Probing the structural requirements for vitamin D3 inhibition of the hedgehog signaling pathway

Article abstract of DOI:10.1016/j.bmcl.2012.05.037

A structure-activity relationship study to elucidate the structural basis for hedgehog (Hh) signaling inhibition by vitamin D3 (VD3) was performed. Functional and non-functional regions of VD3 and VD2 were obtained through straightforward synthetic means

Full text of DOI:10.1016/j.bmcl.2012.05.037

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4859–4863
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Probing the structural requirements for vitamin D3 inhibition
of the hedgehog signaling pathway
⇑
Albert M. DeBerardinis, Upasana Banerjee, Michele Miller, Steven Lemieux, M. Kyle Hadden
Department of Pharmaceutical Sciences, University of Connecticut, 69 N Eagleville Rd., Unit 3092, Storrs, CT 06269-3092, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
A structure–activity relationship study to elucidate the structural basis for hedgehog (Hh) signaling inhi-
bition by vitamin D3 (VD3) was performed. Functional and non-functional regions of VD3 and VD2 were
obtained through straightforward synthetic means and their biological activity was determined in a vari-
ety of cell-based assays. Several of these compounds inhibited Hh signaling at levels comparable to the
parent VD3 with no effects on canonical vitamin D signaling. Most notably, compounds 5 and 9, demon-
strated potent inhibition of the Hh pathway, exhibited no binding affinity for the vitamin D receptor
(VDR), and did not activate VDR in cell culture. In addition, several compounds exhibited anti-
proliferative activity against two human cancer cell lines through a mechanism distinct from the Hh
or VDR pathways, suggesting a new cellular mechanism of action for this class of compounds.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 3 April 2012
Revised 1 May 2012
Accepted 8 May 2012
Available online 23 May 2012
Keywords:
Hedgehog pathway
Vitamin D3
Vitamin D2
Gli1
Cyp24A1
Vitamin D receptor
The hedgehog (Hh) signaling pathway is a developmental path-
way which plays a key role in directing growth and tissue pattern-
ing during embryonic development. Dysregulation of Hh signaling
contributes to the development of a variety of human tumors,
including skin, brain, colon, pancreatic, and lung cancers.1 Consti-
tutive activation of the pathway results in the increased expression
of Hh target genes, including several forms of the glioma-
associated oncogene (Gli) family of signaling proteins and leads
to uncontrolled tumor proliferation. Recent years have seen the
development and characterization of numerous small molecule
inhibitors of Hh signaling as anti-cancer chemotherapeutics; most
notably, cyclopamine and GDC-0449 (Fig. 1).2 Cyclopamine (Cyc) is
a natural product derived from corn lilies that has been used exten-
sively to characterize Hh signaling in vitro and in vivo. An analogue
of Cyc (IPI-926) that demonstrates enhanced potency and
improved pharmacokinetic properties is currently in clinical devel-
opment. GDC-0449, a small molecule inhibitor of Hh signaling,
recently received fast-track approval by the FDA for the treatment
of metastatic basal cell carcinoma. This represents the first Hh
inhibitor to receive FDA approval and validates the clinical rele-
vance of this class of anti-cancer compound.
inhibits pathway signaling in Hh-dependent cell culture as well
as in murine models of basal cell carcinoma (BCC), a skin cancer
widely recognized as Hh-dependent.4,5 To date, the development
of VD3 as an Hh pathway inhibitor with anti-cancer potential has
been underexplored, most probably due to its ability to activate
canonical vitamin
D receptor (VDR) signaling in vitro and
in vivo.5 The VDR pathway plays an essential role in numerous
physiological systems and its activation can result in detrimental
side effects; therefore, VD3 analogues must potently inhibit Hh sig-
naling while exhibiting minimal or no effects on VDR.6
Our initial efforts at identifying structure–activity relationships
for VD3 inhibition of Hh signaling began with straightforward
modifications at the 3-(S)-hydroxyl.4 Results from the evaluation
of these analogues demonstrated that minor changes to the VD3
scaffold, particularly the A-ring, are not sufficient to prevent its cel-
lular activation of VDR. The studies described herein represent a
more detailed approach to probing the structural requirements
for VD3-mediated inhibition of Hh signaling. We sought to deter-
mine which regions of VD3 were essential for Hh inhibition, while
also identifying whether these analogues retain the ability to bind
and activate VDR.
