New Hedgehog/GLI signaling inhibitors from Excoecaria agallocha

Article abstract of DOI:10.1016/j.bmcl.2010.11.126

The inhibition of the Hedgehog (Hh) signaling pathway has emerged as an anti-cancer strategy. Three flavonoid glycosides including 2 new compounds (1-2) were isolated from Excoecaria agallocha as Hedgehog/GLI1-mediated transcriptional inhibitors and exhibited cytotoxicity against human pancreatic (PANC1) and prostate (DU145) cancer cells. Our data revealed that compound 1 clearly inhibited the expression of GLI-related proteins (PTCH and BCL-2) and blocked the translocation of GLI1 transcription factors into the nucleus in PANC1. Deleting the Smoothened (Smo) function in PANC1 treated with 1 led to downregulation of the mRNA expression of Ptch. This study describes the first Hh signaling inhibitor which blocks GLI1 movement into the nucleus without interfering with Smo.

Full text of DOI:10.1016/j.bmcl.2010.11.126

Bioorganic & Medicinal Chemistry Letters 21 (2011) 718–722
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
New Hedgehog/GLI signaling inhibitors from Excoecaria agallocha
Yusnita Rifai a, Midori A. Arai a, Samir K. Sadhu b, Firoj Ahmed b, Masami Ishibashi a,
⇑
a Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 263-8522, Japan
b Pharmacy Discipline, Life Science School, Khulna University, Khulna 9208, Bangladesh
a r t i c l e i n f o
a b s t r a c t
Article history:
The inhibition of the Hedgehog (Hh) signaling pathway has emerged as an anti-cancer strategy. Three
flavonoid glycosides including 2 new compounds (1–2) were isolated from Excoecaria agallocha as
Hedgehog/GLI1-mediated transcriptional inhibitors and exhibited cytotoxicity against human pancreatic
(PANC1) and prostate (DU145) cancer cells. Our data revealed that compound 1 clearly inhibited the
expression of GLI-related proteins (PTCH and BCL-2) and blocked the translocation of GLI1 transcription
factors into the nucleus in PANC1. Deleting the Smoothened (Smo) function in PANC1 treated with 1 led
to downregulation of the mRNA expression of Ptch. This study describes the first Hh signaling inhibitor
which blocks GLI1 movement into the nucleus without interfering with Smo.
Received 26 October 2010
Revised 25 November 2010
Accepted 30 November 2010
Available online 5 December 2010
Keywords:
Excoecaria agallocha
Hedgehog
GLI1-mediated transcriptional activity
Inhibitor
Ó 2010 Elsevier Ltd. All rights reserved.
The Hedgehog signaling pathway is critical for cell growth and
stem cell maintenance. Damage to essential core components of
the Hh pathway often results in congenital birth defects1 whereas
an aberrant activated Hh pathway leads to cancer. In cancer cells,
binding of the Hh protein ligand to receptor patched 1 (PTCH)
abolishes the inhibitory effect of PTCH on proto-oncogene Smooth-
ened (Smo), thereby allowing Smo to activate the GLI family of
transcription factors.2
In recent reports, three sites have been mainly targeted to
identify Hh-signaling inhibitors: Smo protein (cyclopamine,3a,b
SANTs,3c and CUR614143d), GLI protein (GANTs3e) and Hh ligand
(robotnikinin3f). Several studies have suggested that GLI1 is one
of the critical transcriptional effectors closely associated with
tumor formation by direct association with a specific binding site
(50-GACCACCCA-30) in the promoter region of the target genes.4
The regulation of GLI1 in mediating oncogenic Hh signaling is
poorly known. Some reports describe that activated mutations of
Smo induce the nuclear translocation of transcriptional GLI fac-
tors.5a,b The use of small molecule inhibitors of Smo showed poten-
tially limited efficacy. These inhibitors only treat tumors where
pathway activation has occurred upstream or at the level of Smo,
yet cancer with other downstream components is unresponsive
to Smo inhibitors.5c Therefore, searching for small molecule inhib-
itors that target GLI1 in an Smo-independent manner will be of
great importance in response to the urgent need for broadly active
downstream inhibitors of Hh signaling.
