A small-molecule agonist of the Wnt signaling pathway

Article abstract of DOI:10.1002/anie.200462552

(Chemical Equation Presented) A new tool for developmental biology: A screen of combinatorial chemical libraries identified the 2-amino-4,6- disubstituted pyrimidine 1 as a dose-dependent agonist of Wnt signaling. Tadpoles that developed from embryos trea

Full text of DOI:10.1002/anie.200462552

Angewandte
Chemie
Cell Signaling
A Small-Molecule Agonist of the Wnt Signaling
Pathway**
Jun Liu, Xu Wu, Brian Mitchell, Chris Kintner,
Sheng Ding,* and Peter G. Schultz*
The Wnt signaling pathway plays an important role in a
variety of physiological and pathological processes by regu-
lating genes involved in cellular adhesion, proliferation, and
differentiation.[1,2] Many of the key components of the
signaling pathway have been elucidated. The Wnt ligand
activates downstream signal transduction through interaction
with its receptor, Frizzled. This interaction leads to the
inhibition of b-catenin phosphorylation by the axin–APC–
GSK-3b complex and results in the accumulation of b-catenin
in the cytoplasm.[3,4] b-Catenin then translocates to the
nucleus and forms a complex with T-cell factor/lymphoid-
enhancer factor (TCF/LEF), which drives target gene expres-
sion. Because Wnt signaling is essential for many aspects of
early embryonic development, genetic alteration of this
pathway often results in complicated embryonic defects in
animal models, which complicates the interpretation of
experimental results.[1] In contrast, small-molecule agonists
or antagonists offer greater temporal control over Wnt
signaling and provide useful tools for the studies of adult-
tissue homeostasis and regeneration, and embryogenesis in
animal models.[5] Herein we report the identification and
characterization of a small-molecule agonist of Wnt that
activates Wnt signaling in cellular assays as well as in a
Xenopus model.
Combinatorial small-molecule libraries that consist of
100000 heterocyclic compounds (including substituted
purines, pyrimidines, indoles, quinazolines, pyrazines, pyrro-
lopyrimidines, pyrazolopyrimidines, phthalazines, pyrida-
zines, and quinoxalines) were screened for small-molecule
modulators of Wnt signaling.[6–9] A reporter-based cellular
assay was used to measure the transcriptional activity of b-
[*] Dr. J. Liu, Dr. X. Wu,[+] Dr. S. Ding, Dr. P. G. Schultz
Department of Chemistry and
The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 North Torrey Pines Road, SR202, La Jolla, CA 92037 (USA)
Fax.: (+1)858-784-9440
E-mail: sding@scripps.edu
schultz@scripps.edu
Dr. B. Mitchell,[+] Dr. C. Kintner
The Salk Institute for Biological Studies
P.O. Box 85800, La Jolla, California 92186 (USA)
[+] These authors contributed equally to this paper.
[**] This work was supported by the Novartis Research Foundation and
the Skaggs Institute for Chemical Biology (P.G.S.). We thank Dr.
Xiang-ju Gu and Dr. Carsten Andersen for various reagents and
technical assistance in GSK-3b kinase assays. This is manuscript
16945-CH of the Scripps Research Institute.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2005, 44, 1987 –1990
DOI: 10.1002/anie.200462552
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1987