More recently, vitamin D3 (VD3) was identified as an Hh path-
way inhibitor that is thought to exert its effects through direct
binding to Smoothened (SMO), an integral pathway component
that is also the molecular target of Cyc and GDC-0449.3 VD3
The first series of compounds prepared for evaluation focused
on simple semisynthetic derivatives of VD3 (Scheme 1, 1–4). These
compounds were synthesized through known methods to provide
insight into how changing the orientation and/or composition of
the diene linker and elimination of the 3-hydroxyl affected activity.
Triol 1 was prepared through selective conversion of the VD3
C7-C8 olefin to the corresponding diol.7 The two-step protection/
⇑
Corresponding author.
E-mail address: kyle.hadden@uconn.edu (M.K. Hadden).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.05.037

4860
A. M. DeBerardinis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4859–4863
Grundmann’s alcohol 5, which was oxidized to the corresponding
HN
H
HN
ketone 6 with PDC (Scheme 1).9 In addition, the hydroxyl of 5
was eliminated directly upon formation of the mesylate to provide
olefin 7. Exhaustive ozonolysis of VD2 afforded Inhoffen-Lythgoe
diol 8, while oxidative cleavage of the VD2 triol provided the
‘northern’ region of VD2, 9, and the VD3 A-ring, 10 (Scheme 2).9 Fi-
nally, sulfur dioxide protection of the VD2 triene, followed by
selective ozonolysis/reduction of the side chain olefin and removal
of the triene protecting group provides the known TBS-protected
trans-VD3 alcohol 11 (Scheme 3).8 Photoisomerism of 11 and
deprotection of the 3-hydroxyl affords the VD3 diol 12; similarly,
the trans-VD3 diol 13 was prepared for evaluation.
H
H
O
O
H
H
H
H
O
O
H
H
H
S
N
H
HO
IPI-926
Cyclopamine
25
Cl
O
Cl
N
N
H
H
The Hh inhibitory activity of these truncated VD3 analogues
was evaluated through their ability to modulate endogenous Gli1
mRNA in the Hh-dependent C3H10T1/2 cell line.10 Activation of
the Hh pathway in these cells (via exogenous addition of Hh ligand
or oxysterols) results in the robust and reproducible up-regulation
of the Hh target gene Gli1 (ꢀ100-fold) and concomitant adminis-
tration of an Hh pathway inhibitor attenuates this response.
Down-regulation of Gli1 levels in this cellular model is a well-
established method for determining Hh inhibition. For these stud-
O
S
O
GDC-0449
3
1
Vitamin D3 (VD3)
HO
Figure 1. Hh pathway inhibitors.
deprotection of the VD3 triene as the sulfur dioxide adduct yielded
trans-VD3 2.8 The C-3 hydroxyl of VD3 was found to eliminate un-
der modified Mitsunobu conditions to provide VD3 analogue 3.
Finally, hydrogenation of VD3 provided saturated VD3 isomers, 4,
that were evaluated as a mixture. All compounds were character-
ized by 1H and 13C NMR, as well as HRMS analysis prior to under-
going biological evaluation.
Our next series of analogues focused on individually evaluating
the ‘northern’ (CD-ring and side chain) and ‘southern’ (A-ring)
regions of VD3 for residual biological activity. The one-pot,
two-step global ozonolysis/reductive workup of VD3 provided
ies, analogues were evaluated at 5 lM and Gli1 mRNA expression
levels were normalized to oxysterol controls. Several key struc-
ture–activity relationships with respect to different regions of the
VD3 scaffold were observed from this assay (Table 1). First, with
respect to the side chain, VD2 was equipotent at inhibiting the
Hh pathway compared to VD3. By contrast, truncation of the side
chain (analogue 12) resulted in a two-fold loss of Hh inhibitory
activity. Taken together, these data indicate that this region is ame-
nable to minor modifications that retain the general structure and
hydrophobic nature of the natural alkyl side chain. Second, as
6
H
O
HO
H
h
OH
1
g
HO
a
f
7
H
HO
5
H
H
b - c
VD3
2
HO
OH
e
d
H
H
4
3
HO
Scheme 1. Reagents and conditions: (a) KMnO4, H2O (65%); (b) H2SO3, benzene; (c) Na2CO3, EtOH, reflux (80%, two steps); (d) PBu3, DIAD, DPPA (70%); (e) Pd/C, H2, ethanol
(92%) (f) O3, pyr, DCM:MeOH, NaBH4 (86%) (g) NEt3, MsCl, 0 °C (41%); (h) PDC, DCM (75%).

