Identification of vitamin D3-based hedgehog pathway inhibitors that incorporate an aromatic A-ring isostere

Article abstract of DOI:10.1021/ml400014t

Previous structure-activity relationship studies for vitamin D3 (VD3) inhibition of Hedgehog (Hh) signaling directed the design, synthesis, and evaluation of a series of VD3-based analogues that contain an aromatic A-ring mimic. Characterization of these

Full text of DOI:10.1021/ml400014t

Letter
pubs.acs.org/acsmedchemlett
Identification of Vitamin D3-Based Hedgehog Pathway Inhibitors
That Incorporate an Aromatic A‑Ring Isostere
Albert M. DeBerardinis, Upasana Banerjee, and M. Kyle Hadden*
Department of Pharmaceutical Sciences, University of Connecticut, 69 North Eagleville Road, Unit 3092, Storrs, Connecticut
06269-3092, United States
S
* Supporting Information
ABSTRACT: Previous structure−activity relationship studies for
vitamin D3 (VD3) inhibition of Hedgehog (Hh) signaling
directed the design, synthesis, and evaluation of a series of VD3-
based analogues that contain an aromatic A-ring mimic.
Characterization of these compounds in a series of cellular assays
demonstrated their ability to potently and selectively down-
regulate Hh pathway signaling. The most active of these, 17,
inhibited pathway signaling in Hh-dependent mouse fibroblasts
(IC50 = 0.74 0.1 μM) and cultured cancer cells (IC50 values 3.8 0.1 to 5.2 0.2 μM). In addition, 17 demonstrated reduced
activation of the vitamin D receptor (VDR) compared to VD3 in these cellular models. These results suggest that VD3-based
analogues with an aromatic A-ring are a valid scaffold for the development of more selective and potent Hh pathway inhibitors
and identify 17 as an intriguing lead from this class of compounds for further development. In addition, our analysis of Hh
pathway inhibitors in cancer cells suggests that the murine basal cell carcinoma cell line ASZ001 and the human medulloblastoma
cell line DAOY are appropriate in vitro cancer models for early stage evaluation of pathway inhibition.
KEYWORDS: hedgehog signaling pathway, vitamin D3, Gli, basal cell carcinoma, medulloblastoma
Chart 1. Hh Signaling Pathway Antagonists
he Hedgehog (Hh) signaling pathway is a developmental
Tpathway that plays a key role in directing growth and
tissue patterning during embryonic development. Dysregulation
of Hh signaling has been associated with the development of
human tumors; most notably, basal cell carcinoma (BCC) and
medulloblastoma (MB).1 Both forms of cancer are widely
recognized as Hh-dependent and a significant number of BCCs
(70%) and MBs (25%) exhibit detectable genetic mutations in
Hh pathway components.2−4 These mutations result in
constitutive activation of the pathway and increased expression
of Hh target genes, including several forms of the glioma-
associated oncogene (Gli) family of signaling proteins.
Recent years have seen the identification, development, and
evaluation of numerous small molecule inhibitors of Hh
signaling as anticancer chemotherapeutics; most notably, the
natural product cyclopamine (Cyc) and its derivatives, and
GDC-0449 (Chart 1).5 Both of these compounds inhibit the
Hh pathway by directly binding Smoothened (SMO), a key
regulator of pathway signaling and the most druggable target
within the pathway. GDC-0449 recently received fast-track
approval by the FDA for the treatment of metastatic BCC,
representing the first Hh pathway inhibitor to receive FDA
approval and validating the clinical relevance of targeting
aberrant Hh signaling.6 However, a point mutation in SMO
that prevents binding and renders the tumor resistant to GDC-
0449 highlights the continued need for new and improved Hh
signaling inhibitors.7,8
binding to SMO.9 VD3 has been shown to inhibit pathway
signaling in Hh-dependent cell culture9−11 as well as in murine
models of BCC;12 however, VD3 also activates the vitamin D
receptor (VDR),13 a nuclear receptor that plays an essential
role in numerous physiological processes via canonical vitamin
D signaling, in these models. This activation of VDR can result
in detrimental side effects, including hypercalcemia. For this
Received: January 10, 2013
Accepted: May 14, 2013
Recently, vitamin D3 (VD3) was identified as an Hh pathway
inhibitor that was proposed to exert its effects through direct
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
reason, VD3-based Hh pathway inhibitors must exert potent
anti-Hh effects while also exhibiting minimal or no effects on
canonical vitamin D signaling.