Communications
catenin/TCF in 384-well cell culture plates. b-Catenin/TCF
zylamino group and the 2-(4-hydroxyphenyl)ethylamino
group. Representative active and inactive analogues and
their EC50 values are provided in the Supporting Information.
The Wnt signal is transduced through its receptor and
results in the formation of a heterodimeric complex of b-
catenin and TCF that drives downstream target gene expres-
sion.[1,3,4] To determine if the active compounds induce
reporter gene expression in a TCF-dependent manner, a
dominant-negative TCF4 was constructed and transfected
into cells in the absence or presence of compound. The
dominant negative TCF4 contains a DNA-binding domain
but lacks the b-catenin-interaction motif, which leads to the
sequestration of TCF-binding sites and subsequent blockage
of b-catenin- and TCF-dependent transcriptional activity.[10]
The gene encoding the dominant negative TCF4 was gen-
erated by PCR and cloned into the eukaryotic expression
vector pcDNA3. As shown in Figure 2, b-catenin- and TCF-
transcriptional activity is dependent on Wnt signaling and can
be used to monitor the activity of the Wnt signaling path-
way.[1] The reporter construct encodes the luciferase gene
driven by b-catenin/TCF-responsive elements.[10] 293T cells
were transiently transfected with the reporter construct in
T75 cell culture flasks and replated onto 384-well cell culture
plates after overnight culture; 24 h later, cells were treated
with library compounds at a concentration of 10 mm. After
incubation for another 24 h, luciferase activity was measured,
and a number of compounds were identified as potent
modulators of the Wnt/b-catenin signaling pathway. One
such compound, 2-amino-4-[3,4-(methylenedioxy)benzyl-
amino]-6-(3-methoxyphenyl)pyrimidine (1; Figure 1a), indu-
ces b-catenin- and TCF-dependent transcriptional activity in
a dose-dependent manner with an EC50 value of 0.7 mm
(Figure 1b).
Figure 2. Activity of compound 1 blocked by the presence of a domi-
nant-negative TCF4. 293Tcells were treated with DMSO or 1 (0.2–
5 mm) in the presence or absence of a dominant negative TCF4.
Figure 1. a) Structure of compound 1; b) compound 1 induces the
b-catenin/TCF-dependent reporter in a dose-dependent manner.
Irel =relative intensity of the luciferase signal, reflective of the fold
Irel =relative intensity of the luciferase signal, reflective of the fold
activation in reporter gene expression.
activation in reporter gene expression.
To further explore the structure–activity relationship, 124
similar compounds bearing the core 4,6-disubstituted 2-
aminopyrimidine scaffold were analyzed in the same reporter
assay for activation of the Wnt/b-catenin signaling pathway.
Only a fraction of these structurally related compounds
function as Wnt agonists. The active compounds have small
aryl substituents (such as phenyl or methoxyphenyl groups) at
position 6 of the pyrimidine ring. Substitution with bulky aryl
groups at this position dramatically decreases activity. The
introduction of an additional O or NH linkage between the
pyrimidine ring and the aryl groups at position 6 also leads to
a loss in activity. Only two substituents at position 4 were
tolerated in active compounds: the 3,4-(methylenedioxy)ben-
dependent reporter gene expression induced by compound 1
was blocked in the presence of the dominant negative TCF4,
thus suggesting that the active compound functions in the
canonical Wnt pathway through the TCFs.
A number of known agonists of the Wnt pathway act
through inhibition of the activity of the kinase GSK-3b.[7,11,12]
As GSK-3b is involved in multiple signaling pathways other
than Wnt, a small molecule that blocks GSK-3b could have
disparate effects in cellular and organism models. To deter-
mine if compound 1 is an inhibitor of GSK-3b, its activity in
an in vitro GSK-3b kinase assay was measured. As described
previously,[13] purified GSK-3b was incubated with its sub-
strate, a peptide derived from glycogen synthetase, and g-33P-
1988
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 1987 –1990