A. M. DeBerardinis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4859–4863
4861
evidenced by 5 and 9, the ‘northern’ region of the scaffold retains
the inhibitory activity of intact VD3. In addition, the comparable le-
vel of Gli down-regulation exhibited by both 5 and 9 (46.5 3.5
and 28.1 7.4, respectively) further supports the conclusion that
minor changes to the side chain do not affect the overall ability
of the compounds to inhibit Hh signaling. Modification (6) or re-
moval (7) of the C-8 hydroxyl significantly attenuated the Hh inhi-
bition of the ‘northern’ region, suggesting its presence and
orientation is essential for optimal activity. In a similar fashion,
the inability of truncated A-ring analogue 10 to inhibit Hh signal-
ing provides further evidence that the CD-ring and side chain
maintain the residual activity of the VD3 scaffold. Finally, several
analogues that contain the intact ‘northern’ region of VD3 (1–4)
demonstrate reduced Hh inhibition, suggesting that, while not re-
quired for activity, modifications to the seco-B-ring or A-ring can
prevent optimal Hh inhibition.
OH
a
H
H
8
HO
H
VD2
OH
b - d
HO
9
HO
+
HO
10
The next series of assays performed for the VD3 analogues was
designed to evaluate their ability to bind and activate VDR (Table
1). Using a competitive displacement binding assay for VDR,11
none of the analogues displayed affinity for VDR at concentrations
Scheme 2. Reagents and conditions: (a) O3, pyr, DCM:MeOH, NaBH4 (74%); (b)
KMnO4, THF, H2O (65%); (c) Pb(OAc)4, DCM; (d) Vitride™, DCM (9: 69%, 10: 53%,
two steps).
up to 100 lM (Table 1). This result was not surprising as our pre-
vious work with VD3 demonstrated that it and several A-ring ana-
logues do not bindVDR.4 Conversely, VD3 is able to functionally
activate VDR in C3H10T1/2 cell culture (as measured by up-regula-
tion of the well-characterized VDR target gene Cyp24A1) irrespec-
tive of its ability to bind purified VDR; therefore, we also
determined the ability of these compounds to up-regulate
Cyp24A1. The majority of VD3 analogues did not demonstrate
VDR activation compared to DMSO control. In addition, all of the
compounds that induced Cyp24A1 up-regulation (VD3, VD2, 2,
and 12) did so at a level significantly less than that observed for
VD3. These results provided important insight into the regions of
VD3 essential for activating VDR. Primarily, the intact seco-steroid
backbone (A-, seco-B-, and fused CD rings) is essential for a VD3
analogue to significantly activate VDR signaling. This is supported
by the findings that none of the truncated analogues demonstrated
significant Cyp24A1 up-regulation. Taken together, these data pro-
vide evidence that truncated analogues such as 5 and 9 will not
activate canonical VDR signaling and hold promise as selective
Hh pathway inhibitors.