Table 1. Hh Signaling Inhibition in C3H10T1/2 Cells
a
b
c
d
Analogue
Gli1 mRNA (%)
Cyp24A1 mRNA
VDR binding (μM)
DMSO
OHCs
VD3
4
1.0
1.0
ND
100
ND
ND
Scheme 1. Preparation of Protected Aromatic Carboxylic
35.7 0.3
46.4 3.5
116 34
105 12
8336 38
3.6 0.5
1.0 0.1
1.2 0.4
1.2 0.2
1.3 0.1
1.6 0.2
22.2 4.7
6.7 0.5
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
a
Acids
9
13
14
107
104
7
9
15
16
77.5 9.0
1.9 0.5
40.0 14
2.4 0.5
97.8 10
92.6 6.1
17
18
19
24.3
3
20
1.6 0.2
1.6 0.6
3-HBA
a
b
All analogues tested at 5 μM and 24 h. Values represent % Gli1
mRNA expression normalized to DMSO (vehicle) ActinB levels (set
to 1.0). The OHC control Gli1 mRNA levels are set as 100% Hh
pathway activity, and analogue modulation is represented as a %
c
relative to this level. Values represent Cyp24A1 mRNA expression
normalized to ActinB levels treated with DMSO (vehicle control, set
d
to 1.0). Calcitriol is utilized as a positive control for VDR binding.17
The design of this series of analogues focused on the
incorporation of aromatic A-ring isosteres that maintain the
approximate steric and hydrophobic features of the natural
aliphatic A-ring while not being a candidate for metabolism to a
form that could more readily activate VDR. An ester linkage
between the northern CD-ring region and the aromatic A-ring
isostere was chosen due to its ease of formation and because
analogue generation could be performed rapidly based on the
parallel coupling of 4 with a library of commercially available
benzoic acids. Initially, our rationale was to couple 3-
hydroxybenzoic acid (3-HBA) with the northern CD-ring
region (4) for evaluation as a proof-of-concept compound for
this class of analogues. Following the synthesis and evaluation
of promising lead analogue 17, we sought to explore this class
of compounds by establishing structure−activity relationships
(SAR) for the 3-position of the aromatic A-ring.
a
Reagents and conditions: (a) BnBr (3 equiv), K2CO3, acetone, 85%.
(b) KOH (aq., 20%), THF, 75%. (c) (1) NEt3, (1 equiv),
MeOH:H2O; (2) BnBr (1.1 equiv), DMF, 40%. (d) (Boc)2O,
dioxane, NaHCO3 (aq., 5%), 50−80%.
a
Scheme 2. Preparation of Aromatic A-Ring VD3 Analogues
Aromatic carboxylic acids were either purchased commer-
cially or prepared via standard protection protocols (Scheme
1a−c). Acid 5c was prepared by benzyl protection of methyl 3-
hydroxybenzoate (5a) followed by potassium hydroxide-
mediated hydrolysis to provide free carboxylic acid 5c. The
monobenzyl ester of isophthalic acid, 6b, was obtained in
moderate yield following a published protocol.14 Protected
carboxylic acids 7b−8b were also generated in modest yield
using standard conditions.15 Acids were coupled to 4 using
standard esterification conditions followed by appropriate
deprotection (Scheme 2) to afford final analogues in good
yields that were subsequently evaluated for their ability to
selectively inhibit the Hh signaling pathway.