Angewandte
Chemie
labeled ATP with or without compound 1. The kinase activity
of GSK-3b can be blocked by the known inhibitor, stauro-
sporine (Supporting Information).[11] The pyrimidine ana-
logues have no such effect, which suggests that they act
independently of GSK-3b. Although the molecular targets of
the active compounds and their mechanism of action are still
under investigation, current results indicate that the active
compounds activate Wnt without inhibiting GSK-3b.
We next tested the ability of pyrimidine 1 to function in a
whole-organism model. During early Xenopus development,
Wnt antagonism plays an essential role in head specification.
The expression of Wnt inhibitors results in enlarged heads
and forebrains.[14–20] Conversely, inactivation of Wnt inhib-
itors, or overexpression of Wnt activators leads to micro-
cephalic embryos.[21] This developmental process provides an
excellent experimental means to test the ability of a
compound to modulate Wnt signaling in vivo. Embryos
were treated with compound 1 (10 mm) or the vehicle,
DMSO, at stage 10.5 (gastrulation stage) for 24 h, and allowed
to develop until stage 40 (tadpole stage). All the vehicle-
treated embryos (n = 29) appeared normal (Figure 3). All
model. Determination of the molecular mechanism of 1 may
reveal novel components of Wnt signaling, including the
effects of oscillating pathway activities.[22,23] Furthermore,
small-molecule Wnt modulators should be useful tools for
studies of biological processes that involve Wnt signaling,
such as hematopoietic stem cell renewal, mesenchymal
progenitor cell differentiation, and embryonic limb develop-
ment. They may also reveal new physiological effects of Wnt
signaling in model organisms, and might ultimately lead to
novel therapeutic targets.[24]
Experimental Section
1: All chemicals were purchased from Aldrich. 2-Amino-4,6-dichloro-
pyrimidine (100 mg, 0.61 mmol) was dissolved in 1-butanol (5 mL).
Diisopropylethylamine (117 mL, 0.67 mmol) and piperonylamine
(92.2 mg, 0.61 mmol) were then added to the mixture. The reaction
mixture was heated at 708C for 12 h, and the desired product (2-
amino -4-chloro-6-(1,3-benzodioxol-5-ylmethylamino)pyrimidine)
was purified by column chromatography (85% yield). 2-Amino-4-
chloro-6-(1,3-benzodioxol-5-ylmethylamino)pyrimidine
(10 mg,
0.036 mmol), [Pd(PPh3)4] (2.1 mg, 0.0018 mmol), Na2CO3 (15.2 mg,
0.144 mmol), acetonitrile/water (1:1, 2 mL), and 3-methoxyphenyl-
boronic acid (11 mg, 0.072 mmol) were mixed together under Ar.
The reaction mixture was then heated at 1508C for 20 min in a
microwave reactor. The product was then purified by using reversed-
phase HPLC with trifluoroacetic acid (0.1% in H2O) and MeCN as
solvents. A linear gradient of MeCN (5%!95%) over 5 min was
used, in which 1 has a retention time of 3.5 min. The corresponding
fractions were collected and lyophilized to yield pure 1 as a white
powder (80% yield). 1H NMR (400 MHz, [D6]DMSO): d = 3.84 (s,
3H), 4.54 (d, 2H, J = 5.8 Hz), 6.0 (s, 2H), 6.39 (s, 1H), 6.85 (m, 2H),
6.96 (s, 1H), 7.19 (m, 1H), 7.28 (m, 2H), 7.52 (m, 1H), 9.13 ppm (s,
1H); HRMS (ESI-TOF): C19H19N4O3 [M+H]+ calcd: 351.1457,
found: 351.1455.
Cell culture, transfection, and high-throughput screening: Human
embryonic kidney (HEK) 293T cells were cultured in Dulbeccoꢀs
modified Eagleꢀs medium (Invitrogen) supplemented with fetal
bovine serum (10%) and the antibiotics penicillin (100 IUmLꢀ1
)
and streptomycin (100 mgmLꢀ1). Cells were cultured in a humidified
incubator at 378C in CO2 (5%). Cells were transiently transfected
with FuGene 6 (Roche) at a FuGene/DNA ratio of 3:1 according to
the manufacturerꢀs instructions. Luciferase activity was measured
with a BrightGlo or DualGlo luciferase kit (Promega). For high-
throughput screening, 293T cells were transiently transfected with the
reporter construct (Upstate) in T75 cell culture flasks. After over-
night culture, cells were harvested and replated onto 384-well plates
with 1.5 ꢁ 103 cells in each well; 24 h later, compound solution (1 mm,
500 nL) was added to each well. After incubation for another 24 h,
luciferase activity was measured with BrightGlo (Promega).
Figure 3. Compound 1 affected Xenopus embryonic head specification.
Embryos were treated with compound 1 (10 mm) or DMSO from
developmental stage 10.5 to stage 40. All DMSO-treated embryos
appeared normal (n=29). All drug-treated embryos (n=36) had
substantial head defects.
Plasmid construction: A dominant negative TCF4 was generated
by PCR with primer F1: GCGCGC GGATCC GCCGCCACC ATG
ACGA ATCAAAACAG CTCCTC and primer R1: GGGTTAGG-
GATAGGCTTACC TTCTAAAGACTTGGTGACGA. The product
was used in a second round of PCR with primer F1 and primer R2
GCCGCC TCTAGA TTA CGTAGAATCGAGACCGAGGA-
GAGGGTTAGGGATAGGCTTACC to add a V5 epitope tag at
the C terminus. It then was digested with BamH1 and XbaI, inserted
into the pcDNA3 vector (Invitrogen), and sequenced.
embryos treated with compound 1 (n = 36) had substantial
head defects that ranged from significantly diminished heads
and complete loss of eyes, to reduced eyes and heads; the
posterior structures of tadpoles remained largely intact. This
phenotype is similar to that which results from Wnt over-
expression in head specification during early embryonic
development.[21] These results suggest that the small-molecule
Wnt agonist indeed mimics the effects of Wnt at the whole-
organism level.
Received: November 9, 2004
Published online: February 21, 2005
In conclusion, the small-molecule Wnt agonist identified
in this study functions both in cell culture and a Xenopus
Keywords: cellular adhesion · gene expression · high-
throughput screening · pyrimidines · signal transduction
.
Angew. Chem. Int. Ed. 2005, 44, 1987 –1990
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1989