Previous studies have demonstrated that the anti-proliferative
activity of Hh pathway inhibitors (including Cyc, GDC-0449, and
VD3) in cultured cancer cell lines does not correlate with pathway
inhibition in Hh-dependent cells.12,13 However, each of these com-
pounds demonstrates modest anti-proliferative effects in multiple
cell lines, suggesting either Hh signaling does play a role in prevent-
ing cancer cell growth in vitro or the anti-proliferative effects of
these compounds is mediated through unidentified cellular mech-
anisms distinct from Hh inhibition. With respect to VD3, cellular
effects unrelated to Hh signaling most probably result from activa-
tion of VDR signaling; therefore, studying the anti-proliferative
activity of the VD3 analogues could potentially provide more evi-
dence as to their ability to selectively inhibit the Hh pathway. For
these reasons, we sought to determine the anti-proliferative effects
of VD3 and its analogues in two distinct cancer cell lines, U87MG, a
glioblastoma cell line that exhibits anti-proliferative and anti-Hh
effects following Cyc administration,14 and HT-29, a colorectal ade-
nocarcinoma that over-expresses VDR and responds to VDR activa-
tion with a robust, characteristic response.15,16
OH
H
e-f
OH
12
a-d
H
HO
VD2
OH
OTBS
f
11
H
13
OH
Scheme 3. (a) H2SO3, Benzene; (b) TBSCl, imidazole, DCM (95%, 2 steps); (c) O3, pyr,
DCM:MeOH, NaBaH4 (82%); (d) Na2CO3, EtOH, reflux (80%); (e) acridine (50 mol%),
NEt3, benzene, 365 nm (95%); (f) TBAF, DCM (75%).
Table 1
Hh and VDR-related activity of VD3 analogues
Analogue
Gli1 mRNAa (%)
VDR bindingb
Cyp24A1mRNAc
DMSO
Oxy
VD3
VD2
1
2
3
4
5
6
7
8
9
10
12
13
—
100
—
—
1.0
—
35.7 0.3
28.1 7.4
144 41
69.7 8.8
79.4 14.7
82.1 1.1
46.4 3.5
66.1 0.9
111 21
87.6 1.2
32.1 6.3
80.9 2.1
80.0 10.4
66.7 5.7
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
8336 38
2408 584
6.7 1.9
43.5 8.3
3.0 2.4
1.5 0.3
3.6 0.5
3.0 0.4
2.3 0.8
4.1 0.7
2.3 1.2
2.0 0.2
1354 617
12.5 2.4
VD3, VD2 and all of the analogues evaluated were significantly
more active in the U87MG cell line (Table 2). The GI50 values for
VD3 and VD2 (23.2 5.1 and 33.7 0.2 lM, respectively) were
comparable, suggesting the mechanisms responsible for their
anti-proliferative activity in this cell line are similar. By contrast,
VD3 demonstrated significantly reduced anti-proliferative activity
against HT-29 cells and VD2 was inactive in these cells. Several
a
Values represent% Gli mRNA expression relative to oxysterol control (set at
100%). Analogues were evaluated at 5 M and 24 h.
l
b
Analogues were evaluated at 50 and 100 lM in the PolarScreen™ VDR com-
petitor assay, Red (invitrogen) following manufacturer’s protocol.11
c
Values represent Cyp24A1 mRNA expression relative to DMSO control (set to
1.0). Analogues were evaluated at 5 lM and 24 h.

4862
A. M. DeBerardinis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4859–4863
Table 2
of these enzymes plays a role in the proliferation of cancer cells,
Anti-proliferative activity of VD3 analogues17
our goal was to examine whether there was a correlation between
their expression in U87MG and HT-29 cells and the anti-prolifera-
tive activity of the truncated VD3 analogues. U87MG cells demon-
strated significantly higher baseline expression levels for each Cyp
isoform (10- to 100-fold), supporting the hypothesis that VD3 can
be converted to a from that activates VDR in this cell line (Fig. 2). By
contrast, baseline expression of VDR was 40-fold higher in HT-29
cells. The increased relative expression of the vitamin D metaboliz-
ing enzymes in U87MG cells, coupled with the lower GI50 values in
this cell line, provided preliminary evidence to support our
hypothesis that in vitro metabolism of VD3 results increased
anti-proliferation.
Our final series of experiments, designed to correlate VD3 and
analogue metabolism to anti-proliferative activity, was to explore
the ability of these compounds to regulate downstream target
genes of Hh and VDR signaling in both cancer cell lines (Table 3).
Similar to our previous findings,4 treatment of either cell line with
a
b
Analogue
U87MG GI50
(lM)
HT-29 GI50 (lM)
VD3
VD2
1
2
3
4
5
6
7
8
9
10
12
13
23.2 5.1c
33.7 0.2
25.9 4.2
46.0 6.8
48.5 7.4
>100
5.7 1.4
12.1 2.0
>100
>100
19.0 5.0
>100
>100
88.8 14
68.5 10.5
>100
>100
>100
>100
>100
66.5 17.9
>100
>100
>100
49.7 13.9
>100
>100
>100
cValues represent mean SEM for two separate experiments performed in
triplicate.
a
GI50 values measured in U87MG cells (MTS assay).