The Hh inhibitory activity of analogues 9 and 13−20 was
initially evaluated by determining their ability to down-regulate
endogenous Gli1 mRNA levels in the Hh-dependent
C3H10T1/2 cell line, a standard cell-based assay used
previously to evaluate small molecule inhibition of Hh
signaling.10,11 In addition, we have previously demonstrated
that this cell line responds to VDR activation10,11 with
significant up-regulation of Cyp24A1, a well-characterized
target gene of canonical vitamin D signaling.16 For this assay,
a
Reagents and conditions: (a) DCC, DMAP, DCM, 3-(X)-ArCO2H,
60−95%. (b) 9, Pd(OH)2 (10%), H2, MeOH:THF (2:1), 75%. (c) 10,
Pd/C (10%), H2, ethyl acetate, 95%. (d) 11 or 12, TFA:DCM (1:1),
55−70%.
A previous SAR study for VD3 inhibition of pathway
signaling performed in our lab identified the CD-ring alcohol of
VD3 (4, Grundmann’s alcohol) as an Hh inhibitor with activity
equipotent to VD3 (Table 1).11 In addition, 4 did not bind to
purified VDR nor did it activate VDR signaling in cell culture.
On the basis of these results, and the fact that 4 is readily
available from VD3 in a one pot, two-step reaction sequence,
we sought to use this compound as a basis for the development
of improved VD3-based Hh pathway inhibitors. Specifically, we
report herein the design, synthesis, and evaluation of a series of
VD3 analogues that contain an aromatic A-ring mimic with
improved potency and selectivity against Hh signaling.
B
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
a
Table 2. Inhibition of Hh Signaling in Murine Fibroblasts
C3H10T1/2
M2-10B4
Ptch
b
c
Analogue
Gli1
Ptch
Gli1
>10
Cyp24A1
VD3
4.1 0.3
3.1 0.2
0.74 0.1
2.8 0.6
ND
2.9 0.3
2.8 0.4
1.1 0.3
4.5 1.3
ND
>10
274 12
ND
4
ND
ND
c
17
1.1 0.6
ND
1.7 0.8
ND
1.2 0.1
ND
19
d
d
GDC-0449
Cyc
0.2 0.01
0.8 0.1
0.14 0.03
0.43 0.03
1.1 0.4
1.5 0.1
ND
ND
a
Values for Gli1 and Ptch are IC50 values in μM and represent at least two separate experiments performed in triplicate. Gli1 and Ptch mRNA
b
expression were determined in the same manner as described for Table 1. ND = not determined. Values represent Cyp24A1 mRNA expression
normalized to ActinB levels treated with DMSO (vehicle control, set to 1.0). Evaluated at 5 μM and 24 h. Evaluated at 1 μM and 24 h.
c
d
analogues were evaluated at 5 μM and normalized to vehicle
control (DMSO). For evaluation of Hh inhibition, cells treated
only with the positive control [20(S)-hydroxycholeterol:22(S)-
hydroxycholesterol, 5 μM each] were set as 100% Hh pathway
signaling.
that both 17 and 19 are more selective for the Hh pathway
compared to VD3.
Both 17 and 19 inhibited Hh signaling in a dose-dependent
fashion in C3H10T1/2 cells (Table 2). Interestingly, while the
IC50 values for Gli1 and Ptch correlated well for VD3, 4, and
19, analogue 17 was significantly more active against Gli1 (IC50
= 0.74 0.1 μM). To assess whether the potent, selective Hh
inhibitory effects of this class of analogues were dependent on
cell type, 17 was evaluated in M2-10B4 cells, a murine bone
marrow stromal cell line that also responds to Hh pathway
stimulation with a characteristic Gli1 and Ptch mRNA up-
regulation18 (Table 2). While VD3 was found to be less active
in M2-10B4 cells,19 down-regulation of Gli1 and Ptch by 17
was comparable to that determined in the C3H10T1/2 line.
Evaluation of GDC-0449 and Cyc in this model was also
performed to determine whether other Hh pathway inhibitors
differentially affected Gli1 and Ptch levels similar to VD3.