Communications
[1] M. Peifer, P. Polakis, Science 2000, 287, 1606.
[2] A. Giannis, Angew. Chem. 2000, 112, 2349; Angew. Chem. Int.
Ed. 2000, 39, 2261.
[3] M. Bienz, H. Clevers, Cell 2000, 103, 311.
[4] X. He, Dev. Cell 2003, 4, 791.
[5] C. A. Grimes, R. S. Jope, Prog. Neurobiol. 2001, 65, 391.
[6] S. Ding, N. S. Gray, X. Wu, Q. Ding, P. G. Schultz, J. Am. Chem.
Soc. 2002, 124, 1594.
[7] S. Ding, T. Y. Wu, A. Brinker, E. C. Peters, W. Hur, N. S. Gray,
P. G. Schultz, Proc. Natl. Acad. Sci. USA 2003, 100, 7632.
[8] X. Wu, S. Ding, Q. Ding, N. S. Gray, P. G. Schultz, J. Am. Chem.
Soc. 2002, 124, 14520.
[9] X. Wu, S. Ding, Q. Ding, N. S. Gray, P. G. Schultz, J. Am. Chem.
Soc. 2004, 126, 1590.
[10] M. Molenaar, M. van de Wetering, M. Oosterwegel, J. Peterson-
Maduro, S. Godsave, V. Korinek, J. Roose, O. Destree, H.
Clevers, Cell 1996, 86, 391.
[11] J. A. Bertrand, S. Thieffine, A. Vulpetti, C. Cristiani, B.
Valsasina, S. Knapp, H. M. Kalisz, M. Flocco, J. Mol. Biol.
2003, 333, 393.
[12] L. Meijer, A. L. Skaltsounis, P. Magiatis, P. Polychronopoulos, M.
Knockaert, M. Leost, X. P. Ryan, C. A. Vonica, A. Brivanlou, R.
Dajani, C. Crovace, C. Tarricone, A. Musacchio, S. M. Roe, L.
Pearl, P. Greengard, Chem. Biol. 2003, 10, 1255.
[13] Y. Wan, W. Hur, C. Y. Cho, Y. Liu, F. J. Adrian, O. Lozach, S.
Bach, T. Mayer, D. Fabbro, L. Meijer, N. S. Gray, Chem. Biol.
2004, 11, 247.
[14] M. A. Deardorff, C. Tan, L. J. Conrad, P. S. Klein, Development
1998, 125, 2687.
[15] A. Glinka, W. Wu, D. Onichtchouk, C. Blumenstock, C. Niehrs,
Nature 1997, 389, 517.
[16] A. Glinka, W. Wu, H. Delius, A. P. Monaghan, C. Blumenstock,
C. Niehrs, Nature 1998, 391, 357.
[17] S. Hoppler, J. D. Brown, R. T. Moon, Genes Dev. 1996, 10, 2805.
[18] K. Itoh, T. L. Tang, B. G. Neel, S. Y. Sokol, Development 1995,
121, 3979.
[19] L. Leyns, T. Bouwmeester, S. H. Kim, S. Piccolo, E. M.
De Robertis, Cell 1997, 88, 747.
[20] S. B. Pierce, D. Kimelman, Dev. Biol. 1996, 175, 256.
[21] G. Davidson, B. Mao, I. del Barco Barrantes, C. Niehrs, Devel-
opment 2002, 129, 5587.
[22] A. Hoffmann, A. Levchenko, M. L. Scott, D. Baltimore, Science
2002, 298, 1241.
[23] K. Serth, K. Schuster-Gossler, R. Cordes, A. Gossler, Genes Dev.
2003, 17, 912.
[24] T. Reya, A. W. Duncan, L. Ailles, J. Domen, D. C. Scherer, K.
Willert, L. Hintz, R. Nusse, I. L. Weissman, Nature 2003, 423,
409.
1990
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 1987 –1990