Indole-2-amide based biochemical antagonist of Dishevelled PDZ domain interaction down-regulates Dishevelled-driven Tcf transcriptional activity

Article abstract of DOI:10.1016/j.bmcl.2007.12.039

We designed and synthesized a series of indole-2-amide-based compounds that antagonize interaction between the Dishevelled (Dvl) PDZ domain and a peptide derived from the natural PDZ ligand Frizzled-7 (Fz7). These compounds inhibit Tcf-mediated transcript

Full text of DOI:10.1016/j.bmcl.2007.12.039

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 946–949
Indole-2-amide based biochemical antagonist of Dishevelled
PDZ domain interaction down-regulates Dishevelled-driven
Tcf transcriptional activity
Neeraj Mahindroo, Chandanamali Punchihewa, Allison M. Bail and Naoaki Fujii*
Department of Chemical Biology and Therapeutics, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA
Received 15 November 2007; revised 13 December 2007; accepted 17 December 2007
Available online 23 December 2007
Abstract—We designed and synthesized a series of indole-2-amide-based compounds that antagonize interaction between the
Dishevelled (Dvl) PDZ domain and a peptide derived from the natural PDZ ligand Frizzled-7 (Fz7). These compounds inhibit
Tcf-mediated transcription activated by exogenous Dvl via the biochemical antagonism. We confirmed tumor cell-selective activa-
tion of caspases by these compounds.
Ó 2007 Elsevier Ltd. All rights reserved.
Wnt glycoproteins comprise a family of growth factors
that interact with multiple components of the cellular
machinery to stimulate diverse signaling pathways.1
The most significant event in Wnt-promoted cellular
growth is activation of transcription by b-catenin/Tcf
complex (referred to hereafter as Tcf transcription) after
nuclear translocation of hypo-phosphorylated b-cate-
nin. Tcf activates the transcription of several tumor-re-
lated genes, including cyclin-D1,2 c-myc,3 and survivin.4
Dishevelled (Dvl) promotes the Tcf transcriptional acti-
vation by suppressing the GSK3/APC tumor suppressor
complex.5 Several lines of evidence suggest that Dvl
overexpression is sufficient to enhance tumor growth.6,7
Expression of exogenous Dvl is well known to activate
Tcf transcriptional activity without co-treatment with
Wnt.8 These findings suggest that Dvl can induce the
Tcf transcription independently of the upstream compo-
nents of the Wnt signaling pathway. APC is an up-reg-
ulator of caspases, which are crucial mediators of
apoptosis induction.9 Consistently, Wnt signaling pre-
vents the caspase activation.10 Taken together, the avail-
able evidence suggests that inhibition of the Dvl
function in cancer cells may suppress their growth and
trigger their apoptosis by up-regulating the caspases.
The Dvl-family proteins contain a PDZ domain that
interacts with several proteins including Fz7,11 Dap-
per,12 and Idax.13 The functional consequences of the
Dvl PDZ interactions are compromised among these
binding partners, whose expression level depends on
the cellular context. RNAi approaches cannot elucidate
Dvl functions specific to the PDZ domain because Dvl is
a multi-domain, multi-functional molecule. We envi-
sioned that a small molecule antagonist of the Dvl
PDZ domain interactions would be a good tool to inves-
tigate the function of the Dvl PDZ domain in the Tcf
transcription and the caspase activation.
We have reported chemical scaffolds based on indole-
2-14,15 and -3-carbinol16,17 that bind to PDZ domains
by mimicking a peptide sequence of b-strand ligand.
Here, we report a series of new indole-2-amide-based
compounds that are biochemical and functional antago-
nists of Dvl PDZ domain interactions.
Our indole scaffold adopts diversity at the 3- and 2-posi-
tions, which are likely to mimic the side chains of the
second and fourth amino acid residues, respectively, of
the extra-carboxyl terminus of the PDZ domain
ligand.14,16 The second amino acid residue does not
make extensive contact to the PDZ domain.18 We in-
stalled a 2-phenylethyl group at the indole-3-position,
which we have successfully utilized to design a general
scaffold for targeting of PDZ domains.14–17 The fourth
amino acid residue (mimicked by the indole-2-position)
Keywords: Dishevelled; Tcf; PDZ domain; Wnt; Indole; Anti-cancer;
Apoptosis.