Interestingly, Ptch knock-down by both of these well-
characterized Hh pathway inhibitors appeared to be more
potent when compared to Gli1. Treatment of M2-10B4 cells
with 17 (5 μM), GDC-0449 (1 μM), or Cyc (1 μM) did not
result in increased expression of Cyp24A1 compared to DMSO
control, suggesting the enhanced expression seen in the
C3H10T1/2 cells may be cell-type dependent. Cyp24a1 levels
were increased approximately 270-fold in response to treatment
with VD3 (5uM). Finally, 17 did not increase Cyp24A1 levels
in HT-29 cells, a human colon cancer cell line that
overexpresses VDR and responds to vitamin D (calcitriol)
with a robust up-regulation of Cyp24A120,21 (data not shown).
These results suggest that the enhanced expression of Cyp24A1
seen by 17 in the C3H10T1/2 cells is a cell-type dependent
response that is not a general result of Hh pathway inhibition.
Previous studies evaluating Hh pathway inhibition in
cultured cancer cells have suggested that the majority of in
vitro cancer cell lines do not appropriately model in vivo Hh
signaling. With this in mind, we sought to characterize VD3, 17,
and 19 in two cell lines, ASZ001 and DAOY, which have shown
initial promise as in vitro models of oncogenic Hh signaling.
The ASZ001 cell line was grown from a visible BCC tumor
isolated from a Ptch1 mouse.22 These cells demonstrate loss of
the wildtype Ptch1 allele, high baseline expression of Gli1, and
cellular morphology similar to Hh-dependent BCC tumors. In
addition, treatment of these cells with either Cyc or VD3
resulted in Gli1 down-regulation and antiproliferation.12,22
Several recent lines of evidence suggested that DAOY cells may
be a suitable in vitro cancer model of Hh signaling.23 First, the
DNA methylation pattern in DAOY cells is consistent with the
pattern observed in mouse models of MB in which a mutation
in PTCH1 is responsible for MB formation.24,25 Second,
The results of this initial study provided important
preliminary SAR for the ability of this class of VD3-based
analogues to selectively inhibit Hh pathway signaling (Table 1).
First, analogue 17 demonstrated potent inhibition of Hh
signaling as measured by the ability to completely abolish Gli1
up-regulation at 5 μM. By contrast, masking the free phenol as
either a methyl (15) or benzyl ether (9) eliminated activity
from the scaffold. The only other compound that demonstrated
Hh inhibition comparable to 17 was analogue 19, which
contains a primary amine at the 3-position of the aromatic ring.
Similar to SAR for the phenol and ethers, the 3-NHMe
analogue 20 was inactive, further supporting the notion that
masking the alcohol or amine with small alkyl groups in this
region affects Hh inhibitory activity. Interestingly, analogue 18,
which contains a 3-COOH functionality, demonstrated Hh
inhibition similar to VD3, albeit at a level reduced compared to
17. This reduced activity may result from the carboxylic acid
functionality extending its hydrogen bond donating moiety
away from the aromatic ring, reducing key interactions at the
binding site. On the basis of these results, it was not surprising
that neither unsubstituted analogues nor compounds that
contained a more lipophilic moiety were active. Finally, as the
ester linkage could potentially be cleaved by endogenous
esterases, we studied whether the Hh inhibitory activity of 17
could be a result of the individual northern (4) and southern
(3-hydroxybenzoic acid; 3-HBA) regions following potential
cellular cleavage of the ester bond. 3-HBA failed to down-
regulate Gli1 mRNA at 5 μM, suggesting that the inhibitory
activity of this class of analogues is a result of the intact
structure.