We have recently reported naturally occurring Hh/GLI-medi-
ated transcriptional inhibitors.6a–c This Letter reports the structural
elucidation of new Hh/GLI1-mediated transcriptional inhibitors
and demonstrates suppression of the protein level and mRNA
expression by the most potent inhibitor (compound 1), isolated
from Excoecaria agallocha leaves.
The MeOH extract of dried leaves of E. agallocha was partitioned
between hexane, EtOAc and BuOH, followed by repeated column
chromatography and preparative HPLC to yield compounds 1–8,
consisted of two new compounds (1 and 2), and 6 known flavonoid
glycosides (3–8) (Fig. 1); afzelin (3),7a quercitrin (4),7b rutin (5),7c
kaempferol-3-O-(2-O-acetyl-
a-L
-rhamnopyranoside (6),7d kaempfe-
ride 3-O-a-L a-L-
-rhamnopyranoside (7),7e and kaempferol 3-O-
arabinofuranoside (8).7f The structure elucidation of known
compounds was on the basis of comparison with spectral data of
the reported values (Table 1).
Compound 1 was obtained as a yellow powder and had a molec-
ular weight at m/z 485.1162 ([M+Na]+,
D
ꢀ2.1 mmu), correspond-
ing to the molecular formula of C22H22O11 in the HR-FABMS. The IR
absorption bands suggested the presence of hydroxyl (3317 cmꢀ1
(br) and carbonyl (1680 cmꢀ1) groups. The UV absorption maxima
were at UV (MeOH) kmax 261 nm (log 3.8) and 355 nm (log 3.0).
Acid hydrolysis of compound 1 gave quercetin and acofriose (3-O-
)
e
e
methyl rhamnose). Acofriose was determined as having an
L
-form
by comparison of its specific rotation ½a D22
ꢁ
+35 (c 0.1, MeOH) with
those reported in the literature (½a D25
ꢁ
+39).8 The carbon signals at dC
177.7 (C-4), 148.4 (C-30) and 145.2 (C-40) were in agreement with
the quercetin skeleton bearing a carbonyl and two hydroxyl
groups. In addition, four signals of OH hydrogens were observed
at dH 12.63 (1H, s), 10.88 (1H, s), 9.71 (1H, s), and 9.34 (1H, s).
⇑
Corresponding author. Tel./fax: +81 43 290 2913.
E-mail address: mish@p.chiba-u.ac.jp (M. Ishibashi).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.126

Y. Rifai et al. / Bioorg. Med. Chem. Lett. 21 (2011) 718–722
719
Figure 1. Chemical structures of compounds 1–8 from E. agallocha.
system at dH 7.66 (J = 2 Hz), 6.84 (J = 8 Hz), and 7.56 (J = 2 Hz and
8 Hz), confirming that the aglycone was quercetin. The 13C NMR
spectrum of 2 showed 27 resonances, including one methoxy
group (dC 48.5) attached to one of the sugar moieties. Comparison
of NMR spectral data of 2 differed from 1 in the sugar chain.
Further HMBC analysis (Fig. 2) indicated the connections of sugar
units between dH 5.46 (H-100) and dC 133.4 (C-3), and between dH
3.63 (H-500) and dC 101.7 (C-1000). Acid hydrolysis and HPLC isolation
gave quercetin,
anomeric proton of the first sugar (
as b since the coupling constant was smaller (J = 3.3 Hz)10 than that
of
(J = 4.5 Hz).11 Furthermore, the small coupling constant ob-
served for the anomeric proton of the second sugar ( -acofriose, d
5.37, J = 2.0 Hz) suggested its -form of the
configuration.9 The
D
-apiose and
L
-acofriose. The orientation of the
D
-apiose, d 5.46) was deduced
a
L
Figure 2. Key 1H–1H COSY and HMBC correlations observed for 1 and 2.