*
Corresponding author. Tel.: +1 901 495 5854; fax: +1 901 495
5715; e-mail: naoaki.fujii@stjude.org
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.12.039

N. Mahindroo et al. / Bioorg. Med. Chem. Lett. 18 (2008) 946–949
947
defines the preference and selectivity of canonical PDZ
domain interactions.19 However, the PDZ domain of
Dvl has shown to bind to both the extra-carboxyl ter-
mini12 and with internal11,13 sequences, with no clear
preference defined by the fourth amino acid region.
Therefore, we postulated that screening of a chemical li-
brary having diversity at the indole-2-position would be
feasible to investigate the structure–activity relationship
of binding to this PDZ domain, rather than rational de-
sign from the ligand sequences.
we designed a biotin-tagged peptide derived from the
intracellular region of the Fz7 Wnt receptor, which con-
tains the KTXXXW motif responsible for the direct
binding11 of Fz7 and its functional consequence as a
Wnt receptor.21 We chose GST-tagged PDZ domain
of human Dvl for the assay, because it offers higher sig-
nal intensity than does His-tagged PDZ domain. The
signal generated by the interaction of the biotinylated
Fz7 peptide and the GST-fused human Dvl PDZ do-
main was titrated with the test compounds. As the con-
centration of the compound increased, we observed
diminishing signal, signifying competitive antagonism.
Representative competition curves are shown in
Figure 1. The IC50 activity of biochemical antagonism
is summarized in Table 1.
We designed indole-2-carboxylic acid 4 as a key scaffold
for library production (Scheme 1). A pyruvic acid deriv-
ative 2 that contains the 2-phenylethyl moiety was pre-
pared from 3-phenylpropylmagnesium bromide and
dimethyl oxalate. The preparation of 4 proceeded
smoothly via palladium-catalyzed coupling20 of 2 with
iodoaniline 3.15 The indole-2-carboxylic acid 4 was cou-
pled with various amines to prepare the amide library
6a–h. To assess the significance of hydrogen donors on
the indole scaffold, we also prepared methylated variants
of 6e (5e and 7e, Scheme 1).
We began exploring structure–activity relationships
(SAR) by changing the size of the indole-2-amide side
chain of 6a. 6a showed only weak, incomplete antago-
nism (ꢀ30% of full inhibition) even at a concentration
of 1 mM. However, the extension and incorporation of
methyl groups on this side chain increased the biochem-
ical potency of antagonism (compare 6a to 6b), suggest-
ing the presence of a hydrophobic cavity in the PDZ
domain that accommodates a small hydrophobic group.
However, binding was decreased by too-close proximity
of the methyl substitutions to the indole scaffold (com-
pare 6c to 6b). Incorporation of carboxylic acid on this
side chain did not increase the activity (6h). A cyclopro-
pyl group at this position afforded incomplete antago-
nism (6d), similar to that of 6a. Meanwhile, a larger
carbocyclic ring increased the efficacy of antagonism,
as compounds possessing a cyclohexyl group at this po-
sition (6e, 6f) showed increased biochemical antago-
nism. These data clearly show the preference of the
To assay these compounds’ biochemical antagonism of
the Dvl PDZ domain interactions, we utilized Alpha-
Screen energy transfer assays, because they were not af-
fected by auto-fluorescence of the indole scaffold. As a
probe for detecting the Dvl PDZ domain interaction,
I
CO2Me
a, b, c
O
Br
3
Me
H2N
CO2H
2
1
d
e, f
g
CO2Me
Me
5a-5h
CO2Me
Me
R NH
O
HO2C
N
H
N
H
4
R=
6c
6a
6d
6b
CO2H
Me
R NH
O
5e
6e
N
H
6a-6h
HO2C
6f
6h
6g
h, g
5e
CO2H
Me
NH
7e
N
O
Me
Scheme 1. (a) Magnesium turnings, THF, reflux. (b) Dimethyl oxalate
(3 equiv), THF, 0 °C, 1 h. (c) Sodium hydroxide (3 equiv), MeOH,
room temp, 0.5 h. Yield of steps a–c = 52%. (d) 2 (5 equiv), 315 (1
equiv), palladium acetate (0.15 equiv), DABCO (3 equiv), DMF,
100 °C, 18 h. Yield = 64% based on 3. (e) HBTU (2.5 equiv),
diisopropylethylamine (3 equiv), DMF, room temp, 1 h. (f) RNH2 (3
equiv), room temp, 18 h. For preparation of 5h benzyl 3-aminopro-
pionate was used as the RNH2. (g) 10% sodium hydroxide aq (3 equiv),
MeOH, 1, 4-dioxane, 90 °C, 18 h. (h) MeI, K2CO3, DMF, microwave,
150 °C, 15 min. Yield = 95%.