Both 17 and 19 displayed modest up-regulation of Cyp24A1;
however, on the basis of the analysis that neither of these
compounds bound to VDR, it is unclear by what mechanism
they up-regulate Cyp24A1. None of the compounds tested
displaced a fluorescent VDR ligand at concentrations up to 100
μM, suggesting a lack of affinity for VDR.17 It is possible that
17 and 19 are metabolized to a form that can bind and activate
VDR; however, on the basis of the close correlation between
Hh inhibition and Cyp24A1 up-regulation, a more likely
explanation is that inhibition of Hh signaling/Gli1 down-
regulation triggers an increase in Cyp24A1 mRNA, either
through VDR activation or another, as yet unidentified,
mechanism. In either scenario, these results clearly indicate
C
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
expression of REN, a tumor suppressor that negatively regulates
Hh signaling and is commonly mutated in human MB, is
significantly reduced in DAOY cells.26,27 Finally, several reports
in the patent literature have utilized DAOY cells to evaluate
Gli1 down-regulation as a function of Hh inhibition.28 Previous
studies performed in both cell lines demonstrated that Gli1
down-regulation by VD3 (ASZ001)12 and Cyc (DAOY)23 was
more robust and reproducible following a 48 h compound
incubation. Preliminary evaluation in our lab was consistent
with a 48 h incubation period resulting in reproducible Gli1
down-regulation induced by GDC-0449 or Cyc. For this
reason, all data presented from these two cell lines was obtained
at this time point.
Initially, we evaluated the ability of VD3 and several of our
analogues to down-regulate Gli1 mRNA expression in ASZ001
cells in a dose-dependent fashion (Table 3 and Figure 1). While
Table 3. Hh Signaling Inhibition in ASZ001 and DAOY Cells
ASZ001
DAOY
a
b
c
a
b
Analogue
Gli1
Cyp24A1
Gli1
Cyp24a1
d
VD3
2.1 0.1
>10
764 24
>10
>10
>10
149 90
d
d
e
4
2.4 0.9
1.1 0.3
1.9 0.1
2.5 1.4
0.9 0.2
0.8 0.1
0.8 0.1
0.6 0.1
0.4 0.1
0.5 0.1
1.0 0.1
1.0 0.7
d
d
d
d
d
13
>10
d
d
c
17
5.2 0.2
3.7 0.04
9.2 1.7
19
7.1
1
GDC-0449
Cyc
0.04 0.01
0.66 0.02
0.086 0.02
0.16 0.04
c
a
All IC50 values are in μM and represent at least two separate
b
experiments performed in triplicate. Values represent Cyp24A1
mRNA expression after 48 h incubation with drug relative to DMSO
c
d
(set as 1.0). Evaluated at 2.5 μM and 48 h. Evaluated at 5 μM and 48
e
h. Evaluated at 10 μM and 48 h.
the IC50 for VD3 down-regulation of Gli1 correlated well with
that obtained in the C3H10T1/2 fibroblasts, analogues 17 and
19 were less effective in this model (5.2 and 7.1 μM,
respectively). In addition, analogue 4 demonstrated only a
modest reduction of Gli1 mRNA levels (65% relative to
control) at the highest dose evaluated (10 μM). The inactivity
of 4 in ASZ001s provides further evidence that the Hh
antagonism of 17 and 19 results from an intact structure.
Similar to the results from the murine fibroblasts, VD3
demonstrated significant up-regulation of Cyp24A1, while the
analogues had minimal effects. Of note was our finding that, at
higher concentrations (10 and 5 μM), VD3 up-regulated
Cyp24a1 mRNA levels to a lesser degree than at lower
concentrations (≤2.5 μM; Supplementary Figure 1), which
may be an indication that VD3 exhibits significant activity
through off-target effects or toxicity to the cells at higher doses.
To further characterize the ability of ASZ001 cells to function
as an early stage in vitro cancer model of aberrant Hh signaling,
we evaluated both GDC-0449 and Cyc for their dose-
dependent effects on Gli1 expression. Both compounds
completely abolished Gli1 expression at high concentrations
(≥1 μM) and IC50 values correlated well with those previously
published in Hh-dependent murine fibroblasts (Table 3).
Overall, our results suggest that the ASZ001 cell line is a
suitable BCC model for early stage analysis of Hh pathway
inhibition.
