of PRL reveals
highly hydrophobic character with large entrance com-
paring to other PTPs. The inhibitory activities of rhoda-
nine compounds against PRL-3 are enhanced by the
introduction of hydrophobic substituents, which well
agrees with the structural information.
Compound 5e, with an IC50 value of 0.9 lM, showed the
reduced invasion comparable to 30 lg/ml of fumagil-
lin.11 Naphthalene based compound 9d also exhibited
similar result to 5e, both showing higher potency
compared to pentamidine.
The compounds 5e and 9d were evaluated for their
ability to reduce the invasiveness of tumor as shown
fumagillin 30 μg/mL
vehicle
5e 1 μg/mL
5e 10 μg/mL
5e 30 μg/mL
Figure 1. Effect of compound 5e on invasion of B16F10 cells.

J. H. Ahn et al. / Bioorg. Med. Chem. Lett. 16 (2006) 2996–2999
2999
Lengauer, C.; Kinzler, K. W.; Vogelstein, B. Biochem.
Biophys. Res. Commun. 1998, 244, 421.
In conclusion, we have discovered a new series of rhoda-
nine derivatives as inhibitors of PRL-3. Benzylidene
rhodanine derivatives showed good biological activity,
compound 5e was the most active in this series, exhibited
an IC50 value of 0.9 lM and showed a reduction of
invasion in cell. Further studies aimed at improving effi-
cacy using structure-based design are in progress and
will be reported in due course.
4. Matter, W. F.; Estridge, T.; Zhang, C.; Belagaje, R.;
Stancato, L.; Dixon, J.; Johnson, B.; Bloem, L.; Pickard, T.;
Donaghue, M.; Acton, S.; Jeyaseelan, R.; Kadambi, V.;
Vlahos, C. J. Biochem. Biophys. Res. Commun. 2001, 283,
1061.
5. Saha, S.; Bardelli, A.; Buckhaults, P.; Velculescu, V. E.;
Rago, C.; St Croix, B.; Romans, K. E.; Choti, M. A.;
Lengauer, C.; Kinzler, K. W.; Vogelstein, B. Science 2001,
294, 1343.
6. Zeng, Q.; Dong, J. M.; Guo, K.; Li, J.; Tan, H. X.; Koh,
V.; Pallen, C. J.; Manser, E.; Hong, W. Cancer Res. 2003,
63, 2716.
7. Pathak, M. K.; Dhawan, D.; Lindner, D. J.; Borden, E.
C.; Farver, C.; Taolin, Y. Mol. Cancer Ther. 2002, 1,
1255.
Acknowledgments
We thank the cooperation of Korea Chemical Bank to
use chemical library for HTS (http://www.chem-
bank.or.kr). This study was supported by Ministry of
Commerce, Industry and Energy (MOCIE).
8. Jeong, D. G.; Kim, S. J.; Kim, J. H.; Son, J. H.; Park, M.
R.; Lim, S. M.; Yoon, T. S.; Ryu, S. E. J. Mol. Biol. 2005,
345, 401.
References and notes
9. Commercially available.
1. Waldmann, H.; Bialy, L. Angew. Chem. Int. Ed. 2005, 24,
3814.
2. Diamone, R. H.; Cressman, D. E.; Laz, T. M.; Abrams, C.
S.; Taub, R. Mol. Cell. Biol. 1994, 14, 3752.
3. Zeng, S.; Bardelli, A.; Buckhaults, P.; Velculescu, V. E.;
Rago, C.; St Croix, B.; Romans, K. E.; Choti, M. A.;
10. Chen, H.; Fan, Y.-H.; Natarajan, A.; Guo, Y.; Iyasere, J.;
Harbinski, F.; Luus, L.; Christ, W.; Aktas, H.; Halperin,
J. Bioorg. Med. Chem. Lett. 2004, 14, 5401.
11. Bernier, S. G.; Taghizadeh, N.; Thompson, C. D.;
Westlin, W. F.; Hannig, G. J. Cell. Biochem. 2005, 95,
1191.