GI50 values measured in HT-29 cells (MTS assay).
b
VD3 (10 lM) had no effect on Gli mRNA expression levels while
resulting in significant up-regulation of Cyp24A1 (Table 3). Effects
of VD2 treatment were consistent to VD3, providing further evi-
dence that these two sterols exert their anti-proliferative effects
through VDR activation. In addition, up-regulation of Cyp24A1 in
HT-29 cells was significantly greater for both VD3 and VD2 com-
pared to U87MG cells (ꢀ10,000-fold and ꢀ25-fold, respectively).
None of the VD3 analogues modulated Gli expression in either cell
line, suggesting that the anti-proliferative activity of these ana-
logues is unrelated to Hh inhibition. Up-regulation of Cyp24A1
for the VD3 analogues in both cell lines correlated well with the re-
sults obtained in the C3H10T1/2 cell line, i.e. treatment with ana-
logues 2 and 12 resulted in significant up-regulation of Cyp24A1.
Most importantly, the most active analogues in the anti-prolifera-
tion assays, 5 and 9, did not modulate either Hh or VDR signaling,
suggesting these effects may be due to other cellular mechanisms.
The results of these studies provide important SAR information
as to the structural requirements for VD3 inhibition of Hh signal-
ing. First, the A-ring is not essential for Hh inhibition and may be
detrimental, as the only VD3 analogues that activated VDR re-
tained the intact seco-steroid backbone (A-, seco-B-, and fused
CD rings). Second, the alkyl side chain of VD3 appears amenable
to modification provided the hydrophobic nature of the natural lin-
ear alkyl moiety is maintained. Utilizing this knowledge, we are
currently exploring several new series of VD3 analogues that incor-
porate modifications to these regions to further develop improved
VD3-based inhibitors of Hh signaling.
analogues exhibited modest effects against U87MG cells, with the
most active compound being Grundmann’s alcohol,
5
(GI50 = 5.7 1.4 M). Of note, the only truncated VD3 analogues
l
that exhibited anti-proliferative activity in both cell lines, 5 and
9, were those that demonstrated the most robust Gli1 down-
regulation in C3H10T1/2 cells, suggesting a correlation between
Hh inhibition and anti-proliferative activity in these cells. As our
initial hypothesis was that the metabolic conversion of VD3 and
analogues to a form that can more readily activate VDR was
responsible for its anti-proliferative effects, we sought to further
explore the mechanisms that govern the anti-proliferative effects
of these analogues. Specifically, we sought to determine whether
anti-proliferation correlated with the following: (1) expression lev-
els of vitamin D metabolizing enzymes and (2) modulation of Hh-
and VDR-specific target genes.
The major enzymes associated with the metabolism of the vita-
min
D sterols, 25-hydroxylases (Cyp2R1 and Cyp27A1) and
1a-hydroxylases (Cyp27B1) are predominantly expressed in the
liver and kidney, respectively (Scheme 4). However, recent evi-
dence has shown that various types of human cancers, particularly
colon cancer, express increased levels of these metabolic en-
zymes.15,16 While it remains unclear whether the over-expression
OH
Our initial attempts to characterize the mechanisms through
which VD3 and its analogues exert their modest anti-proliferative
H
H
Cyp27A1
Cyp2R1
Cyp27B1
COOH
50
U87MG
HT-29
40
30
HO
HO
20
25(OH)D3
VD3
2.5
2.0
1.5
1.0
0.5
0.0
OH
HO
H
H
Cyp24A1
HO
OH
OH
1α,25(OH)2D3 (calcitriol)
calcitroic acid
Figure 2. Baseline expression (mRNA) of vitamin D-related metabolic enzymes in
Scheme 4. Vitamin D metabolic pathway.
cultured cancer cells.

A. M. DeBerardinis et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4859–4863
4863
Table 3
Supplementary data
Hh- and VDR-modulation in cultured cancer cells
Analoguea
U87MGb
HT-29b
Cyp24A1
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.bmcl. 2012.