Figure 1. Selective inhibition of Hh signaling by VD3 and analogues in
cultured cancer cells. Down-regulation of Gli1 in ASZ001 (A) and
DAOY (B) cell lines. Up-regulation of Cyp24A1 by VD3, 17, and 19
in ASZ001 and DAOY cells (C; DMSO set to 1.0).
down-regulation of Gli1 mRNA expression (Figure 1B).19 By
contrast, VD3 analogues 17 and 19 displayed Hh inhibitory
activity similar to that demonstrated in the ASZ001 cells (IC50
= 3.7 and 9.2 μM, respectively). In addition, neither 4 nor 13
demonstrated the ability to down-regulate Gli1 mRNA
expression in the DAOYs. While the IC50 for GDC-0449 was
comparable to those previously determined, Cyc was
significantly more active in the DAOYs. Finally, only treatment
We next evaluated each of these compounds in the human
MB-derived DAOY cell line (Table 3, Figure 1). Interestingly,
high concentrations of VD3 (10 μM) exhibited only minimal
D
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
with VD3 resulted in up-regulation of Cyp24A1 mRNA
expression.
by Dr. Ervin Epstein (Children’s Hospital Oakland Research
Institute).
To determine whether down-regulation of Gli mRNA in
DAOY cells was a function of Hh inhibition or general
anticancer activity, we evaluated a series of clinically relevant
anticancer agents that work via mechanisms distinct from Hh
inhibition for their ability to affect endogenous Gli1 mRNA.
None of these, including the HDAC inhibitor Vorinostat (1.51
REFERENCES
■
(1) Yang, L.; Xie, G.; Fan, Q.; Xie, J. Activation of the hedgehog-
signaling pathway in human cancer and the clinical implications.
Oncogene 2010, 29, 469−481.
(2) Ling, G.; Ahmadian, A.; Persson, A.; Unden, A. B.; Afink, G.;
Williams, C.; Uhlen, M.; Toftgard, R.; Lundeberg, J.; Ponten, F.
PATCHED and p53 gene alterations in sporadic and hereditary basal
cell cancer. Oncogene 2001, 20, 7770−7778.
(3) Reifenberger, J.; Wolter, M.; Knobbe, C. B.; Kohler, B.;
Schonicke, A.; Scharwachter, C.; Kumar, K.; Blaschke, B.; Ruzicka,
T.; Reifenberger, G. Somatic mutations in the PTCH, SMOH,
SUFUH, and TP53 genes in sporadic basal cell carcinomas. Br. J.
Dermatol. 2005, 152, 43−51.
(4) Klesse, L. J.; Bowers, D. C. Childhood medulloblastoma: Current
status of biology and treatment. CNS Drugs 2010, 24, 285−301.
(5) Peukert, S.; Miller-Moslin, K. Small-molecule inhibitors of the
hedgehog signaling pathway as cancer therapeutics. ChemMedChem
2010, 5, 500−512.
(6) Original Approval: January 30, 2012; FDA Application no.:
(NDA) 203388. Details can be found at http://www.accessdata.fda.
gov/scripts/cder/drugsatfda/index.cfm?fuseaction=Search.
DrugDetails and http://www.curetoday.com/index.cfm/fuseaction/
blog.showIndex/elizabethwhittington/2011/4/4/The-first-patient.
(7) Yauch, R. L.; Dijkgraaf, G. J. P.; Alicke, B.; Januario, T.; Ahn, C.
P.; Holcomb, T.; Pujara, K.; Stinson, J.; Callahan, C. A.; Tang, T.;
Bazan, J. F.; Kan, Z.; Seshagiri, S.; Hann, C. L.; Gould, S. E.; Low, J. A.;
Rudin, C. M.; de Sauvage, F. J. Smoothened mutation confers
resistance to a hedgehog pathway inhibitor in medulloblastoma. Science
2009, 326, 572−574.
(8) Metcalfe, C.; de Sauvage, F. J. Hedgehog fights back: mechanisms
of acquired resistance against smoothened antagonists. Cancer Res.
2011, 71, 5057−5061.
(9) Bijlsma, M. F.; Spek, C. A.; Zivkovic, D.; van de Water, S.;
Rezaee, F.; Peppelenbosch, M. P. Repression of smoothened by
patched-dependent (pro-)vitamin D3 secretion. PloS Biol. 2006, 4,
e232.
(10) Banerjee, U.; Ghosh, M.; Hadden, M. K. Evaluation of vitamin
D3 A-ring analogues as hedgehog pathway inhibitors. Bioorg. Med.