Figure 1. Representative competition curves measured by Alpha-
Screen. The GST-fused Dvl1 PDZ domain protein and biotin-Fz7
peptide were allowed to equilibrate and were then incubated with test
compound to titrate competitive inhibition of the peptide binding. The
binding curve was obtained as the mean of three independent
experiments, and the IC50 (Table 1) was calculated by Prism
(GraphPad software). Compound 6a has shown incomplete antago-
nism. Corresponding % of DMSO control has shown no activity.

948
N. Mahindroo et al. / Bioorg. Med. Chem. Lett. 18 (2008) 946–949
Table 1. Biochemical antagonism (IC50) of the interaction of the Dvl
PDZ domain and the Fz7 PDZ binding region in a competitive
AlphaScreen assay; and inhibitory effect on Dvl1-activated Tcf
reporter activity in HEK293T cells, normalized by cellular viability
and shown as DMSO = 0%
Compound Dvl1 PDZ Dvl3 PDZ Tcf reporter inhibition%
IC50 (lM) IC50 (lM) at 100 lM
6a
6b
6c
6d
6e
6f
>500
159
>500
>500
49
>500
160
>500
>500
23
5>
26
5>
ND
27
157
197
>500
97
309
291
>500
57
16
23
6g
6h
5e
7e
5>
35
78
92
39
Figure 2. Caspase activation in HepG2, Rh30, and BJ cells. The cells
were treated with test compound (100 lM) for 2 days, and caspase
activity was measured by Caspase3/7-Glo (Promega) reagent. The
activity signal is represented as the mean proportion of that of DMSO
control (considered equivalent to 1) in three independent experiments.
Error bars represent standard deviation.
ND, no data.
Dvl PDZ domain to a mid-sized hydrophobe in the in-
dole-2-position.
The indole-5-carboxylic acid mimics the extra carboxyl
terminus of ligand of the PDZ domain,14,16 that is an
essential motif in interacting to canonical PDZ domains
through hydrogen-bond network.18 However, Dvl PDZ
domain binds to both the extra carboxy-terminal ligand
(Dapper12) and internal sequence (Fz711), thus it is ex-
pected that the carboxylic acid is not essential in binding
to this PDZ domain. Consistently, methylation at this
position (5e) only modestly diminished the potency.
Methylation at the indole-1-position also modestly but
not dramatically diminished the potency (7e), suggesting
that the Dvl PDZ domain may have small hydrophobic
cavities in the region that fit the indole-1-methyl. It is
noteworthy that those three indole-2-amide compounds
(5e, 6e, 7e) maintain the activity in spite of lacking a hy-
droxyl group on the indole,14–17 that mimics essential
hydrogen-bond interaction of S/T(À2) residue of class1
PDZ domain ligands. Taken together, these SAR data
suggest that ligand binding of the Dvl PDZ domain is
controlled by hydrophobic interaction, rather than
hydrogen-bond formation.
scription at 100 lM (Table 1) but not at 50 lM. Some
discrepancy observed in between the biochemical antag-
onism potency and transcriptional down-regulation
activity may be due to difference of cellular permeability
or stability in the cell cultures. These results suggest that
inhibition of the Tcf transcription reflects the biochemi-
cal antagonism to the Dvl PDZ domain. No compounds
have significantly reduced the viability signal, suggesting
that the down-regulation of the Tcf transcription is not
cytotoxic effect.