a
D
first sugar connected to aglycone was confirmed by comparing
its specific rotation ½a D21
ꢁ
+6.1 (c 0.03, H2O) with the reported values
The location of the sugar part, according to HMBC analysis (Fig. 2),
was shown to attach to the C-3 of aglycone. The small coupling
of an authentic sample, [
a
]
+5.2 (c 1.1, H2O).12 The downfield
D
location of the C-3000 absorption of the second sugar showed the
constant (1.6 Hz) of the anomeric proton (dH 5.24) indicated its
a
configuration.9
presence of a methoxy group at C-3000. The sugar was identified as
3-O-methyl-rhamnose and was confirmed as having an
the same method as compound 1.
L-form by
Compound 2 was isolated as a yellow amorphous solid and gave
the molecular formula C27H30O15, as deduced from HR-FABMS m/z
617.1584 (calcd for C27H30O15Na, 617.1558). The proton signals
showed two aromatic doublets at dH 6.20 and 6.39, and an ABX
Hh/GLI1 inhibitory effects of all isolated compounds (1–8)
were further evaluated. Compounds 1, and inhibited
2
8

720
Y. Rifai et al. / Bioorg. Med. Chem. Lett. 21 (2011) 718–722
Table 1
viability of compounds was checked simultaneously with lucifer-
ase activity under a fluorimetric microculture cytotoxicity assay
(FMCA) system.13 Of the
active compounds, compound
appeared to be the most potent Hh/GLI-1 inhibitor.
1H and 13C NMR spectral data of 1 and 2
3
1
Position
1
2
a
a
dH (J in Hz)a
dC
dH (J in Hz)a
dC
We further examined the cytotoxicity of active compounds
against PANC1, DU145 and C3H10T1/2 using the FMCA system.
The results revealed that compounds 1, 2, and 8 were cytotoxic
2
3
4
5
6
7
8
9
10
10
20
30
40
50
60
100
200
300
400
156.4
134.2
177.7
161.3
98.6
156.3
133.4
177.5
161.2
98.6
against PANC1 cells (IC50 values of 0.7, 20.0 and 1.8
tively) and DU145 cells (IC50 values of 0.8, 22.0 and 2.4
l
l
M, respec-
M, respec-
6.19 d (2.0)
6.38 d (2.0)
6.20 d (2.0)
6.39 d (2.0)
157.3
93.6
156.2
93.5
tively) but did not affect normal cell lines (Fig. 3B, Table 2). The
cytotoxicity of 1, 2, and 8 against PANC1 and DU145 cells may be
associated with their Hh/GLI transcriptional activity inhibition.
To ensure that the inhibition of Hh signaling by 1 was associ-
ated with the expression of GLI-related proteins (PTCH and
BCL-2) and to identify the effect of 1 on downstream events, we
checked the protein levels of full, cytoplasmic and nuclear PANC1
after treatment with 1. As a result, 1 reduced the expression of
PTCH and BCL-2 proteins in a dose-dependent manner. Our
Western blot result also showed that treatment with 1 at a concen-
156.4
104.1
120.7
115.6
148.4
145.2
115.4
121.1
101.8
70.5
156.1
103.9
121.6
116.2
148.4
144.8
115.2
121.1
100.8
74.0
7.28 d (2.0)
6.86 d (8.0)
7.25 dd (2.0, 8.0)
5.24 d (1.6)
4.92 dd (1.6, 3.0)
3.12 dd (9.0, 3.0)
3.50 d (9.0)
7.66 d (2.0)
6.84 d (8.0)
7.56 dd (2.0, 8.0)
5.46 d (3.3)
4.44 d (3.3)
—
3.57 d (9.6)
3.57 d (9.6)
4.02 d (8.0)
3.63 d (8.0)
71.1
70.0
76.5
77.5
tration of 1.6 lM led to a significant decrease in the protein level of
nuclear GLI1 in PANC1 (Fig. 4A).
500
4.70 m
70.3
17.5
67.9
To further understand the molecular mechanism underlying the
Hh signaling inhibitory effect of 1, we checked the expression of a
GLI-related gene (Ptch) using real-time quantitative RT-PCR. It has
previously been suggested that Ptch is a repressive Hh receptor,
and elevated expression of this gene results in the concomitant
expression of Hh target genes, including GLI1.3e Consistent with
this, our data showed that the downregulation of Ptch mRNA
expression is essential for the inhibition of Hh/GLI1 signaling on
PANC1 treated with 1 (Fig. 4B).
Loss of Ptch function has been reported to cause aberrant Hh
signaling by Smo.14 Most known Hh modulators, including cyclop-
amine, repress the pathway by antagonizing Smo activation; how-
ever, one of the oncogenic Smo mutants (SmoM2) is apparently
resistant to cyclopamine15 and most Smo inhibitors were not effec-
tive against medullablastoma and cancer associated with down-
stream lesions.16
600
0.79 d (6.0)
1000
5.37 d (2.0)
5.07 dd (2.0, 1.2)
3.07 dd (8.0, 1.2)
3.20 d (8.0)
4.95 m
101.7
71.1
73.2
69.9
75.8
17.5
48.5
2000
3000
4000
500
600
0.79 d (6.0)
3.16 s
4000-OCH3
5-OH
7-OH
30-OH
40-OH
3.15 s
12.63 s
10.88 s
9.34 s
9.71 s
48.6
a
In DMSO-d6.
Hh/GLI-mediated transcriptional activity with IC50 values of 0.5,
19.1 and 2.0 M, respectively, whereas compounds 3–7 were inac-
tive (Fig. 3A, Table 2). In this assay, Gli1 protein expression was in-
duced by 12-h treatment with tetracycline. Removal of tetracycline
after another 12 h, followed by the addition of samples eliminated
the undesired production of GLI1 via CMV promoter activation. Cell
To verify the function of Smo in PANC1 during the Hh inhibition
of 1, we knocked down Smo expression in PANC1 cells by siRNA
and performed real-time quantitative RT-PCR on its mRNA expres-
sion. Transfection of a non-targeting siRNA at the same concentra-
tion served as a control. The transfection results, as confirmed by
l
Figure 3. (A) Inhibition of GLI1-mediated transcriptional activity (solid columns) and cell viability (solid curves) of compounds 1, 2, and 8. HaCaT-GLI1-Luc cells were seeded
onto a 96-well plate (2 ꢂ 105 cells per well) and then treated with compounds 12 h after tetracycline addition. Cell viability and luciferase activity were determined at the
same time. (B) Cytotoxicity of compounds 1, 2, and 8 against PANC1, DU145, and C3H10T1/2 cells. Assays were performed at 0.05% DMSO (n = 3). Error bars represent sd.

Y. Rifai et al. / Bioorg. Med. Chem. Lett. 21 (2011) 718–722
721
Table 2
IC50 values (
cells
l
M) of GLI-mediated transcriptional inhibition and cytotoxicity against cancer cells (PANC1, DU145) and C3H10T1/2
Compound
GLI transcriptional inhibition (IC50
,
lM)
Cytotoxicity (IC50
DU145
, lM)
PANC1
C3H10T1/2
1
2
8
0.5
19.1
2.0
0.7
20.0
1.8
0.8
22.0
2.4
>100
>100
>100
Figure 4. (A) Effect of compound 1 on GLI1 and/or GLI-related protein (PTCH and BCL2) levels in full, cytosolic and nuclear PANC1 cells. (B) Inhibition of Ptch mRNA
expression by compound 1 in PANC1 cells. GAPDH was used as an internal control. Assays were performed at 0.05% DMSO (n = 3). Error bars represent sd. (C) Expression of
Ptch mRNA (upper panel) and Smo protein (bottom panel) in PANC1 treated with 1 after siRNA-mediated silencing of Smo.
Western blotting, proved the total depletion of the Smo protein
level after a silencing process (Fig. 4C, bottom panel). Accordingly,
silencing of Smo siRNA significantly reduced the expression of Ptch
mRNA expression in PANC1 cells treated with 1 (Fig. 4C, upper
panel). These results thus indicated that 1 inhibited Hh signaling
in an Smo-independent manner.
that the reduction of Hh signaling was not caused by Smo inter-
ference, but rather by other unrevealed mechanisms related to
the GLI1 inhibitory function. This is the first report of an Hh/
GLI signaling inhibitor which reduces the level of GLI1 in the
nucleus.
Because the inhibitory effect of 1 resulted in the decreased level
of GLI1-related proteins (PTCH and BCL-2) as well as blocking GLI1
transcription factor translocation into the nucleus in PANC1 cells
(Fig. 4A), we may speculate that this compound antagonizes GLI1
activator function. In accordance with this result, several reports
have previously described that GLI function can be modulated
without Smo interference via numerous types of GLI signaling,
such as transforming growth factor-b (TGFb),17a mitogen-activated
protein kinase (MAPK)17b and posphatidylinositol 3-kinase (PI3K)/
Akt signaling.17c
Acknowledgments
We thank Dr. Fritz Aberger and Gerhard Regl (University of
Salzburg) for the kind provision of tetracyclin-regulated HaCaT
cells, and Dr. Rune Toftgård (Karolinska Institute) for the
12GLIRE-TKO luciferase plasmid. This work was supported by a
Grant-in-Aid from the Japan Society for the Promotion of Science
(JSPS).
Supplementary data
In conclusion, we have isolated naturally occurring Hh/GLI
inhibitors including 2 new compounds (1–2) from E. agallocha.
Compound 1 inhibited the translocation of GLI1 transcription fac-
tor into the nucleus of PANC1. Quantitative RT PCR of 1 exhibited
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.11.126.

722
Y. Rifai et al. / Bioorg. Med. Chem. Lett. 21 (2011) 718–722
7. (a) Matthes, H. W. D.; Luu, B.; Ourrison, G. Phytochemistry 1980, 19, 2643; (b)
References and notes
Peng, F. Z.; Strack, D.; Baumert, A.; Subramaniam, R.; Goh, N. K.; Chia, T. F.; Tan,
S. N.; Chia, L. S. Phytochemistry 2003, 62, 219; (c) Wooten, J. B.; Kalengamaliro,
N. E.; Axelson, D. E. Phytochemistry 2009, 70, 940; (d) Masuda, T.; Jitoe, A.; Kato,
S.; Nakatani, N. Phytochemistry 1991, 30, 2391; (e) Bilia, A. R.; Palme, E.; Marsill,
A.; Pistelli, L.; Morelli, I. Phytochemistry 1993, 32, 1078; (f) Kumar, N.; Singh, B.;
Kaul, V. K. Nat. Prod. Commun. 2006, 1, 623.
1. McMahon, A. P.; Ingham, P. W.; Tabin, C. J. Curr. Top. Dev. Biol. 2003, 53, 1.
2. Rubin, L. L.; de Sauvage, F. J. Nat. Rev. Drug Disc. 2006, 5, 1026.
3. (a) Chen, J. K.; Taipale, J.; Cooper, M. K.; Beachy, P. A. Genes Dev. 2002, 16, 2743;
(b) Incardona, J. P.; Gaffield, W.; Kapur, R. P.; Roelink, H. Development 1998, 125,
3553; (c) Chen, J. K.; Taipale, J.; Young, K. E.; Maiti, T.; Beachy, P. A. Proc. Natl.
Acad. Sci. U.S.A. 2002, 99, 14071; (d) Williams, J. A.; Guicherit, O. M.; Zaharian, B.
I.; Xu, Y.; Chai, L.; Wichterle, H.; Kon, C.; Gatchalian, C.; Porter, J. A.; Rubin, L. L.;
Wang, F. Y. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 4616; (e) Lauth, M.; Bergström,
A.; Shimokawa, T.; Toftgård, R. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 8455; (f)
Stanton, B. Z.; Peng, L. F.; Maloof, N.; Nakai, K.; Wang, X.; Duffner, J. L.; Taveras,
K. M.; Hyman, J.; Lee, S. W.; Koehler, A. N.; Chen, J. K.; Fox, J. L.; Mandinova, A.;
Schreiber, S. L. Nat. Chem. Biol. 2009, 5, 154.
8. Muhr, H.; Hunger, A.; Reichstein, T. Helv. Chim. Acta 1954, 31, 403.
9. Yamauchi, T.; Abe, F.; Wan, A. S. C. Chem. Pharm. Bull. 1987, 35, 2744.
10. Kanchanapoom, T.; Kasai, R.; Yamasaki, K. Phytochemistry 2002, 59, 557.
11. Lei, Y.; Wu, L. J.; Shi, H. M.; Tu, P. F. Helv. Chim. Acta 2008, 91, 495.
12. Ho, P. T. Can. J. Chem. 1979, 57, 381.
13. Larsson, R.; Kristensen, J.; Sandberg, C.; Nygren, P. Int. J. Cancer 1992, 50,
177.
14. Carpenter, D.; Stone, D. M.; Brush, J.; Ryan, A.; Armanini, M.; Frantz, G.;
Rosenthal, A.; de Savauge, F. J. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 13630.
15. Taipale, J.; Chen, K. K.; Cooper, M. K.; Wang, B.; Mann, R. K.; Milenkovic, L.;
Scott, M. P.; Beachy, P. A. Nature 2000, 406, 1005.
16. Lee, Y.; Kawagoe, R.; Sasai, K.; Li, Y.; Russell, H. R.; Curran, T.; McKinnon, P. J.
Oncogene 2007, 26, 6442.
17. (a) Dennler, S.; André, J.; Alexaki, I.; Li, A.; Magnaldo, T.; Dijke, P.; Wang, X.;
Verrechia, F.; Muviel, A. Cancer Res. 2007, 67, 6981; (b) Riobo, N. A.; Haines, G.
M., ; Emerson, C. P., Jr. Cancer Res. 2006, 66, 839; (c) Riobo, N. A.; Lu, K.; Ai, X.;
Haines, G. M.; Emerson, C. P., Jr. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 4505.
4. Kinzler, K. W.; Vogelstein, B. Mol. Cell. Biol. 1990, 10, 634.
5. (a) Pasca di Magliano, M.; Hebrok, M. Nat. Rev. Cancer 2003, 3, 903; (b) Jacob, L.;
Lum, L. Science 2007, 318, 66; (c) Taylor, M. D.; Liu, L.; Raffel, C.; Hui, C. C.;
Mainprize, T. G.; Zhang, X.; Agatep, R.; Chiappa, S.; Gao, L.; Lowrance, A.; Hao,
A.; Goldstein, A. M.; Stavrou, T.; Scherer, S. W.; Dura, W. T.; Wainwright, B.;
Squire, J. A.; Rutka, J. T.; Hogg, D. Nat. Genet. 2002, 31, 306.
6. (a) Hosoya, T.; Arai, M. A.; Koyano, T.; Kowithayakorn, T.; Ishibashi, M.
Chembiochem 2008, 9, 1082; (b) Arai, M. A.; Tateno, C.; Koyano, T.;
Kowithayakorn, T.; Ishibashi, M. Bioorg. Med. Chem. 2008, 16, 9420; (c) Rifai,
Y.; Arai, M. A.; Koyano, T.; Kowithayakorn, T.; Ishibashi, M. J. Nat. Prod. 2010, 35,
995.