Chem. Lett. 2012, 22, 1330−1334.
(11) DeBerardinis, A. M.; Banerjee, U.; Miller, M.; Lemieux, S.;
Hadden, M. K. Probing the structural requirements for vitamin D3
inhibition of hedgehog signaling. Bioorg. Med. Chem. Lett. 2012, 22,
4859−4863.
(12) 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. Vitamin D3 inhibits
hedgehog signaling and proliferation in murine basal cell carcinomas.
Cancer Prev. Res. 2011, 4, 744−751.
(13) Haussler, M. R.; Whitfield, G. K.; Haussler, C. A.; Hsieh, J.-C.;
Thompson, P. D.; Selznick, S. H.; Dominguez, C. E.; Jurutka, P. W.
The nuclear vitamin D receptor: biological and molecular regulatory
properties revealed. J. Bone Miner. Res. 1998, 13, 325−349.
(14) Adlington, R. M.; Baldwin, J. E.; Becker, G. W.; Chen, B.;
Cheng, L.; Cooper, S. L.; Hermann, R. B.; Howe, T. J.; McCoull, W.;
McNulty, A. M.; Neubauer, B. L.; Pritchard, G. J. Design, synthesis,
and proposed active site binding analysis of monocyclic 2-azetidinone
inhibitors of prostate specific antigen. J. Med. Chem. 2001, 44, 1491−
1508.
0.1), the DNA intercalator oxaliplatin (1.16
0.3), the
phosphatidylinositol-3-kinase (PI3K) inhibitor wortmannin
(0.92
(0.98
0.2), nor the BCR-ABL kinase inhibitor Gleevec
0.3) significantly modulated Gli1 mRNA expression
compared to DMSO control (1.0) at a concentration of 2.5
μM. Taken together, the results of our studies in DAOYs
indicate this cell line is a suitable in vitro MB model for early
stage evaluation of Hh signaling inhibition.
In summary, we have identified a lead scaffold suitable for
further development as a new class of VD3-based Hh pathway
inhibitors. Replacement of the seco-B and A-ring of VD3 with
an aromatic isostere linked through an ester bond resulted in
compounds with enhanced Hh inhibition and improved
selectivity compared to VD3. A focused series of analogues
provided key preliminary SAR for this class of compounds,
resulting in the identification of 17. As a close mimic to VD3,
lead 17 represents a more potent and selective inhibitor of Hh
signaling in multiple cellular models. The in vitro results
described above suggest the intact analogue structure is
required for the modulation of Hh signaling; however, as the
presence of the ester linkage poses potential stability concerns
for the in vivo evaluation of these compounds, ongoing SAR
studies in the lab are focused on further exploration of the
aromatic A-ring region for this scaffold as well as the design and
synthesis of analogues that incorporate a linker less susceptible
to metabolism. Finally, our evaluation of several Hh pathway
inhibitors in ASZ001 and DAOY cell lines suggests that both
are suitable in vitro cancer models for the early stage analysis of
Hh pathway inhibition.
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic methods, biological assay protocols, and spectro-
scopic data. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*(M.K.H.) Tel: 1-806-486-8446. Fax: 1-860-486-6857. E-mail:
kyle.hadden@uconn.edu.
Author Contributions
A.M.D. synthesized and characterized all analogues and
performed most of the biological assays. U.B. performed
VDR binding assays. All authors contributed to the preparation
of the manuscript.
Notes
The authors declare no competing financial interest.
(15) Norman, M. H.; Rigdon, G. C.; Hall, W. R.; Navas, F., III.
Structure−activity relationships of a series of substituted benzamides:
potent D2/5-HT2 antagonists and 5-HT1a agonists as neuroleptic
agents. J. Med. Chem. 1996, 39, 1172−1188.
(16) Meyer, M. B.; Zella, L. A.; Nerenz, R. D.; Pike, J. W.
Characterizing early events associated with the activation of target
genes by 1,25-dihydroxyvitamin D3 in mouse kidney and intestine in
vivo. J. Biol. Chem. 2007, 282, 22344−22352.