To confirm the apoptosis effect of the indole-2-amide
compounds, we assayed caspase activity in cells after
treatment with the compounds (Fig. 2). In HepG2 hepa-
tocellular carcinoma cells and Rh30 rhabdomyosarcoma
cells, treatment with all compounds except 5e signifi-
cantly increased caspase-3/7 activity. In contrast, such
activation was not observed in hTERT-immortalized
BJ cells23 normal human foreskin fibroblasts, except
those treated with 6f. These results suggest that Tcf acti-
vation of transcriptional down-regulation can induce
apoptosis selectively in cancer cells.
Next, we assayed inhibition of the up-regulation of Tcf
transcription by exogenous Dvl in HEK293T cells,
which are believed to respond normally to the Wnt sig-
naling pathways. As these cells have no mutation or
overexpression that upregulates Tcf transcription, their
intrinsic Tcf transcription activity is at background le-
vel. However, their Tcf activity significantly increases
upon transient expression of Dvl, signifying Dvl-driven
up-regulation of Tcf transcription. We have transiently
transfected HEK293T cells with superTOPflash plas-
mid,22 an optimized reporter for Tcf activation of tran-
scriptional activity, and an expression vector encoding
Dvl1. The cells were then incubated with the test com-
pound for 1 day. Tcf reporter activity was measured
as firefly luciferase and was normalized by the cellular
viability to express activity of Tcf transcription. As ex-
pected, 6a, 6c, and 6h, whose activity in the AlphaScreen
assay was very weak, showed no inhibition. Other com-
pounds with biochemical activity inhibited the Tcf tran-
In summary, we have demonstrated that the indole-2-
amide scaffold can be utilized to functionally antagonize
interactions with the PDZ domain of Dvl, thereby
down-regulating Dvl-driven Tcf activation of transcrip-
tion. The biochemical potency of this antagonism is en-
hanced by incorporating a small hydrophobic group in
the substituent on the 2-amide moiety. Compounds de-
rived from this scaffold can cause apoptosis selectively
in cancer cells.
Acknowledgments
We thank Randall T. Moon (University of Washington)
for his generous gift of superTOPflash plasmid; Liwen
He for providing compound 3; Michele Connelly for cell
culture assistance; and Sharon Naron for scientific edit-

N. Mahindroo et al. / Bioorg. Med. Chem. Lett. 18 (2008) 946–949
949
9. Chen, T.; Yang, I.; Irby, R.; Shain, K. H.; Wang, H. G.;
Quackenbush, J.; Coppola, D.; Cheng, J. Q.; Yeatman, T.
J. Cancer Res. 2003, 63, 4368.
10. You, Z.; Saims, D.; Chen, S.; Zhang, Z.; Guttridge, D. C.;
Guan, K. L.; MacDougald, O. A.; Brown, A. M.; Evan, G.;
Kitajewski, J.; Wang, C. Y. J. Cell Biol. 2002, 157, 429.
11. Wong, H. C.; Bourdelas, A.; Krauss, A.; Lee, H. J.; Shao,
Y.; Wu, D.; Mlodzik, M.; Shi, D. L.; Zheng, J. Mol. Cell
2003, 12, 1251.
ing of the manuscript. Financial support was provided
by the American Lebanese Syrian Associated Charities
(ALSAC) and Pediatric Oncology Education Program
Grant 5 R25 CA23944 from the National Cancer Insti-
tute (AMB).
Supplementary data
12. Cheyette, B. N.; Waxman, J. S.; Miller, J. R.; Takemaru,
K.; Sheldahl, L. C.; Khlebtsova, N.; Fox, E. P.; Earnest,
T.; Moon, R. T. Dev. Cell 2002, 2, 449.
13. London, T. B.; Lee, H. J.; Shao, Y.; Zheng, J. Biochem.
Biophys. Res. Commun. 2004, 322, 326.
14. Fujii, N.; Haresco, J. J.; Novak, K. A.; Gage, R. M.;
Pedemonte, N.; Stokoe, D.; Kuntz, I. D.; Guy, R. K.
Bioorg. Med. Chem. Lett. 2007, 17, 549.
Synthetic procedure, analysis data, protocols for bio-
chemical assay, and cell-based assay. Supplementary
data associated with this article can be found, in the on-
line version. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.bmcl.2007.12.039.
15. Fujii, N.; You, L.; Xu, Z.; Uematsu, K.; Shan, J.; He, B.;
Mikami, I.; Edmondson, L. R.; Neale, G.; Zheng, J.; Guy,
R. K.; Jablons, D. M. Cancer Res. 2007, 67, 573.
16. Fujii, N.; Haresco, J. J.; Novak, K. A.; Stokoe, D.; Kuntz,
I. D.; Guy, R. K. J. Am. Chem. Soc. 2003, 125, 12074.
17. Fujii, N.; Shelat, A.; Hall, R. A.; Guy, R. K. Bioorg. Med.
Chem. Lett. 2007, 17, 546.
References and notes
1. Kikuchi, A.; Yamamoto, H.; Kishida, S. Cell. Signalling
2007, 19, 659.
2. Tetsu, O.; McCormick, F. Nature 1999, 398, 422.
3. He, T. C.; Sparks, A. B.; Rago, C.; Hermeking, H.; Zawel,
L.; da Costa, L. T.; Morin, P. J.; Vogelstein, B.; Kinzler,
K. W. Science 1998, 281, 1509.
4. Kim, P. J.; Plescia, J.; Clevers, H.; Fearon, E. R.; Altieri,
D. C. Lancet 2003, 362, 205.
5. Wharton, K. A., Jr. Dev. Biol. 2003, 253, 1.
6. Uematsu, K.; He, B.; You, L.; Xu, Z.; McCormick, F.;
Jablons, D. M. Oncogene 2003, 22, 7218.
18. Doyle, D. A.; Lee, A.; Lewis, J.; Kim, E.; Sheng, M.;
MacKinnon, R. Cell 1996, 85, 1067.
19. Harris, B. Z.; Lim, W. A. J. Cell Sci. 2001, 114, 3219.
20. Chen Cy, C.; Lieberman, D. R.; Larsen, R. D.; Verho-
even, T. R.; Reider, P. J. J. Org. Chem. 1997, 62, 2676.
21. Umbhauer, M.; Djiane, A.; Goisset, C.; Penzo-Mendez,
A.; Riou, J. F.; Boucaut, J. C.; Shi, D. L. EMBO J. 2000,
19, 4944.
7. Uematsu, K.; Kanazawa, S.; You, L.; He, B.; Xu, Z.; Li,
K.; Peterlin, B. M.; McCormick, F.; Jablons, D. M.
Cancer Res. 2003, 63, 4547.
22. Angers, S.; Thorpe, C. J.; Biechele, T. L.; Goldenberg, S.
J.; Zheng, N.; MacCoss, M. J.; Moon, R. T. Nat. Cell Biol.
2006, 8, 348.
8. Li, L.; Yuan, H.; Xie, W.; Mao, J.; Caruso, A. M.;
McMahon, A.; Sussman, D. J.; Wu, D. J. Biol. Chem.
1999, 274, 129.
23. Jiang, X. R.; Jimenez, G.; Chang, E.; Frolkis, M.; Kusler,
B.; Sage, M.; Beeche, M.; Bodnar, A. G.; Wahl, G. M.;
Tlsty, T. D.; Chiu, C. P. Nat. Genet. 1999, 21, 111.