05.037.
Gli1
Cyp24A1
Gli1
VD3
VD2
1
2
3
4
5
6
7
8
9
10
12
13
0.7 0.02c
0.9 0.01
1.1 0.09
1.1 0.05
1.0 0.3
0.8 0.2
0.7 0.04
0.9 0.01
1.0 0.2
1.0 0.1
0.8 0.3
0.9 0.08
1.0 0.1
0.8 0.1
12.2 4.1
49.6 0.8
0.5 0.06
33.8 5.3
0.9 0.3
0.8 0.2
0.4 0.01
0.5 0.1
0.5 0.09
0.7 0.04
0.3 0.1
0.6 0.03
4.8 1.0
0.8 0.2
1.3 0.3
1.1 0.2
0.9 0.1
1.2 0.01
1.8 0.3
1.4 0.2
1.1 0.1
1.1 0.4
1.1 0.1
1.2 0.1
0.9 0.02
0.9 0.05
2.7 0.4
1.3 0.4
14697 1255
10902 2840
1.2 0.2
27.5 4.3
1.9 0.8
1.4 0.5
1.1 0.1
1.0 0.4
1.1 0.3
1.0 0.2
0.7 0.1
0.7 0.01
69.1 1.9
3.3 1.6
References and notes
1. Yang, L.; Xie, G.; Fan, Q.; Xie, J. Oncogene 2010, 29, 469.
2. Peukert, S.; Miller-Moslin, K. ChemMedChem 2010, 5, 500.
3. Bijlsma, M. F.; Spek, C. A.; Zivkovic, D.; van de Water, S.; Rezaee, F.;
Peppelenbosch, M. P. PloS Biol. 2006, 4, e232.
4. Banerjee, U.; Ghosh, M.; Hadden, M. K. Bioorg. Med. Chem. Lett. 2012, 22, 1330.
5. Tang, J. Y.; Xiao, T. Z.; Oda, Y.; Chang, K. S.; Shpall, E.; Wu, A.; So, P.-L.; Hebert, J.;
Bikle, D.; Epstein, E. H. Cancer Prev. Res. 2011, 4, 744.
6. Haussler, M. R.; Whitfield, G. K.; Haussler, C. A.; Hsieh, J.-C.; Thompson, P. D.;
Selznick, S. H.; Dominguez, C. E.; Jurutka, P. W. J. Bone Min. Res. 1998, 13, 325.
7. Toh, H. T.; Okamura, W. H. J. Org. Chem. 1983, 48, 1414.
8. Andrews, D. R.; Barton, D. H. R.; Hesse, R. H.; Pechet, M. M. J. Org. Chem. 1986,
51, 4819.
cValues represent mean SEM of two separate experiments performed in triplicate.
9. Zhu, G.-D.; Okamura, W. H. Chem. Rev. 1877, 1995, 95.
a
Analogues were evaluated at 10
Values represent Gli1 or Cyp24A1 mRNA expression relative to DMSO (set at 1).
lM and 24 h.
10. General RT-PCR protocol: Following treatment and incubation, RNA was
extracted with TRIZOLÒ reagent following the manufacturer’s instructions.
cDNA synthesis was performed utilizing the high capacity cDNA reverse
b
transcription kit (ABI) per the manufacturer’s instructions on
a
BioRadMyCycler. Quantitative RT-PCR was performed on an ABI 7500 system
using the following Taqman gene expression assays: human GLI1,
Hs00171790_m1; human Cyp24A1, Hs00167999_m1; human ActB,
Hs99999903_ml; mouse GLI1, Mm00494645_m1; mouse Cyp24A1,
Mm00487244_m1; mouse ActB, Mm00607939_s1. Relative gene expression
levels were computed via the DDCt method.
effects have proven contradictory. The increased expression levels
of vitamin D metabolizing enzymes in U87MG cells suggested that
the enhanced anti-proliferative effects in this cell line may result
from the cellular conversion of VD3 to 25(OH)D3 and/or
1a,25(OH)D3 and subsequent activation of VDR; however, up-reg-
11. PolarScreen™ VDR competitor assay, red (invitrogen) per the manufacturer’s
instructions. This assay is a fluorescence polarization-based competition assay
that has been optimized for high-throughput applications and contains all the
required components (including full length human VDR and a proprietary
tight-binding fluorescent VDR ligand).
12. Yauch, R. L.; Gould, S. E.; Scales, S. J.; Tang, T.; Tian, H.; Ahn, C. P.; Marshall, D.;
Fu, L.; Januario, T.; Kallop, D.; Nannini-Pepe, M.; Kotkow, K.; Marsters, J. C.;
Rubin, L. L.; de Sauvage, F. J. Nature 2008, 455, 406.
13. Zhang, X.; Harrington, N.; Moraes, R. C.; Hilsenbeck, S. G.; Lewis, H. T. Breast
Cancer Res. Treat. 2009, 115, 505.
14. Clement, V.; Sanchez, P.; de Tribolet, N.; Radovanovic, I.; Ruiz I Altaba, A. Curr.
Biol. 2007, 17, 165.
ulation of Cyp24A1 is more robust in HT-29 cells. In addition, VD3
analogues that demonstrated anti-proliferation against these cell
lines did not regulate either Hh or VDR signaling, providing initial
evidence that these analogues may exert biological effects through
a third, as yet unidentified, cellular mechanism. Recent studies
have demonstrated that several natural and synthetic cholesterol
metabolites, including oxysterols and bile acids, exhibit anti-prolif-
erative effects through a variety of different mechanisms.18–21 Ste-
roid hormone receptors other than VDR, primarily the estrogen and
androgen receptors, have long been studied as anti-cancer targets
and several selective estrogen receptor modulators (SERMs) are
used clinically for the treatment of breast cancer.22 The core struc-
tural similarities of these steroids and our truncated VD3 ana-
logues, particularly 5 and 9, suggest that the anti-proliferative
activity of these compounds may result from their ability to bind
and modulate a myriad of cellular mechanisms. Therefore, in order
to fully explore the development of VD3 and its analogues as selec-
tive Hh inhibitors, future work must identify and detail all cellular
components that bind VD3 and these analogues, not just those re-
lated to Hh and VDR signaling.
15. Kane, K. F.; Langman, M. J. S.; Williams, G. R. Cancer Res. 1996, 56, 623.
16. Kumagai, T.; O’Kelly, J.; Said, J. W.; Koeffler, P. J. Natl. Cancer Inst. 2003, 95, 896.
17. General anti-proliferation protocol: Cells (3000/well, 100 lL) were seeded in
96-well plates and incubated overnight (37 °C, 5% CO2). Cells were treated in
triplicate with DMSO, VD3, or analogue at varying concentrations (1% DMSO
final concentration). Following 72 h incubation, viable cells were measured
using CellTiter 96Ò aqueous non-radioactive cell proliferation assay (Promega)
per the manufacturer’s instructions. Data was analyzed using GraphPad Prism
and GI50 values represent mean SEM for at least two separate experiments.
18. Bischoff, P. L.; Holl, V.; Coelho, D.; Dufour, P.; Weltin, D.; Luu, B. Curr. Med.
Chem. 2000, 7, 693.
19. Burgett, A. W.; Poulsen, T. B.; Wangkanont, K.; Anderson, D. R.; Kikuchi, C.;
Shimada, K.; Okubo, S.; Fortner, K. C.; Mimaki, Y.; Kuroda, M.; Murphy, J. P.;
Schwalb, D. J.; Petrella, E. C.; Cornella-Taracido, I.; Schirle, M.; Tallarico, J. A.;
Shair, M. D. Nat. Chem. Bio. 2011, 7, 639.
20. Park, S. E.; Choi, H. J.; Yee, S. B.; Chung, H. Y.; Suh, H.; Choi, Y. H.; Yoo, Y. H.;
Kim, N. D. Int. J. Oncol. 2004, 25, 231.
21. Serfaty, L.; Bissonnette, M.; Poupon, R. Gastroenterol. Clin. Biol. 2010, 34, 516.
22. Ahmad, N.; Kumar, R. Cancer Lett. 2011, 300, 1.
Acknowledgements
The authors gratefully acknowledge support of this work by the
American Cancer Society (ACS-IRG-06-002-04) and the V Founda-
tion for Cancer Research (V Scholar, M.K.H.).