ACKNOWLEDGMENTS
■
We gratefully acknowledge support of this work by the V
Foundation for Cancer Research (V Scholar, to M.K.H.), the
Charles H. Hood Foundation, and the University of
Connecticut Research Foundation. ASZ001 cells were provided
E
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
Letter
(17) Compounds evaluated at 50 and 100 μM. The PolarScreen
VDR competitor assay, Red (Invitrogen), was used 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). Calcitriol is utilized as a positive control for VDR binding.
(18) Corman, A.; DeBerardinis, A. M.; Hadden, M. K. Structure−
activity relationships for side chain oxysterol agonists of the hedgehog
signaling pathway. ACS Med. Chem. Lett. 2012, 3, 828−833.
(19) VD3 was dosed up to 10 μM (>10 μM was toxic to cells) and
did not completely knock-down Gli1 or Ptch mRNA levels at 24 h. For
example, 10 μM VD3 treatments down-regulated Gli1 levels between
53−68% relative to OHCs. This was also observed in DAOYs. Thus,
IC50 values are represented as >10 μM for VD3.
(20) Kane, K. F.; Langman, M. J. S.; Williams, G. R. Antiproliferative
responses of two human colon cancer cell lines to vitamin D3 are
differentially modified by 9-cis-retinoic acid. Cancer Res. 1996, 56,
623−632.
(21) Kumagai, T.; O’Kelly, J.; Said, J. W.; Koeffler, P. Vitamin D2
analog 19-nor-1,25-dihydroxyvitamin D2: antitumor activity against
leukemia, myeloma, and colon cancer cells. J. Natl. Cancer Inst. 2003,
95, 896−905.
(22) So, P.-L.; Langston, A. W.; Daniallinia, N.; Hebert, J. L.;
Fujimoto, M. A.; Khaimskiy, Y.; Aszterbaum, M.; Epstein, E. H. Long-
term establishment, characterization and manipulation of cell lines
from mouse basal cell carcinoma tumors. Exp. Dermatol. 2006, 15,
742−750.
(23) Bar, E. E.; Chaudhry, A.; Farah, M. H.; Eberhart, C. G.
Hedgehog signaling promotes medulloblastoma survival via BclII. Am.
J. Pathol. 2007, 170, 347−355.
(24) Diede, S. J.; Guenthoer, J.; Geng, L. N.; Mahoney, S. E.;
Marotta, M.; Olson, J. M.; Tanaka, H.; Tapscott, S. J. DNA
methylation of developmental genes in pediatric medulloblastomas
identified by denaturation analysis of methylation differences. Proc.
Natl. Acad. Sci. 2010, 107, 234−239.
(25) Uhmann, A.; Ferch, U.; Bauer, R.; Tauber, S.; Arziman, Z.;
Chen, C.; Hemmerlein, B.; Wojnowski, L.; Hahn, H. A model for
PTCH1/Ptch1-associated tumors comprising mutational inactivation
and gene silencing. Int. J. Oncol. 2005, 27, 1567−1575.
(26) Di Marcotullio, L.; Ferretti, E.; De Smaele, E.; Argenti, B.;
Mincione, C.; Zazzeroni, F.; Gallo, R.; Masuelli, L.; Napolitano, M.;
Maroder, M.; Modesti, A.; Giangaspero, F.; Screpanti, I.; Alesse, E.;
Gulino, A. RENKCTD11 is a suppressor of hedgehog signaling and is
deleted in human medulloblastoma. Proc. Natl. Acad. Sci. 2004, 101,
10833−10838.
(27) De Smaele, E.; Di Marcotullio, L.; Ferretti, E.; Screpanti, I.;
Alesse, E.; Gulino, A. Chromosome 17p deletion in human
medulloblastoma: a missing checkpoint in the hedgehog pathway.
Cell Cycle 2004, 3, 1263−1266.
(28) Hipskind, P. A.; Wilson, T. U.S. Patent 7,981,892 B2, 2011.
F
dx.doi.org/10.1021/ml400014t | ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX